diff options
Diffstat (limited to 'drivers/gpu/drm')
295 files changed, 10947 insertions, 3109 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 306f75700bf8..5a5f04d0902d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -496,6 +496,7 @@ struct amdgpu_bo_va_mapping { /* bo virtual addresses in a specific vm */ struct amdgpu_bo_va { + struct mutex mutex; /* protected by bo being reserved */ struct list_head bo_list; struct fence *last_pt_update; @@ -538,6 +539,7 @@ struct amdgpu_bo { /* Constant after initialization */ struct amdgpu_device *adev; struct drm_gem_object gem_base; + struct amdgpu_bo *parent; struct ttm_bo_kmap_obj dma_buf_vmap; pid_t pid; @@ -928,8 +930,6 @@ struct amdgpu_vm_id { }; struct amdgpu_vm { - struct mutex mutex; - struct rb_root va; /* protecting invalidated */ @@ -956,6 +956,8 @@ struct amdgpu_vm { struct amdgpu_vm_id ids[AMDGPU_MAX_RINGS]; /* for interval tree */ spinlock_t it_lock; + /* protecting freed */ + spinlock_t freed_lock; }; struct amdgpu_vm_manager { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 3afcf0237c25..4f352ec9dec4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -222,6 +222,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) } p->uf.bo = gem_to_amdgpu_bo(gobj); + amdgpu_bo_ref(p->uf.bo); + drm_gem_object_unreference_unlocked(gobj); p->uf.offset = fence_data->offset; } else { ret = -EINVAL; @@ -487,7 +489,7 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bo amdgpu_ib_free(parser->adev, &parser->ibs[i]); kfree(parser->ibs); if (parser->uf.bo) - drm_gem_object_unreference_unlocked(&parser->uf.bo->gem_base); + amdgpu_bo_unref(&parser->uf.bo); } static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p, @@ -776,7 +778,7 @@ static int amdgpu_cs_free_job(struct amdgpu_job *job) amdgpu_ib_free(job->adev, &job->ibs[i]); kfree(job->ibs); if (job->uf.bo) - drm_gem_object_unreference_unlocked(&job->uf.bo->gem_base); + amdgpu_bo_unref(&job->uf.bo); return 0; } @@ -784,8 +786,6 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { struct amdgpu_device *adev = dev->dev_private; union drm_amdgpu_cs *cs = data; - struct amdgpu_fpriv *fpriv = filp->driver_priv; - struct amdgpu_vm *vm = &fpriv->vm; struct amdgpu_cs_parser parser = {}; bool reserved_buffers = false; int i, r; @@ -803,7 +803,6 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) r = amdgpu_cs_handle_lockup(adev, r); return r; } - mutex_lock(&vm->mutex); r = amdgpu_cs_parser_relocs(&parser); if (r == -ENOMEM) DRM_ERROR("Not enough memory for command submission!\n"); @@ -888,7 +887,6 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) out: amdgpu_cs_parser_fini(&parser, r, reserved_buffers); - mutex_unlock(&vm->mutex); r = amdgpu_cs_handle_lockup(adev, r); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index e173a5a02f0d..acd066d0a805 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -73,6 +73,8 @@ static void amdgpu_flip_work_func(struct work_struct *__work) struct drm_crtc *crtc = &amdgpuCrtc->base; unsigned long flags; unsigned i; + int vpos, hpos, stat, min_udelay; + struct drm_vblank_crtc *vblank = &crtc->dev->vblank[work->crtc_id]; amdgpu_flip_wait_fence(adev, &work->excl); for (i = 0; i < work->shared_count; ++i) @@ -81,6 +83,41 @@ static void amdgpu_flip_work_func(struct work_struct *__work) /* We borrow the event spin lock for protecting flip_status */ spin_lock_irqsave(&crtc->dev->event_lock, flags); + /* If this happens to execute within the "virtually extended" vblank + * interval before the start of the real vblank interval then it needs + * to delay programming the mmio flip until the real vblank is entered. + * This prevents completing a flip too early due to the way we fudge + * our vblank counter and vblank timestamps in order to work around the + * problem that the hw fires vblank interrupts before actual start of + * vblank (when line buffer refilling is done for a frame). It + * complements the fudging logic in amdgpu_get_crtc_scanoutpos() for + * timestamping and amdgpu_get_vblank_counter_kms() for vblank counts. + * + * In practice this won't execute very often unless on very fast + * machines because the time window for this to happen is very small. + */ + for (;;) { + /* GET_DISTANCE_TO_VBLANKSTART returns distance to real vblank + * start in hpos, and to the "fudged earlier" vblank start in + * vpos. + */ + stat = amdgpu_get_crtc_scanoutpos(adev->ddev, work->crtc_id, + GET_DISTANCE_TO_VBLANKSTART, + &vpos, &hpos, NULL, NULL, + &crtc->hwmode); + + if ((stat & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE)) != + (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE) || + !(vpos >= 0 && hpos <= 0)) + break; + + /* Sleep at least until estimated real start of hw vblank */ + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + min_udelay = (-hpos + 1) * max(vblank->linedur_ns / 1000, 5); + usleep_range(min_udelay, 2 * min_udelay); + spin_lock_irqsave(&crtc->dev->event_lock, flags); + }; + /* do the flip (mmio) */ adev->mode_info.funcs->page_flip(adev, work->crtc_id, work->base); /* set the flip status */ @@ -109,7 +146,7 @@ static void amdgpu_unpin_work_func(struct work_struct *__work) } else DRM_ERROR("failed to reserve buffer after flip\n"); - drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base); + amdgpu_bo_unref(&work->old_rbo); kfree(work->shared); kfree(work); } @@ -148,8 +185,8 @@ int amdgpu_crtc_page_flip(struct drm_crtc *crtc, obj = old_amdgpu_fb->obj; /* take a reference to the old object */ - drm_gem_object_reference(obj); work->old_rbo = gem_to_amdgpu_bo(obj); + amdgpu_bo_ref(work->old_rbo); new_amdgpu_fb = to_amdgpu_framebuffer(fb); obj = new_amdgpu_fb->obj; @@ -222,7 +259,7 @@ pflip_cleanup: amdgpu_bo_unreserve(new_rbo); cleanup: - drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base); + amdgpu_bo_unref(&work->old_rbo); fence_put(work->excl); for (i = 0; i < work->shared_count; ++i) fence_put(work->shared[i]); @@ -481,7 +518,7 @@ static const struct drm_framebuffer_funcs amdgpu_fb_funcs = { int amdgpu_framebuffer_init(struct drm_device *dev, struct amdgpu_framebuffer *rfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -498,7 +535,7 @@ amdgpu_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * amdgpu_user_framebuffer_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct amdgpu_framebuffer *amdgpu_fb; @@ -712,6 +749,15 @@ bool amdgpu_crtc_scaling_mode_fixup(struct drm_crtc *crtc, * \param dev Device to query. * \param pipe Crtc to query. * \param flags Flags from caller (DRM_CALLED_FROM_VBLIRQ or 0). + * For driver internal use only also supports these flags: + * + * USE_REAL_VBLANKSTART to use the real start of vblank instead + * of a fudged earlier start of vblank. + * + * GET_DISTANCE_TO_VBLANKSTART to return distance to the + * fudged earlier start of vblank in *vpos and the distance + * to true start of vblank in *hpos. + * * \param *vpos Location where vertical scanout position should be stored. * \param *hpos Location where horizontal scanout position should go. * \param *stime Target location for timestamp taken immediately before @@ -776,10 +822,40 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, vbl_end = 0; } + /* Called from driver internal vblank counter query code? */ + if (flags & GET_DISTANCE_TO_VBLANKSTART) { + /* Caller wants distance from real vbl_start in *hpos */ + *hpos = *vpos - vbl_start; + } + + /* Fudge vblank to start a few scanlines earlier to handle the + * problem that vblank irqs fire a few scanlines before start + * of vblank. Some driver internal callers need the true vblank + * start to be used and signal this via the USE_REAL_VBLANKSTART flag. + * + * The cause of the "early" vblank irq is that the irq is triggered + * by the line buffer logic when the line buffer read position enters + * the vblank, whereas our crtc scanout position naturally lags the + * line buffer read position. + */ + if (!(flags & USE_REAL_VBLANKSTART)) + vbl_start -= adev->mode_info.crtcs[pipe]->lb_vblank_lead_lines; + /* Test scanout position against vblank region. */ if ((*vpos < vbl_start) && (*vpos >= vbl_end)) in_vbl = false; + /* In vblank? */ + if (in_vbl) + ret |= DRM_SCANOUTPOS_IN_VBLANK; + + /* Called from driver internal vblank counter query code? */ + if (flags & GET_DISTANCE_TO_VBLANKSTART) { + /* Caller wants distance from fudged earlier vbl_start */ + *vpos -= vbl_start; + return ret; + } + /* Check if inside vblank area and apply corrective offsets: * vpos will then be >=0 in video scanout area, but negative * within vblank area, counting down the number of lines until @@ -795,32 +871,6 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, /* Correct for shifted end of vbl at vbl_end. */ *vpos = *vpos - vbl_end; - /* In vblank? */ - if (in_vbl) - ret |= DRM_SCANOUTPOS_IN_VBLANK; - - /* Is vpos outside nominal vblank area, but less than - * 1/100 of a frame height away from start of vblank? - * If so, assume this isn't a massively delayed vblank - * interrupt, but a vblank interrupt that fired a few - * microseconds before true start of vblank. Compensate - * by adding a full frame duration to the final timestamp. - * Happens, e.g., on ATI R500, R600. - * - * We only do this if DRM_CALLED_FROM_VBLIRQ. - */ - if ((flags & DRM_CALLED_FROM_VBLIRQ) && !in_vbl) { - vbl_start = mode->crtc_vdisplay; - vtotal = mode->crtc_vtotal; - - if (vbl_start - *vpos < vtotal / 100) { - *vpos -= vtotal; - - /* Signal this correction as "applied". */ - ret |= 0x8; - } - } - return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c index 093a8c618931..6fcbbcc2e99e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c @@ -45,7 +45,6 @@ struct amdgpu_fbdev { struct drm_fb_helper helper; struct amdgpu_framebuffer rfb; - struct list_head fbdev_list; struct amdgpu_device *adev; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 00c5b580f56c..f6ea4b43a60c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -115,12 +115,9 @@ int amdgpu_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_pri struct amdgpu_vm *vm = &fpriv->vm; struct amdgpu_bo_va *bo_va; int r; - mutex_lock(&vm->mutex); r = amdgpu_bo_reserve(rbo, false); - if (r) { - mutex_unlock(&vm->mutex); + if (r) return r; - } bo_va = amdgpu_vm_bo_find(vm, rbo); if (!bo_va) { @@ -129,7 +126,6 @@ int amdgpu_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_pri ++bo_va->ref_count; } amdgpu_bo_unreserve(rbo); - mutex_unlock(&vm->mutex); return 0; } @@ -142,10 +138,8 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj, struct amdgpu_vm *vm = &fpriv->vm; struct amdgpu_bo_va *bo_va; int r; - mutex_lock(&vm->mutex); r = amdgpu_bo_reserve(rbo, true); if (r) { - mutex_unlock(&vm->mutex); dev_err(adev->dev, "leaking bo va because " "we fail to reserve bo (%d)\n", r); return; @@ -157,7 +151,6 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj, } } amdgpu_bo_unreserve(rbo); - mutex_unlock(&vm->mutex); } static int amdgpu_gem_handle_lockup(struct amdgpu_device *adev, int r) @@ -242,8 +235,9 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, AMDGPU_GEM_USERPTR_REGISTER)) return -EINVAL; - if (!(args->flags & AMDGPU_GEM_USERPTR_ANONONLY) || - !(args->flags & AMDGPU_GEM_USERPTR_REGISTER)) { + if (!(args->flags & AMDGPU_GEM_USERPTR_READONLY) && ( + !(args->flags & AMDGPU_GEM_USERPTR_ANONONLY) || + !(args->flags & AMDGPU_GEM_USERPTR_REGISTER))) { /* if we want to write to it we must require anonymous memory and install a MMU notifier */ @@ -553,7 +547,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, gobj = drm_gem_object_lookup(dev, filp, args->handle); if (gobj == NULL) return -ENOENT; - mutex_lock(&fpriv->vm.mutex); rbo = gem_to_amdgpu_bo(gobj); INIT_LIST_HEAD(&list); INIT_LIST_HEAD(&duplicates); @@ -568,7 +561,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, } r = ttm_eu_reserve_buffers(&ticket, &list, true, &duplicates); if (r) { - mutex_unlock(&fpriv->vm.mutex); drm_gem_object_unreference_unlocked(gobj); return r; } @@ -577,7 +569,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, if (!bo_va) { ttm_eu_backoff_reservation(&ticket, &list); drm_gem_object_unreference_unlocked(gobj); - mutex_unlock(&fpriv->vm.mutex); return -ENOENT; } @@ -602,7 +593,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, ttm_eu_backoff_reservation(&ticket, &list); if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE)) amdgpu_gem_va_update_vm(adev, bo_va, args->operation); - mutex_unlock(&fpriv->vm.mutex); + drm_gem_object_unreference_unlocked(gobj); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 1618e2294a16..e23843f4d877 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -611,13 +611,59 @@ void amdgpu_driver_preclose_kms(struct drm_device *dev, u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe) { struct amdgpu_device *adev = dev->dev_private; + int vpos, hpos, stat; + u32 count; if (pipe >= adev->mode_info.num_crtc) { DRM_ERROR("Invalid crtc %u\n", pipe); return -EINVAL; } - return amdgpu_display_vblank_get_counter(adev, pipe); + /* The hw increments its frame counter at start of vsync, not at start + * of vblank, as is required by DRM core vblank counter handling. + * Cook the hw count here to make it appear to the caller as if it + * incremented at start of vblank. We measure distance to start of + * vblank in vpos. vpos therefore will be >= 0 between start of vblank + * and start of vsync, so vpos >= 0 means to bump the hw frame counter + * result by 1 to give the proper appearance to caller. + */ + if (adev->mode_info.crtcs[pipe]) { + /* Repeat readout if needed to provide stable result if + * we cross start of vsync during the queries. + */ + do { + count = amdgpu_display_vblank_get_counter(adev, pipe); + /* Ask amdgpu_get_crtc_scanoutpos to return vpos as + * distance to start of vblank, instead of regular + * vertical scanout pos. + */ + stat = amdgpu_get_crtc_scanoutpos( + dev, pipe, GET_DISTANCE_TO_VBLANKSTART, + &vpos, &hpos, NULL, NULL, + &adev->mode_info.crtcs[pipe]->base.hwmode); + } while (count != amdgpu_display_vblank_get_counter(adev, pipe)); + + if (((stat & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE)) != + (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE))) { + DRM_DEBUG_VBL("Query failed! stat %d\n", stat); + } else { + DRM_DEBUG_VBL("crtc %d: dist from vblank start %d\n", + pipe, vpos); + + /* Bump counter if we are at >= leading edge of vblank, + * but before vsync where vpos would turn negative and + * the hw counter really increments. + */ + if (vpos >= 0) + count++; + } + } else { + /* Fallback to use value as is. */ + count = amdgpu_display_vblank_get_counter(adev, pipe); + DRM_DEBUG_VBL("NULL mode info! Returned count may be wrong.\n"); + } + + return count; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index b62c1710cab6..fdc1be8550da 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -35,6 +35,7 @@ #include <drm/drm_dp_helper.h> #include <drm/drm_fixed.h> #include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_helper.h> #include <drm/drm_plane_helper.h> #include <linux/i2c.h> #include <linux/i2c-algo-bit.h> @@ -407,6 +408,7 @@ struct amdgpu_crtc { u32 line_time; u32 wm_low; u32 wm_high; + u32 lb_vblank_lead_lines; struct drm_display_mode hw_mode; }; @@ -528,6 +530,10 @@ struct amdgpu_framebuffer { #define ENCODER_MODE_IS_DP(em) (((em) == ATOM_ENCODER_MODE_DP) || \ ((em) == ATOM_ENCODER_MODE_DP_MST)) +/* Driver internal use only flags of amdgpu_get_crtc_scanoutpos() */ +#define USE_REAL_VBLANKSTART (1 << 30) +#define GET_DISTANCE_TO_VBLANKSTART (1 << 31) + void amdgpu_link_encoder_connector(struct drm_device *dev); struct drm_connector * @@ -551,7 +557,7 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, int amdgpu_framebuffer_init(struct drm_device *dev, struct amdgpu_framebuffer *rfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); int amdgpufb_remove(struct drm_device *dev, struct drm_framebuffer *fb); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 0d524384ff79..c3ce103b6a33 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -100,6 +100,7 @@ static void amdgpu_ttm_bo_destroy(struct ttm_buffer_object *tbo) list_del_init(&bo->list); mutex_unlock(&bo->adev->gem.mutex); drm_gem_object_release(&bo->gem_base); + amdgpu_bo_unref(&bo->parent); kfree(bo->metadata); kfree(bo); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index d4bac5f49939..8a1752ff3d8e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -587,9 +587,13 @@ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm, uint32_t flags = amdgpu_ttm_tt_pte_flags(gtt->adev, ttm, bo_mem); int r; - if (gtt->userptr) - amdgpu_ttm_tt_pin_userptr(ttm); - + if (gtt->userptr) { + r = amdgpu_ttm_tt_pin_userptr(ttm); + if (r) { + DRM_ERROR("failed to pin userptr\n"); + return r; + } + } gtt->offset = (unsigned long)(bo_mem->start << PAGE_SHIFT); if (!ttm->num_pages) { WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n", @@ -797,11 +801,12 @@ uint32_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, if (mem && mem->mem_type != TTM_PL_SYSTEM) flags |= AMDGPU_PTE_VALID; - if (mem && mem->mem_type == TTM_PL_TT) + if (mem && mem->mem_type == TTM_PL_TT) { flags |= AMDGPU_PTE_SYSTEM; - if (!ttm || ttm->caching_state == tt_cached) - flags |= AMDGPU_PTE_SNOOPED; + if (ttm->caching_state == tt_cached) + flags |= AMDGPU_PTE_SNOOPED; + } if (adev->asic_type >= CHIP_TOPAZ) flags |= AMDGPU_PTE_EXECUTABLE; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 03f0c3bae516..a745eeeb5d82 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -392,7 +392,10 @@ int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, ib->ptr[ib->length_dw++] = 0x00000001; /* session cmd */ ib->ptr[ib->length_dw++] = handle; - ib->ptr[ib->length_dw++] = 0x00000030; /* len */ + if ((ring->adev->vce.fw_version >> 24) >= 52) + ib->ptr[ib->length_dw++] = 0x00000040; /* len */ + else + ib->ptr[ib->length_dw++] = 0x00000030; /* len */ ib->ptr[ib->length_dw++] = 0x01000001; /* create cmd */ ib->ptr[ib->length_dw++] = 0x00000000; ib->ptr[ib->length_dw++] = 0x00000042; @@ -404,6 +407,12 @@ int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, ib->ptr[ib->length_dw++] = 0x00000100; ib->ptr[ib->length_dw++] = 0x0000000c; ib->ptr[ib->length_dw++] = 0x00000000; + if ((ring->adev->vce.fw_version >> 24) >= 52) { + ib->ptr[ib->length_dw++] = 0x00000000; + ib->ptr[ib->length_dw++] = 0x00000000; + ib->ptr[ib->length_dw++] = 0x00000000; + ib->ptr[ib->length_dw++] = 0x00000000; + } ib->ptr[ib->length_dw++] = 0x00000014; /* len */ ib->ptr[ib->length_dw++] = 0x05000005; /* feedback buffer */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 159ce54bbd8d..b53d273eb7a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -885,17 +885,21 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev, struct amdgpu_bo_va_mapping *mapping; int r; + spin_lock(&vm->freed_lock); while (!list_empty(&vm->freed)) { mapping = list_first_entry(&vm->freed, struct amdgpu_bo_va_mapping, list); list_del(&mapping->list); - + spin_unlock(&vm->freed_lock); r = amdgpu_vm_bo_update_mapping(adev, vm, mapping, 0, 0, NULL); kfree(mapping); if (r) return r; + spin_lock(&vm->freed_lock); } + spin_unlock(&vm->freed_lock); + return 0; } @@ -922,8 +926,9 @@ int amdgpu_vm_clear_invalids(struct amdgpu_device *adev, bo_va = list_first_entry(&vm->invalidated, struct amdgpu_bo_va, vm_status); spin_unlock(&vm->status_lock); - + mutex_lock(&bo_va->mutex); r = amdgpu_vm_bo_update(adev, bo_va, NULL); + mutex_unlock(&bo_va->mutex); if (r) return r; @@ -967,7 +972,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev, INIT_LIST_HEAD(&bo_va->valids); INIT_LIST_HEAD(&bo_va->invalids); INIT_LIST_HEAD(&bo_va->vm_status); - + mutex_init(&bo_va->mutex); list_add_tail(&bo_va->bo_list, &bo->va); return bo_va; @@ -1045,7 +1050,9 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, mapping->offset = offset; mapping->flags = flags; + mutex_lock(&bo_va->mutex); list_add(&mapping->list, &bo_va->invalids); + mutex_unlock(&bo_va->mutex); spin_lock(&vm->it_lock); interval_tree_insert(&mapping->it, &vm->va); spin_unlock(&vm->it_lock); @@ -1076,6 +1083,11 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, if (r) goto error_free; + /* Keep a reference to the page table to avoid freeing + * them up in the wrong order. + */ + pt->parent = amdgpu_bo_ref(vm->page_directory); + r = amdgpu_vm_clear_bo(adev, pt); if (r) { amdgpu_bo_unref(&pt); @@ -1121,7 +1133,7 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev, bool valid = true; saddr /= AMDGPU_GPU_PAGE_SIZE; - + mutex_lock(&bo_va->mutex); list_for_each_entry(mapping, &bo_va->valids, list) { if (mapping->it.start == saddr) break; @@ -1135,20 +1147,25 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev, break; } - if (&mapping->list == &bo_va->invalids) + if (&mapping->list == &bo_va->invalids) { + mutex_unlock(&bo_va->mutex); return -ENOENT; + } } - + mutex_unlock(&bo_va->mutex); list_del(&mapping->list); spin_lock(&vm->it_lock); interval_tree_remove(&mapping->it, &vm->va); spin_unlock(&vm->it_lock); trace_amdgpu_vm_bo_unmap(bo_va, mapping); - if (valid) + if (valid) { + spin_lock(&vm->freed_lock); list_add(&mapping->list, &vm->freed); - else + spin_unlock(&vm->freed_lock); + } else { kfree(mapping); + } return 0; } @@ -1181,7 +1198,9 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, interval_tree_remove(&mapping->it, &vm->va); spin_unlock(&vm->it_lock); trace_amdgpu_vm_bo_unmap(bo_va, mapping); + spin_lock(&vm->freed_lock); list_add(&mapping->list, &vm->freed); + spin_unlock(&vm->freed_lock); } list_for_each_entry_safe(mapping, next, &bo_va->invalids, list) { list_del(&mapping->list); @@ -1190,8 +1209,8 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, spin_unlock(&vm->it_lock); kfree(mapping); } - fence_put(bo_va->last_pt_update); + mutex_destroy(&bo_va->mutex); kfree(bo_va); } @@ -1236,13 +1255,13 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) vm->ids[i].id = 0; vm->ids[i].flushed_updates = NULL; } - mutex_init(&vm->mutex); vm->va = RB_ROOT; spin_lock_init(&vm->status_lock); INIT_LIST_HEAD(&vm->invalidated); INIT_LIST_HEAD(&vm->cleared); INIT_LIST_HEAD(&vm->freed); spin_lock_init(&vm->it_lock); + spin_lock_init(&vm->freed_lock); pd_size = amdgpu_vm_directory_size(adev); pd_entries = amdgpu_vm_num_pdes(adev); @@ -1320,7 +1339,6 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) fence_put(vm->ids[i].flushed_updates); } - mutex_destroy(&vm->mutex); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index cb0f7747e3dc..093599aba64b 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -1250,7 +1250,7 @@ static void dce_v10_0_program_watermarks(struct amdgpu_device *adev, u32 pixel_period; u32 line_time = 0; u32 latency_watermark_a = 0, latency_watermark_b = 0; - u32 tmp, wm_mask; + u32 tmp, wm_mask, lb_vblank_lead_lines = 0; if (amdgpu_crtc->base.enabled && num_heads && mode) { pixel_period = 1000000 / (u32)mode->clock; @@ -1333,6 +1333,7 @@ static void dce_v10_0_program_watermarks(struct amdgpu_device *adev, (adev->mode_info.disp_priority == 2)) { DRM_DEBUG_KMS("force priority to high\n"); } + lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay); } /* select wm A */ @@ -1357,6 +1358,8 @@ static void dce_v10_0_program_watermarks(struct amdgpu_device *adev, amdgpu_crtc->line_time = line_time; amdgpu_crtc->wm_high = latency_watermark_a; amdgpu_crtc->wm_low = latency_watermark_b; + /* Save number of lines the linebuffer leads before the scanout */ + amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines; } /** @@ -3726,7 +3729,7 @@ static void dce_v10_0_encoder_add(struct amdgpu_device *adev, case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1: case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2: drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &dce_v10_0_dac_helper_funcs); break; case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1: @@ -3737,15 +3740,15 @@ static void dce_v10_0_encoder_add(struct amdgpu_device *adev, if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { amdgpu_encoder->rmx_type = RMX_FULL; drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_lcd_info(amdgpu_encoder); } else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) { drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder); } else { drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder); } drm_encoder_helper_add(encoder, &dce_v10_0_dig_helper_funcs); @@ -3763,13 +3766,13 @@ static void dce_v10_0_encoder_add(struct amdgpu_device *adev, amdgpu_encoder->is_ext_encoder = true; if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); else drm_encoder_init(dev, encoder, &dce_v10_0_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &dce_v10_0_ext_helper_funcs); break; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 5af3721851d6..8701661a8868 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -1238,7 +1238,7 @@ static void dce_v11_0_program_watermarks(struct amdgpu_device *adev, u32 pixel_period; u32 line_time = 0; u32 latency_watermark_a = 0, latency_watermark_b = 0; - u32 tmp, wm_mask; + u32 tmp, wm_mask, lb_vblank_lead_lines = 0; if (amdgpu_crtc->base.enabled && num_heads && mode) { pixel_period = 1000000 / (u32)mode->clock; @@ -1321,6 +1321,7 @@ static void dce_v11_0_program_watermarks(struct amdgpu_device *adev, (adev->mode_info.disp_priority == 2)) { DRM_DEBUG_KMS("force priority to high\n"); } + lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay); } /* select wm A */ @@ -1345,6 +1346,8 @@ static void dce_v11_0_program_watermarks(struct amdgpu_device *adev, amdgpu_crtc->line_time = line_time; amdgpu_crtc->wm_high = latency_watermark_a; amdgpu_crtc->wm_low = latency_watermark_b; + /* Save number of lines the linebuffer leads before the scanout */ + amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines; } /** @@ -3719,7 +3722,7 @@ static void dce_v11_0_encoder_add(struct amdgpu_device *adev, case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1: case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2: drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &dce_v11_0_dac_helper_funcs); break; case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1: @@ -3730,15 +3733,15 @@ static void dce_v11_0_encoder_add(struct amdgpu_device *adev, if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { amdgpu_encoder->rmx_type = RMX_FULL; drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_lcd_info(amdgpu_encoder); } else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) { drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder); } else { drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder); } drm_encoder_helper_add(encoder, &dce_v11_0_dig_helper_funcs); @@ -3756,13 +3759,13 @@ static void dce_v11_0_encoder_add(struct amdgpu_device *adev, amdgpu_encoder->is_ext_encoder = true; if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); else drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &dce_v11_0_ext_helper_funcs); break; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 4f7b49a6dc50..d0e128c24813 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -1193,7 +1193,7 @@ static void dce_v8_0_program_watermarks(struct amdgpu_device *adev, u32 pixel_period; u32 line_time = 0; u32 latency_watermark_a = 0, latency_watermark_b = 0; - u32 tmp, wm_mask; + u32 tmp, wm_mask, lb_vblank_lead_lines = 0; if (amdgpu_crtc->base.enabled && num_heads && mode) { pixel_period = 1000000 / (u32)mode->clock; @@ -1276,6 +1276,7 @@ static void dce_v8_0_program_watermarks(struct amdgpu_device *adev, (adev->mode_info.disp_priority == 2)) { DRM_DEBUG_KMS("force priority to high\n"); } + lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay); } /* select wm A */ @@ -1302,6 +1303,8 @@ static void dce_v8_0_program_watermarks(struct amdgpu_device *adev, amdgpu_crtc->line_time = line_time; amdgpu_crtc->wm_high = latency_watermark_a; amdgpu_crtc->wm_low = latency_watermark_b; + /* Save number of lines the linebuffer leads before the scanout */ + amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines; } /** @@ -3656,7 +3659,7 @@ static void dce_v8_0_encoder_add(struct amdgpu_device *adev, case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1: case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2: drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &dce_v8_0_dac_helper_funcs); break; case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1: @@ -3667,15 +3670,15 @@ static void dce_v8_0_encoder_add(struct amdgpu_device *adev, if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { amdgpu_encoder->rmx_type = RMX_FULL; drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_lcd_info(amdgpu_encoder); } else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) { drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder); } else { drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder); } drm_encoder_helper_add(encoder, &dce_v8_0_dig_helper_funcs); @@ -3693,13 +3696,13 @@ static void dce_v8_0_encoder_add(struct amdgpu_device *adev, amdgpu_encoder->is_ext_encoder = true; if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); else drm_encoder_init(dev, encoder, &dce_v8_0_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &dce_v8_0_ext_helper_funcs); break; } diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 7427d8cd4c43..ed8abb58a785 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -513,7 +513,7 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev) WREG32(mmVM_L2_CNTL3, tmp); /* setup context0 */ WREG32(mmVM_CONTEXT0_PAGE_TABLE_START_ADDR, adev->mc.gtt_start >> 12); - WREG32(mmVM_CONTEXT0_PAGE_TABLE_END_ADDR, (adev->mc.gtt_end >> 12) - 1); + WREG32(mmVM_CONTEXT0_PAGE_TABLE_END_ADDR, adev->mc.gtt_end >> 12); WREG32(mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR, adev->gart.table_addr >> 12); WREG32(mmVM_CONTEXT0_PROTECTION_FAULT_DEFAULT_ADDR, (u32)(adev->dummy_page.addr >> 12)); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index cb0e50ebb528..d39028440814 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -657,7 +657,7 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev) WREG32(mmVM_L2_CNTL4, tmp); /* setup context0 */ WREG32(mmVM_CONTEXT0_PAGE_TABLE_START_ADDR, adev->mc.gtt_start >> 12); - WREG32(mmVM_CONTEXT0_PAGE_TABLE_END_ADDR, (adev->mc.gtt_end >> 12) - 1); + WREG32(mmVM_CONTEXT0_PAGE_TABLE_END_ADDR, adev->mc.gtt_end >> 12); WREG32(mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR, adev->gart.table_addr >> 12); WREG32(mmVM_CONTEXT0_PROTECTION_FAULT_DEFAULT_ADDR, (u32)(adev->dummy_page.addr >> 12)); diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index 6a52db6ad8d7..370c6c9d81c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -40,6 +40,9 @@ #define GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT 0x04 #define GRBM_GFX_INDEX__VCE_INSTANCE_MASK 0x10 +#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR0 0x8616 +#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR1 0x8617 +#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR2 0x8618 #define VCE_V3_0_FW_SIZE (384 * 1024) #define VCE_V3_0_STACK_SIZE (64 * 1024) @@ -130,9 +133,11 @@ static int vce_v3_0_start(struct amdgpu_device *adev) /* set BUSY flag */ WREG32_P(mmVCE_STATUS, 1, ~1); - - WREG32_P(mmVCE_VCPU_CNTL, VCE_VCPU_CNTL__CLK_EN_MASK, - ~VCE_VCPU_CNTL__CLK_EN_MASK); + if (adev->asic_type >= CHIP_STONEY) + WREG32_P(mmVCE_VCPU_CNTL, 1, ~0x200001); + else + WREG32_P(mmVCE_VCPU_CNTL, VCE_VCPU_CNTL__CLK_EN_MASK, + ~VCE_VCPU_CNTL__CLK_EN_MASK); WREG32_P(mmVCE_SOFT_RESET, VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK, @@ -391,8 +396,12 @@ static void vce_v3_0_mc_resume(struct amdgpu_device *adev, int idx) WREG32(mmVCE_LMI_SWAP_CNTL, 0); WREG32(mmVCE_LMI_SWAP_CNTL1, 0); WREG32(mmVCE_LMI_VM_CTRL, 0); - - WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR, (adev->vce.gpu_addr >> 8)); + if (adev->asic_type >= CHIP_STONEY) { + WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR0, (adev->vce.gpu_addr >> 8)); + WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR1, (adev->vce.gpu_addr >> 8)); + WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR2, (adev->vce.gpu_addr >> 8)); + } else + WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR, (adev->vce.gpu_addr >> 8)); offset = AMDGPU_VCE_FIRMWARE_OFFSET; size = VCE_V3_0_FW_SIZE; WREG32(mmVCE_VCPU_CACHE_OFFSET0, offset & 0x7fffffff); @@ -576,6 +585,11 @@ static int vce_v3_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { DRM_DEBUG("IH: VCE\n"); + + WREG32_P(mmVCE_SYS_INT_STATUS, + VCE_SYS_INT_STATUS__VCE_SYS_INT_TRAP_INTERRUPT_INT_MASK, + ~VCE_SYS_INT_STATUS__VCE_SYS_INT_TRAP_INTERRUPT_INT_MASK); + switch (entry->src_data) { case 0: amdgpu_fence_process(&adev->vce.ring[0]); diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index ea30d6ad4c13..3a4820e863ec 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -30,8 +30,7 @@ #define CREATE_TRACE_POINTS #include "gpu_sched_trace.h" -static struct amd_sched_job * -amd_sched_entity_pop_job(struct amd_sched_entity *entity); +static bool amd_sched_entity_is_ready(struct amd_sched_entity *entity); static void amd_sched_wakeup(struct amd_gpu_scheduler *sched); struct kmem_cache *sched_fence_slab; @@ -64,36 +63,36 @@ static void amd_sched_rq_remove_entity(struct amd_sched_rq *rq, } /** - * Select next job from a specified run queue with round robin policy. - * Return NULL if nothing available. + * Select an entity which could provide a job to run + * + * @rq The run queue to check. + * + * Try to find a ready entity, returns NULL if none found. */ -static struct amd_sched_job * -amd_sched_rq_select_job(struct amd_sched_rq *rq) +static struct amd_sched_entity * +amd_sched_rq_select_entity(struct amd_sched_rq *rq) { struct amd_sched_entity *entity; - struct amd_sched_job *sched_job; spin_lock(&rq->lock); entity = rq->current_entity; if (entity) { list_for_each_entry_continue(entity, &rq->entities, list) { - sched_job = amd_sched_entity_pop_job(entity); - if (sched_job) { + if (amd_sched_entity_is_ready(entity)) { rq->current_entity = entity; spin_unlock(&rq->lock); - return sched_job; + return entity; } } } list_for_each_entry(entity, &rq->entities, list) { - sched_job = amd_sched_entity_pop_job(entity); - if (sched_job) { + if (amd_sched_entity_is_ready(entity)) { rq->current_entity = entity; spin_unlock(&rq->lock); - return sched_job; + return entity; } if (entity == rq->current_entity) @@ -177,6 +176,24 @@ static bool amd_sched_entity_is_idle(struct amd_sched_entity *entity) } /** + * Check if entity is ready + * + * @entity The pointer to a valid scheduler entity + * + * Return true if entity could provide a job. + */ +static bool amd_sched_entity_is_ready(struct amd_sched_entity *entity) +{ + if (kfifo_is_empty(&entity->job_queue)) + return false; + + if (ACCESS_ONCE(entity->dependency)) + return false; + + return true; +} + +/** * Destroy a context entity * * @sched Pointer to scheduler instance @@ -211,32 +228,53 @@ static void amd_sched_entity_wakeup(struct fence *f, struct fence_cb *cb) amd_sched_wakeup(entity->sched); } +static bool amd_sched_entity_add_dependency_cb(struct amd_sched_entity *entity) +{ + struct amd_gpu_scheduler *sched = entity->sched; + struct fence * fence = entity->dependency; + struct amd_sched_fence *s_fence; + + if (fence->context == entity->fence_context) { + /* We can ignore fences from ourself */ + fence_put(entity->dependency); + return false; + } + + s_fence = to_amd_sched_fence(fence); + if (s_fence && s_fence->sched == sched) { + /* Fence is from the same scheduler */ + if (test_bit(AMD_SCHED_FENCE_SCHEDULED_BIT, &fence->flags)) { + /* Ignore it when it is already scheduled */ + fence_put(entity->dependency); + return false; + } + + /* Wait for fence to be scheduled */ + entity->cb.func = amd_sched_entity_wakeup; + list_add_tail(&entity->cb.node, &s_fence->scheduled_cb); + return true; + } + + if (!fence_add_callback(entity->dependency, &entity->cb, + amd_sched_entity_wakeup)) + return true; + + fence_put(entity->dependency); + return false; +} + static struct amd_sched_job * amd_sched_entity_pop_job(struct amd_sched_entity *entity) { struct amd_gpu_scheduler *sched = entity->sched; struct amd_sched_job *sched_job; - if (ACCESS_ONCE(entity->dependency)) - return NULL; - if (!kfifo_out_peek(&entity->job_queue, &sched_job, sizeof(sched_job))) return NULL; - while ((entity->dependency = sched->ops->dependency(sched_job))) { - - if (entity->dependency->context == entity->fence_context) { - /* We can ignore fences from ourself */ - fence_put(entity->dependency); - continue; - } - - if (fence_add_callback(entity->dependency, &entity->cb, - amd_sched_entity_wakeup)) - fence_put(entity->dependency); - else + while ((entity->dependency = sched->ops->dependency(sched_job))) + if (amd_sched_entity_add_dependency_cb(entity)) return NULL; - } return sched_job; } @@ -250,6 +288,7 @@ amd_sched_entity_pop_job(struct amd_sched_entity *entity) */ static bool amd_sched_entity_in(struct amd_sched_job *sched_job) { + struct amd_gpu_scheduler *sched = sched_job->sched; struct amd_sched_entity *entity = sched_job->s_entity; bool added, first = false; @@ -264,7 +303,7 @@ static bool amd_sched_entity_in(struct amd_sched_job *sched_job) /* first job wakes up scheduler */ if (first) - amd_sched_wakeup(sched_job->sched); + amd_sched_wakeup(sched); return added; } @@ -280,9 +319,9 @@ void amd_sched_entity_push_job(struct amd_sched_job *sched_job) { struct amd_sched_entity *entity = sched_job->s_entity; + trace_amd_sched_job(sched_job); wait_event(entity->sched->job_scheduled, amd_sched_entity_in(sched_job)); - trace_amd_sched_job(sched_job); } /** @@ -304,22 +343,22 @@ static void amd_sched_wakeup(struct amd_gpu_scheduler *sched) } /** - * Select next to run + * Select next entity to process */ -static struct amd_sched_job * -amd_sched_select_job(struct amd_gpu_scheduler *sched) +static struct amd_sched_entity * +amd_sched_select_entity(struct amd_gpu_scheduler *sched) { - struct amd_sched_job *sched_job; + struct amd_sched_entity *entity; if (!amd_sched_ready(sched)) return NULL; /* Kernel run queue has higher priority than normal run queue*/ - sched_job = amd_sched_rq_select_job(&sched->kernel_rq); - if (sched_job == NULL) - sched_job = amd_sched_rq_select_job(&sched->sched_rq); + entity = amd_sched_rq_select_entity(&sched->kernel_rq); + if (entity == NULL) + entity = amd_sched_rq_select_entity(&sched->sched_rq); - return sched_job; + return entity; } static void amd_sched_process_job(struct fence *f, struct fence_cb *cb) @@ -381,13 +420,16 @@ static int amd_sched_main(void *param) unsigned long flags; wait_event_interruptible(sched->wake_up_worker, - kthread_should_stop() || - (sched_job = amd_sched_select_job(sched))); + (entity = amd_sched_select_entity(sched)) || + kthread_should_stop()); + if (!entity) + continue; + + sched_job = amd_sched_entity_pop_job(entity); if (!sched_job) continue; - entity = sched_job->s_entity; s_fence = sched_job->s_fence; if (sched->timeout != MAX_SCHEDULE_TIMEOUT) { @@ -400,6 +442,7 @@ static int amd_sched_main(void *param) atomic_inc(&sched->hw_rq_count); fence = sched->ops->run_job(sched_job); + amd_sched_fence_scheduled(s_fence); if (fence) { r = fence_add_callback(fence, &s_fence->cb, amd_sched_process_job); diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index 939692b14f4b..a0f0ae53aacd 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -27,6 +27,8 @@ #include <linux/kfifo.h> #include <linux/fence.h> +#define AMD_SCHED_FENCE_SCHEDULED_BIT FENCE_FLAG_USER_BITS + struct amd_gpu_scheduler; struct amd_sched_rq; @@ -68,6 +70,7 @@ struct amd_sched_rq { struct amd_sched_fence { struct fence base; struct fence_cb cb; + struct list_head scheduled_cb; struct amd_gpu_scheduler *sched; spinlock_t lock; void *owner; @@ -134,7 +137,7 @@ void amd_sched_entity_push_job(struct amd_sched_job *sched_job); struct amd_sched_fence *amd_sched_fence_create( struct amd_sched_entity *s_entity, void *owner); +void amd_sched_fence_scheduled(struct amd_sched_fence *fence); void amd_sched_fence_signal(struct amd_sched_fence *fence); - #endif diff --git a/drivers/gpu/drm/amd/scheduler/sched_fence.c b/drivers/gpu/drm/amd/scheduler/sched_fence.c index 8d2130b9ff05..87c78eecea64 100644 --- a/drivers/gpu/drm/amd/scheduler/sched_fence.c +++ b/drivers/gpu/drm/amd/scheduler/sched_fence.c @@ -35,6 +35,8 @@ struct amd_sched_fence *amd_sched_fence_create(struct amd_sched_entity *s_entity fence = kmem_cache_zalloc(sched_fence_slab, GFP_KERNEL); if (fence == NULL) return NULL; + + INIT_LIST_HEAD(&fence->scheduled_cb); fence->owner = owner; fence->sched = s_entity->sched; spin_lock_init(&fence->lock); @@ -55,6 +57,17 @@ void amd_sched_fence_signal(struct amd_sched_fence *fence) FENCE_TRACE(&fence->base, "was already signaled\n"); } +void amd_sched_fence_scheduled(struct amd_sched_fence *s_fence) +{ + struct fence_cb *cur, *tmp; + + set_bit(AMD_SCHED_FENCE_SCHEDULED_BIT, &s_fence->base.flags); + list_for_each_entry_safe(cur, tmp, &s_fence->scheduled_cb, node) { + list_del_init(&cur->node); + cur->func(&s_fence->base, cur); + } +} + static const char *amd_sched_fence_get_driver_name(struct fence *fence) { return "amd_sched"; diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c index cebcab560626..9bdc28cf927e 100644 --- a/drivers/gpu/drm/armada/armada_crtc.c +++ b/drivers/gpu/drm/armada/armada_crtc.c @@ -1216,14 +1216,14 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev, &armada_primary_plane_funcs, armada_primary_formats, ARRAY_SIZE(armada_primary_formats), - DRM_PLANE_TYPE_PRIMARY); + DRM_PLANE_TYPE_PRIMARY, NULL); if (ret) { kfree(primary); return ret; } ret = drm_crtc_init_with_planes(drm, &dcrtc->crtc, &primary->base, NULL, - &armada_crtc_funcs); + &armada_crtc_funcs, NULL); if (ret) goto err_crtc_init; diff --git a/drivers/gpu/drm/armada/armada_fb.c b/drivers/gpu/drm/armada/armada_fb.c index 1c90969def3e..5fa4bf20b232 100644 --- a/drivers/gpu/drm/armada/armada_fb.c +++ b/drivers/gpu/drm/armada/armada_fb.c @@ -35,7 +35,7 @@ static const struct drm_framebuffer_funcs armada_fb_funcs = { }; struct armada_framebuffer *armada_framebuffer_create(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode, struct armada_gem_object *obj) + const struct drm_mode_fb_cmd2 *mode, struct armada_gem_object *obj) { struct armada_framebuffer *dfb; uint8_t format, config; @@ -101,7 +101,7 @@ struct armada_framebuffer *armada_framebuffer_create(struct drm_device *dev, } static struct drm_framebuffer *armada_fb_create(struct drm_device *dev, - struct drm_file *dfile, struct drm_mode_fb_cmd2 *mode) + struct drm_file *dfile, const struct drm_mode_fb_cmd2 *mode) { struct armada_gem_object *obj; struct armada_framebuffer *dfb; diff --git a/drivers/gpu/drm/armada/armada_fb.h b/drivers/gpu/drm/armada/armada_fb.h index ce3f12ebfc53..48073c4f54d8 100644 --- a/drivers/gpu/drm/armada/armada_fb.h +++ b/drivers/gpu/drm/armada/armada_fb.h @@ -19,6 +19,6 @@ struct armada_framebuffer { #define drm_fb_obj(fb) drm_fb_to_armada_fb(fb)->obj struct armada_framebuffer *armada_framebuffer_create(struct drm_device *, - struct drm_mode_fb_cmd2 *, struct armada_gem_object *); + const struct drm_mode_fb_cmd2 *, struct armada_gem_object *); #endif diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c index 5c22b380f8f3..148e8a42b2c6 100644 --- a/drivers/gpu/drm/armada/armada_overlay.c +++ b/drivers/gpu/drm/armada/armada_overlay.c @@ -460,7 +460,7 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs) &armada_ovl_plane_funcs, armada_ovl_formats, ARRAY_SIZE(armada_ovl_formats), - DRM_PLANE_TYPE_OVERLAY); + DRM_PLANE_TYPE_OVERLAY, NULL); if (ret) { kfree(dplane); return ret; diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 05f6522c0457..eb5715994ac2 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -256,7 +256,6 @@ struct ast_framebuffer { struct ast_fbdev { struct drm_fb_helper helper; struct ast_framebuffer afb; - struct list_head fbdev_list; void *sysram; int size; struct ttm_bo_kmap_obj mapping; @@ -309,7 +308,7 @@ extern void ast_mode_fini(struct drm_device *dev); int ast_framebuffer_init(struct drm_device *dev, struct ast_framebuffer *ast_fb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); int ast_fbdev_init(struct drm_device *dev); diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c index a37e7ea4a00c..5320f8c57884 100644 --- a/drivers/gpu/drm/ast/ast_fb.c +++ b/drivers/gpu/drm/ast/ast_fb.c @@ -163,7 +163,7 @@ static struct fb_ops astfb_ops = { }; static int astfb_create_object(struct ast_fbdev *afbdev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **gobj_p) { struct drm_device *dev = afbdev->helper.dev; diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 541a610667ad..9759009d1da3 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -309,7 +309,7 @@ static const struct drm_framebuffer_funcs ast_fb_funcs = { int ast_framebuffer_init(struct drm_device *dev, struct ast_framebuffer *ast_fb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -327,7 +327,7 @@ int ast_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * ast_user_framebuffer_create(struct drm_device *dev, struct drm_file *filp, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct ast_framebuffer *ast_fb; diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 69d19f3304a5..0123458cbd83 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -751,7 +751,7 @@ static int ast_encoder_init(struct drm_device *dev) return -ENOMEM; drm_encoder_init(dev, &ast_encoder->base, &ast_enc_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(&ast_encoder->base, &ast_enc_helper_funcs); ast_encoder->base.possible_crtcs = 1; diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c index 9f6e234e7029..468a14f266a7 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c @@ -344,7 +344,7 @@ int atmel_hlcdc_crtc_create(struct drm_device *dev) ret = drm_crtc_init_with_planes(dev, &crtc->base, &planes->primary->base, planes->cursor ? &planes->cursor->base : NULL, - &atmel_hlcdc_crtc_funcs); + &atmel_hlcdc_crtc_funcs, NULL); if (ret < 0) goto fail; diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c index 244df0a440b7..816895447155 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c @@ -402,7 +402,7 @@ static irqreturn_t atmel_hlcdc_dc_irq_handler(int irq, void *data) } static struct drm_framebuffer *atmel_hlcdc_fb_create(struct drm_device *dev, - struct drm_file *file_priv, struct drm_mode_fb_cmd2 *mode_cmd) + struct drm_file *file_priv, const struct drm_mode_fb_cmd2 *mode_cmd) { return drm_fb_cma_create(dev, file_priv, mode_cmd); } diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c index 067e4c144bd6..d1129000c5cf 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_output.c @@ -256,7 +256,7 @@ static int atmel_hlcdc_create_panel_output(struct drm_device *dev, &atmel_hlcdc_panel_encoder_helper_funcs); ret = drm_encoder_init(dev, &panel->base.encoder, &atmel_hlcdc_panel_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); if (ret) return ret; diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c index d0299aed517e..1ffe9c329c46 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c @@ -941,7 +941,7 @@ atmel_hlcdc_plane_create(struct drm_device *dev, ret = drm_universal_plane_init(dev, &plane->base, 0, &layer_plane_funcs, desc->formats->formats, - desc->formats->nformats, type); + desc->formats->nformats, type, NULL); if (ret) return ERR_PTR(ret); diff --git a/drivers/gpu/drm/bochs/bochs.h b/drivers/gpu/drm/bochs/bochs.h index 71f2687fc3cc..19b5adaebe24 100644 --- a/drivers/gpu/drm/bochs/bochs.h +++ b/drivers/gpu/drm/bochs/bochs.h @@ -149,7 +149,7 @@ int bochs_dumb_mmap_offset(struct drm_file *file, struct drm_device *dev, int bochs_framebuffer_init(struct drm_device *dev, struct bochs_framebuffer *gfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); int bochs_bo_pin(struct bochs_bo *bo, u32 pl_flag, u64 *gpu_addr); int bochs_bo_unpin(struct bochs_bo *bo); diff --git a/drivers/gpu/drm/bochs/bochs_fbdev.c b/drivers/gpu/drm/bochs/bochs_fbdev.c index 09a0637aab3e..7520bf81fc25 100644 --- a/drivers/gpu/drm/bochs/bochs_fbdev.c +++ b/drivers/gpu/drm/bochs/bochs_fbdev.c @@ -34,7 +34,7 @@ static struct fb_ops bochsfb_ops = { }; static int bochsfb_create_object(struct bochs_device *bochs, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **gobj_p) { struct drm_device *dev = bochs->dev; diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c index 26bcd03a8cb6..a88be6dd34a4 100644 --- a/drivers/gpu/drm/bochs/bochs_kms.c +++ b/drivers/gpu/drm/bochs/bochs_kms.c @@ -196,7 +196,7 @@ static void bochs_encoder_init(struct drm_device *dev) encoder->possible_crtcs = 0x1; drm_encoder_init(dev, encoder, &bochs_encoder_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &bochs_encoder_helper_funcs); } diff --git a/drivers/gpu/drm/bochs/bochs_mm.c b/drivers/gpu/drm/bochs/bochs_mm.c index f69e6bf9bb0e..d812ad014da5 100644 --- a/drivers/gpu/drm/bochs/bochs_mm.c +++ b/drivers/gpu/drm/bochs/bochs_mm.c @@ -484,7 +484,7 @@ static const struct drm_framebuffer_funcs bochs_fb_funcs = { int bochs_framebuffer_init(struct drm_device *dev, struct bochs_framebuffer *gfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -502,7 +502,7 @@ int bochs_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * bochs_user_framebuffer_create(struct drm_device *dev, struct drm_file *filp, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct bochs_framebuffer *bochs_fb; diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.h b/drivers/gpu/drm/cirrus/cirrus_drv.h index 705061537a27..b774d637a00f 100644 --- a/drivers/gpu/drm/cirrus/cirrus_drv.h +++ b/drivers/gpu/drm/cirrus/cirrus_drv.h @@ -153,7 +153,6 @@ struct cirrus_device { struct cirrus_fbdev { struct drm_fb_helper helper; struct cirrus_framebuffer gfb; - struct list_head fbdev_list; void *sysram; int size; int x1, y1, x2, y2; /* dirty rect */ @@ -207,7 +206,7 @@ int cirrus_dumb_create(struct drm_file *file, int cirrus_framebuffer_init(struct drm_device *dev, struct cirrus_framebuffer *gfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); bool cirrus_check_framebuffer(struct cirrus_device *cdev, int width, int height, diff --git a/drivers/gpu/drm/cirrus/cirrus_fbdev.c b/drivers/gpu/drm/cirrus/cirrus_fbdev.c index 589103bcc06c..3b5be7272357 100644 --- a/drivers/gpu/drm/cirrus/cirrus_fbdev.c +++ b/drivers/gpu/drm/cirrus/cirrus_fbdev.c @@ -135,7 +135,7 @@ static struct fb_ops cirrusfb_ops = { }; static int cirrusfb_create_object(struct cirrus_fbdev *afbdev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **gobj_p) { struct drm_device *dev = afbdev->helper.dev; diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c index 055fd86ba717..0907715e90fd 100644 --- a/drivers/gpu/drm/cirrus/cirrus_main.c +++ b/drivers/gpu/drm/cirrus/cirrus_main.c @@ -29,7 +29,7 @@ static const struct drm_framebuffer_funcs cirrus_fb_funcs = { int cirrus_framebuffer_init(struct drm_device *dev, struct cirrus_framebuffer *gfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -47,7 +47,7 @@ int cirrus_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * cirrus_user_framebuffer_create(struct drm_device *dev, struct drm_file *filp, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct cirrus_device *cdev = dev->dev_private; struct drm_gem_object *obj; diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c index 61385f2298bf..276719e52153 100644 --- a/drivers/gpu/drm/cirrus/cirrus_mode.c +++ b/drivers/gpu/drm/cirrus/cirrus_mode.c @@ -489,7 +489,7 @@ static struct drm_encoder *cirrus_encoder_init(struct drm_device *dev) encoder->possible_crtcs = 0x1; drm_encoder_init(dev, encoder, &cirrus_encoder_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &cirrus_encoder_helper_funcs); return encoder; diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index aeee083c7f95..6a21e5c378c1 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -288,8 +288,8 @@ drm_atomic_get_crtc_state(struct drm_atomic_state *state, state->crtcs[index] = crtc; crtc_state->state = state; - DRM_DEBUG_ATOMIC("Added [CRTC:%d] %p state to %p\n", - crtc->base.id, crtc_state, state); + DRM_DEBUG_ATOMIC("Added [CRTC:%d:%s] %p state to %p\n", + crtc->base.id, crtc->name, crtc_state, state); return crtc_state; } @@ -316,8 +316,7 @@ int drm_atomic_set_mode_for_crtc(struct drm_crtc_state *state, if (mode && memcmp(&state->mode, mode, sizeof(*mode)) == 0) return 0; - if (state->mode_blob) - drm_property_unreference_blob(state->mode_blob); + drm_property_unreference_blob(state->mode_blob); state->mode_blob = NULL; if (mode) { @@ -363,8 +362,7 @@ int drm_atomic_set_mode_prop_for_crtc(struct drm_crtc_state *state, if (blob == state->mode_blob) return 0; - if (state->mode_blob) - drm_property_unreference_blob(state->mode_blob); + drm_property_unreference_blob(state->mode_blob); state->mode_blob = NULL; if (blob) { @@ -419,8 +417,7 @@ int drm_atomic_crtc_set_property(struct drm_crtc *crtc, struct drm_property_blob *mode = drm_property_lookup_blob(dev, val); ret = drm_atomic_set_mode_prop_for_crtc(state, mode); - if (mode) - drm_property_unreference_blob(mode); + drm_property_unreference_blob(mode); return ret; } else if (crtc->funcs->atomic_set_property) @@ -432,11 +429,20 @@ int drm_atomic_crtc_set_property(struct drm_crtc *crtc, } EXPORT_SYMBOL(drm_atomic_crtc_set_property); -/* +/** + * drm_atomic_crtc_get_property - get property value from CRTC state + * @crtc: the drm CRTC to set a property on + * @state: the state object to get the property value from + * @property: the property to set + * @val: return location for the property value + * * This function handles generic/core properties and calls out to * driver's ->atomic_get_property() for driver properties. To ensure * consistent behavior you must call this function rather than the * driver hook directly. + * + * RETURNS: + * Zero on success, error code on failure */ static int drm_atomic_crtc_get_property(struct drm_crtc *crtc, @@ -480,8 +486,8 @@ static int drm_atomic_crtc_check(struct drm_crtc *crtc, */ if (state->active && !state->enable) { - DRM_DEBUG_ATOMIC("[CRTC:%d] active without enabled\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] active without enabled\n", + crtc->base.id, crtc->name); return -EINVAL; } @@ -490,15 +496,15 @@ static int drm_atomic_crtc_check(struct drm_crtc *crtc, * be able to trigger. */ if (drm_core_check_feature(crtc->dev, DRIVER_ATOMIC) && WARN_ON(state->enable && !state->mode_blob)) { - DRM_DEBUG_ATOMIC("[CRTC:%d] enabled without mode blob\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] enabled without mode blob\n", + crtc->base.id, crtc->name); return -EINVAL; } if (drm_core_check_feature(crtc->dev, DRIVER_ATOMIC) && WARN_ON(!state->enable && state->mode_blob)) { - DRM_DEBUG_ATOMIC("[CRTC:%d] disabled with mode blob\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] disabled with mode blob\n", + crtc->base.id, crtc->name); return -EINVAL; } @@ -543,8 +549,8 @@ drm_atomic_get_plane_state(struct drm_atomic_state *state, state->planes[index] = plane; plane_state->state = state; - DRM_DEBUG_ATOMIC("Added [PLANE:%d] %p state to %p\n", - plane->base.id, plane_state, state); + DRM_DEBUG_ATOMIC("Added [PLANE:%d:%s] %p state to %p\n", + plane->base.id, plane->name, plane_state, state); if (plane_state->crtc) { struct drm_crtc_state *crtc_state; @@ -619,11 +625,20 @@ int drm_atomic_plane_set_property(struct drm_plane *plane, } EXPORT_SYMBOL(drm_atomic_plane_set_property); -/* +/** + * drm_atomic_plane_get_property - get property value from plane state + * @plane: the drm plane to set a property on + * @state: the state object to get the property value from + * @property: the property to set + * @val: return location for the property value + * * This function handles generic/core properties and calls out to * driver's ->atomic_get_property() for driver properties. To ensure * consistent behavior you must call this function rather than the * driver hook directly. + * + * RETURNS: + * Zero on success, error code on failure */ static int drm_atomic_plane_get_property(struct drm_plane *plane, @@ -755,8 +770,8 @@ static int drm_atomic_plane_check(struct drm_plane *plane, } if (plane_switching_crtc(state->state, plane, state)) { - DRM_DEBUG_ATOMIC("[PLANE:%d] switching CRTC directly\n", - plane->base.id); + DRM_DEBUG_ATOMIC("[PLANE:%d:%s] switching CRTC directly\n", + plane->base.id, plane->name); return -EINVAL; } @@ -875,11 +890,20 @@ int drm_atomic_connector_set_property(struct drm_connector *connector, } EXPORT_SYMBOL(drm_atomic_connector_set_property); -/* +/** + * drm_atomic_connector_get_property - get property value from connector state + * @connector: the drm connector to set a property on + * @state: the state object to get the property value from + * @property: the property to set + * @val: return location for the property value + * * This function handles generic/core properties and calls out to * driver's ->atomic_get_property() for driver properties. To ensure * consistent behavior you must call this function rather than the * driver hook directly. + * + * RETURNS: + * Zero on success, error code on failure */ static int drm_atomic_connector_get_property(struct drm_connector *connector, @@ -980,8 +1004,8 @@ drm_atomic_set_crtc_for_plane(struct drm_plane_state *plane_state, } if (crtc) - DRM_DEBUG_ATOMIC("Link plane state %p to [CRTC:%d]\n", - plane_state, crtc->base.id); + DRM_DEBUG_ATOMIC("Link plane state %p to [CRTC:%d:%s]\n", + plane_state, crtc->base.id, crtc->name); else DRM_DEBUG_ATOMIC("Link plane state %p to [NOCRTC]\n", plane_state); @@ -1048,8 +1072,8 @@ drm_atomic_set_crtc_for_connector(struct drm_connector_state *conn_state, conn_state->crtc = crtc; if (crtc) - DRM_DEBUG_ATOMIC("Link connector state %p to [CRTC:%d]\n", - conn_state, crtc->base.id); + DRM_DEBUG_ATOMIC("Link connector state %p to [CRTC:%d:%s]\n", + conn_state, crtc->base.id, crtc->name); else DRM_DEBUG_ATOMIC("Link connector state %p to [NOCRTC]\n", conn_state); @@ -1088,8 +1112,8 @@ drm_atomic_add_affected_connectors(struct drm_atomic_state *state, if (ret) return ret; - DRM_DEBUG_ATOMIC("Adding all current connectors for [CRTC:%d] to %p\n", - crtc->base.id, state); + DRM_DEBUG_ATOMIC("Adding all current connectors for [CRTC:%d:%s] to %p\n", + crtc->base.id, crtc->name, state); /* * Changed connectors are already in @state, so only need to look at the @@ -1169,8 +1193,9 @@ drm_atomic_connectors_for_crtc(struct drm_atomic_state *state, num_connected_connectors++; } - DRM_DEBUG_ATOMIC("State %p has %i connectors for [CRTC:%d]\n", - state, num_connected_connectors, crtc->base.id); + DRM_DEBUG_ATOMIC("State %p has %i connectors for [CRTC:%d:%s]\n", + state, num_connected_connectors, + crtc->base.id, crtc->name); return num_connected_connectors; } @@ -1191,12 +1216,7 @@ void drm_atomic_legacy_backoff(struct drm_atomic_state *state) retry: drm_modeset_backoff(state->acquire_ctx); - ret = drm_modeset_lock(&state->dev->mode_config.connection_mutex, - state->acquire_ctx); - if (ret) - goto retry; - ret = drm_modeset_lock_all_crtcs(state->dev, - state->acquire_ctx); + ret = drm_modeset_lock_all_ctx(state->dev, state->acquire_ctx); if (ret) goto retry; } @@ -1228,8 +1248,8 @@ int drm_atomic_check_only(struct drm_atomic_state *state) for_each_plane_in_state(state, plane, plane_state, i) { ret = drm_atomic_plane_check(plane, plane_state); if (ret) { - DRM_DEBUG_ATOMIC("[PLANE:%d] atomic core check failed\n", - plane->base.id); + DRM_DEBUG_ATOMIC("[PLANE:%d:%s] atomic core check failed\n", + plane->base.id, plane->name); return ret; } } @@ -1237,8 +1257,8 @@ int drm_atomic_check_only(struct drm_atomic_state *state) for_each_crtc_in_state(state, crtc, crtc_state, i) { ret = drm_atomic_crtc_check(crtc, crtc_state); if (ret) { - DRM_DEBUG_ATOMIC("[CRTC:%d] atomic core check failed\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] atomic core check failed\n", + crtc->base.id, crtc->name); return ret; } } @@ -1249,8 +1269,8 @@ int drm_atomic_check_only(struct drm_atomic_state *state) if (!state->allow_modeset) { for_each_crtc_in_state(state, crtc, crtc_state, i) { if (drm_atomic_crtc_needs_modeset(crtc_state)) { - DRM_DEBUG_ATOMIC("[CRTC:%d] requires full modeset\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] requires full modeset\n", + crtc->base.id, crtc->name); return -EINVAL; } } @@ -1433,7 +1453,7 @@ static int atomic_set_prop(struct drm_atomic_state *state, } /** - * drm_atomic_update_old_fb -- Unset old_fb pointers and set plane->fb pointers. + * drm_atomic_clean_old_fb -- Unset old_fb pointers and set plane->fb pointers. * * @dev: drm device to check. * @plane_mask: plane mask for planes that were updated. diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index e5aec45bf985..63f925b75357 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -52,6 +52,12 @@ * drm_atomic_helper_disable_plane(), drm_atomic_helper_disable_plane() and the * various functions to implement set_property callbacks. New drivers must not * implement these functions themselves but must use the provided helpers. + * + * The atomic helper uses the same function table structures as all other + * modesetting helpers. See the documentation for struct &drm_crtc_helper_funcs, + * struct &drm_encoder_helper_funcs and struct &drm_connector_helper_funcs. It + * also shares the struct &drm_plane_helper_funcs function table with the plane + * helpers. */ static void drm_atomic_helper_plane_changed(struct drm_atomic_state *state, @@ -80,6 +86,27 @@ drm_atomic_helper_plane_changed(struct drm_atomic_state *state, } } +static bool +check_pending_encoder_assignment(struct drm_atomic_state *state, + struct drm_encoder *new_encoder, + struct drm_connector *new_connector) +{ + struct drm_connector *connector; + struct drm_connector_state *conn_state; + int i; + + for_each_connector_in_state(state, connector, conn_state, i) { + if (conn_state->best_encoder != new_encoder) + continue; + + /* encoder already assigned and we're trying to re-steal it! */ + if (connector->state->best_encoder != conn_state->best_encoder) + return false; + } + + return true; +} + static struct drm_crtc * get_current_crtc_for_encoder(struct drm_device *dev, struct drm_encoder *encoder) @@ -116,9 +143,9 @@ steal_encoder(struct drm_atomic_state *state, */ WARN_ON(!drm_modeset_is_locked(&config->connection_mutex)); - DRM_DEBUG_ATOMIC("[ENCODER:%d:%s] in use on [CRTC:%d], stealing it\n", + DRM_DEBUG_ATOMIC("[ENCODER:%d:%s] in use on [CRTC:%d:%s], stealing it\n", encoder->base.id, encoder->name, - encoder_crtc->base.id); + encoder_crtc->base.id, encoder_crtc->name); crtc_state = drm_atomic_get_crtc_state(state, encoder_crtc); if (IS_ERR(crtc_state)) @@ -219,16 +246,24 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx) } if (new_encoder == connector_state->best_encoder) { - DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] keeps [ENCODER:%d:%s], now on [CRTC:%d]\n", + DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] keeps [ENCODER:%d:%s], now on [CRTC:%d:%s]\n", connector->base.id, connector->name, new_encoder->base.id, new_encoder->name, - connector_state->crtc->base.id); + connector_state->crtc->base.id, + connector_state->crtc->name); return 0; } + if (!check_pending_encoder_assignment(state, new_encoder, connector)) { + DRM_DEBUG_ATOMIC("Encoder for [CONNECTOR:%d:%s] already assigned\n", + connector->base.id, + connector->name); + return -EINVAL; + } + encoder_crtc = get_current_crtc_for_encoder(state->dev, new_encoder); @@ -251,12 +286,13 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx) crtc_state = state->crtc_states[idx]; crtc_state->connectors_changed = true; - DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] using [ENCODER:%d:%s] on [CRTC:%d]\n", + DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] using [ENCODER:%d:%s] on [CRTC:%d:%s]\n", connector->base.id, connector->name, new_encoder->base.id, new_encoder->name, - connector_state->crtc->base.id); + connector_state->crtc->base.id, + connector_state->crtc->name); return 0; } @@ -340,8 +376,8 @@ mode_fixup(struct drm_atomic_state *state) ret = funcs->mode_fixup(crtc, &crtc_state->mode, &crtc_state->adjusted_mode); if (!ret) { - DRM_DEBUG_ATOMIC("[CRTC:%d] fixup failed\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] fixup failed\n", + crtc->base.id, crtc->name); return -EINVAL; } } @@ -388,14 +424,14 @@ drm_atomic_helper_check_modeset(struct drm_device *dev, for_each_crtc_in_state(state, crtc, crtc_state, i) { if (!drm_mode_equal(&crtc->state->mode, &crtc_state->mode)) { - DRM_DEBUG_ATOMIC("[CRTC:%d] mode changed\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] mode changed\n", + crtc->base.id, crtc->name); crtc_state->mode_changed = true; } if (crtc->state->enable != crtc_state->enable) { - DRM_DEBUG_ATOMIC("[CRTC:%d] enable changed\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] enable changed\n", + crtc->base.id, crtc->name); /* * For clarity this assignment is done here, but @@ -436,18 +472,18 @@ drm_atomic_helper_check_modeset(struct drm_device *dev, * a full modeset because update_connector_routing force that. */ if (crtc->state->active != crtc_state->active) { - DRM_DEBUG_ATOMIC("[CRTC:%d] active changed\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] active changed\n", + crtc->base.id, crtc->name); crtc_state->active_changed = true; } if (!drm_atomic_crtc_needs_modeset(crtc_state)) continue; - DRM_DEBUG_ATOMIC("[CRTC:%d] needs all connectors, enable: %c, active: %c\n", - crtc->base.id, + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] needs all connectors, enable: %c, active: %c\n", + crtc->base.id, crtc->name, crtc_state->enable ? 'y' : 'n', - crtc_state->active ? 'y' : 'n'); + crtc_state->active ? 'y' : 'n'); ret = drm_atomic_add_affected_connectors(state, crtc); if (ret != 0) @@ -461,8 +497,8 @@ drm_atomic_helper_check_modeset(struct drm_device *dev, crtc); if (crtc_state->enable != !!num_connectors) { - DRM_DEBUG_ATOMIC("[CRTC:%d] enabled/connectors mismatch\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] enabled/connectors mismatch\n", + crtc->base.id, crtc->name); return -EINVAL; } @@ -509,8 +545,8 @@ drm_atomic_helper_check_planes(struct drm_device *dev, ret = funcs->atomic_check(plane, plane_state); if (ret) { - DRM_DEBUG_ATOMIC("[PLANE:%d] atomic driver check failed\n", - plane->base.id); + DRM_DEBUG_ATOMIC("[PLANE:%d:%s] atomic driver check failed\n", + plane->base.id, plane->name); return ret; } } @@ -525,8 +561,8 @@ drm_atomic_helper_check_planes(struct drm_device *dev, ret = funcs->atomic_check(crtc, state->crtc_states[i]); if (ret) { - DRM_DEBUG_ATOMIC("[CRTC:%d] atomic driver check failed\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("[CRTC:%d:%s] atomic driver check failed\n", + crtc->base.id, crtc->name); return ret; } } @@ -639,8 +675,8 @@ disable_outputs(struct drm_device *dev, struct drm_atomic_state *old_state) funcs = crtc->helper_private; - DRM_DEBUG_ATOMIC("disabling [CRTC:%d]\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("disabling [CRTC:%d:%s]\n", + crtc->base.id, crtc->name); /* Right function depends upon target state. */ @@ -751,8 +787,8 @@ crtc_set_mode(struct drm_device *dev, struct drm_atomic_state *old_state) funcs = crtc->helper_private; if (crtc->state->enable && funcs->mode_set_nofb) { - DRM_DEBUG_ATOMIC("modeset on [CRTC:%d]\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("modeset on [CRTC:%d:%s]\n", + crtc->base.id, crtc->name); funcs->mode_set_nofb(crtc); } @@ -851,8 +887,8 @@ void drm_atomic_helper_commit_modeset_enables(struct drm_device *dev, funcs = crtc->helper_private; if (crtc->state->enable) { - DRM_DEBUG_ATOMIC("enabling [CRTC:%d]\n", - crtc->base.id); + DRM_DEBUG_ATOMIC("enabling [CRTC:%d:%s]\n", + crtc->base.id, crtc->name); if (funcs->enable) funcs->enable(crtc); @@ -1342,6 +1378,49 @@ drm_atomic_helper_commit_planes_on_crtc(struct drm_crtc_state *old_crtc_state) EXPORT_SYMBOL(drm_atomic_helper_commit_planes_on_crtc); /** + * drm_atomic_helper_disable_planes_on_crtc - helper to disable CRTC's planes + * @crtc: CRTC + * @atomic: if set, synchronize with CRTC's atomic_begin/flush hooks + * + * Disables all planes associated with the given CRTC. This can be + * used for instance in the CRTC helper disable callback to disable + * all planes before shutting down the display pipeline. + * + * If the atomic-parameter is set the function calls the CRTC's + * atomic_begin hook before and atomic_flush hook after disabling the + * planes. + * + * It is a bug to call this function without having implemented the + * ->atomic_disable() plane hook. + */ +void drm_atomic_helper_disable_planes_on_crtc(struct drm_crtc *crtc, + bool atomic) +{ + const struct drm_crtc_helper_funcs *crtc_funcs = + crtc->helper_private; + struct drm_plane *plane; + + if (atomic && crtc_funcs && crtc_funcs->atomic_begin) + crtc_funcs->atomic_begin(crtc, NULL); + + drm_for_each_plane(plane, crtc->dev) { + const struct drm_plane_helper_funcs *plane_funcs = + plane->helper_private; + + if (plane->state->crtc != crtc || !plane_funcs) + continue; + + WARN_ON(!plane_funcs->atomic_disable); + if (plane_funcs->atomic_disable) + plane_funcs->atomic_disable(plane, NULL); + } + + if (atomic && crtc_funcs && crtc_funcs->atomic_flush) + crtc_funcs->atomic_flush(crtc, NULL); +} +EXPORT_SYMBOL(drm_atomic_helper_disable_planes_on_crtc); + +/** * drm_atomic_helper_cleanup_planes - cleanup plane resources after commit * @dev: DRM device * @old_state: atomic state object with old state structures @@ -1485,12 +1564,12 @@ retry: drm_atomic_set_fb_for_plane(plane_state, fb); plane_state->crtc_x = crtc_x; plane_state->crtc_y = crtc_y; - plane_state->crtc_h = crtc_h; plane_state->crtc_w = crtc_w; + plane_state->crtc_h = crtc_h; plane_state->src_x = src_x; plane_state->src_y = src_y; - plane_state->src_h = src_h; plane_state->src_w = src_w; + plane_state->src_h = src_h; if (plane == crtc->cursor) state->legacy_cursor_update = true; @@ -1609,12 +1688,12 @@ int __drm_atomic_helper_disable_plane(struct drm_plane *plane, drm_atomic_set_fb_for_plane(plane_state, NULL); plane_state->crtc_x = 0; plane_state->crtc_y = 0; - plane_state->crtc_h = 0; plane_state->crtc_w = 0; + plane_state->crtc_h = 0; plane_state->src_x = 0; plane_state->src_y = 0; - plane_state->src_h = 0; plane_state->src_w = 0; + plane_state->src_h = 0; return 0; } @@ -1797,16 +1876,16 @@ int __drm_atomic_helper_set_config(struct drm_mode_set *set, drm_atomic_set_fb_for_plane(primary_state, set->fb); primary_state->crtc_x = 0; primary_state->crtc_y = 0; - primary_state->crtc_h = vdisplay; primary_state->crtc_w = hdisplay; + primary_state->crtc_h = vdisplay; primary_state->src_x = set->x << 16; primary_state->src_y = set->y << 16; if (primary_state->rotation & (BIT(DRM_ROTATE_90) | BIT(DRM_ROTATE_270))) { - primary_state->src_h = hdisplay << 16; primary_state->src_w = vdisplay << 16; + primary_state->src_h = hdisplay << 16; } else { - primary_state->src_h = vdisplay << 16; primary_state->src_w = hdisplay << 16; + primary_state->src_h = vdisplay << 16; } commit: @@ -1818,6 +1897,161 @@ commit: } /** + * drm_atomic_helper_disable_all - disable all currently active outputs + * @dev: DRM device + * @ctx: lock acquisition context + * + * Loops through all connectors, finding those that aren't turned off and then + * turns them off by setting their DPMS mode to OFF and deactivating the CRTC + * that they are connected to. + * + * This is used for example in suspend/resume to disable all currently active + * functions when suspending. + * + * Note that if callers haven't already acquired all modeset locks this might + * return -EDEADLK, which must be handled by calling drm_modeset_backoff(). + * + * Returns: + * 0 on success or a negative error code on failure. + * + * See also: + * drm_atomic_helper_suspend(), drm_atomic_helper_resume() + */ +int drm_atomic_helper_disable_all(struct drm_device *dev, + struct drm_modeset_acquire_ctx *ctx) +{ + struct drm_atomic_state *state; + struct drm_connector *conn; + int err; + + state = drm_atomic_state_alloc(dev); + if (!state) + return -ENOMEM; + + state->acquire_ctx = ctx; + + drm_for_each_connector(conn, dev) { + struct drm_crtc *crtc = conn->state->crtc; + struct drm_crtc_state *crtc_state; + + if (!crtc || conn->dpms != DRM_MODE_DPMS_ON) + continue; + + crtc_state = drm_atomic_get_crtc_state(state, crtc); + if (IS_ERR(crtc_state)) { + err = PTR_ERR(crtc_state); + goto free; + } + + crtc_state->active = false; + } + + err = drm_atomic_commit(state); + +free: + if (err < 0) + drm_atomic_state_free(state); + + return err; +} +EXPORT_SYMBOL(drm_atomic_helper_disable_all); + +/** + * drm_atomic_helper_suspend - subsystem-level suspend helper + * @dev: DRM device + * + * Duplicates the current atomic state, disables all active outputs and then + * returns a pointer to the original atomic state to the caller. Drivers can + * pass this pointer to the drm_atomic_helper_resume() helper upon resume to + * restore the output configuration that was active at the time the system + * entered suspend. + * + * Note that it is potentially unsafe to use this. The atomic state object + * returned by this function is assumed to be persistent. Drivers must ensure + * that this holds true. Before calling this function, drivers must make sure + * to suspend fbdev emulation so that nothing can be using the device. + * + * Returns: + * A pointer to a copy of the state before suspend on success or an ERR_PTR()- + * encoded error code on failure. Drivers should store the returned atomic + * state object and pass it to the drm_atomic_helper_resume() helper upon + * resume. + * + * See also: + * drm_atomic_helper_duplicate_state(), drm_atomic_helper_disable_all(), + * drm_atomic_helper_resume() + */ +struct drm_atomic_state *drm_atomic_helper_suspend(struct drm_device *dev) +{ + struct drm_modeset_acquire_ctx ctx; + struct drm_atomic_state *state; + int err; + + drm_modeset_acquire_init(&ctx, 0); + +retry: + err = drm_modeset_lock_all_ctx(dev, &ctx); + if (err < 0) { + state = ERR_PTR(err); + goto unlock; + } + + state = drm_atomic_helper_duplicate_state(dev, &ctx); + if (IS_ERR(state)) + goto unlock; + + err = drm_atomic_helper_disable_all(dev, &ctx); + if (err < 0) { + drm_atomic_state_free(state); + state = ERR_PTR(err); + goto unlock; + } + +unlock: + if (PTR_ERR(state) == -EDEADLK) { + drm_modeset_backoff(&ctx); + goto retry; + } + + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); + return state; +} +EXPORT_SYMBOL(drm_atomic_helper_suspend); + +/** + * drm_atomic_helper_resume - subsystem-level resume helper + * @dev: DRM device + * @state: atomic state to resume to + * + * Calls drm_mode_config_reset() to synchronize hardware and software states, + * grabs all modeset locks and commits the atomic state object. This can be + * used in conjunction with the drm_atomic_helper_suspend() helper to + * implement suspend/resume for drivers that support atomic mode-setting. + * + * Returns: + * 0 on success or a negative error code on failure. + * + * See also: + * drm_atomic_helper_suspend() + */ +int drm_atomic_helper_resume(struct drm_device *dev, + struct drm_atomic_state *state) +{ + struct drm_mode_config *config = &dev->mode_config; + int err; + + drm_mode_config_reset(dev); + drm_modeset_lock_all(dev); + state->acquire_ctx = config->acquire_ctx; + err = drm_atomic_commit(state); + drm_modeset_unlock_all(dev); + + return err; +} +EXPORT_SYMBOL(drm_atomic_helper_resume); + +/** * drm_atomic_helper_crtc_set_property - helper for crtc properties * @crtc: DRM crtc * @property: DRM property @@ -2173,6 +2407,12 @@ EXPORT_SYMBOL(drm_atomic_helper_connector_dpms); * The simpler solution is to just reset the software state to everything off, * which is easiest to do by calling drm_mode_config_reset(). To facilitate this * the atomic helpers provide default reset implementations for all hooks. + * + * On the upside the precise state tracking of atomic simplifies system suspend + * and resume a lot. For drivers using drm_mode_config_reset() a complete recipe + * is implemented in drm_atomic_helper_suspend() and drm_atomic_helper_resume(). + * For other drivers the building blocks are split out, see the documentation + * for these functions. */ /** @@ -2184,7 +2424,7 @@ EXPORT_SYMBOL(drm_atomic_helper_connector_dpms); */ void drm_atomic_helper_crtc_reset(struct drm_crtc *crtc) { - if (crtc->state && crtc->state->mode_blob) + if (crtc->state) drm_property_unreference_blob(crtc->state->mode_blob); kfree(crtc->state); crtc->state = kzalloc(sizeof(*crtc->state), GFP_KERNEL); @@ -2252,8 +2492,7 @@ EXPORT_SYMBOL(drm_atomic_helper_crtc_duplicate_state); void __drm_atomic_helper_crtc_destroy_state(struct drm_crtc *crtc, struct drm_crtc_state *state) { - if (state->mode_blob) - drm_property_unreference_blob(state->mode_blob); + drm_property_unreference_blob(state->mode_blob); } EXPORT_SYMBOL(__drm_atomic_helper_crtc_destroy_state); @@ -2430,7 +2669,9 @@ EXPORT_SYMBOL(drm_atomic_helper_connector_duplicate_state); * @ctx: lock acquisition context * * Makes a copy of the current atomic state by looping over all objects and - * duplicating their respective states. + * duplicating their respective states. This is used for example by suspend/ + * resume support code to save the state prior to suspend such that it can + * be restored upon resume. * * Note that this treats atomic state as persistent between save and restore. * Drivers must make sure that this is possible and won't result in confusion @@ -2442,6 +2683,9 @@ EXPORT_SYMBOL(drm_atomic_helper_connector_duplicate_state); * Returns: * A pointer to the copy of the atomic state object on success or an * ERR_PTR()-encoded error code on failure. + * + * See also: + * drm_atomic_helper_suspend(), drm_atomic_helper_resume() */ struct drm_atomic_state * drm_atomic_helper_duplicate_state(struct drm_device *dev, diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c index 6b8f7211e543..bd93453afa61 100644 --- a/drivers/gpu/drm/drm_bridge.c +++ b/drivers/gpu/drm/drm_bridge.c @@ -31,14 +31,14 @@ /** * DOC: overview * - * drm_bridge represents a device that hangs on to an encoder. These are handy - * when a regular drm_encoder entity isn't enough to represent the entire + * struct &drm_bridge represents a device that hangs on to an encoder. These are + * handy when a regular &drm_encoder entity isn't enough to represent the entire * encoder chain. * - * A bridge is always associated to a single drm_encoder at a time, but can be + * A bridge is always attached to a single &drm_encoder at a time, but can be * either connected to it directly, or through an intermediate bridge: * - * encoder ---> bridge B ---> bridge A + * encoder ---> bridge B ---> bridge A * * Here, the output of the encoder feeds to bridge B, and that furthers feeds to * bridge A. @@ -46,11 +46,16 @@ * The driver using the bridge is responsible to make the associations between * the encoder and bridges. Once these links are made, the bridges will * participate along with encoder functions to perform mode_set/enable/disable - * through the ops provided in drm_bridge_funcs. + * through the ops provided in &drm_bridge_funcs. * * drm_bridge, like drm_panel, aren't drm_mode_object entities like planes, - * crtcs, encoders or connectors. They just provide additional hooks to get the - * desired output at the end of the encoder chain. + * CRTCs, encoders or connectors and hence are not visible to userspace. They + * just provide additional hooks to get the desired output at the end of the + * encoder chain. + * + * Bridges can also be chained up using the next pointer in struct &drm_bridge. + * + * Both legacy CRTC helpers and the new atomic modeset helpers support bridges. */ static DEFINE_MUTEX(bridge_lock); @@ -122,34 +127,12 @@ EXPORT_SYMBOL(drm_bridge_attach); /** * DOC: bridge callbacks * - * The drm_bridge_funcs ops are populated by the bridge driver. The drm - * internals(atomic and crtc helpers) use the helpers defined in drm_bridge.c - * These helpers call a specific drm_bridge_funcs op for all the bridges + * The &drm_bridge_funcs ops are populated by the bridge driver. The DRM + * internals (atomic and CRTC helpers) use the helpers defined in drm_bridge.c + * These helpers call a specific &drm_bridge_funcs op for all the bridges * during encoder configuration. * - * When creating a bridge driver, one can implement drm_bridge_funcs op with - * the help of these rough rules: - * - * pre_enable: this contains things needed to be done for the bridge before - * its clock and timings are enabled by its source. For a bridge, its source - * is generally the encoder or bridge just before it in the encoder chain. - * - * enable: this contains things needed to be done for the bridge once its - * source is enabled. In other words, enable is called once the source is - * ready with clock and timing needed by the bridge. - * - * disable: this contains things needed to be done for the bridge assuming - * that its source is still enabled, i.e. clock and timings are still on. - * - * post_disable: this contains things needed to be done for the bridge once - * its source is disabled, i.e. once clocks and timings are off. - * - * mode_fixup: this should fixup the given mode for the bridge. It is called - * after the encoder's mode fixup. mode_fixup can also reject a mode completely - * if it's unsuitable for the hardware. - * - * mode_set: this sets up the mode for the bridge. It assumes that its source - * (an encoder or a bridge) has set the mode too. + * For detailed specification of the bridge callbacks see &drm_bridge_funcs. */ /** @@ -159,7 +142,7 @@ EXPORT_SYMBOL(drm_bridge_attach); * @mode: desired mode to be set for the bridge * @adjusted_mode: updated mode that works for this bridge * - * Calls 'mode_fixup' drm_bridge_funcs op for all the bridges in the + * Calls ->mode_fixup() &drm_bridge_funcs op for all the bridges in the * encoder chain, starting from the first bridge to the last. * * Note: the bridge passed should be the one closest to the encoder @@ -186,11 +169,11 @@ bool drm_bridge_mode_fixup(struct drm_bridge *bridge, EXPORT_SYMBOL(drm_bridge_mode_fixup); /** - * drm_bridge_disable - calls 'disable' drm_bridge_funcs op for all + * drm_bridge_disable - calls ->disable() &drm_bridge_funcs op for all * bridges in the encoder chain. * @bridge: bridge control structure * - * Calls 'disable' drm_bridge_funcs op for all the bridges in the encoder + * Calls ->disable() &drm_bridge_funcs op for all the bridges in the encoder * chain, starting from the last bridge to the first. These are called before * calling the encoder's prepare op. * @@ -208,11 +191,11 @@ void drm_bridge_disable(struct drm_bridge *bridge) EXPORT_SYMBOL(drm_bridge_disable); /** - * drm_bridge_post_disable - calls 'post_disable' drm_bridge_funcs op for + * drm_bridge_post_disable - calls ->post_disable() &drm_bridge_funcs op for * all bridges in the encoder chain. * @bridge: bridge control structure * - * Calls 'post_disable' drm_bridge_funcs op for all the bridges in the + * Calls ->post_disable() &drm_bridge_funcs op for all the bridges in the * encoder chain, starting from the first bridge to the last. These are called * after completing the encoder's prepare op. * @@ -236,7 +219,7 @@ EXPORT_SYMBOL(drm_bridge_post_disable); * @mode: desired mode to be set for the bridge * @adjusted_mode: updated mode that works for this bridge * - * Calls 'mode_set' drm_bridge_funcs op for all the bridges in the + * Calls ->mode_set() &drm_bridge_funcs op for all the bridges in the * encoder chain, starting from the first bridge to the last. * * Note: the bridge passed should be the one closest to the encoder @@ -256,11 +239,11 @@ void drm_bridge_mode_set(struct drm_bridge *bridge, EXPORT_SYMBOL(drm_bridge_mode_set); /** - * drm_bridge_pre_enable - calls 'pre_enable' drm_bridge_funcs op for all + * drm_bridge_pre_enable - calls ->pre_enable() &drm_bridge_funcs op for all * bridges in the encoder chain. * @bridge: bridge control structure * - * Calls 'pre_enable' drm_bridge_funcs op for all the bridges in the encoder + * Calls ->pre_enable() &drm_bridge_funcs op for all the bridges in the encoder * chain, starting from the last bridge to the first. These are called * before calling the encoder's commit op. * @@ -278,11 +261,11 @@ void drm_bridge_pre_enable(struct drm_bridge *bridge) EXPORT_SYMBOL(drm_bridge_pre_enable); /** - * drm_bridge_enable - calls 'enable' drm_bridge_funcs op for all bridges + * drm_bridge_enable - calls ->enable() &drm_bridge_funcs op for all bridges * in the encoder chain. * @bridge: bridge control structure * - * Calls 'enable' drm_bridge_funcs op for all the bridges in the encoder + * Calls ->enable() &drm_bridge_funcs op for all the bridges in the encoder * chain, starting from the first bridge to the last. These are called * after completing the encoder's commit op. * diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 24c5434abd1c..62fa95fa5471 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -45,7 +45,7 @@ static struct drm_framebuffer * internal_framebuffer_create(struct drm_device *dev, - struct drm_mode_fb_cmd2 *r, + const struct drm_mode_fb_cmd2 *r, struct drm_file *file_priv); /* Avoid boilerplate. I'm tired of typing. */ @@ -649,6 +649,18 @@ EXPORT_SYMBOL(drm_framebuffer_remove); DEFINE_WW_CLASS(crtc_ww_class); +static unsigned int drm_num_crtcs(struct drm_device *dev) +{ + unsigned int num = 0; + struct drm_crtc *tmp; + + drm_for_each_crtc(tmp, dev) { + num++; + } + + return num; +} + /** * drm_crtc_init_with_planes - Initialise a new CRTC object with * specified primary and cursor planes. @@ -657,6 +669,7 @@ DEFINE_WW_CLASS(crtc_ww_class); * @primary: Primary plane for CRTC * @cursor: Cursor plane for CRTC * @funcs: callbacks for the new CRTC + * @name: printf style format string for the CRTC name, or NULL for default name * * Inits a new object created as base part of a driver crtc object. * @@ -666,7 +679,8 @@ DEFINE_WW_CLASS(crtc_ww_class); int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc, struct drm_plane *primary, struct drm_plane *cursor, - const struct drm_crtc_funcs *funcs) + const struct drm_crtc_funcs *funcs, + const char *name, ...) { struct drm_mode_config *config = &dev->mode_config; int ret; @@ -682,6 +696,21 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc, if (ret) return ret; + if (name) { + va_list ap; + + va_start(ap, name); + crtc->name = kvasprintf(GFP_KERNEL, name, ap); + va_end(ap); + } else { + crtc->name = kasprintf(GFP_KERNEL, "crtc-%d", + drm_num_crtcs(dev)); + } + if (!crtc->name) { + drm_mode_object_put(dev, &crtc->base); + return -ENOMEM; + } + crtc->base.properties = &crtc->properties; list_add_tail(&crtc->head, &config->crtc_list); @@ -728,6 +757,8 @@ void drm_crtc_cleanup(struct drm_crtc *crtc) if (crtc->state && crtc->funcs->atomic_destroy_state) crtc->funcs->atomic_destroy_state(crtc, crtc->state); + kfree(crtc->name); + memset(crtc, 0, sizeof(*crtc)); } EXPORT_SYMBOL(drm_crtc_cleanup); @@ -1075,6 +1106,7 @@ EXPORT_SYMBOL(drm_connector_unplug_all); * @encoder: the encoder to init * @funcs: callbacks for this encoder * @encoder_type: user visible type of the encoder + * @name: printf style format string for the encoder name, or NULL for default name * * Initialises a preallocated encoder. Encoder should be * subclassed as part of driver encoder objects. @@ -1085,7 +1117,7 @@ EXPORT_SYMBOL(drm_connector_unplug_all); int drm_encoder_init(struct drm_device *dev, struct drm_encoder *encoder, const struct drm_encoder_funcs *funcs, - int encoder_type) + int encoder_type, const char *name, ...) { int ret; @@ -1098,9 +1130,17 @@ int drm_encoder_init(struct drm_device *dev, encoder->dev = dev; encoder->encoder_type = encoder_type; encoder->funcs = funcs; - encoder->name = kasprintf(GFP_KERNEL, "%s-%d", - drm_encoder_enum_list[encoder_type].name, - encoder->base.id); + if (name) { + va_list ap; + + va_start(ap, name); + encoder->name = kvasprintf(GFP_KERNEL, name, ap); + va_end(ap); + } else { + encoder->name = kasprintf(GFP_KERNEL, "%s-%d", + drm_encoder_enum_list[encoder_type].name, + encoder->base.id); + } if (!encoder->name) { ret = -ENOMEM; goto out_put; @@ -1141,6 +1181,18 @@ void drm_encoder_cleanup(struct drm_encoder *encoder) } EXPORT_SYMBOL(drm_encoder_cleanup); +static unsigned int drm_num_planes(struct drm_device *dev) +{ + unsigned int num = 0; + struct drm_plane *tmp; + + drm_for_each_plane(tmp, dev) { + num++; + } + + return num; +} + /** * drm_universal_plane_init - Initialize a new universal plane object * @dev: DRM device @@ -1150,6 +1202,7 @@ EXPORT_SYMBOL(drm_encoder_cleanup); * @formats: array of supported formats (%DRM_FORMAT_*) * @format_count: number of elements in @formats * @type: type of plane (overlay, primary, cursor) + * @name: printf style format string for the plane name, or NULL for default name * * Initializes a plane object of type @type. * @@ -1160,7 +1213,8 @@ int drm_universal_plane_init(struct drm_device *dev, struct drm_plane *plane, unsigned long possible_crtcs, const struct drm_plane_funcs *funcs, const uint32_t *formats, unsigned int format_count, - enum drm_plane_type type) + enum drm_plane_type type, + const char *name, ...) { struct drm_mode_config *config = &dev->mode_config; int ret; @@ -1182,6 +1236,22 @@ int drm_universal_plane_init(struct drm_device *dev, struct drm_plane *plane, return -ENOMEM; } + if (name) { + va_list ap; + + va_start(ap, name); + plane->name = kvasprintf(GFP_KERNEL, name, ap); + va_end(ap); + } else { + plane->name = kasprintf(GFP_KERNEL, "plane-%d", + drm_num_planes(dev)); + } + if (!plane->name) { + kfree(plane->format_types); + drm_mode_object_put(dev, &plane->base); + return -ENOMEM; + } + memcpy(plane->format_types, formats, format_count * sizeof(uint32_t)); plane->format_count = format_count; plane->possible_crtcs = possible_crtcs; @@ -1240,7 +1310,7 @@ int drm_plane_init(struct drm_device *dev, struct drm_plane *plane, type = is_primary ? DRM_PLANE_TYPE_PRIMARY : DRM_PLANE_TYPE_OVERLAY; return drm_universal_plane_init(dev, plane, possible_crtcs, funcs, - formats, format_count, type); + formats, format_count, type, NULL); } EXPORT_SYMBOL(drm_plane_init); @@ -1272,6 +1342,8 @@ void drm_plane_cleanup(struct drm_plane *plane) if (plane->state && plane->funcs->atomic_destroy_state) plane->funcs->atomic_destroy_state(plane, plane->state); + kfree(plane->name); + memset(plane, 0, sizeof(*plane)); } EXPORT_SYMBOL(drm_plane_cleanup); @@ -1801,7 +1873,8 @@ int drm_mode_getresources(struct drm_device *dev, void *data, copied = 0; crtc_id = (uint32_t __user *)(unsigned long)card_res->crtc_id_ptr; drm_for_each_crtc(crtc, dev) { - DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id); + DRM_DEBUG_KMS("[CRTC:%d:%s]\n", + crtc->base.id, crtc->name); if (put_user(crtc->base.id, crtc_id + copied)) { ret = -EFAULT; goto out; @@ -2646,7 +2719,7 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data, ret = -ENOENT; goto out; } - DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id); + DRM_DEBUG_KMS("[CRTC:%d:%s]\n", crtc->base.id, crtc->name); if (crtc_req->mode_valid) { /* If we have a mode we need a framebuffer. */ @@ -3235,7 +3308,7 @@ static int framebuffer_check(const struct drm_mode_fb_cmd2 *r) static struct drm_framebuffer * internal_framebuffer_create(struct drm_device *dev, - struct drm_mode_fb_cmd2 *r, + const struct drm_mode_fb_cmd2 *r, struct drm_file *file_priv) { struct drm_mode_config *config = &dev->mode_config; @@ -4785,9 +4858,7 @@ static int drm_mode_connector_set_obj_prop(struct drm_mode_object *obj, /* Do DPMS ourselves */ if (property == connector->dev->mode_config.dpms_property) { - ret = 0; - if (connector->funcs->dpms) - ret = (*connector->funcs->dpms)(connector, (int)value); + ret = (*connector->funcs->dpms)(connector, (int)value); } else if (connector->funcs->set_property) ret = connector->funcs->set_property(connector, property, value); diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index ef534758a02c..a02a7f9a6a9d 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -51,6 +51,11 @@ * the same callbacks which drivers can use to e.g. restore the modeset * configuration on resume with drm_helper_resume_force_mode(). * + * Note that this helper library doesn't track the current power state of CRTCs + * and encoders. It can call callbacks like ->dpms() even though the hardware is + * already in the desired state. This deficiency has been fixed in the atomic + * helpers. + * * The driver callbacks are mostly compatible with the atomic modeset helpers, * except for the handling of the primary plane: Atomic helpers require that the * primary plane is implemented as a real standalone plane and not directly tied @@ -62,6 +67,11 @@ * converting to the plane helpers). New drivers must not use these functions * but need to implement the atomic interface instead, potentially using the * atomic helpers for that. + * + * These legacy modeset helpers use the same function table structures as + * all other modesetting helpers. See the documentation for struct + * &drm_crtc_helper_funcs, struct &drm_encoder_helper_funcs and struct + * &drm_connector_helper_funcs. */ MODULE_AUTHOR("David Airlie, Jesse Barnes"); MODULE_DESCRIPTION("DRM KMS helper"); @@ -206,8 +216,8 @@ static void __drm_helper_disable_unused_functions(struct drm_device *dev) * @dev: DRM device * * This function walks through the entire mode setting configuration of @dev. It - * will remove any crtc links of unused encoders and encoder links of - * disconnected connectors. Then it will disable all unused encoders and crtcs + * will remove any CRTC links of unused encoders and encoder links of + * disconnected connectors. Then it will disable all unused encoders and CRTCs * either by calling their disable callback if available or by calling their * dpms callback with DRM_MODE_DPMS_OFF. */ @@ -329,7 +339,7 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc, DRM_DEBUG_KMS("CRTC fixup failed\n"); goto done; } - DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id); + DRM_DEBUG_KMS("[CRTC:%d:%s]\n", crtc->base.id, crtc->name); crtc->hwmode = *adjusted_mode; @@ -445,11 +455,36 @@ drm_crtc_helper_disable(struct drm_crtc *crtc) * drm_crtc_helper_set_config - set a new config from userspace * @set: mode set configuration * - * Setup a new configuration, provided by the upper layers (either an ioctl call - * from userspace or internally e.g. from the fbdev support code) in @set, and - * enable it. This is the main helper functions for drivers that implement - * kernel mode setting with the crtc helper functions and the assorted - * ->prepare(), ->modeset() and ->commit() helper callbacks. + * The drm_crtc_helper_set_config() helper function implements the set_config + * callback of struct &drm_crtc_funcs for drivers using the legacy CRTC helpers. + * + * It first tries to locate the best encoder for each connector by calling the + * connector ->best_encoder() (struct &drm_connector_helper_funcs) helper + * operation. + * + * After locating the appropriate encoders, the helper function will call the + * mode_fixup encoder and CRTC helper operations to adjust the requested mode, + * or reject it completely in which case an error will be returned to the + * application. If the new configuration after mode adjustment is identical to + * the current configuration the helper function will return without performing + * any other operation. + * + * If the adjusted mode is identical to the current mode but changes to the + * frame buffer need to be applied, the drm_crtc_helper_set_config() function + * will call the CRTC ->mode_set_base() (struct &drm_crtc_helper_funcs) helper + * operation. + * + * If the adjusted mode differs from the current mode, or if the + * ->mode_set_base() helper operation is not provided, the helper function + * performs a full mode set sequence by calling the ->prepare(), ->mode_set() + * and ->commit() CRTC and encoder helper operations, in that order. + * Alternatively it can also use the dpms and disable helper operations. For + * details see struct &drm_crtc_helper_funcs and struct + * &drm_encoder_helper_funcs. + * + * This function is deprecated. New drivers must implement atomic modeset + * support, for which this function is unsuitable. Instead drivers should use + * drm_atomic_helper_set_config(). * * Returns: * Returns 0 on success, negative errno numbers on failure. @@ -484,11 +519,13 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) set->fb = NULL; if (set->fb) { - DRM_DEBUG_KMS("[CRTC:%d] [FB:%d] #connectors=%d (x y) (%i %i)\n", - set->crtc->base.id, set->fb->base.id, - (int)set->num_connectors, set->x, set->y); + DRM_DEBUG_KMS("[CRTC:%d:%s] [FB:%d] #connectors=%d (x y) (%i %i)\n", + set->crtc->base.id, set->crtc->name, + set->fb->base.id, + (int)set->num_connectors, set->x, set->y); } else { - DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id); + DRM_DEBUG_KMS("[CRTC:%d:%s] [NOFB]\n", + set->crtc->base.id, set->crtc->name); drm_crtc_helper_disable(set->crtc); return 0; } @@ -628,12 +665,12 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) connector->encoder->crtc = new_crtc; } if (new_crtc) { - DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n", - connector->base.id, connector->name, - new_crtc->base.id); + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d:%s]\n", + connector->base.id, connector->name, + new_crtc->base.id, new_crtc->name); } else { DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n", - connector->base.id, connector->name); + connector->base.id, connector->name); } } @@ -650,8 +687,8 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) if (!drm_crtc_helper_set_mode(set->crtc, set->mode, set->x, set->y, save_set.fb)) { - DRM_ERROR("failed to set mode on [CRTC:%d]\n", - set->crtc->base.id); + DRM_ERROR("failed to set mode on [CRTC:%d:%s]\n", + set->crtc->base.id, set->crtc->name); set->crtc->primary->fb = save_set.fb; ret = -EINVAL; goto fail; @@ -758,10 +795,18 @@ static int drm_helper_choose_crtc_dpms(struct drm_crtc *crtc) * @connector: affected connector * @mode: DPMS mode * - * This is the main helper function provided by the crtc helper framework for + * The drm_helper_connector_dpms() helper function implements the ->dpms() + * callback of struct &drm_connector_funcs for drivers using the legacy CRTC helpers. + * + * This is the main helper function provided by the CRTC helper framework for * implementing the DPMS connector attribute. It computes the new desired DPMS - * state for all encoders and crtcs in the output mesh and calls the ->dpms() - * callback provided by the driver appropriately. + * state for all encoders and CRTCs in the output mesh and calls the ->dpms() + * callbacks provided by the driver in struct &drm_crtc_helper_funcs and struct + * &drm_encoder_helper_funcs appropriately. + * + * This function is deprecated. New drivers must implement atomic modeset + * support, for which this function is unsuitable. Instead drivers should use + * drm_atomic_helper_connector_dpms(). * * Returns: * Always returns 0. @@ -818,7 +863,7 @@ EXPORT_SYMBOL(drm_helper_connector_dpms); * metadata fields. */ void drm_helper_mode_fill_fb_struct(struct drm_framebuffer *fb, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { int i; @@ -855,6 +900,12 @@ EXPORT_SYMBOL(drm_helper_mode_fill_fb_struct); * due to slight differences in allocating shared resources when the * configuration is restored in a different order than when userspace set it up) * need to use their own restore logic. + * + * This function is deprecated. New drivers should implement atomic mode- + * setting and use the atomic suspend/resume helpers. + * + * See also: + * drm_atomic_helper_suspend(), drm_atomic_helper_resume() */ void drm_helper_resume_force_mode(struct drm_device *dev) { @@ -913,9 +964,9 @@ EXPORT_SYMBOL(drm_helper_resume_force_mode); * @old_fb: previous framebuffer * * This function implements a callback useable as the ->mode_set callback - * required by the crtc helpers. Besides the atomic plane helper functions for + * required by the CRTC helpers. Besides the atomic plane helper functions for * the primary plane the driver must also provide the ->mode_set_nofb callback - * to set up the crtc. + * to set up the CRTC. * * This is a transitional helper useful for converting drivers to the atomic * interfaces. @@ -979,7 +1030,7 @@ EXPORT_SYMBOL(drm_helper_crtc_mode_set); * @old_fb: previous framebuffer * * This function implements a callback useable as the ->mode_set_base used - * required by the crtc helpers. The driver must provide the atomic plane helper + * required by the CRTC helpers. The driver must provide the atomic plane helper * functions for the primary plane. * * This is a transitional helper useful for converting drivers to the atomic diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 9362609df38a..7dd6728dd092 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -160,6 +160,11 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data, goto out_unlock; } + if (!file_priv->allowed_master) { + ret = drm_new_set_master(dev, file_priv); + goto out_unlock; + } + file_priv->minor->master = drm_master_get(file_priv->master); file_priv->is_master = 1; if (dev->driver->master_set) { diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index d5d2c03fd136..c214f1246cb4 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -2545,6 +2545,33 @@ cea_mode_alternate_clock(const struct drm_display_mode *cea_mode) return clock; } +static u8 drm_match_cea_mode_clock_tolerance(const struct drm_display_mode *to_match, + unsigned int clock_tolerance) +{ + u8 mode; + + if (!to_match->clock) + return 0; + + for (mode = 0; mode < ARRAY_SIZE(edid_cea_modes); mode++) { + const struct drm_display_mode *cea_mode = &edid_cea_modes[mode]; + unsigned int clock1, clock2; + + /* Check both 60Hz and 59.94Hz */ + clock1 = cea_mode->clock; + clock2 = cea_mode_alternate_clock(cea_mode); + + if (abs(to_match->clock - clock1) > clock_tolerance && + abs(to_match->clock - clock2) > clock_tolerance) + continue; + + if (drm_mode_equal_no_clocks(to_match, cea_mode)) + return mode + 1; + } + + return 0; +} + /** * drm_match_cea_mode - look for a CEA mode matching given mode * @to_match: display mode @@ -2609,6 +2636,33 @@ hdmi_mode_alternate_clock(const struct drm_display_mode *hdmi_mode) return cea_mode_alternate_clock(hdmi_mode); } +static u8 drm_match_hdmi_mode_clock_tolerance(const struct drm_display_mode *to_match, + unsigned int clock_tolerance) +{ + u8 mode; + + if (!to_match->clock) + return 0; + + for (mode = 0; mode < ARRAY_SIZE(edid_4k_modes); mode++) { + const struct drm_display_mode *hdmi_mode = &edid_4k_modes[mode]; + unsigned int clock1, clock2; + + /* Make sure to also match alternate clocks */ + clock1 = hdmi_mode->clock; + clock2 = hdmi_mode_alternate_clock(hdmi_mode); + + if (abs(to_match->clock - clock1) > clock_tolerance && + abs(to_match->clock - clock2) > clock_tolerance) + continue; + + if (drm_mode_equal_no_clocks(to_match, hdmi_mode)) + return mode + 1; + } + + return 0; +} + /* * drm_match_hdmi_mode - look for a HDMI mode matching given mode * @to_match: display mode @@ -3119,14 +3173,18 @@ static void fixup_detailed_cea_mode_clock(struct drm_display_mode *mode) u8 mode_idx; const char *type; - mode_idx = drm_match_cea_mode(mode) - 1; + /* + * allow 5kHz clock difference either way to account for + * the 10kHz clock resolution limit of detailed timings. + */ + mode_idx = drm_match_cea_mode_clock_tolerance(mode, 5) - 1; if (mode_idx < ARRAY_SIZE(edid_cea_modes)) { type = "CEA"; cea_mode = &edid_cea_modes[mode_idx]; clock1 = cea_mode->clock; clock2 = cea_mode_alternate_clock(cea_mode); } else { - mode_idx = drm_match_hdmi_mode(mode) - 1; + mode_idx = drm_match_hdmi_mode_clock_tolerance(mode, 5) - 1; if (mode_idx < ARRAY_SIZE(edid_4k_modes)) { type = "HDMI"; cea_mode = &edid_4k_modes[mode_idx]; diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c index c19a62561183..b7d5b848d2f8 100644 --- a/drivers/gpu/drm/drm_fb_cma_helper.c +++ b/drivers/gpu/drm/drm_fb_cma_helper.c @@ -74,7 +74,7 @@ static struct drm_framebuffer_funcs drm_fb_cma_funcs = { }; static struct drm_fb_cma *drm_fb_cma_alloc(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_cma_object **obj, + const const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_cma_object **obj, unsigned int num_planes) { struct drm_fb_cma *fb_cma; @@ -107,7 +107,7 @@ static struct drm_fb_cma *drm_fb_cma_alloc(struct drm_device *dev, * checked before calling this function. */ struct drm_framebuffer *drm_fb_cma_create(struct drm_device *dev, - struct drm_file *file_priv, struct drm_mode_fb_cmd2 *mode_cmd) + struct drm_file *file_priv, const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_fb_cma *fb_cma; struct drm_gem_cma_object *objs[4]; diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c index c59ce4d0ef75..1ea8790e5090 100644 --- a/drivers/gpu/drm/drm_fops.c +++ b/drivers/gpu/drm/drm_fops.c @@ -126,6 +126,60 @@ static int drm_cpu_valid(void) } /** + * drm_new_set_master - Allocate a new master object and become master for the + * associated master realm. + * + * @dev: The associated device. + * @fpriv: File private identifying the client. + * + * This function must be called with dev::struct_mutex held. + * Returns negative error code on failure. Zero on success. + */ +int drm_new_set_master(struct drm_device *dev, struct drm_file *fpriv) +{ + struct drm_master *old_master; + int ret; + + lockdep_assert_held_once(&dev->master_mutex); + + /* create a new master */ + fpriv->minor->master = drm_master_create(fpriv->minor); + if (!fpriv->minor->master) + return -ENOMEM; + + /* take another reference for the copy in the local file priv */ + old_master = fpriv->master; + fpriv->master = drm_master_get(fpriv->minor->master); + + if (dev->driver->master_create) { + ret = dev->driver->master_create(dev, fpriv->master); + if (ret) + goto out_err; + } + if (dev->driver->master_set) { + ret = dev->driver->master_set(dev, fpriv, true); + if (ret) + goto out_err; + } + + fpriv->is_master = 1; + fpriv->allowed_master = 1; + fpriv->authenticated = 1; + if (old_master) + drm_master_put(&old_master); + + return 0; + +out_err: + /* drop both references and restore old master on failure */ + drm_master_put(&fpriv->minor->master); + drm_master_put(&fpriv->master); + fpriv->master = old_master; + + return ret; +} + +/** * Called whenever a process opens /dev/drm. * * \param filp file pointer. @@ -172,6 +226,8 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor) init_waitqueue_head(&priv->event_wait); priv->event_space = 4096; /* set aside 4k for event buffer */ + mutex_init(&priv->event_read_lock); + if (drm_core_check_feature(dev, DRIVER_GEM)) drm_gem_open(dev, priv); @@ -189,35 +245,9 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor) mutex_lock(&dev->master_mutex); if (drm_is_primary_client(priv) && !priv->minor->master) { /* create a new master */ - priv->minor->master = drm_master_create(priv->minor); - if (!priv->minor->master) { - ret = -ENOMEM; + ret = drm_new_set_master(dev, priv); + if (ret) goto out_close; - } - - priv->is_master = 1; - /* take another reference for the copy in the local file priv */ - priv->master = drm_master_get(priv->minor->master); - priv->authenticated = 1; - - if (dev->driver->master_create) { - ret = dev->driver->master_create(dev, priv->master); - if (ret) { - /* drop both references if this fails */ - drm_master_put(&priv->minor->master); - drm_master_put(&priv->master); - goto out_close; - } - } - if (dev->driver->master_set) { - ret = dev->driver->master_set(dev, priv, true); - if (ret) { - /* drop both references if this fails */ - drm_master_put(&priv->minor->master); - drm_master_put(&priv->master); - goto out_close; - } - } } else if (drm_is_primary_client(priv)) { /* get a reference to the master */ priv->master = drm_master_get(priv->minor->master); @@ -483,14 +513,28 @@ ssize_t drm_read(struct file *filp, char __user *buffer, { struct drm_file *file_priv = filp->private_data; struct drm_device *dev = file_priv->minor->dev; - ssize_t ret = 0; + ssize_t ret; if (!access_ok(VERIFY_WRITE, buffer, count)) return -EFAULT; - spin_lock_irq(&dev->event_lock); + ret = mutex_lock_interruptible(&file_priv->event_read_lock); + if (ret) + return ret; + for (;;) { - if (list_empty(&file_priv->event_list)) { + struct drm_pending_event *e = NULL; + + spin_lock_irq(&dev->event_lock); + if (!list_empty(&file_priv->event_list)) { + e = list_first_entry(&file_priv->event_list, + struct drm_pending_event, link); + file_priv->event_space += e->event->length; + list_del(&e->link); + } + spin_unlock_irq(&dev->event_lock); + + if (e == NULL) { if (ret) break; @@ -499,36 +543,36 @@ ssize_t drm_read(struct file *filp, char __user *buffer, break; } - spin_unlock_irq(&dev->event_lock); + mutex_unlock(&file_priv->event_read_lock); ret = wait_event_interruptible(file_priv->event_wait, !list_empty(&file_priv->event_list)); - spin_lock_irq(&dev->event_lock); - if (ret < 0) - break; - - ret = 0; + if (ret >= 0) + ret = mutex_lock_interruptible(&file_priv->event_read_lock); + if (ret) + return ret; } else { - struct drm_pending_event *e; - - e = list_first_entry(&file_priv->event_list, - struct drm_pending_event, link); - if (e->event->length + ret > count) + unsigned length = e->event->length; + + if (length > count - ret) { +put_back_event: + spin_lock_irq(&dev->event_lock); + file_priv->event_space -= length; + list_add(&e->link, &file_priv->event_list); + spin_unlock_irq(&dev->event_lock); break; + } - if (__copy_to_user_inatomic(buffer + ret, - e->event, e->event->length)) { + if (copy_to_user(buffer + ret, e->event, length)) { if (ret == 0) ret = -EFAULT; - break; + goto put_back_event; } - file_priv->event_space += e->event->length; - ret += e->event->length; - list_del(&e->link); + ret += length; e->destroy(e); } } - spin_unlock_irq(&dev->event_lock); + mutex_unlock(&file_priv->event_read_lock); return ret; } diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index c7de454e8e88..2e10bba4468b 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -244,8 +244,9 @@ drm_gem_object_handle_unreference_unlocked(struct drm_gem_object *obj) * @filp: drm file-private structure to use for the handle look up * @handle: userspace handle to delete * - * Removes the GEM handle from the @filp lookup table and if this is the last - * handle also cleans up linked resources like GEM names. + * Removes the GEM handle from the @filp lookup table which has been added with + * drm_gem_handle_create(). If this is the last handle also cleans up linked + * resources like GEM names. */ int drm_gem_handle_delete(struct drm_file *filp, u32 handle) @@ -314,6 +315,10 @@ EXPORT_SYMBOL(drm_gem_dumb_destroy); * This expects the dev->object_name_lock to be held already and will drop it * before returning. Used to avoid races in establishing new handles when * importing an object from either an flink name or a dma-buf. + * + * Handles must be release again through drm_gem_handle_delete(). This is done + * when userspace closes @file_priv for all attached handles, or through the + * GEM_CLOSE ioctl for individual handles. */ int drm_gem_handle_create_tail(struct drm_file *file_priv, @@ -541,7 +546,17 @@ void drm_gem_put_pages(struct drm_gem_object *obj, struct page **pages, } EXPORT_SYMBOL(drm_gem_put_pages); -/** Returns a reference to the object named by the handle. */ +/** + * drm_gem_object_lookup - look up a GEM object from it's handle + * @dev: DRM device + * @filp: DRM file private date + * @handle: userspace handle + * + * Returns: + * + * A reference to the object named by the handle if such exists on @filp, NULL + * otherwise. + */ struct drm_gem_object * drm_gem_object_lookup(struct drm_device *dev, struct drm_file *filp, u32 handle) @@ -774,6 +789,13 @@ drm_gem_object_free(struct kref *kref) } EXPORT_SYMBOL(drm_gem_object_free); +/** + * drm_gem_vm_open - vma->ops->open implementation for GEM + * @vma: VM area structure + * + * This function implements the #vm_operations_struct open() callback for GEM + * drivers. This must be used together with drm_gem_vm_close(). + */ void drm_gem_vm_open(struct vm_area_struct *vma) { struct drm_gem_object *obj = vma->vm_private_data; @@ -782,6 +804,13 @@ void drm_gem_vm_open(struct vm_area_struct *vma) } EXPORT_SYMBOL(drm_gem_vm_open); +/** + * drm_gem_vm_close - vma->ops->close implementation for GEM + * @vma: VM area structure + * + * This function implements the #vm_operations_struct close() callback for GEM + * drivers. This must be used together with drm_gem_vm_open(). + */ void drm_gem_vm_close(struct vm_area_struct *vma) { struct drm_gem_object *obj = vma->vm_private_data; diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c index e109b49cd25d..0f7b00ba57da 100644 --- a/drivers/gpu/drm/drm_gem_cma_helper.c +++ b/drivers/gpu/drm/drm_gem_cma_helper.c @@ -59,11 +59,13 @@ __drm_gem_cma_create(struct drm_device *drm, size_t size) struct drm_gem_object *gem_obj; int ret; - cma_obj = kzalloc(sizeof(*cma_obj), GFP_KERNEL); - if (!cma_obj) + if (drm->driver->gem_create_object) + gem_obj = drm->driver->gem_create_object(drm, size); + else + gem_obj = kzalloc(sizeof(*cma_obj), GFP_KERNEL); + if (!gem_obj) return ERR_PTR(-ENOMEM); - - gem_obj = &cma_obj->base; + cma_obj = container_of(gem_obj, struct drm_gem_cma_object, base); ret = drm_gem_object_init(drm, gem_obj, size); if (ret) diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 2151ea551d3b..607f493ae801 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -980,7 +980,8 @@ static void send_vblank_event(struct drm_device *dev, struct drm_pending_vblank_event *e, unsigned long seq, struct timeval *now) { - WARN_ON_SMP(!spin_is_locked(&dev->event_lock)); + assert_spin_locked(&dev->event_lock); + e->event.sequence = seq; e->event.tv_sec = now->tv_sec; e->event.tv_usec = now->tv_usec; @@ -993,6 +994,57 @@ static void send_vblank_event(struct drm_device *dev, } /** + * drm_arm_vblank_event - arm vblank event after pageflip + * @dev: DRM device + * @pipe: CRTC index + * @e: the event to prepare to send + * + * A lot of drivers need to generate vblank events for the very next vblank + * interrupt. For example when the page flip interrupt happens when the page + * flip gets armed, but not when it actually executes within the next vblank + * period. This helper function implements exactly the required vblank arming + * behaviour. + * + * Caller must hold event lock. Caller must also hold a vblank reference for + * the event @e, which will be dropped when the next vblank arrives. + * + * This is the legacy version of drm_crtc_arm_vblank_event(). + */ +void drm_arm_vblank_event(struct drm_device *dev, unsigned int pipe, + struct drm_pending_vblank_event *e) +{ + assert_spin_locked(&dev->event_lock); + + e->pipe = pipe; + e->event.sequence = drm_vblank_count(dev, pipe); + list_add_tail(&e->base.link, &dev->vblank_event_list); +} +EXPORT_SYMBOL(drm_arm_vblank_event); + +/** + * drm_crtc_arm_vblank_event - arm vblank event after pageflip + * @crtc: the source CRTC of the vblank event + * @e: the event to send + * + * A lot of drivers need to generate vblank events for the very next vblank + * interrupt. For example when the page flip interrupt happens when the page + * flip gets armed, but not when it actually executes within the next vblank + * period. This helper function implements exactly the required vblank arming + * behaviour. + * + * Caller must hold event lock. Caller must also hold a vblank reference for + * the event @e, which will be dropped when the next vblank arrives. + * + * This is the native KMS version of drm_arm_vblank_event(). + */ +void drm_crtc_arm_vblank_event(struct drm_crtc *crtc, + struct drm_pending_vblank_event *e) +{ + drm_arm_vblank_event(crtc->dev, drm_crtc_index(crtc), e); +} +EXPORT_SYMBOL(drm_crtc_arm_vblank_event); + +/** * drm_send_vblank_event - helper to send vblank event after pageflip * @dev: DRM device * @pipe: CRTC index diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c index cd74a0953f42..5a8a78d5e960 100644 --- a/drivers/gpu/drm/drm_modes.c +++ b/drivers/gpu/drm/drm_modes.c @@ -708,7 +708,8 @@ void drm_mode_set_name(struct drm_display_mode *mode) } EXPORT_SYMBOL(drm_mode_set_name); -/** drm_mode_hsync - get the hsync of a mode +/** + * drm_mode_hsync - get the hsync of a mode * @mode: mode * * Returns: @@ -917,13 +918,30 @@ bool drm_mode_equal(const struct drm_display_mode *mode1, const struct drm_displ } else if (mode1->clock != mode2->clock) return false; + return drm_mode_equal_no_clocks(mode1, mode2); +} +EXPORT_SYMBOL(drm_mode_equal); + +/** + * drm_mode_equal_no_clocks - test modes for equality + * @mode1: first mode + * @mode2: second mode + * + * Check to see if @mode1 and @mode2 are equivalent, but + * don't check the pixel clocks. + * + * Returns: + * True if the modes are equal, false otherwise. + */ +bool drm_mode_equal_no_clocks(const struct drm_display_mode *mode1, const struct drm_display_mode *mode2) +{ if ((mode1->flags & DRM_MODE_FLAG_3D_MASK) != (mode2->flags & DRM_MODE_FLAG_3D_MASK)) return false; return drm_mode_equal_no_clocks_no_stereo(mode1, mode2); } -EXPORT_SYMBOL(drm_mode_equal); +EXPORT_SYMBOL(drm_mode_equal_no_clocks); /** * drm_mode_equal_no_clocks_no_stereo - test modes for equality @@ -1056,7 +1074,7 @@ static const char * const drm_mode_status_names[] = { MODE_STATUS(ONE_SIZE), MODE_STATUS(NO_REDUCED), MODE_STATUS(NO_STEREO), - MODE_STATUS(UNVERIFIED), + MODE_STATUS(STALE), MODE_STATUS(BAD), MODE_STATUS(ERROR), }; @@ -1154,7 +1172,6 @@ EXPORT_SYMBOL(drm_mode_sort); /** * drm_mode_connector_list_update - update the mode list for the connector * @connector: the connector to update - * @merge_type_bits: whether to merge or overwrite type bits * * This moves the modes from the @connector probed_modes list * to the actual mode list. It compares the probed mode against the current @@ -1163,33 +1180,48 @@ EXPORT_SYMBOL(drm_mode_sort); * This is just a helper functions doesn't validate any modes itself and also * doesn't prune any invalid modes. Callers need to do that themselves. */ -void drm_mode_connector_list_update(struct drm_connector *connector, - bool merge_type_bits) +void drm_mode_connector_list_update(struct drm_connector *connector) { - struct drm_display_mode *mode; struct drm_display_mode *pmode, *pt; - int found_it; WARN_ON(!mutex_is_locked(&connector->dev->mode_config.mutex)); - list_for_each_entry_safe(pmode, pt, &connector->probed_modes, - head) { - found_it = 0; + list_for_each_entry_safe(pmode, pt, &connector->probed_modes, head) { + struct drm_display_mode *mode; + bool found_it = false; + /* go through current modes checking for the new probed mode */ list_for_each_entry(mode, &connector->modes, head) { - if (drm_mode_equal(pmode, mode)) { - found_it = 1; - /* if equal delete the probed mode */ - mode->status = pmode->status; - /* Merge type bits together */ - if (merge_type_bits) - mode->type |= pmode->type; - else - mode->type = pmode->type; - list_del(&pmode->head); - drm_mode_destroy(connector->dev, pmode); - break; + if (!drm_mode_equal(pmode, mode)) + continue; + + found_it = true; + + /* + * If the old matching mode is stale (ie. left over + * from a previous probe) just replace it outright. + * Otherwise just merge the type bits between all + * equal probed modes. + * + * If two probed modes are considered equal, pick the + * actual timings from the one that's marked as + * preferred (in case the match isn't 100%). If + * multiple or zero preferred modes are present, favor + * the mode added to the probed_modes list first. + */ + if (mode->status == MODE_STALE) { + drm_mode_copy(mode, pmode); + } else if ((mode->type & DRM_MODE_TYPE_PREFERRED) == 0 && + (pmode->type & DRM_MODE_TYPE_PREFERRED) != 0) { + pmode->type |= mode->type; + drm_mode_copy(mode, pmode); + } else { + mode->type |= pmode->type; } + + list_del(&pmode->head); + drm_mode_destroy(connector->dev, pmode); + break; } if (!found_it) { diff --git a/drivers/gpu/drm/drm_modeset_lock.c b/drivers/gpu/drm/drm_modeset_lock.c index 6675b1428410..c2f5971146ba 100644 --- a/drivers/gpu/drm/drm_modeset_lock.c +++ b/drivers/gpu/drm/drm_modeset_lock.c @@ -57,11 +57,18 @@ /** * drm_modeset_lock_all - take all modeset locks - * @dev: drm device + * @dev: DRM device * * This function takes all modeset locks, suitable where a more fine-grained - * scheme isn't (yet) implemented. Locks must be dropped with - * drm_modeset_unlock_all. + * scheme isn't (yet) implemented. Locks must be dropped by calling the + * drm_modeset_unlock_all() function. + * + * This function is deprecated. It allocates a lock acquisition context and + * stores it in the DRM device's ->mode_config. This facilitate conversion of + * existing code because it removes the need to manually deal with the + * acquisition context, but it is also brittle because the context is global + * and care must be taken not to nest calls. New code should use the + * drm_modeset_lock_all_ctx() function and pass in the context explicitly. */ void drm_modeset_lock_all(struct drm_device *dev) { @@ -78,39 +85,43 @@ void drm_modeset_lock_all(struct drm_device *dev) drm_modeset_acquire_init(ctx, 0); retry: - ret = drm_modeset_lock(&config->connection_mutex, ctx); - if (ret) - goto fail; - ret = drm_modeset_lock_all_crtcs(dev, ctx); - if (ret) - goto fail; + ret = drm_modeset_lock_all_ctx(dev, ctx); + if (ret < 0) { + if (ret == -EDEADLK) { + drm_modeset_backoff(ctx); + goto retry; + } + + drm_modeset_acquire_fini(ctx); + kfree(ctx); + return; + } WARN_ON(config->acquire_ctx); - /* now we hold the locks, so now that it is safe, stash the - * ctx for drm_modeset_unlock_all(): + /* + * We hold the locks now, so it is safe to stash the acquisition + * context for drm_modeset_unlock_all(). */ config->acquire_ctx = ctx; drm_warn_on_modeset_not_all_locked(dev); - - return; - -fail: - if (ret == -EDEADLK) { - drm_modeset_backoff(ctx); - goto retry; - } - - kfree(ctx); } EXPORT_SYMBOL(drm_modeset_lock_all); /** * drm_modeset_unlock_all - drop all modeset locks - * @dev: device + * @dev: DRM device * - * This function drop all modeset locks taken by drm_modeset_lock_all. + * This function drops all modeset locks taken by a previous call to the + * drm_modeset_lock_all() function. + * + * This function is deprecated. It uses the lock acquisition context stored + * in the DRM device's ->mode_config. This facilitates conversion of existing + * code because it removes the need to manually deal with the acquisition + * context, but it is also brittle because the context is global and care must + * be taken not to nest calls. New code should pass the acquisition context + * directly to the drm_modeset_drop_locks() function. */ void drm_modeset_unlock_all(struct drm_device *dev) { @@ -431,14 +442,34 @@ void drm_modeset_unlock(struct drm_modeset_lock *lock) } EXPORT_SYMBOL(drm_modeset_unlock); -/* In some legacy codepaths it's convenient to just grab all the crtc and plane - * related locks. */ -int drm_modeset_lock_all_crtcs(struct drm_device *dev, - struct drm_modeset_acquire_ctx *ctx) +/** + * drm_modeset_lock_all_ctx - take all modeset locks + * @dev: DRM device + * @ctx: lock acquisition context + * + * This function takes all modeset locks, suitable where a more fine-grained + * scheme isn't (yet) implemented. + * + * Unlike drm_modeset_lock_all(), it doesn't take the dev->mode_config.mutex + * since that lock isn't required for modeset state changes. Callers which + * need to grab that lock too need to do so outside of the acquire context + * @ctx. + * + * Locks acquired with this function should be released by calling the + * drm_modeset_drop_locks() function on @ctx. + * + * Returns: 0 on success or a negative error-code on failure. + */ +int drm_modeset_lock_all_ctx(struct drm_device *dev, + struct drm_modeset_acquire_ctx *ctx) { struct drm_crtc *crtc; struct drm_plane *plane; - int ret = 0; + int ret; + + ret = drm_modeset_lock(&dev->mode_config.connection_mutex, ctx); + if (ret) + return ret; drm_for_each_crtc(crtc, dev) { ret = drm_modeset_lock(&crtc->mutex, ctx); @@ -454,4 +485,4 @@ int drm_modeset_lock_all_crtcs(struct drm_device *dev, return 0; } -EXPORT_SYMBOL(drm_modeset_lock_all_crtcs); +EXPORT_SYMBOL(drm_modeset_lock_all_ctx); diff --git a/drivers/gpu/drm/drm_plane_helper.c b/drivers/gpu/drm/drm_plane_helper.c index d384ebcf0aaf..369d2898ff9e 100644 --- a/drivers/gpu/drm/drm_plane_helper.c +++ b/drivers/gpu/drm/drm_plane_helper.c @@ -57,6 +57,10 @@ * by the atomic helpers. * * Again drivers are strongly urged to switch to the new interfaces. + * + * The plane helpers share the function table structures with other helpers, + * specifically also the atomic helpers. See struct &drm_plane_helper_funcs for + * the details. */ /* @@ -164,6 +168,8 @@ int drm_plane_helper_check_update(struct drm_plane *plane, vscale = drm_rect_calc_vscale(src, dest, min_scale, max_scale); if (hscale < 0 || vscale < 0) { DRM_DEBUG_KMS("Invalid scaling of plane\n"); + drm_rect_debug_print("src: ", src, true); + drm_rect_debug_print("dst: ", dest, false); return -ERANGE; } @@ -180,6 +186,8 @@ int drm_plane_helper_check_update(struct drm_plane *plane, if (!can_position && !drm_rect_equals(dest, clip)) { DRM_DEBUG_KMS("Plane must cover entire CRTC\n"); + drm_rect_debug_print("dst: ", dest, false); + drm_rect_debug_print("clip: ", clip, false); return -EINVAL; } @@ -367,7 +375,7 @@ static struct drm_plane *create_primary_plane(struct drm_device *dev) &drm_primary_helper_funcs, safe_modeset_formats, ARRAY_SIZE(safe_modeset_formats), - DRM_PLANE_TYPE_PRIMARY); + DRM_PLANE_TYPE_PRIMARY, NULL); if (ret) { kfree(primary); primary = NULL; @@ -394,7 +402,8 @@ int drm_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, struct drm_plane *primary; primary = create_primary_plane(dev); - return drm_crtc_init_with_planes(dev, crtc, primary, NULL, funcs); + return drm_crtc_init_with_planes(dev, crtc, primary, NULL, funcs, + NULL); } EXPORT_SYMBOL(drm_crtc_init); diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index a18164f2f6d2..e714b5a7955f 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -53,6 +53,9 @@ * This helper library can be used independently of the modeset helper library. * Drivers can also overwrite different parts e.g. use their own hotplug * handling code to avoid probing unrelated outputs. + * + * The probe helpers share the function table structures with other display + * helper libraries. See struct &drm_connector_helper_funcs for the details. */ static bool drm_kms_helper_poll = true; @@ -126,9 +129,64 @@ void drm_kms_helper_poll_enable_locked(struct drm_device *dev) } EXPORT_SYMBOL(drm_kms_helper_poll_enable_locked); - -static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connector *connector, - uint32_t maxX, uint32_t maxY, bool merge_type_bits) +/** + * drm_helper_probe_single_connector_modes - get complete set of display modes + * @connector: connector to probe + * @maxX: max width for modes + * @maxY: max height for modes + * + * Based on the helper callbacks implemented by @connector in struct + * &drm_connector_helper_funcs try to detect all valid modes. Modes will first + * be added to the connector's probed_modes list, then culled (based on validity + * and the @maxX, @maxY parameters) and put into the normal modes list. + * + * Intended to be used as a generic implementation of the ->fill_modes() + * @connector vfunc for drivers that use the CRTC helpers for output mode + * filtering and detection. + * + * The basic procedure is as follows + * + * 1. All modes currently on the connector's modes list are marked as stale + * + * 2. New modes are added to the connector's probed_modes list with + * drm_mode_probed_add(). New modes start their life with status as OK. + * Modes are added from a single source using the following priority order. + * + * - debugfs 'override_edid' (used for testing only) + * - firmware EDID (drm_load_edid_firmware()) + * - connector helper ->get_modes() vfunc + * - if the connector status is connector_status_connected, standard + * VESA DMT modes up to 1024x768 are automatically added + * (drm_add_modes_noedid()) + * + * Finally modes specified via the kernel command line (video=...) are + * added in addition to what the earlier probes produced + * (drm_helper_probe_add_cmdline_mode()). These modes are generated + * using the VESA GTF/CVT formulas. + * + * 3. Modes are moved from the probed_modes list to the modes list. Potential + * duplicates are merged together (see drm_mode_connector_list_update()). + * After this step the probed_modes list will be empty again. + * + * 4. Any non-stale mode on the modes list then undergoes validation + * + * - drm_mode_validate_basic() performs basic sanity checks + * - drm_mode_validate_size() filters out modes larger than @maxX and @maxY + * (if specified) + * - drm_mode_validate_flag() checks the modes againt basic connector + * capabilites (interlace_allowed,doublescan_allowed,stereo_allowed) + * - the optional connector ->mode_valid() helper can perform driver and/or + * hardware specific checks + * + * 5. Any mode whose status is not OK is pruned from the connector's modes list, + * accompanied by a debug message indicating the reason for the mode's + * rejection (see drm_mode_prune_invalid()). + * + * Returns: + * The number of modes found on @connector. + */ +int drm_helper_probe_single_connector_modes(struct drm_connector *connector, + uint32_t maxX, uint32_t maxY) { struct drm_device *dev = connector->dev; struct drm_display_mode *mode; @@ -143,9 +201,11 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id, connector->name); - /* set all modes to the unverified state */ + /* set all old modes to the stale state */ list_for_each_entry(mode, &connector->modes, head) - mode->status = MODE_UNVERIFIED; + mode->status = MODE_STALE; + + old_status = connector->status; if (connector->force) { if (connector->force == DRM_FORCE_ON || @@ -156,33 +216,32 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect if (connector->funcs->force) connector->funcs->force(connector); } else { - old_status = connector->status; - connector->status = connector->funcs->detect(connector, true); + } + + /* + * Normally either the driver's hpd code or the poll loop should + * pick up any changes and fire the hotplug event. But if + * userspace sneaks in a probe, we might miss a change. Hence + * check here, and if anything changed start the hotplug code. + */ + if (old_status != connector->status) { + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n", + connector->base.id, + connector->name, + drm_get_connector_status_name(old_status), + drm_get_connector_status_name(connector->status)); /* - * Normally either the driver's hpd code or the poll loop should - * pick up any changes and fire the hotplug event. But if - * userspace sneaks in a probe, we might miss a change. Hence - * check here, and if anything changed start the hotplug code. + * The hotplug event code might call into the fb + * helpers, and so expects that we do not hold any + * locks. Fire up the poll struct instead, it will + * disable itself again. */ - if (old_status != connector->status) { - DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %d to %d\n", - connector->base.id, - connector->name, - old_status, connector->status); - - /* - * The hotplug event code might call into the fb - * helpers, and so expects that we do not hold any - * locks. Fire up the poll struct instead, it will - * disable itself again. - */ - dev->mode_config.delayed_event = true; - if (dev->mode_config.poll_enabled) - schedule_delayed_work(&dev->mode_config.output_poll_work, - 0); - } + dev->mode_config.delayed_event = true; + if (dev->mode_config.poll_enabled) + schedule_delayed_work(&dev->mode_config.output_poll_work, + 0); } /* Re-enable polling in case the global poll config changed. */ @@ -199,17 +258,16 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect goto prune; } + if (connector->override_edid) { + struct edid *edid = (struct edid *) connector->edid_blob_ptr->data; + + count = drm_add_edid_modes(connector, edid); + drm_edid_to_eld(connector, edid); + } else { #ifdef CONFIG_DRM_LOAD_EDID_FIRMWARE - count = drm_load_edid_firmware(connector); - if (count == 0) + count = drm_load_edid_firmware(connector); + if (count == 0) #endif - { - if (connector->override_edid) { - struct edid *edid = (struct edid *) connector->edid_blob_ptr->data; - - count = drm_add_edid_modes(connector, edid); - drm_edid_to_eld(connector, edid); - } else count = (*connector_funcs->get_modes)(connector); } @@ -219,7 +277,7 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect if (count == 0) goto prune; - drm_mode_connector_list_update(connector, merge_type_bits); + drm_mode_connector_list_update(connector); if (connector->interlace_allowed) mode_flags |= DRM_MODE_FLAG_INTERLACE; @@ -229,7 +287,8 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect mode_flags |= DRM_MODE_FLAG_3D_MASK; list_for_each_entry(mode, &connector->modes, head) { - mode->status = drm_mode_validate_basic(mode); + if (mode->status == MODE_OK) + mode->status = drm_mode_validate_basic(mode); if (mode->status == MODE_OK) mode->status = drm_mode_validate_size(mode, maxX, maxY); @@ -262,49 +321,9 @@ prune: return count; } - -/** - * drm_helper_probe_single_connector_modes - get complete set of display modes - * @connector: connector to probe - * @maxX: max width for modes - * @maxY: max height for modes - * - * Based on the helper callbacks implemented by @connector try to detect all - * valid modes. Modes will first be added to the connector's probed_modes list, - * then culled (based on validity and the @maxX, @maxY parameters) and put into - * the normal modes list. - * - * Intended to be use as a generic implementation of the ->fill_modes() - * @connector vfunc for drivers that use the crtc helpers for output mode - * filtering and detection. - * - * Returns: - * The number of modes found on @connector. - */ -int drm_helper_probe_single_connector_modes(struct drm_connector *connector, - uint32_t maxX, uint32_t maxY) -{ - return drm_helper_probe_single_connector_modes_merge_bits(connector, maxX, maxY, true); -} EXPORT_SYMBOL(drm_helper_probe_single_connector_modes); /** - * drm_helper_probe_single_connector_modes_nomerge - get complete set of display modes - * @connector: connector to probe - * @maxX: max width for modes - * @maxY: max height for modes - * - * This operates like drm_hehlper_probe_single_connector_modes except it - * replaces the mode bits instead of merging them for preferred modes. - */ -int drm_helper_probe_single_connector_modes_nomerge(struct drm_connector *connector, - uint32_t maxX, uint32_t maxY) -{ - return drm_helper_probe_single_connector_modes_merge_bits(connector, maxX, maxY, false); -} -EXPORT_SYMBOL(drm_helper_probe_single_connector_modes_nomerge); - -/** * drm_kms_helper_hotplug_event - fire off KMS hotplug events * @dev: drm_device whose connector state changed * diff --git a/drivers/gpu/drm/drm_rect.c b/drivers/gpu/drm/drm_rect.c index 531ac4cc9756..a8e2c8603945 100644 --- a/drivers/gpu/drm/drm_rect.c +++ b/drivers/gpu/drm/drm_rect.c @@ -275,22 +275,23 @@ EXPORT_SYMBOL(drm_rect_calc_vscale_relaxed); /** * drm_rect_debug_print - print the rectangle information + * @prefix: prefix string * @r: rectangle to print * @fixed_point: rectangle is in 16.16 fixed point format */ -void drm_rect_debug_print(const struct drm_rect *r, bool fixed_point) +void drm_rect_debug_print(const char *prefix, const struct drm_rect *r, bool fixed_point) { int w = drm_rect_width(r); int h = drm_rect_height(r); if (fixed_point) - DRM_DEBUG_KMS("%d.%06ux%d.%06u%+d.%06u%+d.%06u\n", + DRM_DEBUG_KMS("%s%d.%06ux%d.%06u%+d.%06u%+d.%06u\n", prefix, w >> 16, ((w & 0xffff) * 15625) >> 10, h >> 16, ((h & 0xffff) * 15625) >> 10, r->x1 >> 16, ((r->x1 & 0xffff) * 15625) >> 10, r->y1 >> 16, ((r->y1 & 0xffff) * 15625) >> 10); else - DRM_DEBUG_KMS("%dx%d%+d%+d\n", w, h, r->x1, r->y1); + DRM_DEBUG_KMS("%s%dx%d%+d%+d\n", prefix, w, h, r->x1, r->y1); } EXPORT_SYMBOL(drm_rect_debug_print); diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c index 615b7e667320..0ca64106a97b 100644 --- a/drivers/gpu/drm/drm_sysfs.c +++ b/drivers/gpu/drm/drm_sysfs.c @@ -167,47 +167,35 @@ static ssize_t status_store(struct device *device, { struct drm_connector *connector = to_drm_connector(device); struct drm_device *dev = connector->dev; - enum drm_connector_status old_status; + enum drm_connector_force old_force; int ret; ret = mutex_lock_interruptible(&dev->mode_config.mutex); if (ret) return ret; - old_status = connector->status; + old_force = connector->force; - if (sysfs_streq(buf, "detect")) { + if (sysfs_streq(buf, "detect")) connector->force = 0; - connector->status = connector->funcs->detect(connector, true); - } else if (sysfs_streq(buf, "on")) { + else if (sysfs_streq(buf, "on")) connector->force = DRM_FORCE_ON; - } else if (sysfs_streq(buf, "on-digital")) { + else if (sysfs_streq(buf, "on-digital")) connector->force = DRM_FORCE_ON_DIGITAL; - } else if (sysfs_streq(buf, "off")) { + else if (sysfs_streq(buf, "off")) connector->force = DRM_FORCE_OFF; - } else + else ret = -EINVAL; - if (ret == 0 && connector->force) { - if (connector->force == DRM_FORCE_ON || - connector->force == DRM_FORCE_ON_DIGITAL) - connector->status = connector_status_connected; - else - connector->status = connector_status_disconnected; - if (connector->funcs->force) - connector->funcs->force(connector); - } - - if (old_status != connector->status) { - DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %d to %d\n", + if (old_force != connector->force || !connector->force) { + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] force updated from %d to %d or reprobing\n", connector->base.id, connector->name, - old_status, connector->status); + old_force, connector->force); - dev->mode_config.delayed_event = true; - if (dev->mode_config.poll_enabled) - schedule_delayed_work(&dev->mode_config.output_poll_work, - 0); + connector->funcs->fill_modes(connector, + dev->mode_config.max_width, + dev->mode_config.max_height); } mutex_unlock(&dev->mode_config.mutex); @@ -256,23 +244,29 @@ static ssize_t edid_show(struct file *filp, struct kobject *kobj, struct drm_connector *connector = to_drm_connector(connector_dev); unsigned char *edid; size_t size; + ssize_t ret = 0; + mutex_lock(&connector->dev->mode_config.mutex); if (!connector->edid_blob_ptr) - return 0; + goto unlock; edid = connector->edid_blob_ptr->data; size = connector->edid_blob_ptr->length; if (!edid) - return 0; + goto unlock; if (off >= size) - return 0; + goto unlock; if (off + count > size) count = size - off; memcpy(buf, edid + off, count); - return count; + ret = count; +unlock: + mutex_unlock(&connector->dev->mode_config.mutex); + + return ret; } static ssize_t modes_show(struct device *device, @@ -283,10 +277,12 @@ static ssize_t modes_show(struct device *device, struct drm_display_mode *mode; int written = 0; + mutex_lock(&connector->dev->mode_config.mutex); list_for_each_entry(mode, &connector->modes, head) { written += snprintf(buf + written, PAGE_SIZE - written, "%s\n", mode->name); } + mutex_unlock(&connector->dev->mode_config.mutex); return written; } diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig index 96e86cf4455b..83efca941388 100644 --- a/drivers/gpu/drm/exynos/Kconfig +++ b/drivers/gpu/drm/exynos/Kconfig @@ -118,7 +118,7 @@ config DRM_EXYNOS_ROTATOR config DRM_EXYNOS_GSC bool "GScaler" - depends on DRM_EXYNOS_IPP && ARCH_EXYNOS5 && !ARCH_MULTIPLATFORM + depends on DRM_EXYNOS_IPP && ARCH_EXYNOS5 && !VIDEO_SAMSUNG_EXYNOS_GSC help Choose this option if you want to use Exynos GSC for DRM. diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index fbe1b3174f75..c7362b99ce28 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -21,11 +21,11 @@ #include "exynos_drm_drv.h" #include "exynos_drm_crtc.h" +#include "exynos_drm_fb.h" #include "exynos_drm_plane.h" #include "exynos_drm_iommu.h" #define WINDOWS_NR 3 -#define CURSOR_WIN 2 #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 static const char * const decon_clks_name[] = { @@ -56,6 +56,7 @@ struct decon_context { struct drm_device *drm_dev; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; + struct exynos_drm_plane_config configs[WINDOWS_NR]; void __iomem *addr; struct clk *clks[ARRAY_SIZE(decon_clks_name)]; int pipe; @@ -71,6 +72,12 @@ static const uint32_t decon_formats[] = { DRM_FORMAT_ARGB8888, }; +static const enum drm_plane_type decon_win_types[WINDOWS_NR] = { + DRM_PLANE_TYPE_PRIMARY, + DRM_PLANE_TYPE_OVERLAY, + DRM_PLANE_TYPE_CURSOR, +}; + static inline void decon_set_bits(struct decon_context *ctx, u32 reg, u32 mask, u32 val) { @@ -259,21 +266,24 @@ static void decon_atomic_begin(struct exynos_drm_crtc *crtc, static void decon_update_plane(struct exynos_drm_crtc *crtc, struct exynos_drm_plane *plane) { + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane->base.state); struct decon_context *ctx = crtc->ctx; - struct drm_plane_state *state = plane->base.state; + struct drm_framebuffer *fb = state->base.fb; unsigned int win = plane->zpos; - unsigned int bpp = state->fb->bits_per_pixel >> 3; - unsigned int pitch = state->fb->pitches[0]; + unsigned int bpp = fb->bits_per_pixel >> 3; + unsigned int pitch = fb->pitches[0]; + dma_addr_t dma_addr = exynos_drm_fb_dma_addr(fb, 0); u32 val; if (test_bit(BIT_SUSPENDED, &ctx->flags)) return; - val = COORDINATE_X(plane->crtc_x) | COORDINATE_Y(plane->crtc_y); + val = COORDINATE_X(state->crtc.x) | COORDINATE_Y(state->crtc.y); writel(val, ctx->addr + DECON_VIDOSDxA(win)); - val = COORDINATE_X(plane->crtc_x + plane->crtc_w - 1) | - COORDINATE_Y(plane->crtc_y + plane->crtc_h - 1); + val = COORDINATE_X(state->crtc.x + state->crtc.w - 1) | + COORDINATE_Y(state->crtc.y + state->crtc.h - 1); writel(val, ctx->addr + DECON_VIDOSDxB(win)); val = VIDOSD_Wx_ALPHA_R_F(0x0) | VIDOSD_Wx_ALPHA_G_F(0x0) | @@ -284,20 +294,20 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc, VIDOSD_Wx_ALPHA_B_F(0x0); writel(val, ctx->addr + DECON_VIDOSDxD(win)); - writel(plane->dma_addr[0], ctx->addr + DECON_VIDW0xADD0B0(win)); + writel(dma_addr, ctx->addr + DECON_VIDW0xADD0B0(win)); - val = plane->dma_addr[0] + pitch * plane->crtc_h; + val = dma_addr + pitch * state->src.h; writel(val, ctx->addr + DECON_VIDW0xADD1B0(win)); if (ctx->out_type != IFTYPE_HDMI) - val = BIT_VAL(pitch - plane->crtc_w * bpp, 27, 14) - | BIT_VAL(plane->crtc_w * bpp, 13, 0); + val = BIT_VAL(pitch - state->crtc.w * bpp, 27, 14) + | BIT_VAL(state->crtc.w * bpp, 13, 0); else - val = BIT_VAL(pitch - plane->crtc_w * bpp, 29, 15) - | BIT_VAL(plane->crtc_w * bpp, 14, 0); + val = BIT_VAL(pitch - state->crtc.w * bpp, 29, 15) + | BIT_VAL(state->crtc.w * bpp, 14, 0); writel(val, ctx->addr + DECON_VIDW0xADD2(win)); - decon_win_set_pixfmt(ctx, win, state->fb); + decon_win_set_pixfmt(ctx, win, fb); /* window enable */ decon_set_bits(ctx, DECON_WINCONx(win), WINCONx_ENWIN_F, ~0); @@ -377,20 +387,12 @@ static void decon_swreset(struct decon_context *ctx) static void decon_enable(struct exynos_drm_crtc *crtc) { struct decon_context *ctx = crtc->ctx; - int ret; - int i; if (!test_and_clear_bit(BIT_SUSPENDED, &ctx->flags)) return; pm_runtime_get_sync(ctx->dev); - for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) { - ret = clk_prepare_enable(ctx->clks[i]); - if (ret < 0) - goto err; - } - set_bit(BIT_CLKS_ENABLED, &ctx->flags); /* if vblank was enabled status, enable it again. */ @@ -399,11 +401,6 @@ static void decon_enable(struct exynos_drm_crtc *crtc) decon_commit(ctx->crtc); - return; -err: - while (--i >= 0) - clk_disable_unprepare(ctx->clks[i]); - set_bit(BIT_SUSPENDED, &ctx->flags); } @@ -425,9 +422,6 @@ static void decon_disable(struct exynos_drm_crtc *crtc) decon_swreset(ctx); - for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) - clk_disable_unprepare(ctx->clks[i]); - clear_bit(BIT_CLKS_ENABLED, &ctx->flags); pm_runtime_put_sync(ctx->dev); @@ -478,7 +472,6 @@ err: static struct exynos_drm_crtc_ops decon_crtc_ops = { .enable = decon_enable, .disable = decon_disable, - .commit = decon_commit, .enable_vblank = decon_enable_vblank, .disable_vblank = decon_disable_vblank, .atomic_begin = decon_atomic_begin, @@ -495,7 +488,6 @@ static int decon_bind(struct device *dev, struct device *master, void *data) struct exynos_drm_private *priv = drm_dev->dev_private; struct exynos_drm_plane *exynos_plane; enum exynos_drm_output_type out_type; - enum drm_plane_type type; unsigned int win; int ret; @@ -505,10 +497,13 @@ static int decon_bind(struct device *dev, struct device *master, void *data) for (win = ctx->first_win; win < WINDOWS_NR; win++) { int tmp = (win == ctx->first_win) ? 0 : win; - type = exynos_plane_get_type(tmp, CURSOR_WIN); + ctx->configs[win].pixel_formats = decon_formats; + ctx->configs[win].num_pixel_formats = ARRAY_SIZE(decon_formats); + ctx->configs[win].zpos = win; + ctx->configs[win].type = decon_win_types[tmp]; + ret = exynos_plane_init(drm_dev, &ctx->planes[win], - 1 << ctx->pipe, type, decon_formats, - ARRAY_SIZE(decon_formats), win); + 1 << ctx->pipe, &ctx->configs[win]); if (ret) return ret; } @@ -581,6 +576,44 @@ out: return IRQ_HANDLED; } +#ifdef CONFIG_PM +static int exynos5433_decon_suspend(struct device *dev) +{ + struct decon_context *ctx = dev_get_drvdata(dev); + int i; + + for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) + clk_disable_unprepare(ctx->clks[i]); + + return 0; +} + +static int exynos5433_decon_resume(struct device *dev) +{ + struct decon_context *ctx = dev_get_drvdata(dev); + int i, ret; + + for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) { + ret = clk_prepare_enable(ctx->clks[i]); + if (ret < 0) + goto err; + } + + return 0; + +err: + while (--i >= 0) + clk_disable_unprepare(ctx->clks[i]); + + return ret; +} +#endif + +static const struct dev_pm_ops exynos5433_decon_pm_ops = { + SET_RUNTIME_PM_OPS(exynos5433_decon_suspend, exynos5433_decon_resume, + NULL) +}; + static const struct of_device_id exynos5433_decon_driver_dt_match[] = { { .compatible = "samsung,exynos5433-decon", @@ -684,6 +717,7 @@ struct platform_driver exynos5433_decon_driver = { .remove = exynos5433_decon_remove, .driver = { .name = "exynos5433-decon", + .pm = &exynos5433_decon_pm_ops, .of_match_table = exynos5433_decon_driver_dt_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index ead2b16e237d..c47f9af8170b 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -30,6 +30,7 @@ #include "exynos_drm_crtc.h" #include "exynos_drm_plane.h" #include "exynos_drm_drv.h" +#include "exynos_drm_fb.h" #include "exynos_drm_fbdev.h" #include "exynos_drm_iommu.h" @@ -40,13 +41,13 @@ #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 #define WINDOWS_NR 2 -#define CURSOR_WIN 1 struct decon_context { struct device *dev; struct drm_device *drm_dev; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; + struct exynos_drm_plane_config configs[WINDOWS_NR]; struct clk *pclk; struct clk *aclk; struct clk *eclk; @@ -81,6 +82,11 @@ static const uint32_t decon_formats[] = { DRM_FORMAT_BGRA8888, }; +static const enum drm_plane_type decon_win_types[WINDOWS_NR] = { + DRM_PLANE_TYPE_PRIMARY, + DRM_PLANE_TYPE_CURSOR, +}; + static void decon_wait_for_vblank(struct exynos_drm_crtc *crtc) { struct decon_context *ctx = crtc->ctx; @@ -119,13 +125,8 @@ static void decon_clear_channels(struct exynos_drm_crtc *crtc) } /* Wait for vsync, as disable channel takes effect at next vsync */ - if (ch_enabled) { - unsigned int state = ctx->suspended; - - ctx->suspended = 0; + if (ch_enabled) decon_wait_for_vblank(ctx->crtc); - ctx->suspended = state; - } } static int decon_ctx_initialize(struct decon_context *ctx, @@ -398,16 +399,17 @@ static void decon_atomic_begin(struct exynos_drm_crtc *crtc, static void decon_update_plane(struct exynos_drm_crtc *crtc, struct exynos_drm_plane *plane) { + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane->base.state); struct decon_context *ctx = crtc->ctx; - struct drm_display_mode *mode = &crtc->base.state->adjusted_mode; - struct drm_plane_state *state = plane->base.state; + struct drm_framebuffer *fb = state->base.fb; int padding; unsigned long val, alpha; unsigned int last_x; unsigned int last_y; unsigned int win = plane->zpos; - unsigned int bpp = state->fb->bits_per_pixel >> 3; - unsigned int pitch = state->fb->pitches[0]; + unsigned int bpp = fb->bits_per_pixel >> 3; + unsigned int pitch = fb->pitches[0]; if (ctx->suspended) return; @@ -423,41 +425,32 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc, */ /* buffer start address */ - val = (unsigned long)plane->dma_addr[0]; + val = (unsigned long)exynos_drm_fb_dma_addr(fb, 0); writel(val, ctx->regs + VIDW_BUF_START(win)); - padding = (pitch / bpp) - state->fb->width; + padding = (pitch / bpp) - fb->width; /* buffer size */ - writel(state->fb->width + padding, ctx->regs + VIDW_WHOLE_X(win)); - writel(state->fb->height, ctx->regs + VIDW_WHOLE_Y(win)); + writel(fb->width + padding, ctx->regs + VIDW_WHOLE_X(win)); + writel(fb->height, ctx->regs + VIDW_WHOLE_Y(win)); /* offset from the start of the buffer to read */ - writel(plane->src_x, ctx->regs + VIDW_OFFSET_X(win)); - writel(plane->src_y, ctx->regs + VIDW_OFFSET_Y(win)); + writel(state->src.x, ctx->regs + VIDW_OFFSET_X(win)); + writel(state->src.y, ctx->regs + VIDW_OFFSET_Y(win)); DRM_DEBUG_KMS("start addr = 0x%lx\n", (unsigned long)val); DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n", - plane->crtc_w, plane->crtc_h); + state->crtc.w, state->crtc.h); - /* - * OSD position. - * In case the window layout goes of LCD layout, DECON fails. - */ - if ((plane->crtc_x + plane->crtc_w) > mode->hdisplay) - plane->crtc_x = mode->hdisplay - plane->crtc_w; - if ((plane->crtc_y + plane->crtc_h) > mode->vdisplay) - plane->crtc_y = mode->vdisplay - plane->crtc_h; - - val = VIDOSDxA_TOPLEFT_X(plane->crtc_x) | - VIDOSDxA_TOPLEFT_Y(plane->crtc_y); + val = VIDOSDxA_TOPLEFT_X(state->crtc.x) | + VIDOSDxA_TOPLEFT_Y(state->crtc.y); writel(val, ctx->regs + VIDOSD_A(win)); - last_x = plane->crtc_x + plane->crtc_w; + last_x = state->crtc.x + state->crtc.w; if (last_x) last_x--; - last_y = plane->crtc_y + plane->crtc_h; + last_y = state->crtc.y + state->crtc.h; if (last_y) last_y--; @@ -466,7 +459,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc, writel(val, ctx->regs + VIDOSD_B(win)); DRM_DEBUG_KMS("osd pos: tx = %d, ty = %d, bx = %d, by = %d\n", - plane->crtc_x, plane->crtc_y, last_x, last_y); + state->crtc.x, state->crtc.y, last_x, last_y); /* OSD alpha */ alpha = VIDOSDxC_ALPHA0_R_F(0x0) | @@ -481,7 +474,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc, writel(alpha, ctx->regs + VIDOSD_D(win)); - decon_win_set_pixfmt(ctx, win, state->fb); + decon_win_set_pixfmt(ctx, win, fb); /* hardware window 0 doesn't support color key. */ if (win != 0) @@ -555,39 +548,12 @@ static void decon_init(struct decon_context *ctx) static void decon_enable(struct exynos_drm_crtc *crtc) { struct decon_context *ctx = crtc->ctx; - int ret; if (!ctx->suspended) return; - ctx->suspended = false; - pm_runtime_get_sync(ctx->dev); - ret = clk_prepare_enable(ctx->pclk); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the pclk [%d]\n", ret); - return; - } - - ret = clk_prepare_enable(ctx->aclk); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the aclk [%d]\n", ret); - return; - } - - ret = clk_prepare_enable(ctx->eclk); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the eclk [%d]\n", ret); - return; - } - - ret = clk_prepare_enable(ctx->vclk); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the vclk [%d]\n", ret); - return; - } - decon_init(ctx); /* if vblank was enabled status, enable it again. */ @@ -595,6 +561,8 @@ static void decon_enable(struct exynos_drm_crtc *crtc) decon_enable_vblank(ctx->crtc); decon_commit(ctx->crtc); + + ctx->suspended = false; } static void decon_disable(struct exynos_drm_crtc *crtc) @@ -613,11 +581,6 @@ static void decon_disable(struct exynos_drm_crtc *crtc) for (i = 0; i < WINDOWS_NR; i++) decon_disable_plane(crtc, &ctx->planes[i]); - clk_disable_unprepare(ctx->vclk); - clk_disable_unprepare(ctx->eclk); - clk_disable_unprepare(ctx->aclk); - clk_disable_unprepare(ctx->pclk); - pm_runtime_put_sync(ctx->dev); ctx->suspended = true; @@ -679,8 +642,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data) struct decon_context *ctx = dev_get_drvdata(dev); struct drm_device *drm_dev = data; struct exynos_drm_plane *exynos_plane; - enum drm_plane_type type; - unsigned int zpos; + unsigned int i; int ret; ret = decon_ctx_initialize(ctx, drm_dev); @@ -689,11 +651,14 @@ static int decon_bind(struct device *dev, struct device *master, void *data) return ret; } - for (zpos = 0; zpos < WINDOWS_NR; zpos++) { - type = exynos_plane_get_type(zpos, CURSOR_WIN); - ret = exynos_plane_init(drm_dev, &ctx->planes[zpos], - 1 << ctx->pipe, type, decon_formats, - ARRAY_SIZE(decon_formats), zpos); + for (i = 0; i < WINDOWS_NR; i++) { + ctx->configs[i].pixel_formats = decon_formats; + ctx->configs[i].num_pixel_formats = ARRAY_SIZE(decon_formats); + ctx->configs[i].zpos = i; + ctx->configs[i].type = decon_win_types[i]; + + ret = exynos_plane_init(drm_dev, &ctx->planes[i], + 1 << ctx->pipe, &ctx->configs[i]); if (ret) return ret; } @@ -843,11 +808,63 @@ static int decon_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM +static int exynos7_decon_suspend(struct device *dev) +{ + struct decon_context *ctx = dev_get_drvdata(dev); + + clk_disable_unprepare(ctx->vclk); + clk_disable_unprepare(ctx->eclk); + clk_disable_unprepare(ctx->aclk); + clk_disable_unprepare(ctx->pclk); + + return 0; +} + +static int exynos7_decon_resume(struct device *dev) +{ + struct decon_context *ctx = dev_get_drvdata(dev); + int ret; + + ret = clk_prepare_enable(ctx->pclk); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the pclk [%d]\n", ret); + return ret; + } + + ret = clk_prepare_enable(ctx->aclk); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the aclk [%d]\n", ret); + return ret; + } + + ret = clk_prepare_enable(ctx->eclk); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the eclk [%d]\n", ret); + return ret; + } + + ret = clk_prepare_enable(ctx->vclk); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the vclk [%d]\n", ret); + return ret; + } + + return 0; +} +#endif + +static const struct dev_pm_ops exynos7_decon_pm_ops = { + SET_RUNTIME_PM_OPS(exynos7_decon_suspend, exynos7_decon_resume, + NULL) +}; + struct platform_driver decon_driver = { .probe = decon_probe, .remove = decon_remove, .driver = { .name = "exynos-decon", + .pm = &exynos7_decon_pm_ops, .of_match_table = decon_driver_dt_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c index 124fb9a56f02..793e4977fcf7 100644 --- a/drivers/gpu/drm/exynos/exynos_dp_core.c +++ b/drivers/gpu/drm/exynos/exynos_dp_core.c @@ -1009,9 +1009,9 @@ static int exynos_drm_attach_lcd_bridge(struct exynos_dp_device *dp, { int ret; - encoder->bridge = dp->bridge; - dp->bridge->encoder = encoder; - ret = drm_bridge_attach(encoder->dev, dp->bridge); + encoder->bridge->next = dp->ptn_bridge; + dp->ptn_bridge->encoder = encoder; + ret = drm_bridge_attach(encoder->dev, dp->ptn_bridge); if (ret) { DRM_ERROR("Failed to attach bridge to drm\n"); return ret; @@ -1020,14 +1020,15 @@ static int exynos_drm_attach_lcd_bridge(struct exynos_dp_device *dp, return 0; } -static int exynos_dp_create_connector(struct drm_encoder *encoder) +static int exynos_dp_bridge_attach(struct drm_bridge *bridge) { - struct exynos_dp_device *dp = encoder_to_dp(encoder); + struct exynos_dp_device *dp = bridge->driver_private; + struct drm_encoder *encoder = &dp->encoder; struct drm_connector *connector = &dp->connector; int ret; /* Pre-empt DP connector creation if there's a bridge */ - if (dp->bridge) { + if (dp->ptn_bridge) { ret = exynos_drm_attach_lcd_bridge(dp, encoder); if (!ret) return 0; @@ -1052,27 +1053,16 @@ static int exynos_dp_create_connector(struct drm_encoder *encoder) return ret; } -static bool exynos_dp_mode_fixup(struct drm_encoder *encoder, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - return true; -} - -static void exynos_dp_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ -} - -static void exynos_dp_enable(struct drm_encoder *encoder) +static void exynos_dp_bridge_enable(struct drm_bridge *bridge) { - struct exynos_dp_device *dp = encoder_to_dp(encoder); + struct exynos_dp_device *dp = bridge->driver_private; struct exynos_drm_crtc *crtc = dp_to_crtc(dp); if (dp->dpms_mode == DRM_MODE_DPMS_ON) return; + pm_runtime_get_sync(dp->dev); + if (dp->panel) { if (drm_panel_prepare(dp->panel)) { DRM_ERROR("failed to setup the panel\n"); @@ -1083,7 +1073,6 @@ static void exynos_dp_enable(struct drm_encoder *encoder) if (crtc->ops->clock_enable) crtc->ops->clock_enable(dp_to_crtc(dp), true); - clk_prepare_enable(dp->clock); phy_power_on(dp->phy); exynos_dp_init_dp(dp); enable_irq(dp->irq); @@ -1092,9 +1081,9 @@ static void exynos_dp_enable(struct drm_encoder *encoder) dp->dpms_mode = DRM_MODE_DPMS_ON; } -static void exynos_dp_disable(struct drm_encoder *encoder) +static void exynos_dp_bridge_disable(struct drm_bridge *bridge) { - struct exynos_dp_device *dp = encoder_to_dp(encoder); + struct exynos_dp_device *dp = bridge->driver_private; struct exynos_drm_crtc *crtc = dp_to_crtc(dp); if (dp->dpms_mode != DRM_MODE_DPMS_ON) @@ -1110,7 +1099,6 @@ static void exynos_dp_disable(struct drm_encoder *encoder) disable_irq(dp->irq); flush_work(&dp->hotplug_work); phy_power_off(dp->phy); - clk_disable_unprepare(dp->clock); if (crtc->ops->clock_enable) crtc->ops->clock_enable(dp_to_crtc(dp), false); @@ -1120,9 +1108,74 @@ static void exynos_dp_disable(struct drm_encoder *encoder) DRM_ERROR("failed to turnoff the panel\n"); } + pm_runtime_put_sync(dp->dev); + dp->dpms_mode = DRM_MODE_DPMS_OFF; } +static void exynos_dp_bridge_nop(struct drm_bridge *bridge) +{ + /* do nothing */ +} + +static const struct drm_bridge_funcs exynos_dp_bridge_funcs = { + .enable = exynos_dp_bridge_enable, + .disable = exynos_dp_bridge_disable, + .pre_enable = exynos_dp_bridge_nop, + .post_disable = exynos_dp_bridge_nop, + .attach = exynos_dp_bridge_attach, +}; + +static int exynos_dp_create_connector(struct drm_encoder *encoder) +{ + struct exynos_dp_device *dp = encoder_to_dp(encoder); + struct drm_device *drm_dev = dp->drm_dev; + struct drm_bridge *bridge; + int ret; + + bridge = devm_kzalloc(drm_dev->dev, sizeof(*bridge), GFP_KERNEL); + if (!bridge) { + DRM_ERROR("failed to allocate for drm bridge\n"); + return -ENOMEM; + } + + dp->bridge = bridge; + + encoder->bridge = bridge; + bridge->driver_private = dp; + bridge->encoder = encoder; + bridge->funcs = &exynos_dp_bridge_funcs; + + ret = drm_bridge_attach(drm_dev, bridge); + if (ret) { + DRM_ERROR("failed to attach drm bridge\n"); + return -EINVAL; + } + + return 0; +} + +static bool exynos_dp_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + return true; +} + +static void exynos_dp_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ +} + +static void exynos_dp_enable(struct drm_encoder *encoder) +{ +} + +static void exynos_dp_disable(struct drm_encoder *encoder) +{ +} + static struct drm_encoder_helper_funcs exynos_dp_encoder_helper_funcs = { .mode_fixup = exynos_dp_mode_fixup, .mode_set = exynos_dp_mode_set, @@ -1238,7 +1291,7 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data) } } - if (!dp->panel && !dp->bridge) { + if (!dp->panel && !dp->ptn_bridge) { ret = exynos_dp_dt_parse_panel(dp); if (ret) return ret; @@ -1289,10 +1342,6 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data) INIT_WORK(&dp->hotplug_work, exynos_dp_hotplug); - phy_power_on(dp->phy); - - exynos_dp_init_dp(dp); - ret = devm_request_irq(&pdev->dev, dp->irq, exynos_dp_irq_handler, irq_flags, "exynos-dp", dp); if (ret) { @@ -1313,7 +1362,7 @@ static int exynos_dp_bind(struct device *dev, struct device *master, void *data) DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs); drm_encoder_init(drm_dev, encoder, &exynos_dp_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &exynos_dp_encoder_helper_funcs); @@ -1343,8 +1392,9 @@ static const struct component_ops exynos_dp_ops = { static int exynos_dp_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; - struct device_node *panel_node, *bridge_node, *endpoint; + struct device_node *panel_node = NULL, *bridge_node, *endpoint = NULL; struct exynos_dp_device *dp; + int ret; dp = devm_kzalloc(&pdev->dev, sizeof(struct exynos_dp_device), GFP_KERNEL); @@ -1353,36 +1403,96 @@ static int exynos_dp_probe(struct platform_device *pdev) platform_set_drvdata(pdev, dp); + /* This is for the backward compatibility. */ panel_node = of_parse_phandle(dev->of_node, "panel", 0); if (panel_node) { dp->panel = of_drm_find_panel(panel_node); of_node_put(panel_node); if (!dp->panel) return -EPROBE_DEFER; + } else { + endpoint = of_graph_get_next_endpoint(dev->of_node, NULL); + if (endpoint) { + panel_node = of_graph_get_remote_port_parent(endpoint); + if (panel_node) { + dp->panel = of_drm_find_panel(panel_node); + of_node_put(panel_node); + if (!dp->panel) + return -EPROBE_DEFER; + } else { + DRM_ERROR("no port node for panel device.\n"); + return -EINVAL; + } + } } + if (endpoint) + goto out; + endpoint = of_graph_get_next_endpoint(dev->of_node, NULL); if (endpoint) { bridge_node = of_graph_get_remote_port_parent(endpoint); if (bridge_node) { - dp->bridge = of_drm_find_bridge(bridge_node); + dp->ptn_bridge = of_drm_find_bridge(bridge_node); of_node_put(bridge_node); - if (!dp->bridge) + if (!dp->ptn_bridge) return -EPROBE_DEFER; } else return -EPROBE_DEFER; } - return component_add(&pdev->dev, &exynos_dp_ops); +out: + pm_runtime_enable(dev); + + ret = component_add(&pdev->dev, &exynos_dp_ops); + if (ret) + goto err_disable_pm_runtime; + + return ret; + +err_disable_pm_runtime: + pm_runtime_disable(dev); + + return ret; } static int exynos_dp_remove(struct platform_device *pdev) { + pm_runtime_disable(&pdev->dev); component_del(&pdev->dev, &exynos_dp_ops); return 0; } +#ifdef CONFIG_PM +static int exynos_dp_suspend(struct device *dev) +{ + struct exynos_dp_device *dp = dev_get_drvdata(dev); + + clk_disable_unprepare(dp->clock); + + return 0; +} + +static int exynos_dp_resume(struct device *dev) +{ + struct exynos_dp_device *dp = dev_get_drvdata(dev); + int ret; + + ret = clk_prepare_enable(dp->clock); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the clock clk [%d]\n", ret); + return ret; + } + + return 0; +} +#endif + +static const struct dev_pm_ops exynos_dp_pm_ops = { + SET_RUNTIME_PM_OPS(exynos_dp_suspend, exynos_dp_resume, NULL) +}; + static const struct of_device_id exynos_dp_match[] = { { .compatible = "samsung,exynos5-dp" }, {}, @@ -1395,6 +1505,7 @@ struct platform_driver dp_driver = { .driver = { .name = "exynos-dp", .owner = THIS_MODULE, + .pm = &exynos_dp_pm_ops, .of_match_table = exynos_dp_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.h b/drivers/gpu/drm/exynos/exynos_dp_core.h index e413b6f7b0e7..66eec4b2d5c6 100644 --- a/drivers/gpu/drm/exynos/exynos_dp_core.h +++ b/drivers/gpu/drm/exynos/exynos_dp_core.h @@ -153,6 +153,7 @@ struct exynos_dp_device { struct drm_connector connector; struct drm_panel *panel; struct drm_bridge *bridge; + struct drm_bridge *ptn_bridge; struct clk *clock; unsigned int irq; void __iomem *reg_base; diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index b3ba27fd9a6b..9d30a0fa3248 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -150,7 +150,7 @@ struct exynos_drm_crtc *exynos_drm_crtc_create(struct drm_device *drm_dev, private->crtc[pipe] = crtc; ret = drm_crtc_init_with_planes(drm_dev, crtc, plane, NULL, - &exynos_crtc_funcs); + &exynos_crtc_funcs, NULL); if (ret < 0) goto err_crtc; diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c index c748b8790de3..1dbf8dca2d6b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c @@ -309,7 +309,7 @@ int exynos_dpi_bind(struct drm_device *dev, struct drm_encoder *encoder) DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs); drm_encoder_init(dev, encoder, &exynos_dpi_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &exynos_dpi_encoder_helper_funcs); diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 2c6019d6a205..9756797a15a5 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -304,45 +304,6 @@ int exynos_atomic_commit(struct drm_device *dev, struct drm_atomic_state *state, return 0; } -#ifdef CONFIG_PM_SLEEP -static int exynos_drm_suspend(struct drm_device *dev, pm_message_t state) -{ - struct drm_connector *connector; - - drm_modeset_lock_all(dev); - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - int old_dpms = connector->dpms; - - if (connector->funcs->dpms) - connector->funcs->dpms(connector, DRM_MODE_DPMS_OFF); - - /* Set the old mode back to the connector for resume */ - connector->dpms = old_dpms; - } - drm_modeset_unlock_all(dev); - - return 0; -} - -static int exynos_drm_resume(struct drm_device *dev) -{ - struct drm_connector *connector; - - drm_modeset_lock_all(dev); - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - if (connector->funcs->dpms) { - int dpms = connector->dpms; - - connector->dpms = DRM_MODE_DPMS_OFF; - connector->funcs->dpms(connector, dpms); - } - } - drm_modeset_unlock_all(dev); - - return 0; -} -#endif - static int exynos_drm_open(struct drm_device *dev, struct drm_file *file) { struct drm_exynos_file_private *file_priv; @@ -476,31 +437,54 @@ static struct drm_driver exynos_drm_driver = { }; #ifdef CONFIG_PM_SLEEP -static int exynos_drm_sys_suspend(struct device *dev) +static int exynos_drm_suspend(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); - pm_message_t message; + struct drm_connector *connector; if (pm_runtime_suspended(dev) || !drm_dev) return 0; - message.event = PM_EVENT_SUSPEND; - return exynos_drm_suspend(drm_dev, message); + drm_modeset_lock_all(drm_dev); + drm_for_each_connector(connector, drm_dev) { + int old_dpms = connector->dpms; + + if (connector->funcs->dpms) + connector->funcs->dpms(connector, DRM_MODE_DPMS_OFF); + + /* Set the old mode back to the connector for resume */ + connector->dpms = old_dpms; + } + drm_modeset_unlock_all(drm_dev); + + return 0; } -static int exynos_drm_sys_resume(struct device *dev) +static int exynos_drm_resume(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); + struct drm_connector *connector; if (pm_runtime_suspended(dev) || !drm_dev) return 0; - return exynos_drm_resume(drm_dev); + drm_modeset_lock_all(drm_dev); + drm_for_each_connector(connector, drm_dev) { + if (connector->funcs->dpms) { + int dpms = connector->dpms; + + connector->dpms = DRM_MODE_DPMS_OFF; + connector->funcs->dpms(connector, dpms); + } + } + drm_modeset_unlock_all(drm_dev); + + return 0; } #endif static const struct dev_pm_ops exynos_drm_pm_ops = { - SET_SYSTEM_SLEEP_PM_OPS(exynos_drm_sys_suspend, exynos_drm_sys_resume) + SET_SYSTEM_SLEEP_PM_OPS(exynos_drm_suspend, exynos_drm_resume) }; /* forward declaration */ diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index f1eda7fa4e3c..82bbd7f4b316 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -38,24 +38,44 @@ enum exynos_drm_output_type { EXYNOS_DISPLAY_TYPE_VIDI, }; +struct exynos_drm_rect { + unsigned int x, y; + unsigned int w, h; +}; + /* - * Exynos drm common overlay structure. + * Exynos drm plane state structure. * - * @base: plane object - * @src_x: offset x on a framebuffer to be displayed. - * - the unit is screen coordinates. - * @src_y: offset y on a framebuffer to be displayed. - * - the unit is screen coordinates. - * @src_w: width of a partial image to be displayed from framebuffer. - * @src_h: height of a partial image to be displayed from framebuffer. - * @crtc_x: offset x on hardware screen. - * @crtc_y: offset y on hardware screen. - * @crtc_w: window width to be displayed (hardware screen). - * @crtc_h: window height to be displayed (hardware screen). + * @base: plane_state object (contains drm_framebuffer pointer) + * @src: rectangle of the source image data to be displayed (clipped to + * visible part). + * @crtc: rectangle of the target image position on hardware screen + * (clipped to visible part). * @h_ratio: horizontal scaling ratio, 16.16 fixed point * @v_ratio: vertical scaling ratio, 16.16 fixed point - * @dma_addr: array of bus(accessed by dma) address to the memory region - * allocated for a overlay. + * + * this structure consists plane state data that will be applied to hardware + * specific overlay info. + */ + +struct exynos_drm_plane_state { + struct drm_plane_state base; + struct exynos_drm_rect crtc; + struct exynos_drm_rect src; + unsigned int h_ratio; + unsigned int v_ratio; +}; + +static inline struct exynos_drm_plane_state * +to_exynos_plane_state(struct drm_plane_state *state) +{ + return container_of(state, struct exynos_drm_plane_state, base); +} + +/* + * Exynos drm common overlay structure. + * + * @base: plane object * @zpos: order of overlay layer(z position). * * this structure is common to exynos SoC and its contents would be copied @@ -64,21 +84,32 @@ enum exynos_drm_output_type { struct exynos_drm_plane { struct drm_plane base; - unsigned int src_x; - unsigned int src_y; - unsigned int src_w; - unsigned int src_h; - unsigned int crtc_x; - unsigned int crtc_y; - unsigned int crtc_w; - unsigned int crtc_h; - unsigned int h_ratio; - unsigned int v_ratio; - dma_addr_t dma_addr[MAX_FB_BUFFER]; + const struct exynos_drm_plane_config *config; unsigned int zpos; struct drm_framebuffer *pending_fb; }; +#define EXYNOS_DRM_PLANE_CAP_DOUBLE (1 << 0) +#define EXYNOS_DRM_PLANE_CAP_SCALE (1 << 1) + +/* + * Exynos DRM plane configuration structure. + * + * @zpos: z-position of the plane. + * @type: type of the plane (primary, cursor or overlay). + * @pixel_formats: supported pixel formats. + * @num_pixel_formats: number of elements in 'pixel_formats'. + * @capabilities: supported features (see EXYNOS_DRM_PLANE_CAP_*) + */ + +struct exynos_drm_plane_config { + unsigned int zpos; + enum drm_plane_type type; + const uint32_t *pixel_formats; + unsigned int num_pixel_formats; + unsigned int capabilities; +}; + /* * Exynos drm crtc ops * diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index 12b03b364703..bc09bba3132a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -1458,66 +1458,6 @@ static const struct mipi_dsi_host_ops exynos_dsi_ops = { .transfer = exynos_dsi_host_transfer, }; -static int exynos_dsi_poweron(struct exynos_dsi *dsi) -{ - struct exynos_dsi_driver_data *driver_data = dsi->driver_data; - int ret, i; - - ret = regulator_bulk_enable(ARRAY_SIZE(dsi->supplies), dsi->supplies); - if (ret < 0) { - dev_err(dsi->dev, "cannot enable regulators %d\n", ret); - return ret; - } - - for (i = 0; i < driver_data->num_clks; i++) { - ret = clk_prepare_enable(dsi->clks[i]); - if (ret < 0) - goto err_clk; - } - - ret = phy_power_on(dsi->phy); - if (ret < 0) { - dev_err(dsi->dev, "cannot enable phy %d\n", ret); - goto err_clk; - } - - return 0; - -err_clk: - while (--i > -1) - clk_disable_unprepare(dsi->clks[i]); - regulator_bulk_disable(ARRAY_SIZE(dsi->supplies), dsi->supplies); - - return ret; -} - -static void exynos_dsi_poweroff(struct exynos_dsi *dsi) -{ - struct exynos_dsi_driver_data *driver_data = dsi->driver_data; - int ret, i; - - usleep_range(10000, 20000); - - if (dsi->state & DSIM_STATE_INITIALIZED) { - dsi->state &= ~DSIM_STATE_INITIALIZED; - - exynos_dsi_disable_clock(dsi); - - exynos_dsi_disable_irq(dsi); - } - - dsi->state &= ~DSIM_STATE_CMD_LPM; - - phy_power_off(dsi->phy); - - for (i = driver_data->num_clks - 1; i > -1; i--) - clk_disable_unprepare(dsi->clks[i]); - - ret = regulator_bulk_disable(ARRAY_SIZE(dsi->supplies), dsi->supplies); - if (ret < 0) - dev_err(dsi->dev, "cannot disable regulators %d\n", ret); -} - static void exynos_dsi_enable(struct drm_encoder *encoder) { struct exynos_dsi *dsi = encoder_to_dsi(encoder); @@ -1526,16 +1466,14 @@ static void exynos_dsi_enable(struct drm_encoder *encoder) if (dsi->state & DSIM_STATE_ENABLED) return; - ret = exynos_dsi_poweron(dsi); - if (ret < 0) - return; + pm_runtime_get_sync(dsi->dev); dsi->state |= DSIM_STATE_ENABLED; ret = drm_panel_prepare(dsi->panel); if (ret < 0) { dsi->state &= ~DSIM_STATE_ENABLED; - exynos_dsi_poweroff(dsi); + pm_runtime_put_sync(dsi->dev); return; } @@ -1547,7 +1485,7 @@ static void exynos_dsi_enable(struct drm_encoder *encoder) dsi->state &= ~DSIM_STATE_ENABLED; exynos_dsi_set_display_enable(dsi, false); drm_panel_unprepare(dsi->panel); - exynos_dsi_poweroff(dsi); + pm_runtime_put_sync(dsi->dev); return; } @@ -1569,7 +1507,7 @@ static void exynos_dsi_disable(struct drm_encoder *encoder) dsi->state &= ~DSIM_STATE_ENABLED; - exynos_dsi_poweroff(dsi); + pm_runtime_put_sync(dsi->dev); } static enum drm_connector_status @@ -1797,13 +1735,13 @@ static int exynos_dsi_parse_dt(struct exynos_dsi *dsi) ep = of_graph_get_next_endpoint(node, NULL); if (!ep) { - ret = -ENXIO; + ret = -EINVAL; goto end; } dsi->bridge_node = of_graph_get_remote_port_parent(ep); if (!dsi->bridge_node) { - ret = -ENXIO; + ret = -EINVAL; goto end; } end: @@ -1831,7 +1769,7 @@ static int exynos_dsi_bind(struct device *dev, struct device *master, DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs); drm_encoder_init(drm_dev, encoder, &exynos_dsi_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &exynos_dsi_encoder_helper_funcs); @@ -1954,22 +1892,99 @@ static int exynos_dsi_probe(struct platform_device *pdev) platform_set_drvdata(pdev, &dsi->encoder); + pm_runtime_enable(dev); + return component_add(dev, &exynos_dsi_component_ops); } static int exynos_dsi_remove(struct platform_device *pdev) { + pm_runtime_disable(&pdev->dev); + component_del(&pdev->dev, &exynos_dsi_component_ops); return 0; } +#ifdef CONFIG_PM +static int exynos_dsi_suspend(struct device *dev) +{ + struct drm_encoder *encoder = dev_get_drvdata(dev); + struct exynos_dsi *dsi = encoder_to_dsi(encoder); + struct exynos_dsi_driver_data *driver_data = dsi->driver_data; + int ret, i; + + usleep_range(10000, 20000); + + if (dsi->state & DSIM_STATE_INITIALIZED) { + dsi->state &= ~DSIM_STATE_INITIALIZED; + + exynos_dsi_disable_clock(dsi); + + exynos_dsi_disable_irq(dsi); + } + + dsi->state &= ~DSIM_STATE_CMD_LPM; + + phy_power_off(dsi->phy); + + for (i = driver_data->num_clks - 1; i > -1; i--) + clk_disable_unprepare(dsi->clks[i]); + + ret = regulator_bulk_disable(ARRAY_SIZE(dsi->supplies), dsi->supplies); + if (ret < 0) + dev_err(dsi->dev, "cannot disable regulators %d\n", ret); + + return 0; +} + +static int exynos_dsi_resume(struct device *dev) +{ + struct drm_encoder *encoder = dev_get_drvdata(dev); + struct exynos_dsi *dsi = encoder_to_dsi(encoder); + struct exynos_dsi_driver_data *driver_data = dsi->driver_data; + int ret, i; + + ret = regulator_bulk_enable(ARRAY_SIZE(dsi->supplies), dsi->supplies); + if (ret < 0) { + dev_err(dsi->dev, "cannot enable regulators %d\n", ret); + return ret; + } + + for (i = 0; i < driver_data->num_clks; i++) { + ret = clk_prepare_enable(dsi->clks[i]); + if (ret < 0) + goto err_clk; + } + + ret = phy_power_on(dsi->phy); + if (ret < 0) { + dev_err(dsi->dev, "cannot enable phy %d\n", ret); + goto err_clk; + } + + return 0; + +err_clk: + while (--i > -1) + clk_disable_unprepare(dsi->clks[i]); + regulator_bulk_disable(ARRAY_SIZE(dsi->supplies), dsi->supplies); + + return ret; +} +#endif + +static const struct dev_pm_ops exynos_dsi_pm_ops = { + SET_RUNTIME_PM_OPS(exynos_dsi_suspend, exynos_dsi_resume, NULL) +}; + struct platform_driver dsi_driver = { .probe = exynos_dsi_probe, .remove = exynos_dsi_remove, .driver = { .name = "exynos-dsi", .owner = THIS_MODULE, + .pm = &exynos_dsi_pm_ops, .of_match_table = exynos_dsi_of_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c index fcea28bdbc42..f6bdb0d6f142 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c @@ -37,6 +37,7 @@ struct exynos_drm_fb { struct drm_framebuffer fb; struct exynos_drm_gem *exynos_gem[MAX_FB_BUFFER]; + dma_addr_t dma_addr[MAX_FB_BUFFER]; }; static int check_fb_gem_memory_type(struct drm_device *drm_dev, @@ -117,7 +118,7 @@ static struct drm_framebuffer_funcs exynos_drm_fb_funcs = { struct drm_framebuffer * exynos_drm_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct exynos_drm_gem **exynos_gem, int count) { @@ -135,6 +136,8 @@ exynos_drm_framebuffer_init(struct drm_device *dev, goto err; exynos_fb->exynos_gem[i] = exynos_gem[i]; + exynos_fb->dma_addr[i] = exynos_gem[i]->dma_addr + + mode_cmd->offsets[i]; } drm_helper_mode_fill_fb_struct(&exynos_fb->fb, mode_cmd); @@ -154,7 +157,7 @@ err: static struct drm_framebuffer * exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct exynos_drm_gem *exynos_gem[MAX_FB_BUFFER]; struct drm_gem_object *obj; @@ -189,21 +192,14 @@ err: return ERR_PTR(ret); } -struct exynos_drm_gem *exynos_drm_fb_gem(struct drm_framebuffer *fb, int index) +dma_addr_t exynos_drm_fb_dma_addr(struct drm_framebuffer *fb, int index) { struct exynos_drm_fb *exynos_fb = to_exynos_fb(fb); - struct exynos_drm_gem *exynos_gem; if (index >= MAX_FB_BUFFER) - return NULL; + return DMA_ERROR_CODE; - exynos_gem = exynos_fb->exynos_gem[index]; - if (!exynos_gem) - return NULL; - - DRM_DEBUG_KMS("dma_addr: 0x%lx\n", (unsigned long)exynos_gem->dma_addr); - - return exynos_gem; + return exynos_fb->dma_addr[index]; } static void exynos_drm_output_poll_changed(struct drm_device *dev) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.h b/drivers/gpu/drm/exynos/exynos_drm_fb.h index 726a2d44371f..4aae9dd2b0d1 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.h +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.h @@ -18,12 +18,11 @@ struct drm_framebuffer * exynos_drm_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct exynos_drm_gem **exynos_gem, int count); -/* get gem object of a drm framebuffer */ -struct exynos_drm_gem *exynos_drm_fb_gem(struct drm_framebuffer *fb, int index); +dma_addr_t exynos_drm_fb_dma_addr(struct drm_framebuffer *fb, int index); void exynos_drm_mode_config_init(struct drm_device *dev); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index bd75c1531cac..2e2247126581 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -29,6 +29,7 @@ #include <drm/exynos_drm.h> #include "exynos_drm_drv.h" +#include "exynos_drm_fb.h" #include "exynos_drm_fbdev.h" #include "exynos_drm_crtc.h" #include "exynos_drm_plane.h" @@ -87,7 +88,6 @@ /* FIMD has totally five hardware windows. */ #define WINDOWS_NR 5 -#define CURSOR_WIN 4 struct fimd_driver_data { unsigned int timing_base; @@ -150,6 +150,7 @@ struct fimd_context { struct drm_device *drm_dev; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; + struct exynos_drm_plane_config configs[WINDOWS_NR]; struct clk *bus_clk; struct clk *lcd_clk; void __iomem *regs; @@ -187,6 +188,14 @@ static const struct of_device_id fimd_driver_dt_match[] = { }; MODULE_DEVICE_TABLE(of, fimd_driver_dt_match); +static const enum drm_plane_type fimd_win_types[WINDOWS_NR] = { + DRM_PLANE_TYPE_PRIMARY, + DRM_PLANE_TYPE_OVERLAY, + DRM_PLANE_TYPE_OVERLAY, + DRM_PLANE_TYPE_OVERLAY, + DRM_PLANE_TYPE_CURSOR, +}; + static const uint32_t fimd_formats[] = { DRM_FORMAT_C8, DRM_FORMAT_XRGB1555, @@ -478,7 +487,7 @@ static void fimd_commit(struct exynos_drm_crtc *crtc) static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, - struct drm_framebuffer *fb) + uint32_t pixel_format, int width) { unsigned long val; @@ -489,11 +498,11 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, * So the request format is ARGB8888 then change it to XRGB8888. */ if (ctx->driver_data->has_limited_fmt && !win) { - if (fb->pixel_format == DRM_FORMAT_ARGB8888) - fb->pixel_format = DRM_FORMAT_XRGB8888; + if (pixel_format == DRM_FORMAT_ARGB8888) + pixel_format = DRM_FORMAT_XRGB8888; } - switch (fb->pixel_format) { + switch (pixel_format) { case DRM_FORMAT_C8: val |= WINCON0_BPPMODE_8BPP_PALETTE; val |= WINCONx_BURSTLEN_8WORD; @@ -529,17 +538,15 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, break; } - DRM_DEBUG_KMS("bpp = %d\n", fb->bits_per_pixel); - /* - * In case of exynos, setting dma-burst to 16Word causes permanent - * tearing for very small buffers, e.g. cursor buffer. Burst Mode - * switching which is based on plane size is not recommended as - * plane size varies alot towards the end of the screen and rapid - * movement causes unstable DMA which results into iommu crash/tear. + * Setting dma-burst to 16Word causes permanent tearing for very small + * buffers, e.g. cursor buffer. Burst Mode switching which based on + * plane size is not recommended as plane size varies alot towards the + * end of the screen and rapid movement causes unstable DMA, but it is + * still better to change dma-burst than displaying garbage. */ - if (fb->width < MIN_FB_WIDTH_FOR_16WORD_BURST) { + if (width < MIN_FB_WIDTH_FOR_16WORD_BURST) { val &= ~WINCONx_BURSTLEN_MASK; val |= WINCONx_BURSTLEN_4WORD; } @@ -640,39 +647,41 @@ static void fimd_atomic_flush(struct exynos_drm_crtc *crtc, static void fimd_update_plane(struct exynos_drm_crtc *crtc, struct exynos_drm_plane *plane) { + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane->base.state); struct fimd_context *ctx = crtc->ctx; - struct drm_plane_state *state = plane->base.state; + struct drm_framebuffer *fb = state->base.fb; dma_addr_t dma_addr; unsigned long val, size, offset; unsigned int last_x, last_y, buf_offsize, line_size; unsigned int win = plane->zpos; - unsigned int bpp = state->fb->bits_per_pixel >> 3; - unsigned int pitch = state->fb->pitches[0]; + unsigned int bpp = fb->bits_per_pixel >> 3; + unsigned int pitch = fb->pitches[0]; if (ctx->suspended) return; - offset = plane->src_x * bpp; - offset += plane->src_y * pitch; + offset = state->src.x * bpp; + offset += state->src.y * pitch; /* buffer start address */ - dma_addr = plane->dma_addr[0] + offset; + dma_addr = exynos_drm_fb_dma_addr(fb, 0) + offset; val = (unsigned long)dma_addr; writel(val, ctx->regs + VIDWx_BUF_START(win, 0)); /* buffer end address */ - size = pitch * plane->crtc_h; + size = pitch * state->crtc.h; val = (unsigned long)(dma_addr + size); writel(val, ctx->regs + VIDWx_BUF_END(win, 0)); DRM_DEBUG_KMS("start addr = 0x%lx, end addr = 0x%lx, size = 0x%lx\n", (unsigned long)dma_addr, val, size); DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n", - plane->crtc_w, plane->crtc_h); + state->crtc.w, state->crtc.h); /* buffer size */ - buf_offsize = pitch - (plane->crtc_w * bpp); - line_size = plane->crtc_w * bpp; + buf_offsize = pitch - (state->crtc.w * bpp); + line_size = state->crtc.w * bpp; val = VIDW_BUF_SIZE_OFFSET(buf_offsize) | VIDW_BUF_SIZE_PAGEWIDTH(line_size) | VIDW_BUF_SIZE_OFFSET_E(buf_offsize) | @@ -680,16 +689,16 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc, writel(val, ctx->regs + VIDWx_BUF_SIZE(win, 0)); /* OSD position */ - val = VIDOSDxA_TOPLEFT_X(plane->crtc_x) | - VIDOSDxA_TOPLEFT_Y(plane->crtc_y) | - VIDOSDxA_TOPLEFT_X_E(plane->crtc_x) | - VIDOSDxA_TOPLEFT_Y_E(plane->crtc_y); + val = VIDOSDxA_TOPLEFT_X(state->crtc.x) | + VIDOSDxA_TOPLEFT_Y(state->crtc.y) | + VIDOSDxA_TOPLEFT_X_E(state->crtc.x) | + VIDOSDxA_TOPLEFT_Y_E(state->crtc.y); writel(val, ctx->regs + VIDOSD_A(win)); - last_x = plane->crtc_x + plane->crtc_w; + last_x = state->crtc.x + state->crtc.w; if (last_x) last_x--; - last_y = plane->crtc_y + plane->crtc_h; + last_y = state->crtc.y + state->crtc.h; if (last_y) last_y--; @@ -699,20 +708,20 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc, writel(val, ctx->regs + VIDOSD_B(win)); DRM_DEBUG_KMS("osd pos: tx = %d, ty = %d, bx = %d, by = %d\n", - plane->crtc_x, plane->crtc_y, last_x, last_y); + state->crtc.x, state->crtc.y, last_x, last_y); /* OSD size */ if (win != 3 && win != 4) { u32 offset = VIDOSD_D(win); if (win == 0) offset = VIDOSD_C(win); - val = plane->crtc_w * plane->crtc_h; + val = state->crtc.w * state->crtc.h; writel(val, ctx->regs + offset); DRM_DEBUG_KMS("osd size = 0x%x\n", (unsigned int)val); } - fimd_win_set_pixfmt(ctx, win, state->fb); + fimd_win_set_pixfmt(ctx, win, fb->pixel_format, state->src.w); /* hardware window 0 doesn't support color key. */ if (win != 0) @@ -745,7 +754,6 @@ static void fimd_disable_plane(struct exynos_drm_crtc *crtc, static void fimd_enable(struct exynos_drm_crtc *crtc) { struct fimd_context *ctx = crtc->ctx; - int ret; if (!ctx->suspended) return; @@ -754,18 +762,6 @@ static void fimd_enable(struct exynos_drm_crtc *crtc) pm_runtime_get_sync(ctx->dev); - ret = clk_prepare_enable(ctx->bus_clk); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the bus clk [%d]\n", ret); - return; - } - - ret = clk_prepare_enable(ctx->lcd_clk); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the lcd clk [%d]\n", ret); - return; - } - /* if vblank was enabled status, enable it again. */ if (test_and_clear_bit(0, &ctx->irq_flags)) fimd_enable_vblank(ctx->crtc); @@ -795,11 +791,7 @@ static void fimd_disable(struct exynos_drm_crtc *crtc) writel(0, ctx->regs + VIDCON0); - clk_disable_unprepare(ctx->lcd_clk); - clk_disable_unprepare(ctx->bus_clk); - pm_runtime_put_sync(ctx->dev); - ctx->suspended = true; } @@ -941,18 +933,19 @@ static int fimd_bind(struct device *dev, struct device *master, void *data) struct drm_device *drm_dev = data; struct exynos_drm_private *priv = drm_dev->dev_private; struct exynos_drm_plane *exynos_plane; - enum drm_plane_type type; - unsigned int zpos; + unsigned int i; int ret; ctx->drm_dev = drm_dev; ctx->pipe = priv->pipe++; - for (zpos = 0; zpos < WINDOWS_NR; zpos++) { - type = exynos_plane_get_type(zpos, CURSOR_WIN); - ret = exynos_plane_init(drm_dev, &ctx->planes[zpos], - 1 << ctx->pipe, type, fimd_formats, - ARRAY_SIZE(fimd_formats), zpos); + for (i = 0; i < WINDOWS_NR; i++) { + ctx->configs[i].pixel_formats = fimd_formats; + ctx->configs[i].num_pixel_formats = ARRAY_SIZE(fimd_formats); + ctx->configs[i].zpos = i; + ctx->configs[i].type = fimd_win_types[i]; + ret = exynos_plane_init(drm_dev, &ctx->planes[i], + 1 << ctx->pipe, &ctx->configs[i]); if (ret) return ret; } @@ -1121,12 +1114,49 @@ static int fimd_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM +static int exynos_fimd_suspend(struct device *dev) +{ + struct fimd_context *ctx = dev_get_drvdata(dev); + + clk_disable_unprepare(ctx->lcd_clk); + clk_disable_unprepare(ctx->bus_clk); + + return 0; +} + +static int exynos_fimd_resume(struct device *dev) +{ + struct fimd_context *ctx = dev_get_drvdata(dev); + int ret; + + ret = clk_prepare_enable(ctx->bus_clk); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the bus clk [%d]\n", ret); + return ret; + } + + ret = clk_prepare_enable(ctx->lcd_clk); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the lcd clk [%d]\n", ret); + return ret; + } + + return 0; +} +#endif + +static const struct dev_pm_ops exynos_fimd_pm_ops = { + SET_RUNTIME_PM_OPS(exynos_fimd_suspend, exynos_fimd_resume, NULL) +}; + struct platform_driver fimd_driver = { .probe = fimd_probe, .remove = fimd_remove, .driver = { .name = "exynos4-fb", .owner = THIS_MODULE, + .pm = &exynos_fimd_pm_ops, .of_match_table = fimd_driver_dt_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index 37ab8b282db6..9ca5047959ec 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -55,8 +55,6 @@ struct exynos_drm_gem { struct sg_table *sgt; }; -struct page **exynos_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask); - /* destroy a buffer with gem object */ void exynos_drm_gem_destroy(struct exynos_drm_gem *exynos_gem); @@ -91,10 +89,6 @@ void exynos_drm_gem_put_dma_addr(struct drm_device *dev, unsigned int gem_handle, struct drm_file *filp); -/* map user space allocated by malloc to pages. */ -int exynos_drm_gem_userptr_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv); - /* get buffer information to memory region allocated by gem. */ int exynos_drm_gem_get_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); @@ -123,28 +117,6 @@ int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf); /* set vm_flags and we can change the vm attribute to other one at here. */ int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma); -static inline int vma_is_io(struct vm_area_struct *vma) -{ - return !!(vma->vm_flags & (VM_IO | VM_PFNMAP)); -} - -/* get a copy of a virtual memory region. */ -struct vm_area_struct *exynos_gem_get_vma(struct vm_area_struct *vma); - -/* release a userspace virtual memory area. */ -void exynos_gem_put_vma(struct vm_area_struct *vma); - -/* get pages from user space. */ -int exynos_gem_get_pages_from_userptr(unsigned long start, - unsigned int npages, - struct page **pages, - struct vm_area_struct *vma); - -/* drop the reference to pages. */ -void exynos_gem_put_pages_to_userptr(struct page **pages, - unsigned int npages, - struct vm_area_struct *vma); - /* map sgt with dma region. */ int exynos_gem_map_sgt_with_dma(struct drm_device *drm_dev, struct sg_table *sgt, diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 11b87d2a7913..7aecd23cfa11 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -15,7 +15,8 @@ #include <linux/platform_device.h> #include <linux/clk.h> #include <linux/pm_runtime.h> -#include <plat/map-base.h> +#include <linux/mfd/syscon.h> +#include <linux/regmap.h> #include <drm/drmP.h> #include <drm/exynos_drm.h> @@ -126,6 +127,7 @@ struct gsc_capability { * @ippdrv: prepare initialization using ippdrv. * @regs_res: register resources. * @regs: memory mapped io registers. + * @sysreg: handle to SYSREG block regmap. * @lock: locking of operations. * @gsc_clk: gsc gate clock. * @sc: scaler infomations. @@ -138,6 +140,7 @@ struct gsc_context { struct exynos_drm_ippdrv ippdrv; struct resource *regs_res; void __iomem *regs; + struct regmap *sysreg; struct mutex lock; struct clk *gsc_clk; struct gsc_scaler sc; @@ -437,9 +440,12 @@ static int gsc_sw_reset(struct gsc_context *ctx) static void gsc_set_gscblk_fimd_wb(struct gsc_context *ctx, bool enable) { - u32 gscblk_cfg; + unsigned int gscblk_cfg; - gscblk_cfg = readl(SYSREG_GSCBLK_CFG1); + if (!ctx->sysreg) + return; + + regmap_read(ctx->sysreg, SYSREG_GSCBLK_CFG1, &gscblk_cfg); if (enable) gscblk_cfg |= GSC_BLK_DISP1WB_DEST(ctx->id) | @@ -448,7 +454,7 @@ static void gsc_set_gscblk_fimd_wb(struct gsc_context *ctx, bool enable) else gscblk_cfg |= GSC_BLK_PXLASYNC_LO_MASK_WB(ctx->id); - writel(gscblk_cfg, SYSREG_GSCBLK_CFG1); + regmap_write(ctx->sysreg, SYSREG_GSCBLK_CFG1, gscblk_cfg); } static void gsc_handle_irq(struct gsc_context *ctx, bool enable, @@ -1215,10 +1221,10 @@ static int gsc_clk_ctrl(struct gsc_context *ctx, bool enable) DRM_DEBUG_KMS("enable[%d]\n", enable); if (enable) { - clk_enable(ctx->gsc_clk); + clk_prepare_enable(ctx->gsc_clk); ctx->suspended = false; } else { - clk_disable(ctx->gsc_clk); + clk_disable_unprepare(ctx->gsc_clk); ctx->suspended = true; } @@ -1663,6 +1669,15 @@ static int gsc_probe(struct platform_device *pdev) if (!ctx) return -ENOMEM; + if (dev->of_node) { + ctx->sysreg = syscon_regmap_lookup_by_phandle(dev->of_node, + "samsung,sysreg"); + if (IS_ERR(ctx->sysreg)) { + dev_warn(dev, "failed to get system register.\n"); + ctx->sysreg = NULL; + } + } + /* clock control */ ctx->gsc_clk = devm_clk_get(dev, "gscl"); if (IS_ERR(ctx->gsc_clk)) { @@ -1713,7 +1728,6 @@ static int gsc_probe(struct platform_device *pdev) mutex_init(&ctx->lock); platform_set_drvdata(pdev, ctx); - pm_runtime_set_active(dev); pm_runtime_enable(dev); ret = exynos_drm_ippdrv_register(ippdrv); @@ -1797,6 +1811,12 @@ static const struct dev_pm_ops gsc_pm_ops = { SET_RUNTIME_PM_OPS(gsc_runtime_suspend, gsc_runtime_resume, NULL) }; +static const struct of_device_id exynos_drm_gsc_of_match[] = { + { .compatible = "samsung,exynos5-gsc" }, + { }, +}; +MODULE_DEVICE_TABLE(of, exynos_drm_gsc_of_match); + struct platform_driver gsc_driver = { .probe = gsc_probe, .remove = gsc_remove, @@ -1804,6 +1824,7 @@ struct platform_driver gsc_driver = { .name = "exynos-drm-gsc", .owner = THIS_MODULE, .pm = &gsc_pm_ops, + .of_match_table = of_match_ptr(exynos_drm_gsc_of_match), }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c index 179311760bb7..e668fcdbcafc 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_plane.c +++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c @@ -56,93 +56,170 @@ static int exynos_plane_get_size(int start, unsigned length, unsigned last) return size; } -static void exynos_plane_mode_set(struct drm_plane *plane, - struct drm_crtc *crtc, - struct drm_framebuffer *fb, - int crtc_x, int crtc_y, - unsigned int crtc_w, unsigned int crtc_h, - uint32_t src_x, uint32_t src_y, - uint32_t src_w, uint32_t src_h) +static void exynos_plane_mode_set(struct exynos_drm_plane_state *exynos_state) + { - struct exynos_drm_plane *exynos_plane = to_exynos_plane(plane); + struct drm_plane_state *state = &exynos_state->base; + struct drm_crtc *crtc = exynos_state->base.crtc; struct drm_display_mode *mode = &crtc->state->adjusted_mode; + int crtc_x, crtc_y; + unsigned int crtc_w, crtc_h; + unsigned int src_x, src_y; + unsigned int src_w, src_h; unsigned int actual_w; unsigned int actual_h; + /* + * The original src/dest coordinates are stored in exynos_state->base, + * but we want to keep another copy internal to our driver that we can + * clip/modify ourselves. + */ + + crtc_x = state->crtc_x; + crtc_y = state->crtc_y; + crtc_w = state->crtc_w; + crtc_h = state->crtc_h; + + src_x = state->src_x >> 16; + src_y = state->src_y >> 16; + src_w = state->src_w >> 16; + src_h = state->src_h >> 16; + + /* set ratio */ + exynos_state->h_ratio = (src_w << 16) / crtc_w; + exynos_state->v_ratio = (src_h << 16) / crtc_h; + + /* clip to visible area */ actual_w = exynos_plane_get_size(crtc_x, crtc_w, mode->hdisplay); actual_h = exynos_plane_get_size(crtc_y, crtc_h, mode->vdisplay); if (crtc_x < 0) { if (actual_w) - src_x -= crtc_x; + src_x += ((-crtc_x) * exynos_state->h_ratio) >> 16; crtc_x = 0; } if (crtc_y < 0) { if (actual_h) - src_y -= crtc_y; + src_y += ((-crtc_y) * exynos_state->v_ratio) >> 16; crtc_y = 0; } - /* set ratio */ - exynos_plane->h_ratio = (src_w << 16) / crtc_w; - exynos_plane->v_ratio = (src_h << 16) / crtc_h; - /* set drm framebuffer data. */ - exynos_plane->src_x = src_x; - exynos_plane->src_y = src_y; - exynos_plane->src_w = (actual_w * exynos_plane->h_ratio) >> 16; - exynos_plane->src_h = (actual_h * exynos_plane->v_ratio) >> 16; + exynos_state->src.x = src_x; + exynos_state->src.y = src_y; + exynos_state->src.w = (actual_w * exynos_state->h_ratio) >> 16; + exynos_state->src.h = (actual_h * exynos_state->v_ratio) >> 16; /* set plane range to be displayed. */ - exynos_plane->crtc_x = crtc_x; - exynos_plane->crtc_y = crtc_y; - exynos_plane->crtc_w = actual_w; - exynos_plane->crtc_h = actual_h; + exynos_state->crtc.x = crtc_x; + exynos_state->crtc.y = crtc_y; + exynos_state->crtc.w = actual_w; + exynos_state->crtc.h = actual_h; DRM_DEBUG_KMS("plane : offset_x/y(%d,%d), width/height(%d,%d)", - exynos_plane->crtc_x, exynos_plane->crtc_y, - exynos_plane->crtc_w, exynos_plane->crtc_h); + exynos_state->crtc.x, exynos_state->crtc.y, + exynos_state->crtc.w, exynos_state->crtc.h); +} + +static void exynos_drm_plane_reset(struct drm_plane *plane) +{ + struct exynos_drm_plane_state *exynos_state; + + if (plane->state) { + exynos_state = to_exynos_plane_state(plane->state); + if (exynos_state->base.fb) + drm_framebuffer_unreference(exynos_state->base.fb); + kfree(exynos_state); + plane->state = NULL; + } + + exynos_state = kzalloc(sizeof(*exynos_state), GFP_KERNEL); + if (exynos_state) { + plane->state = &exynos_state->base; + plane->state->plane = plane; + } +} + +static struct drm_plane_state * +exynos_drm_plane_duplicate_state(struct drm_plane *plane) +{ + struct exynos_drm_plane_state *exynos_state; + struct exynos_drm_plane_state *copy; + + exynos_state = to_exynos_plane_state(plane->state); + copy = kzalloc(sizeof(*exynos_state), GFP_KERNEL); + if (!copy) + return NULL; + + __drm_atomic_helper_plane_duplicate_state(plane, ©->base); + return ©->base; +} - plane->crtc = crtc; +static void exynos_drm_plane_destroy_state(struct drm_plane *plane, + struct drm_plane_state *old_state) +{ + struct exynos_drm_plane_state *old_exynos_state = + to_exynos_plane_state(old_state); + __drm_atomic_helper_plane_destroy_state(plane, old_state); + kfree(old_exynos_state); } static struct drm_plane_funcs exynos_plane_funcs = { .update_plane = drm_atomic_helper_update_plane, .disable_plane = drm_atomic_helper_disable_plane, .destroy = drm_plane_cleanup, - .reset = drm_atomic_helper_plane_reset, - .atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state, - .atomic_destroy_state = drm_atomic_helper_plane_destroy_state, + .reset = exynos_drm_plane_reset, + .atomic_duplicate_state = exynos_drm_plane_duplicate_state, + .atomic_destroy_state = exynos_drm_plane_destroy_state, }; +static int +exynos_drm_plane_check_size(const struct exynos_drm_plane_config *config, + struct exynos_drm_plane_state *state) +{ + bool width_ok = false, height_ok = false; + + if (config->capabilities & EXYNOS_DRM_PLANE_CAP_SCALE) + return 0; + + if (state->src.w == state->crtc.w) + width_ok = true; + + if (state->src.h == state->crtc.h) + height_ok = true; + + if ((config->capabilities & EXYNOS_DRM_PLANE_CAP_DOUBLE) && + state->h_ratio == (1 << 15)) + width_ok = true; + + if ((config->capabilities & EXYNOS_DRM_PLANE_CAP_DOUBLE) && + state->v_ratio == (1 << 15)) + height_ok = true; + + if (width_ok & height_ok) + return 0; + + DRM_DEBUG_KMS("scaling mode is not supported"); + return -ENOTSUPP; +} + static int exynos_plane_atomic_check(struct drm_plane *plane, struct drm_plane_state *state) { struct exynos_drm_plane *exynos_plane = to_exynos_plane(plane); - int nr; - int i; + struct exynos_drm_plane_state *exynos_state = + to_exynos_plane_state(state); + int ret = 0; - if (!state->fb) + if (!state->crtc || !state->fb) return 0; - nr = drm_format_num_planes(state->fb->pixel_format); - for (i = 0; i < nr; i++) { - struct exynos_drm_gem *exynos_gem = - exynos_drm_fb_gem(state->fb, i); - if (!exynos_gem) { - DRM_DEBUG_KMS("gem object is null\n"); - return -EFAULT; - } - - exynos_plane->dma_addr[i] = exynos_gem->dma_addr + - state->fb->offsets[i]; - - DRM_DEBUG_KMS("buffer: %d, dma_addr = 0x%lx\n", - i, (unsigned long)exynos_plane->dma_addr[i]); - } + /* translate state into exynos_state */ + exynos_plane_mode_set(exynos_state); - return 0; + ret = exynos_drm_plane_check_size(exynos_plane->config, exynos_state); + return ret; } static void exynos_plane_atomic_update(struct drm_plane *plane, @@ -155,12 +232,7 @@ static void exynos_plane_atomic_update(struct drm_plane *plane, if (!state->crtc) return; - exynos_plane_mode_set(plane, state->crtc, state->fb, - state->crtc_x, state->crtc_y, - state->crtc_w, state->crtc_h, - state->src_x >> 16, state->src_y >> 16, - state->src_w >> 16, state->src_h >> 16); - + plane->crtc = state->crtc; exynos_plane->pending_fb = state->fb; if (exynos_crtc->ops->update_plane) @@ -177,8 +249,7 @@ static void exynos_plane_atomic_disable(struct drm_plane *plane, return; if (exynos_crtc->ops->disable_plane) - exynos_crtc->ops->disable_plane(exynos_crtc, - exynos_plane); + exynos_crtc->ops->disable_plane(exynos_crtc, exynos_plane); } static const struct drm_plane_helper_funcs plane_helper_funcs = { @@ -207,28 +278,19 @@ static void exynos_plane_attach_zpos_property(struct drm_plane *plane, drm_object_attach_property(&plane->base, prop, zpos); } -enum drm_plane_type exynos_plane_get_type(unsigned int zpos, - unsigned int cursor_win) -{ - if (zpos == DEFAULT_WIN) - return DRM_PLANE_TYPE_PRIMARY; - else if (zpos == cursor_win) - return DRM_PLANE_TYPE_CURSOR; - else - return DRM_PLANE_TYPE_OVERLAY; -} - int exynos_plane_init(struct drm_device *dev, struct exynos_drm_plane *exynos_plane, - unsigned long possible_crtcs, enum drm_plane_type type, - const uint32_t *formats, unsigned int fcount, - unsigned int zpos) + unsigned long possible_crtcs, + const struct exynos_drm_plane_config *config) { int err; - err = drm_universal_plane_init(dev, &exynos_plane->base, possible_crtcs, - &exynos_plane_funcs, formats, fcount, - type); + err = drm_universal_plane_init(dev, &exynos_plane->base, + possible_crtcs, + &exynos_plane_funcs, + config->pixel_formats, + config->num_pixel_formats, + config->type, NULL); if (err) { DRM_ERROR("failed to initialize plane\n"); return err; @@ -236,10 +298,12 @@ int exynos_plane_init(struct drm_device *dev, drm_plane_helper_add(&exynos_plane->base, &plane_helper_funcs); - exynos_plane->zpos = zpos; + exynos_plane->zpos = config->zpos; + exynos_plane->config = config; - if (type == DRM_PLANE_TYPE_OVERLAY) - exynos_plane_attach_zpos_property(&exynos_plane->base, zpos); + if (config->type == DRM_PLANE_TYPE_OVERLAY) + exynos_plane_attach_zpos_property(&exynos_plane->base, + config->zpos); return 0; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.h b/drivers/gpu/drm/exynos/exynos_drm_plane.h index abb641e64c23..0dd096548284 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_plane.h +++ b/drivers/gpu/drm/exynos/exynos_drm_plane.h @@ -9,10 +9,7 @@ * */ -enum drm_plane_type exynos_plane_get_type(unsigned int zpos, - unsigned int cursor_win); int exynos_plane_init(struct drm_device *dev, struct exynos_drm_plane *exynos_plane, - unsigned long possible_crtcs, enum drm_plane_type type, - const uint32_t *formats, unsigned int fcount, - unsigned int zpos); + unsigned long possible_crtcs, + const struct exynos_drm_plane_config *config); diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index 2f5c118f4c8e..bea0f7826d30 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -790,10 +790,10 @@ static int rotator_remove(struct platform_device *pdev) static int rotator_clk_crtl(struct rot_context *rot, bool enable) { if (enable) { - clk_enable(rot->clock); + clk_prepare_enable(rot->clock); rot->suspended = false; } else { - clk_disable(rot->clock); + clk_disable_unprepare(rot->clock); rot->suspended = true; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c index 669362c53f49..319aa31954d1 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c @@ -24,12 +24,12 @@ #include "exynos_drm_drv.h" #include "exynos_drm_crtc.h" +#include "exynos_drm_fb.h" #include "exynos_drm_plane.h" #include "exynos_drm_vidi.h" /* vidi has totally three virtual windows. */ #define WINDOWS_NR 3 -#define CURSOR_WIN 2 #define ctx_from_connector(c) container_of(c, struct vidi_context, \ connector) @@ -89,6 +89,12 @@ static const uint32_t formats[] = { DRM_FORMAT_NV12, }; +static const enum drm_plane_type vidi_win_types[WINDOWS_NR] = { + DRM_PLANE_TYPE_PRIMARY, + DRM_PLANE_TYPE_OVERLAY, + DRM_PLANE_TYPE_CURSOR, +}; + static int vidi_enable_vblank(struct exynos_drm_crtc *crtc) { struct vidi_context *ctx = crtc->ctx; @@ -125,12 +131,15 @@ static void vidi_disable_vblank(struct exynos_drm_crtc *crtc) static void vidi_update_plane(struct exynos_drm_crtc *crtc, struct exynos_drm_plane *plane) { + struct drm_plane_state *state = plane->base.state; struct vidi_context *ctx = crtc->ctx; + dma_addr_t addr; if (ctx->suspended) return; - DRM_DEBUG_KMS("dma_addr = %pad\n", plane->dma_addr); + addr = exynos_drm_fb_dma_addr(state->fb, 0); + DRM_DEBUG_KMS("dma_addr = %pad\n", &addr); if (ctx->vblank_on) schedule_work(&ctx->work); @@ -439,17 +448,21 @@ static int vidi_bind(struct device *dev, struct device *master, void *data) struct drm_device *drm_dev = data; struct drm_encoder *encoder = &ctx->encoder; struct exynos_drm_plane *exynos_plane; - enum drm_plane_type type; - unsigned int zpos; + struct exynos_drm_plane_config plane_config = { 0 }; + unsigned int i; int pipe, ret; vidi_ctx_initialize(ctx, drm_dev); - for (zpos = 0; zpos < WINDOWS_NR; zpos++) { - type = exynos_plane_get_type(zpos, CURSOR_WIN); - ret = exynos_plane_init(drm_dev, &ctx->planes[zpos], - 1 << ctx->pipe, type, formats, - ARRAY_SIZE(formats), zpos); + plane_config.pixel_formats = formats; + plane_config.num_pixel_formats = ARRAY_SIZE(formats); + + for (i = 0; i < WINDOWS_NR; i++) { + plane_config.zpos = i; + plane_config.type = vidi_win_types[i]; + + ret = exynos_plane_init(drm_dev, &ctx->planes[i], + 1 << ctx->pipe, &plane_config); if (ret) return ret; } @@ -473,7 +486,7 @@ static int vidi_bind(struct device *dev, struct device *master, void *data) DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs); drm_encoder_init(drm_dev, encoder, &exynos_vidi_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &exynos_vidi_encoder_helper_funcs); diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 57b675563e94..7d5ca6ca4efe 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -113,7 +113,7 @@ struct hdmi_context { void __iomem *regs_hdmiphy; struct i2c_client *hdmiphy_port; struct i2c_adapter *ddc_adpt; - struct gpio_desc *hpd_gpio; + struct gpio_desc *hpd_gpio; int irq; struct regmap *pmureg; struct clk *hdmi; @@ -1588,8 +1588,6 @@ static void hdmi_enable(struct drm_encoder *encoder) if (hdata->powered) return; - hdata->powered = true; - pm_runtime_get_sync(hdata->dev); if (regulator_bulk_enable(ARRAY_SIZE(supply), hdata->regul_bulk)) @@ -1599,10 +1597,9 @@ static void hdmi_enable(struct drm_encoder *encoder) regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL, PMU_HDMI_PHY_ENABLE_BIT, 1); - clk_prepare_enable(hdata->hdmi); - clk_prepare_enable(hdata->sclk_hdmi); - hdmi_conf_apply(hdata); + + hdata->powered = true; } static void hdmi_disable(struct drm_encoder *encoder) @@ -1633,9 +1630,6 @@ static void hdmi_disable(struct drm_encoder *encoder) cancel_delayed_work(&hdata->hotplug_work); - clk_disable_unprepare(hdata->sclk_hdmi); - clk_disable_unprepare(hdata->hdmi); - /* reset pmu hdmiphy control bit to disable hdmiphy */ regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL, PMU_HDMI_PHY_ENABLE_BIT, 0); @@ -1793,7 +1787,7 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data) DRM_DEBUG_KMS("possible_crtcs = 0x%x\n", encoder->possible_crtcs); drm_encoder_init(drm_dev, encoder, &exynos_hdmi_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &exynos_hdmi_encoder_helper_funcs); @@ -1978,12 +1972,49 @@ static int hdmi_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM +static int exynos_hdmi_suspend(struct device *dev) +{ + struct hdmi_context *hdata = dev_get_drvdata(dev); + + clk_disable_unprepare(hdata->sclk_hdmi); + clk_disable_unprepare(hdata->hdmi); + + return 0; +} + +static int exynos_hdmi_resume(struct device *dev) +{ + struct hdmi_context *hdata = dev_get_drvdata(dev); + int ret; + + ret = clk_prepare_enable(hdata->hdmi); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the hdmi clk [%d]\n", ret); + return ret; + } + ret = clk_prepare_enable(hdata->sclk_hdmi); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the sclk_mixer clk [%d]\n", + ret); + return ret; + } + + return 0; +} +#endif + +static const struct dev_pm_ops exynos_hdmi_pm_ops = { + SET_RUNTIME_PM_OPS(exynos_hdmi_suspend, exynos_hdmi_resume, NULL) +}; + struct platform_driver hdmi_driver = { .probe = hdmi_probe, .remove = hdmi_remove, .driver = { .name = "exynos-hdmi", .owner = THIS_MODULE, + .pm = &exynos_hdmi_pm_ops, .of_match_table = hdmi_match_types, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index d09f8f9a8939..dfb35e2da4db 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -37,12 +37,12 @@ #include "exynos_drm_drv.h" #include "exynos_drm_crtc.h" +#include "exynos_drm_fb.h" #include "exynos_drm_plane.h" #include "exynos_drm_iommu.h" #define MIXER_WIN_NR 3 #define VP_DEFAULT_WIN 2 -#define CURSOR_WIN 1 /* The pixelformats that are natively supported by the mixer. */ #define MXR_FORMAT_RGB565 4 @@ -111,6 +111,28 @@ struct mixer_drv_data { bool has_sclk; }; +static const struct exynos_drm_plane_config plane_configs[MIXER_WIN_NR] = { + { + .zpos = 0, + .type = DRM_PLANE_TYPE_PRIMARY, + .pixel_formats = mixer_formats, + .num_pixel_formats = ARRAY_SIZE(mixer_formats), + .capabilities = EXYNOS_DRM_PLANE_CAP_DOUBLE, + }, { + .zpos = 1, + .type = DRM_PLANE_TYPE_CURSOR, + .pixel_formats = mixer_formats, + .num_pixel_formats = ARRAY_SIZE(mixer_formats), + .capabilities = EXYNOS_DRM_PLANE_CAP_DOUBLE, + }, { + .zpos = 2, + .type = DRM_PLANE_TYPE_OVERLAY, + .pixel_formats = vp_formats, + .num_pixel_formats = ARRAY_SIZE(vp_formats), + .capabilities = EXYNOS_DRM_PLANE_CAP_SCALE, + }, +}; + static const u8 filter_y_horiz_tap8[] = { 0, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 0, 0, 0, @@ -399,10 +421,11 @@ static void mixer_stop(struct mixer_context *ctx) static void vp_video_buffer(struct mixer_context *ctx, struct exynos_drm_plane *plane) { + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane->base.state); + struct drm_display_mode *mode = &state->base.crtc->state->adjusted_mode; struct mixer_resources *res = &ctx->mixer_res; - struct drm_plane_state *state = plane->base.state; - struct drm_framebuffer *fb = state->fb; - struct drm_display_mode *mode = &state->crtc->mode; + struct drm_framebuffer *fb = state->base.fb; unsigned long flags; dma_addr_t luma_addr[2], chroma_addr[2]; bool tiled_mode = false; @@ -422,8 +445,8 @@ static void vp_video_buffer(struct mixer_context *ctx, return; } - luma_addr[0] = plane->dma_addr[0]; - chroma_addr[0] = plane->dma_addr[1]; + luma_addr[0] = exynos_drm_fb_dma_addr(fb, 0); + chroma_addr[0] = exynos_drm_fb_dma_addr(fb, 1); if (mode->flags & DRM_MODE_FLAG_INTERLACE) { ctx->interlace = true; @@ -459,24 +482,24 @@ static void vp_video_buffer(struct mixer_context *ctx, vp_reg_write(res, VP_IMG_SIZE_C, VP_IMG_HSIZE(fb->pitches[0]) | VP_IMG_VSIZE(fb->height / 2)); - vp_reg_write(res, VP_SRC_WIDTH, plane->src_w); - vp_reg_write(res, VP_SRC_HEIGHT, plane->src_h); + vp_reg_write(res, VP_SRC_WIDTH, state->src.w); + vp_reg_write(res, VP_SRC_HEIGHT, state->src.h); vp_reg_write(res, VP_SRC_H_POSITION, - VP_SRC_H_POSITION_VAL(plane->src_x)); - vp_reg_write(res, VP_SRC_V_POSITION, plane->src_y); + VP_SRC_H_POSITION_VAL(state->src.x)); + vp_reg_write(res, VP_SRC_V_POSITION, state->src.y); - vp_reg_write(res, VP_DST_WIDTH, plane->crtc_w); - vp_reg_write(res, VP_DST_H_POSITION, plane->crtc_x); + vp_reg_write(res, VP_DST_WIDTH, state->crtc.w); + vp_reg_write(res, VP_DST_H_POSITION, state->crtc.x); if (ctx->interlace) { - vp_reg_write(res, VP_DST_HEIGHT, plane->crtc_h / 2); - vp_reg_write(res, VP_DST_V_POSITION, plane->crtc_y / 2); + vp_reg_write(res, VP_DST_HEIGHT, state->crtc.h / 2); + vp_reg_write(res, VP_DST_V_POSITION, state->crtc.y / 2); } else { - vp_reg_write(res, VP_DST_HEIGHT, plane->crtc_h); - vp_reg_write(res, VP_DST_V_POSITION, plane->crtc_y); + vp_reg_write(res, VP_DST_HEIGHT, state->crtc.h); + vp_reg_write(res, VP_DST_V_POSITION, state->crtc.y); } - vp_reg_write(res, VP_H_RATIO, plane->h_ratio); - vp_reg_write(res, VP_V_RATIO, plane->v_ratio); + vp_reg_write(res, VP_H_RATIO, state->h_ratio); + vp_reg_write(res, VP_V_RATIO, state->v_ratio); vp_reg_write(res, VP_ENDIAN_MODE, VP_ENDIAN_MODE_LITTLE); @@ -505,37 +528,14 @@ static void mixer_layer_update(struct mixer_context *ctx) mixer_reg_writemask(res, MXR_CFG, ~0, MXR_CFG_LAYER_UPDATE); } -static int mixer_setup_scale(const struct exynos_drm_plane *plane, - unsigned int *x_ratio, unsigned int *y_ratio) -{ - if (plane->crtc_w != plane->src_w) { - if (plane->crtc_w == 2 * plane->src_w) - *x_ratio = 1; - else - goto fail; - } - - if (plane->crtc_h != plane->src_h) { - if (plane->crtc_h == 2 * plane->src_h) - *y_ratio = 1; - else - goto fail; - } - - return 0; - -fail: - DRM_DEBUG_KMS("only 2x width/height scaling of plane supported\n"); - return -ENOTSUPP; -} - static void mixer_graph_buffer(struct mixer_context *ctx, struct exynos_drm_plane *plane) { + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane->base.state); + struct drm_display_mode *mode = &state->base.crtc->state->adjusted_mode; struct mixer_resources *res = &ctx->mixer_res; - struct drm_plane_state *state = plane->base.state; - struct drm_framebuffer *fb = state->fb; - struct drm_display_mode *mode = &state->crtc->mode; + struct drm_framebuffer *fb = state->base.fb; unsigned long flags; unsigned int win = plane->zpos; unsigned int x_ratio = 0, y_ratio = 0; @@ -567,17 +567,17 @@ static void mixer_graph_buffer(struct mixer_context *ctx, return; } - /* check if mixer supports requested scaling setup */ - if (mixer_setup_scale(plane, &x_ratio, &y_ratio)) - return; + /* ratio is already checked by common plane code */ + x_ratio = state->h_ratio == (1 << 15); + y_ratio = state->v_ratio == (1 << 15); - dst_x_offset = plane->crtc_x; - dst_y_offset = plane->crtc_y; + dst_x_offset = state->crtc.x; + dst_y_offset = state->crtc.y; /* converting dma address base and source offset */ - dma_addr = plane->dma_addr[0] - + (plane->src_x * fb->bits_per_pixel >> 3) - + (plane->src_y * fb->pitches[0]); + dma_addr = exynos_drm_fb_dma_addr(fb, 0) + + (state->src.x * fb->bits_per_pixel >> 3) + + (state->src.y * fb->pitches[0]); src_x_offset = 0; src_y_offset = 0; @@ -605,8 +605,8 @@ static void mixer_graph_buffer(struct mixer_context *ctx, mixer_reg_write(res, MXR_RESOLUTION, val); } - val = MXR_GRP_WH_WIDTH(plane->src_w); - val |= MXR_GRP_WH_HEIGHT(plane->src_h); + val = MXR_GRP_WH_WIDTH(state->src.w); + val |= MXR_GRP_WH_HEIGHT(state->src.h); val |= MXR_GRP_WH_H_SCALE(x_ratio); val |= MXR_GRP_WH_V_SCALE(y_ratio); mixer_reg_write(res, MXR_GRAPHIC_WH(win), val); @@ -1020,43 +1020,12 @@ static void mixer_enable(struct exynos_drm_crtc *crtc) { struct mixer_context *ctx = crtc->ctx; struct mixer_resources *res = &ctx->mixer_res; - int ret; if (test_bit(MXR_BIT_POWERED, &ctx->flags)) return; pm_runtime_get_sync(ctx->dev); - ret = clk_prepare_enable(res->mixer); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the mixer clk [%d]\n", ret); - return; - } - ret = clk_prepare_enable(res->hdmi); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the hdmi clk [%d]\n", ret); - return; - } - if (ctx->vp_enabled) { - ret = clk_prepare_enable(res->vp); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the vp clk [%d]\n", - ret); - return; - } - if (ctx->has_sclk) { - ret = clk_prepare_enable(res->sclk_mixer); - if (ret < 0) { - DRM_ERROR("Failed to prepare_enable the " \ - "sclk_mixer clk [%d]\n", - ret); - return; - } - } - } - - set_bit(MXR_BIT_POWERED, &ctx->flags); - mixer_reg_writemask(res, MXR_STATUS, ~0, MXR_STATUS_SOFT_RESET); if (test_bit(MXR_BIT_VSYNC, &ctx->flags)) { @@ -1064,12 +1033,13 @@ static void mixer_enable(struct exynos_drm_crtc *crtc) mixer_reg_writemask(res, MXR_INT_EN, ~0, MXR_INT_EN_VSYNC); } mixer_win_reset(ctx); + + set_bit(MXR_BIT_POWERED, &ctx->flags); } static void mixer_disable(struct exynos_drm_crtc *crtc) { struct mixer_context *ctx = crtc->ctx; - struct mixer_resources *res = &ctx->mixer_res; int i; if (!test_bit(MXR_BIT_POWERED, &ctx->flags)) @@ -1081,17 +1051,9 @@ static void mixer_disable(struct exynos_drm_crtc *crtc) for (i = 0; i < MIXER_WIN_NR; i++) mixer_disable_plane(crtc, &ctx->planes[i]); - clear_bit(MXR_BIT_POWERED, &ctx->flags); + pm_runtime_put(ctx->dev); - clk_disable_unprepare(res->hdmi); - clk_disable_unprepare(res->mixer); - if (ctx->vp_enabled) { - clk_disable_unprepare(res->vp); - if (ctx->has_sclk) - clk_disable_unprepare(res->sclk_mixer); - } - - pm_runtime_put_sync(ctx->dev); + clear_bit(MXR_BIT_POWERED, &ctx->flags); } /* Only valid for Mixer version 16.0.33.0 */ @@ -1187,30 +1149,19 @@ static int mixer_bind(struct device *dev, struct device *manager, void *data) struct mixer_context *ctx = dev_get_drvdata(dev); struct drm_device *drm_dev = data; struct exynos_drm_plane *exynos_plane; - unsigned int zpos; + unsigned int i; int ret; ret = mixer_initialize(ctx, drm_dev); if (ret) return ret; - for (zpos = 0; zpos < MIXER_WIN_NR; zpos++) { - enum drm_plane_type type; - const uint32_t *formats; - unsigned int fcount; - - if (zpos < VP_DEFAULT_WIN) { - formats = mixer_formats; - fcount = ARRAY_SIZE(mixer_formats); - } else { - formats = vp_formats; - fcount = ARRAY_SIZE(vp_formats); - } + for (i = 0; i < MIXER_WIN_NR; i++) { + if (i == VP_DEFAULT_WIN && !ctx->vp_enabled) + continue; - type = exynos_plane_get_type(zpos, CURSOR_WIN); - ret = exynos_plane_init(drm_dev, &ctx->planes[zpos], - 1 << ctx->pipe, type, formats, fcount, - zpos); + ret = exynos_plane_init(drm_dev, &ctx->planes[i], + 1 << ctx->pipe, &plane_configs[i]); if (ret) return ret; } @@ -1293,10 +1244,70 @@ static int mixer_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM_SLEEP +static int exynos_mixer_suspend(struct device *dev) +{ + struct mixer_context *ctx = dev_get_drvdata(dev); + struct mixer_resources *res = &ctx->mixer_res; + + clk_disable_unprepare(res->hdmi); + clk_disable_unprepare(res->mixer); + if (ctx->vp_enabled) { + clk_disable_unprepare(res->vp); + if (ctx->has_sclk) + clk_disable_unprepare(res->sclk_mixer); + } + + return 0; +} + +static int exynos_mixer_resume(struct device *dev) +{ + struct mixer_context *ctx = dev_get_drvdata(dev); + struct mixer_resources *res = &ctx->mixer_res; + int ret; + + ret = clk_prepare_enable(res->mixer); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the mixer clk [%d]\n", ret); + return ret; + } + ret = clk_prepare_enable(res->hdmi); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the hdmi clk [%d]\n", ret); + return ret; + } + if (ctx->vp_enabled) { + ret = clk_prepare_enable(res->vp); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the vp clk [%d]\n", + ret); + return ret; + } + if (ctx->has_sclk) { + ret = clk_prepare_enable(res->sclk_mixer); + if (ret < 0) { + DRM_ERROR("Failed to prepare_enable the " \ + "sclk_mixer clk [%d]\n", + ret); + return ret; + } + } + } + + return 0; +} +#endif + +static const struct dev_pm_ops exynos_mixer_pm_ops = { + SET_RUNTIME_PM_OPS(exynos_mixer_suspend, exynos_mixer_resume, NULL) +}; + struct platform_driver mixer_driver = { .driver = { .name = "exynos-mixer", .owner = THIS_MODULE, + .pm = &exynos_mixer_pm_ops, .of_match_table = mixer_match_types, }, .probe = mixer_probe, diff --git a/drivers/gpu/drm/exynos/regs-gsc.h b/drivers/gpu/drm/exynos/regs-gsc.h index 9ad592707aaf..4704a993cbb7 100644 --- a/drivers/gpu/drm/exynos/regs-gsc.h +++ b/drivers/gpu/drm/exynos/regs-gsc.h @@ -273,12 +273,12 @@ #define GSC_CLK_GATE_MODE_SNOOP_CNT(x) ((x) << 0) /* SYSCON. GSCBLK_CFG */ -#define SYSREG_GSCBLK_CFG1 (S3C_VA_SYS + 0x0224) +#define SYSREG_GSCBLK_CFG1 0x0224 #define GSC_BLK_DISP1WB_DEST(x) (x << 10) #define GSC_BLK_SW_RESET_WB_DEST(x) (1 << (18 + x)) #define GSC_BLK_PXLASYNC_LO_MASK_WB(x) (0 << (14 + x)) #define GSC_BLK_GSCL_WB_IN_SRC_SEL(x) (1 << (2 * x)) -#define SYSREG_GSCBLK_CFG2 (S3C_VA_SYS + 0x2000) +#define SYSREG_GSCBLK_CFG2 0x2000 #define PXLASYNC_LO_MASK_CAMIF_GSCL(x) (1 << (x)) #endif /* EXYNOS_REGS_GSC_H_ */ diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c index 82a3d311e164..d8ab8f0af10c 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c @@ -175,7 +175,7 @@ int fsl_dcu_drm_crtc_create(struct fsl_dcu_drm_device *fsl_dev) primary = fsl_dcu_drm_primary_create_plane(fsl_dev->drm); ret = drm_crtc_init_with_planes(fsl_dev->drm, crtc, primary, NULL, - &fsl_dcu_drm_crtc_funcs); + &fsl_dcu_drm_crtc_funcs, NULL); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c index 51daaea40b4d..4b13cf919575 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c @@ -249,7 +249,7 @@ struct drm_plane *fsl_dcu_drm_primary_create_plane(struct drm_device *dev) &fsl_dcu_drm_plane_funcs, fsl_dcu_drm_plane_formats, ARRAY_SIZE(fsl_dcu_drm_plane_formats), - DRM_PLANE_TYPE_PRIMARY); + DRM_PLANE_TYPE_PRIMARY, NULL); if (ret) { kfree(primary); primary = NULL; diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c index fe8ab5da04fb..8780deba5e8a 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c @@ -57,7 +57,7 @@ int fsl_dcu_drm_encoder_create(struct fsl_dcu_drm_device *fsl_dev, encoder->possible_crtcs = 1; ret = drm_encoder_init(fsl_dev->drm, encoder, &encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/gma500/cdv_device.c b/drivers/gpu/drm/gma500/cdv_device.c index 3531f90e53d0..8745971a7680 100644 --- a/drivers/gpu/drm/gma500/cdv_device.c +++ b/drivers/gpu/drm/gma500/cdv_device.c @@ -619,6 +619,8 @@ const struct psb_ops cdv_chip_ops = { .init_pm = cdv_init_pm, .save_regs = cdv_save_display_registers, .restore_regs = cdv_restore_display_registers, + .save_crtc = gma_crtc_save, + .restore_crtc = gma_crtc_restore, .power_down = cdv_power_down, .power_up = cdv_power_up, .update_wm = cdv_update_wm, diff --git a/drivers/gpu/drm/gma500/cdv_intel_crt.c b/drivers/gpu/drm/gma500/cdv_intel_crt.c index 248c33a35ebf..d0717a85c7ec 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_crt.c +++ b/drivers/gpu/drm/gma500/cdv_intel_crt.c @@ -273,7 +273,7 @@ void cdv_intel_crt_init(struct drm_device *dev, encoder = &gma_encoder->base; drm_encoder_init(dev, encoder, - &cdv_intel_crt_enc_funcs, DRM_MODE_ENCODER_DAC); + &cdv_intel_crt_enc_funcs, DRM_MODE_ENCODER_DAC, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); diff --git a/drivers/gpu/drm/gma500/cdv_intel_display.c b/drivers/gpu/drm/gma500/cdv_intel_display.c index 7d47b3d5cc0d..6126546295e9 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_display.c +++ b/drivers/gpu/drm/gma500/cdv_intel_display.c @@ -983,8 +983,6 @@ const struct drm_crtc_helper_funcs cdv_intel_helper_funcs = { }; const struct drm_crtc_funcs cdv_intel_crtc_funcs = { - .save = gma_crtc_save, - .restore = gma_crtc_restore, .cursor_set = gma_crtc_cursor_set, .cursor_move = gma_crtc_cursor_move, .gamma_set = gma_crtc_gamma_set, diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c index 17cea400ae32..7bb1f1aff932 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_dp.c +++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c @@ -2020,7 +2020,8 @@ cdv_intel_dp_init(struct drm_device *dev, struct psb_intel_mode_device *mode_dev encoder = &gma_encoder->base; drm_connector_init(dev, connector, &cdv_intel_dp_connector_funcs, type); - drm_encoder_init(dev, encoder, &cdv_intel_dp_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &cdv_intel_dp_enc_funcs, + DRM_MODE_ENCODER_TMDS, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); diff --git a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c index 6b1d3340ba14..ddf2d7700759 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c +++ b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c @@ -270,8 +270,6 @@ static const struct drm_connector_helper_funcs static const struct drm_connector_funcs cdv_hdmi_connector_funcs = { .dpms = drm_helper_connector_dpms, - .save = cdv_hdmi_save, - .restore = cdv_hdmi_restore, .detect = cdv_hdmi_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = cdv_hdmi_set_property, @@ -306,13 +304,16 @@ void cdv_hdmi_init(struct drm_device *dev, connector = &gma_connector->base; connector->polled = DRM_CONNECTOR_POLL_HPD; + gma_connector->save = cdv_hdmi_save; + gma_connector->restore = cdv_hdmi_restore; + encoder = &gma_encoder->base; drm_connector_init(dev, connector, &cdv_hdmi_connector_funcs, DRM_MODE_CONNECTOR_DVID); drm_encoder_init(dev, encoder, &psb_intel_lvds_enc_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); gma_encoder->type = INTEL_OUTPUT_HDMI; diff --git a/drivers/gpu/drm/gma500/cdv_intel_lvds.c b/drivers/gpu/drm/gma500/cdv_intel_lvds.c index 211069b2b951..813ef23a8054 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_lvds.c +++ b/drivers/gpu/drm/gma500/cdv_intel_lvds.c @@ -530,8 +530,6 @@ static const struct drm_connector_helper_funcs static const struct drm_connector_funcs cdv_intel_lvds_connector_funcs = { .dpms = drm_helper_connector_dpms, - .save = cdv_intel_lvds_save, - .restore = cdv_intel_lvds_restore, .detect = cdv_intel_lvds_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = cdv_intel_lvds_set_property, @@ -643,6 +641,8 @@ void cdv_intel_lvds_init(struct drm_device *dev, gma_encoder->dev_priv = lvds_priv; connector = &gma_connector->base; + gma_connector->save = cdv_intel_lvds_save; + gma_connector->restore = cdv_intel_lvds_restore; encoder = &gma_encoder->base; @@ -652,7 +652,7 @@ void cdv_intel_lvds_init(struct drm_device *dev, drm_encoder_init(dev, encoder, &cdv_intel_lvds_enc_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c index 2eaf1b31c7bd..ee95c03a8c54 100644 --- a/drivers/gpu/drm/gma500/framebuffer.c +++ b/drivers/gpu/drm/gma500/framebuffer.c @@ -241,7 +241,7 @@ static struct fb_ops psbfb_unaccel_ops = { */ static int psb_framebuffer_init(struct drm_device *dev, struct psb_framebuffer *fb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct gtt_range *gt) { u32 bpp, depth; @@ -284,7 +284,7 @@ static int psb_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer *psb_framebuffer_create (struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct gtt_range *gt) { struct psb_framebuffer *fb; @@ -406,8 +406,6 @@ static int psbfb_create(struct psb_fbdev *fbdev, memset(dev_priv->vram_addr + backing->offset, 0, size); - mutex_lock(&dev->struct_mutex); - info = drm_fb_helper_alloc_fbi(&fbdev->psb_fb_helper); if (IS_ERR(info)) { ret = PTR_ERR(info); @@ -463,17 +461,15 @@ static int psbfb_create(struct psb_fbdev *fbdev, dev_dbg(dev->dev, "allocated %dx%d fb\n", psbfb->base.width, psbfb->base.height); - mutex_unlock(&dev->struct_mutex); return 0; out_unref: if (backing->stolen) psb_gtt_free_range(dev, backing); else - drm_gem_object_unreference(&backing->gem); + drm_gem_object_unreference_unlocked(&backing->gem); drm_fb_helper_release_fbi(&fbdev->psb_fb_helper); out_err1: - mutex_unlock(&dev->struct_mutex); psb_gtt_free_range(dev, backing); return ret; } @@ -488,7 +484,7 @@ out_err1: */ static struct drm_framebuffer *psb_user_framebuffer_create (struct drm_device *dev, struct drm_file *filp, - struct drm_mode_fb_cmd2 *cmd) + const struct drm_mode_fb_cmd2 *cmd) { struct gtt_range *r; struct drm_gem_object *obj; @@ -569,7 +565,7 @@ static int psb_fbdev_destroy(struct drm_device *dev, struct psb_fbdev *fbdev) drm_framebuffer_cleanup(&psbfb->base); if (psbfb->gtt) - drm_gem_object_unreference(&psbfb->gtt->gem); + drm_gem_object_unreference_unlocked(&psbfb->gtt->gem); return 0; } @@ -784,12 +780,8 @@ void psb_modeset_cleanup(struct drm_device *dev) { struct drm_psb_private *dev_priv = dev->dev_private; if (dev_priv->modeset) { - mutex_lock(&dev->struct_mutex); - drm_kms_helper_poll_fini(dev); psb_fbdev_fini(dev); drm_mode_config_cleanup(dev); - - mutex_unlock(&dev->struct_mutex); } } diff --git a/drivers/gpu/drm/gma500/gem.c b/drivers/gpu/drm/gma500/gem.c index c707fa6fca85..506224b3a0ad 100644 --- a/drivers/gpu/drm/gma500/gem.c +++ b/drivers/gpu/drm/gma500/gem.c @@ -62,15 +62,10 @@ int psb_gem_dumb_map_gtt(struct drm_file *file, struct drm_device *dev, int ret = 0; struct drm_gem_object *obj; - mutex_lock(&dev->struct_mutex); - /* GEM does all our handle to object mapping */ obj = drm_gem_object_lookup(dev, file, handle); - if (obj == NULL) { - ret = -ENOENT; - goto unlock; - } - /* What validation is needed here ? */ + if (obj == NULL) + return -ENOENT; /* Make it mmapable */ ret = drm_gem_create_mmap_offset(obj); @@ -78,9 +73,7 @@ int psb_gem_dumb_map_gtt(struct drm_file *file, struct drm_device *dev, goto out; *offset = drm_vma_node_offset_addr(&obj->vma_node); out: - drm_gem_object_unreference(obj); -unlock: - mutex_unlock(&dev->struct_mutex); + drm_gem_object_unreference_unlocked(obj); return ret; } @@ -130,7 +123,7 @@ int psb_gem_create(struct drm_file *file, struct drm_device *dev, u64 size, return ret; } /* We have the initial and handle reference but need only one now */ - drm_gem_object_unreference(&r->gem); + drm_gem_object_unreference_unlocked(&r->gem); *handlep = handle; return 0; } @@ -189,7 +182,7 @@ int psb_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) /* Make sure we don't parallel update on a fault, nor move or remove something from beneath our feet */ - mutex_lock(&dev->struct_mutex); + mutex_lock(&dev_priv->mmap_mutex); /* For now the mmap pins the object and it stays pinned. As things stand that will do us no harm */ @@ -215,7 +208,7 @@ int psb_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ret = vm_insert_pfn(vma, (unsigned long)vmf->virtual_address, pfn); fail: - mutex_unlock(&dev->struct_mutex); + mutex_unlock(&dev_priv->mmap_mutex); switch (ret) { case 0: case -ERESTARTSYS: diff --git a/drivers/gpu/drm/gma500/gma_display.c b/drivers/gpu/drm/gma500/gma_display.c index 001b450b27b3..ff17af4cfc64 100644 --- a/drivers/gpu/drm/gma500/gma_display.c +++ b/drivers/gpu/drm/gma500/gma_display.c @@ -349,8 +349,6 @@ int gma_crtc_cursor_set(struct drm_crtc *crtc, /* If we didn't get a handle then turn the cursor off */ if (!handle) { temp = CURSOR_MODE_DISABLE; - mutex_lock(&dev->struct_mutex); - if (gma_power_begin(dev, false)) { REG_WRITE(control, temp); REG_WRITE(base, 0); @@ -362,11 +360,9 @@ int gma_crtc_cursor_set(struct drm_crtc *crtc, gt = container_of(gma_crtc->cursor_obj, struct gtt_range, gem); psb_gtt_unpin(gt); - drm_gem_object_unreference(gma_crtc->cursor_obj); + drm_gem_object_unreference_unlocked(gma_crtc->cursor_obj); gma_crtc->cursor_obj = NULL; } - - mutex_unlock(&dev->struct_mutex); return 0; } @@ -376,7 +372,6 @@ int gma_crtc_cursor_set(struct drm_crtc *crtc, return -EINVAL; } - mutex_lock(&dev->struct_mutex); obj = drm_gem_object_lookup(dev, file_priv, handle); if (!obj) { ret = -ENOENT; @@ -441,17 +436,15 @@ int gma_crtc_cursor_set(struct drm_crtc *crtc, if (gma_crtc->cursor_obj) { gt = container_of(gma_crtc->cursor_obj, struct gtt_range, gem); psb_gtt_unpin(gt); - drm_gem_object_unreference(gma_crtc->cursor_obj); + drm_gem_object_unreference_unlocked(gma_crtc->cursor_obj); } gma_crtc->cursor_obj = obj; unlock: - mutex_unlock(&dev->struct_mutex); return ret; unref_cursor: - drm_gem_object_unreference(obj); - mutex_unlock(&dev->struct_mutex); + drm_gem_object_unreference_unlocked(obj); return ret; } diff --git a/drivers/gpu/drm/gma500/gtt.c b/drivers/gpu/drm/gma500/gtt.c index ce015db59dc6..8f69225ce2b4 100644 --- a/drivers/gpu/drm/gma500/gtt.c +++ b/drivers/gpu/drm/gma500/gtt.c @@ -425,6 +425,7 @@ int psb_gtt_init(struct drm_device *dev, int resume) if (!resume) { mutex_init(&dev_priv->gtt_mutex); + mutex_init(&dev_priv->mmap_mutex); psb_gtt_alloc(dev); } diff --git a/drivers/gpu/drm/gma500/mdfld_device.c b/drivers/gpu/drm/gma500/mdfld_device.c index 265ad0de44a6..e2ab858122f9 100644 --- a/drivers/gpu/drm/gma500/mdfld_device.c +++ b/drivers/gpu/drm/gma500/mdfld_device.c @@ -546,6 +546,8 @@ const struct psb_ops mdfld_chip_ops = { .save_regs = mdfld_save_registers, .restore_regs = mdfld_restore_registers, + .save_crtc = gma_crtc_save, + .restore_crtc = gma_crtc_restore, .power_down = mdfld_power_down, .power_up = mdfld_power_up, }; diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c index d4813e03f5ee..1a1acd3cb049 100644 --- a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c +++ b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c @@ -994,7 +994,7 @@ struct mdfld_dsi_encoder *mdfld_dsi_dpi_init(struct drm_device *dev, drm_encoder_init(dev, encoder, p_funcs->encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); drm_encoder_helper_add(encoder, p_funcs->encoder_helper_funcs); diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_output.c b/drivers/gpu/drm/gma500/mdfld_dsi_output.c index 89f705c3a5eb..d758f4cc6805 100644 --- a/drivers/gpu/drm/gma500/mdfld_dsi_output.c +++ b/drivers/gpu/drm/gma500/mdfld_dsi_output.c @@ -405,8 +405,6 @@ static struct drm_encoder *mdfld_dsi_connector_best_encoder( /*DSI connector funcs*/ static const struct drm_connector_funcs mdfld_dsi_connector_funcs = { .dpms = /*drm_helper_connector_dpms*/mdfld_dsi_connector_dpms, - .save = mdfld_dsi_connector_save, - .restore = mdfld_dsi_connector_restore, .detect = mdfld_dsi_connector_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = mdfld_dsi_connector_set_property, @@ -563,6 +561,9 @@ void mdfld_dsi_output_init(struct drm_device *dev, connector = &dsi_connector->base.base; + dsi_connector->base.save = mdfld_dsi_connector_save; + dsi_connector->base.restore = mdfld_dsi_connector_restore; + drm_connector_init(dev, connector, &mdfld_dsi_connector_funcs, DRM_MODE_CONNECTOR_LVDS); drm_connector_helper_add(connector, &mdfld_dsi_connector_helper_funcs); diff --git a/drivers/gpu/drm/gma500/oaktrail_device.c b/drivers/gpu/drm/gma500/oaktrail_device.c index 368a03ae3010..ba30b43a3412 100644 --- a/drivers/gpu/drm/gma500/oaktrail_device.c +++ b/drivers/gpu/drm/gma500/oaktrail_device.c @@ -568,6 +568,8 @@ const struct psb_ops oaktrail_chip_ops = { .save_regs = oaktrail_save_display_registers, .restore_regs = oaktrail_restore_display_registers, + .save_crtc = gma_crtc_save, + .restore_crtc = gma_crtc_restore, .power_down = oaktrail_power_down, .power_up = oaktrail_power_up, diff --git a/drivers/gpu/drm/gma500/oaktrail_hdmi.c b/drivers/gpu/drm/gma500/oaktrail_hdmi.c index 2310d879cdc2..2d18499d6060 100644 --- a/drivers/gpu/drm/gma500/oaktrail_hdmi.c +++ b/drivers/gpu/drm/gma500/oaktrail_hdmi.c @@ -654,7 +654,7 @@ void oaktrail_hdmi_init(struct drm_device *dev, drm_encoder_init(dev, encoder, &oaktrail_hdmi_enc_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); diff --git a/drivers/gpu/drm/gma500/oaktrail_lvds.c b/drivers/gpu/drm/gma500/oaktrail_lvds.c index 83bbc271bcfb..f7038f12ac76 100644 --- a/drivers/gpu/drm/gma500/oaktrail_lvds.c +++ b/drivers/gpu/drm/gma500/oaktrail_lvds.c @@ -323,7 +323,7 @@ void oaktrail_lvds_init(struct drm_device *dev, DRM_MODE_CONNECTOR_LVDS); drm_encoder_init(dev, encoder, &psb_intel_lvds_enc_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); gma_encoder->type = INTEL_OUTPUT_LVDS; diff --git a/drivers/gpu/drm/gma500/psb_device.c b/drivers/gpu/drm/gma500/psb_device.c index 07df7d4eea72..dc0f8527570c 100644 --- a/drivers/gpu/drm/gma500/psb_device.c +++ b/drivers/gpu/drm/gma500/psb_device.c @@ -181,7 +181,7 @@ static int psb_save_display_registers(struct drm_device *dev) { struct drm_psb_private *dev_priv = dev->dev_private; struct drm_crtc *crtc; - struct drm_connector *connector; + struct gma_connector *connector; struct psb_state *regs = &dev_priv->regs.psb; /* Display arbitration control + watermarks */ @@ -198,12 +198,12 @@ static int psb_save_display_registers(struct drm_device *dev) drm_modeset_lock_all(dev); list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { if (drm_helper_crtc_in_use(crtc)) - crtc->funcs->save(crtc); + dev_priv->ops->save_crtc(crtc); } - list_for_each_entry(connector, &dev->mode_config.connector_list, head) - if (connector->funcs->save) - connector->funcs->save(connector); + list_for_each_entry(connector, &dev->mode_config.connector_list, base.head) + if (connector->save) + connector->save(&connector->base); drm_modeset_unlock_all(dev); return 0; @@ -219,7 +219,7 @@ static int psb_restore_display_registers(struct drm_device *dev) { struct drm_psb_private *dev_priv = dev->dev_private; struct drm_crtc *crtc; - struct drm_connector *connector; + struct gma_connector *connector; struct psb_state *regs = &dev_priv->regs.psb; /* Display arbitration + watermarks */ @@ -238,11 +238,11 @@ static int psb_restore_display_registers(struct drm_device *dev) drm_modeset_lock_all(dev); list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) if (drm_helper_crtc_in_use(crtc)) - crtc->funcs->restore(crtc); + dev_priv->ops->restore_crtc(crtc); - list_for_each_entry(connector, &dev->mode_config.connector_list, head) - if (connector->funcs->restore) - connector->funcs->restore(connector); + list_for_each_entry(connector, &dev->mode_config.connector_list, base.head) + if (connector->restore) + connector->restore(&connector->base); drm_modeset_unlock_all(dev); return 0; @@ -354,6 +354,8 @@ const struct psb_ops psb_chip_ops = { .init_pm = psb_init_pm, .save_regs = psb_save_display_registers, .restore_regs = psb_restore_display_registers, + .save_crtc = gma_crtc_save, + .restore_crtc = gma_crtc_restore, .power_down = psb_power_down, .power_up = psb_power_up, }; diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h index e21726ecac32..b74372760d7f 100644 --- a/drivers/gpu/drm/gma500/psb_drv.h +++ b/drivers/gpu/drm/gma500/psb_drv.h @@ -465,6 +465,8 @@ struct drm_psb_private { struct mutex gtt_mutex; struct resource *gtt_mem; /* Our PCI resource */ + struct mutex mmap_mutex; + struct psb_mmu_driver *mmu; struct psb_mmu_pd *pf_pd; @@ -651,6 +653,8 @@ struct psb_ops { void (*init_pm)(struct drm_device *dev); int (*save_regs)(struct drm_device *dev); int (*restore_regs)(struct drm_device *dev); + void (*save_crtc)(struct drm_crtc *crtc); + void (*restore_crtc)(struct drm_crtc *crtc); int (*power_up)(struct drm_device *dev); int (*power_down)(struct drm_device *dev); void (*update_wm)(struct drm_device *dev, struct drm_crtc *crtc); diff --git a/drivers/gpu/drm/gma500/psb_intel_display.c b/drivers/gpu/drm/gma500/psb_intel_display.c index 6659da88fe5b..dcdbc37e55e1 100644 --- a/drivers/gpu/drm/gma500/psb_intel_display.c +++ b/drivers/gpu/drm/gma500/psb_intel_display.c @@ -439,8 +439,6 @@ const struct drm_crtc_helper_funcs psb_intel_helper_funcs = { }; const struct drm_crtc_funcs psb_intel_crtc_funcs = { - .save = gma_crtc_save, - .restore = gma_crtc_restore, .cursor_set = gma_crtc_cursor_set, .cursor_move = gma_crtc_cursor_move, .gamma_set = gma_crtc_gamma_set, diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h index 860dd2177ca1..2a3b7c684db2 100644 --- a/drivers/gpu/drm/gma500/psb_intel_drv.h +++ b/drivers/gpu/drm/gma500/psb_intel_drv.h @@ -140,6 +140,9 @@ struct gma_encoder { struct gma_connector { struct drm_connector base; struct gma_encoder *encoder; + + void (*save)(struct drm_connector *connector); + void (*restore)(struct drm_connector *connector); }; struct psb_intel_crtc_state { diff --git a/drivers/gpu/drm/gma500/psb_intel_lvds.c b/drivers/gpu/drm/gma500/psb_intel_lvds.c index ce0645d0c1e5..b1b93317d054 100644 --- a/drivers/gpu/drm/gma500/psb_intel_lvds.c +++ b/drivers/gpu/drm/gma500/psb_intel_lvds.c @@ -653,8 +653,6 @@ const struct drm_connector_helper_funcs const struct drm_connector_funcs psb_intel_lvds_connector_funcs = { .dpms = drm_helper_connector_dpms, - .save = psb_intel_lvds_save, - .restore = psb_intel_lvds_restore, .detect = psb_intel_lvds_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = psb_intel_lvds_set_property, @@ -715,6 +713,9 @@ void psb_intel_lvds_init(struct drm_device *dev, gma_encoder->dev_priv = lvds_priv; connector = &gma_connector->base; + gma_connector->save = psb_intel_lvds_save; + gma_connector->restore = psb_intel_lvds_restore; + encoder = &gma_encoder->base; drm_connector_init(dev, connector, &psb_intel_lvds_connector_funcs, @@ -722,7 +723,7 @@ void psb_intel_lvds_init(struct drm_device *dev, drm_encoder_init(dev, encoder, &psb_intel_lvds_enc_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); gma_connector_attach_encoder(gma_connector, gma_encoder); gma_encoder->type = INTEL_OUTPUT_LVDS; diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c index 58529cea575d..e787d376ba67 100644 --- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c +++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c @@ -1837,8 +1837,6 @@ static const struct drm_encoder_helper_funcs psb_intel_sdvo_helper_funcs = { static const struct drm_connector_funcs psb_intel_sdvo_connector_funcs = { .dpms = drm_helper_connector_dpms, - .save = psb_intel_sdvo_save, - .restore = psb_intel_sdvo_restore, .detect = psb_intel_sdvo_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = psb_intel_sdvo_set_property, @@ -2021,6 +2019,9 @@ psb_intel_sdvo_connector_init(struct psb_intel_sdvo_connector *connector, connector->base.base.doublescan_allowed = 0; connector->base.base.display_info.subpixel_order = SubPixelHorizontalRGB; + connector->base.save = psb_intel_sdvo_save; + connector->base.restore = psb_intel_sdvo_restore; + gma_connector_attach_encoder(&connector->base, &encoder->base); drm_connector_register(&connector->base.base); } @@ -2525,7 +2526,8 @@ bool psb_intel_sdvo_init(struct drm_device *dev, int sdvo_reg) /* encoder type will be decided later */ gma_encoder = &psb_intel_sdvo->base; gma_encoder->type = INTEL_OUTPUT_SDVO; - drm_encoder_init(dev, &gma_encoder->base, &psb_intel_sdvo_enc_funcs, 0); + drm_encoder_init(dev, &gma_encoder->base, &psb_intel_sdvo_enc_funcs, + 0, NULL); /* Read the regs to test if we can talk to the device */ for (i = 0; i < 0x40; i++) { diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index 896b6aaf8c4d..a46248f0c9c3 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -855,18 +855,6 @@ static void tda998x_encoder_dpms(struct drm_encoder *encoder, int mode) priv->dpms = mode; } -static void -tda998x_encoder_save(struct drm_encoder *encoder) -{ - DBG(""); -} - -static void -tda998x_encoder_restore(struct drm_encoder *encoder) -{ - DBG(""); -} - static bool tda998x_encoder_mode_fixup(struct drm_encoder *encoder, const struct drm_display_mode *mode, @@ -1351,8 +1339,6 @@ static void tda998x_encoder_commit(struct drm_encoder *encoder) static const struct drm_encoder_helper_funcs tda998x_encoder_helper_funcs = { .dpms = tda998x_encoder_dpms, - .save = tda998x_encoder_save, - .restore = tda998x_encoder_restore, .mode_fixup = tda998x_encoder_mode_fixup, .prepare = tda998x_encoder_prepare, .commit = tda998x_encoder_commit, @@ -1437,7 +1423,7 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data) drm_encoder_helper_add(&priv->encoder, &tda998x_encoder_helper_funcs); ret = drm_encoder_init(drm, &priv->encoder, &tda998x_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); if (ret) goto err_encoder; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 5b3830ef44df..f1a8a53e9e30 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -286,7 +286,7 @@ struct i915_hotplug { list_for_each_entry(intel_plane, \ &(dev)->mode_config.plane_list, \ base.head) \ - if ((intel_plane)->pipe == (intel_crtc)->pipe) + for_each_if ((intel_plane)->pipe == (intel_crtc)->pipe) #define for_each_intel_crtc(dev, intel_crtc) \ list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) @@ -303,15 +303,15 @@ struct i915_hotplug { #define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \ list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \ - if ((intel_encoder)->base.crtc == (__crtc)) + for_each_if ((intel_encoder)->base.crtc == (__crtc)) #define for_each_connector_on_encoder(dev, __encoder, intel_connector) \ list_for_each_entry((intel_connector), &(dev)->mode_config.connector_list, base.head) \ - if ((intel_connector)->base.encoder == (__encoder)) + for_each_if ((intel_connector)->base.encoder == (__encoder)) #define for_each_power_domain(domain, mask) \ for ((domain) = 0; (domain) < POWER_DOMAIN_NUM; (domain)++) \ - if ((1 << (domain)) & (mask)) + for_each_if ((1 << (domain)) & (mask)) struct drm_i915_private; struct i915_mm_struct; @@ -730,7 +730,7 @@ struct intel_uncore { for ((i__) = 0, (domain__) = &(dev_priv__)->uncore.fw_domain[0]; \ (i__) < FW_DOMAIN_ID_COUNT; \ (i__)++, (domain__) = &(dev_priv__)->uncore.fw_domain[i__]) \ - if (((mask__) & (dev_priv__)->uncore.fw_domains) & (1 << (i__))) + for_each_if (((mask__) & (dev_priv__)->uncore.fw_domains) & (1 << (i__))) #define for_each_fw_domain(domain__, dev_priv__, i__) \ for_each_fw_domain_mask(domain__, FORCEWAKE_ALL, dev_priv__, i__) @@ -1968,7 +1968,7 @@ static inline struct drm_i915_private *guc_to_i915(struct intel_guc *guc) /* Iterate over initialised rings */ #define for_each_ring(ring__, dev_priv__, i__) \ for ((i__) = 0; (i__) < I915_NUM_RINGS; (i__)++) \ - if (((ring__) = &(dev_priv__)->ring[(i__)]), intel_ring_initialized((ring__))) + for_each_if ((((ring__) = &(dev_priv__)->ring[(i__)]), intel_ring_initialized((ring__)))) enum hdmi_force_audio { HDMI_AUDIO_OFF_DVI = -2, /* no aux data for HDMI-DVI converter */ diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index a531cb83295c..b7d7cecdddf6 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1210,8 +1210,16 @@ int __i915_wait_request(struct drm_i915_gem_request *req, if (i915_gem_request_completed(req, true)) return 0; - timeout_expire = timeout ? - jiffies + nsecs_to_jiffies_timeout((u64)*timeout) : 0; + timeout_expire = 0; + if (timeout) { + if (WARN_ON(*timeout < 0)) + return -EINVAL; + + if (*timeout == 0) + return -ETIME; + + timeout_expire = jiffies + nsecs_to_jiffies_timeout(*timeout); + } if (INTEL_INFO(dev_priv)->gen >= 6) gen6_rps_boost(dev_priv, rps, req->emitted_jiffies); @@ -2941,6 +2949,10 @@ i915_gem_idle_work_handler(struct work_struct *work) if (!list_empty(&ring->request_list)) return; + /* we probably should sync with hangcheck here, using cancel_work_sync. + * Also locking seems to be fubar here, ring->request_list is protected + * by dev->struct_mutex. */ + intel_mark_idle(dev); if (mutex_trylock(&dev->struct_mutex)) { diff --git a/drivers/gpu/drm/i915/i915_gem_fence.c b/drivers/gpu/drm/i915/i915_gem_fence.c index b80d0456fe03..598198543dcd 100644 --- a/drivers/gpu/drm/i915/i915_gem_fence.c +++ b/drivers/gpu/drm/i915/i915_gem_fence.c @@ -642,11 +642,10 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev) } /* check for L-shaped memory aka modified enhanced addressing */ - if (IS_GEN4(dev)) { - uint32_t ddc2 = I915_READ(DCC2); - - if (!(ddc2 & DCC2_MODIFIED_ENHANCED_DISABLE)) - dev_priv->quirks |= QUIRK_PIN_SWIZZLED_PAGES; + if (IS_GEN4(dev) && + !(I915_READ(DCC2) & DCC2_MODIFIED_ENHANCED_DISABLE)) { + swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN; + swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN; } if (dcc == 0xffffffff) { @@ -675,16 +674,35 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev) * matching, which was the case for the swizzling required in * the table above, or from the 1-ch value being less than * the minimum size of a rank. + * + * Reports indicate that the swizzling actually + * varies depending upon page placement inside the + * channels, i.e. we see swizzled pages where the + * banks of memory are paired and unswizzled on the + * uneven portion, so leave that as unknown. */ - if (I915_READ16(C0DRB3) != I915_READ16(C1DRB3)) { - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - } else { + if (I915_READ16(C0DRB3) == I915_READ16(C1DRB3)) { swizzle_x = I915_BIT_6_SWIZZLE_9_10; swizzle_y = I915_BIT_6_SWIZZLE_9; } } + if (swizzle_x == I915_BIT_6_SWIZZLE_UNKNOWN || + swizzle_y == I915_BIT_6_SWIZZLE_UNKNOWN) { + /* Userspace likes to explode if it sees unknown swizzling, + * so lie. We will finish the lie when reporting through + * the get-tiling-ioctl by reporting the physical swizzle + * mode as unknown instead. + * + * As we don't strictly know what the swizzling is, it may be + * bit17 dependent, and so we need to also prevent the pages + * from being moved. + */ + dev_priv->quirks |= QUIRK_PIN_SWIZZLED_PAGES; + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + } + dev_priv->mm.bit_6_swizzle_x = swizzle_x; dev_priv->mm.bit_6_swizzle_y = swizzle_y; } diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 12008af797bd..9285fc1e64ee 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -824,7 +824,7 @@ void intel_crt_init(struct drm_device *dev) &intel_crt_connector_funcs, DRM_MODE_CONNECTOR_VGA); drm_encoder_init(dev, &crt->base.base, &intel_crt_enc_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); intel_connector_attach_encoder(intel_connector, &crt->base); diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 7f618cf5289c..4afb3103eb96 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -3284,7 +3284,7 @@ void intel_ddi_init(struct drm_device *dev, enum port port) encoder = &intel_encoder->base; drm_encoder_init(dev, encoder, &intel_ddi_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); intel_encoder->compute_config = intel_ddi_compute_config; intel_encoder->enable = intel_enable_ddi; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 3a13a498722a..bda6b9c82e66 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -12427,7 +12427,7 @@ static bool intel_fuzzy_clock_check(int clock1, int clock2) list_for_each_entry((intel_crtc), \ &(dev)->mode_config.crtc_list, \ base.head) \ - if (mask & (1 <<(intel_crtc)->pipe)) + for_each_if (mask & (1 <<(intel_crtc)->pipe)) static bool intel_compare_m_n(unsigned int m, unsigned int n, @@ -12606,7 +12606,6 @@ intel_pipe_config_compare(struct drm_device *dev, if (INTEL_INFO(dev)->gen < 8) { PIPE_CONF_CHECK_M_N(dp_m_n); - PIPE_CONF_CHECK_I(has_drrs); if (current_config->has_drrs) PIPE_CONF_CHECK_M_N(dp_m2_n2); } else @@ -13949,7 +13948,7 @@ static struct drm_plane *intel_primary_plane_create(struct drm_device *dev, drm_universal_plane_init(dev, &primary->base, 0, &intel_plane_funcs, intel_primary_formats, num_formats, - DRM_PLANE_TYPE_PRIMARY); + DRM_PLANE_TYPE_PRIMARY, NULL); if (INTEL_INFO(dev)->gen >= 4) intel_create_rotation_property(dev, primary); @@ -14088,7 +14087,7 @@ static struct drm_plane *intel_cursor_plane_create(struct drm_device *dev, &intel_plane_funcs, intel_cursor_formats, ARRAY_SIZE(intel_cursor_formats), - DRM_PLANE_TYPE_CURSOR); + DRM_PLANE_TYPE_CURSOR, NULL); if (INTEL_INFO(dev)->gen >= 4) { if (!dev->mode_config.rotation_property) @@ -14165,7 +14164,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe) goto fail; ret = drm_crtc_init_with_planes(dev, &intel_crtc->base, primary, - cursor, &intel_crtc_funcs); + cursor, &intel_crtc_funcs, NULL); if (ret) goto fail; @@ -14701,7 +14700,7 @@ static int intel_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * intel_user_framebuffer_create(struct drm_device *dev, struct drm_file *filp, - struct drm_mode_fb_cmd2 *user_mode_cmd) + const struct drm_mode_fb_cmd2 *user_mode_cmd) { struct drm_framebuffer *fb; struct drm_i915_gem_object *obj; diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 7e3ffc30a00e..0f0573aa1b0d 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -4974,7 +4974,8 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd) enum intel_display_power_domain power_domain; enum irqreturn ret = IRQ_NONE; - if (intel_dig_port->base.type != INTEL_OUTPUT_EDP) + if (intel_dig_port->base.type != INTEL_OUTPUT_EDP && + intel_dig_port->base.type != INTEL_OUTPUT_HDMI) intel_dig_port->base.type = INTEL_OUTPUT_DISPLAYPORT; if (long_hpd && intel_dig_port->base.type == INTEL_OUTPUT_EDP) { @@ -5988,7 +5989,7 @@ intel_dp_init(struct drm_device *dev, encoder = &intel_encoder->base; drm_encoder_init(dev, &intel_encoder->base, &intel_dp_enc_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); intel_encoder->compute_config = intel_dp_compute_config; intel_encoder->disable = intel_disable_dp; diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 8c4e7dfe304c..e8d369d0a713 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -536,7 +536,7 @@ intel_dp_create_fake_mst_encoder(struct intel_digital_port *intel_dig_port, enum intel_mst->primary = intel_dig_port; drm_encoder_init(dev, &intel_encoder->base, &intel_dp_mst_enc_funcs, - DRM_MODE_ENCODER_DPMST); + DRM_MODE_ENCODER_DPMST, NULL); intel_encoder->type = INTEL_OUTPUT_DP_MST; intel_encoder->crtc_mask = 0x7; diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 3517cd1ce266..50f83d220249 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -123,8 +123,6 @@ struct intel_framebuffer { struct intel_fbdev { struct drm_fb_helper helper; struct intel_framebuffer *fb; - struct list_head fbdev_list; - struct drm_display_mode *our_mode; int preferred_bpp; }; diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c index 230957f9f663..fff9a66c32a1 100644 --- a/drivers/gpu/drm/i915/intel_dsi.c +++ b/drivers/gpu/drm/i915/intel_dsi.c @@ -1152,7 +1152,8 @@ void intel_dsi_init(struct drm_device *dev) connector = &intel_connector->base; - drm_encoder_init(dev, encoder, &intel_dsi_funcs, DRM_MODE_ENCODER_DSI); + drm_encoder_init(dev, encoder, &intel_dsi_funcs, DRM_MODE_ENCODER_DSI, + NULL); intel_encoder->compute_config = intel_dsi_compute_config; intel_encoder->pre_enable = intel_dsi_pre_enable; diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h index e6cb25239941..02551ff228c2 100644 --- a/drivers/gpu/drm/i915/intel_dsi.h +++ b/drivers/gpu/drm/i915/intel_dsi.h @@ -117,7 +117,7 @@ static inline struct intel_dsi_host *to_intel_dsi_host(struct mipi_dsi_host *h) #define for_each_dsi_port(__port, __ports_mask) \ for ((__port) = PORT_A; (__port) < I915_MAX_PORTS; (__port)++) \ - if ((__ports_mask) & (1 << (__port))) + for_each_if ((__ports_mask) & (1 << (__port))) static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder) { diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c index 7161deb2aed8..286baec979c8 100644 --- a/drivers/gpu/drm/i915/intel_dvo.c +++ b/drivers/gpu/drm/i915/intel_dvo.c @@ -429,7 +429,7 @@ void intel_dvo_init(struct drm_device *dev) intel_encoder = &intel_dvo->base; drm_encoder_init(dev, &intel_encoder->base, - &intel_dvo_enc_funcs, encoder_type); + &intel_dvo_enc_funcs, encoder_type, NULL); intel_encoder->disable = intel_disable_dvo; intel_encoder->enable = intel_enable_dvo; diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index c3978bad5ca0..00d065fee506 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -2164,7 +2164,7 @@ void intel_hdmi_init(struct drm_device *dev, intel_encoder = &intel_dig_port->base; drm_encoder_init(dev, &intel_encoder->base, &intel_hdmi_enc_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); intel_encoder->compute_config = intel_hdmi_compute_config; if (HAS_PCH_SPLIT(dev)) { diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index 61f1145f6579..0da0240caf81 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -1025,7 +1025,7 @@ void intel_lvds_init(struct drm_device *dev) DRM_MODE_CONNECTOR_LVDS); drm_encoder_init(dev, &intel_encoder->base, &intel_lvds_enc_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); intel_encoder->enable = intel_enable_lvds; intel_encoder->pre_enable = intel_pre_enable_lvds; diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index ab833efb5f0c..2c2151f1c47e 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -54,13 +54,13 @@ i < (power_domains)->power_well_count && \ ((power_well) = &(power_domains)->power_wells[i]); \ i++) \ - if ((power_well)->domains & (domain_mask)) + for_each_if ((power_well)->domains & (domain_mask)) #define for_each_power_well_rev(i, power_well, domain_mask, power_domains) \ for (i = (power_domains)->power_well_count - 1; \ i >= 0 && ((power_well) = &(power_domains)->power_wells[i]);\ i--) \ - if ((power_well)->domains & (domain_mask)) + for_each_if ((power_well)->domains & (domain_mask)) bool intel_display_power_well_is_enabled(struct drm_i915_private *dev_priv, int power_well_id); diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 06679f164b3e..2e1da060b0e1 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -2978,7 +2978,8 @@ bool intel_sdvo_init(struct drm_device *dev, /* encoder type will be decided later */ intel_encoder = &intel_sdvo->base; intel_encoder->type = INTEL_OUTPUT_SDVO; - drm_encoder_init(dev, &intel_encoder->base, &intel_sdvo_enc_funcs, 0); + drm_encoder_init(dev, &intel_encoder->base, &intel_sdvo_enc_funcs, 0, + NULL); /* Read the regs to test if we can talk to the device */ for (i = 0; i < 0x40; i++) { diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index a2c15f811a33..dbf421351b5c 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -817,8 +817,8 @@ intel_check_sprite_plane(struct drm_plane *plane, hscale = drm_rect_calc_hscale(src, dst, min_scale, max_scale); if (hscale < 0) { DRM_DEBUG_KMS("Horizontal scaling factor out of limits\n"); - drm_rect_debug_print(src, true); - drm_rect_debug_print(dst, false); + drm_rect_debug_print("src: ", src, true); + drm_rect_debug_print("dst: ", dst, false); return hscale; } @@ -826,8 +826,8 @@ intel_check_sprite_plane(struct drm_plane *plane, vscale = drm_rect_calc_vscale(src, dst, min_scale, max_scale); if (vscale < 0) { DRM_DEBUG_KMS("Vertical scaling factor out of limits\n"); - drm_rect_debug_print(src, true); - drm_rect_debug_print(dst, false); + drm_rect_debug_print("src: ", src, true); + drm_rect_debug_print("dst: ", dst, false); return vscale; } @@ -1123,7 +1123,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane) ret = drm_universal_plane_init(dev, &intel_plane->base, possible_crtcs, &intel_plane_funcs, plane_formats, num_plane_formats, - DRM_PLANE_TYPE_OVERLAY); + DRM_PLANE_TYPE_OVERLAY, NULL); if (ret) { kfree(intel_plane); goto out; diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index 6bea78944cd6..948cbff6c62e 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -1645,7 +1645,7 @@ intel_tv_init(struct drm_device *dev) DRM_MODE_CONNECTOR_SVIDEO); drm_encoder_init(dev, &intel_encoder->base, &intel_tv_enc_funcs, - DRM_MODE_ENCODER_TVDAC); + DRM_MODE_ENCODER_TVDAC, NULL); intel_encoder->compute_config = intel_tv_compute_config; intel_encoder->get_config = intel_tv_get_config; diff --git a/drivers/gpu/drm/imx/Kconfig b/drivers/gpu/drm/imx/Kconfig index 2b81a417cf29..35ca4f007839 100644 --- a/drivers/gpu/drm/imx/Kconfig +++ b/drivers/gpu/drm/imx/Kconfig @@ -10,15 +10,6 @@ config DRM_IMX help enable i.MX graphics support -config DRM_IMX_FB_HELPER - tristate "provide legacy framebuffer /dev/fb0" - select DRM_KMS_CMA_HELPER - depends on DRM_IMX - help - The DRM framework can provide a legacy /dev/fb0 framebuffer - for your device. This is necessary to get a framebuffer console - and also for applications using the legacy framebuffer API - config DRM_IMX_PARALLEL_DISPLAY tristate "Support for parallel displays" select DRM_PANEL diff --git a/drivers/gpu/drm/imx/dw_hdmi-imx.c b/drivers/gpu/drm/imx/dw_hdmi-imx.c index 98605ea2ad9d..35fcf6b84537 100644 --- a/drivers/gpu/drm/imx/dw_hdmi-imx.c +++ b/drivers/gpu/drm/imx/dw_hdmi-imx.c @@ -251,7 +251,7 @@ static int dw_hdmi_imx_bind(struct device *dev, struct device *master, drm_encoder_helper_add(encoder, &dw_hdmi_imx_encoder_helper_funcs); drm_encoder_init(drm, encoder, &dw_hdmi_imx_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); return dw_hdmi_bind(dev, master, data, encoder, iores, irq, plat_data); } diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c index 64f16ea779ef..09e20ea69419 100644 --- a/drivers/gpu/drm/imx/imx-drm-core.c +++ b/drivers/gpu/drm/imx/imx-drm-core.c @@ -49,8 +49,10 @@ struct imx_drm_crtc { struct imx_drm_crtc_helper_funcs imx_drm_helper_funcs; }; +#if IS_ENABLED(CONFIG_DRM_FBDEV_EMULATION) static int legacyfb_depth = 16; module_param(legacyfb_depth, int, 0444); +#endif int imx_drm_crtc_id(struct imx_drm_crtc *crtc) { @@ -60,26 +62,19 @@ EXPORT_SYMBOL_GPL(imx_drm_crtc_id); static void imx_drm_driver_lastclose(struct drm_device *drm) { -#if IS_ENABLED(CONFIG_DRM_IMX_FB_HELPER) struct imx_drm_device *imxdrm = drm->dev_private; - if (imxdrm->fbhelper) - drm_fbdev_cma_restore_mode(imxdrm->fbhelper); -#endif + drm_fbdev_cma_restore_mode(imxdrm->fbhelper); } static int imx_drm_driver_unload(struct drm_device *drm) { -#if IS_ENABLED(CONFIG_DRM_IMX_FB_HELPER) struct imx_drm_device *imxdrm = drm->dev_private; -#endif drm_kms_helper_poll_fini(drm); -#if IS_ENABLED(CONFIG_DRM_IMX_FB_HELPER) if (imxdrm->fbhelper) drm_fbdev_cma_fini(imxdrm->fbhelper); -#endif component_unbind_all(drm->dev, drm); @@ -215,11 +210,9 @@ EXPORT_SYMBOL_GPL(imx_drm_encoder_destroy); static void imx_drm_output_poll_changed(struct drm_device *drm) { -#if IS_ENABLED(CONFIG_DRM_IMX_FB_HELPER) struct imx_drm_device *imxdrm = drm->dev_private; drm_fbdev_cma_hotplug_event(imxdrm->fbhelper); -#endif } static struct drm_mode_config_funcs imx_drm_mode_config_funcs = { @@ -308,7 +301,7 @@ static int imx_drm_driver_load(struct drm_device *drm, unsigned long flags) * The fb helper takes copies of key hardware information, so the * crtcs/connectors/encoders must not change after this point. */ -#if IS_ENABLED(CONFIG_DRM_IMX_FB_HELPER) +#if IS_ENABLED(CONFIG_DRM_FBDEV_EMULATION) if (legacyfb_depth != 16 && legacyfb_depth != 32) { dev_warn(drm->dev, "Invalid legacyfb_depth. Defaulting to 16bpp\n"); legacyfb_depth = 16; @@ -340,7 +333,7 @@ err_kms: * imx_drm_add_crtc - add a new crtc */ int imx_drm_add_crtc(struct drm_device *drm, struct drm_crtc *crtc, - struct imx_drm_crtc **new_crtc, + struct imx_drm_crtc **new_crtc, struct drm_plane *primary_plane, const struct imx_drm_crtc_helper_funcs *imx_drm_helper_funcs, struct device_node *port) { @@ -379,8 +372,8 @@ int imx_drm_add_crtc(struct drm_device *drm, struct drm_crtc *crtc, drm_crtc_helper_add(crtc, imx_drm_crtc->imx_drm_helper_funcs.crtc_helper_funcs); - drm_crtc_init(drm, crtc, - imx_drm_crtc->imx_drm_helper_funcs.crtc_funcs); + drm_crtc_init_with_planes(drm, crtc, primary_plane, NULL, + imx_drm_crtc->imx_drm_helper_funcs.crtc_funcs, NULL); return 0; diff --git a/drivers/gpu/drm/imx/imx-drm.h b/drivers/gpu/drm/imx/imx-drm.h index 28e776d8d9d2..83284b4d4be1 100644 --- a/drivers/gpu/drm/imx/imx-drm.h +++ b/drivers/gpu/drm/imx/imx-drm.h @@ -9,6 +9,7 @@ struct drm_display_mode; struct drm_encoder; struct drm_fbdev_cma; struct drm_framebuffer; +struct drm_plane; struct imx_drm_crtc; struct platform_device; @@ -24,7 +25,7 @@ struct imx_drm_crtc_helper_funcs { }; int imx_drm_add_crtc(struct drm_device *drm, struct drm_crtc *crtc, - struct imx_drm_crtc **new_crtc, + struct imx_drm_crtc **new_crtc, struct drm_plane *primary_plane, const struct imx_drm_crtc_helper_funcs *imx_helper_funcs, struct device_node *port); int imx_drm_remove_crtc(struct imx_drm_crtc *); diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c index abacc8f67469..c79a61b67ded 100644 --- a/drivers/gpu/drm/imx/imx-ldb.c +++ b/drivers/gpu/drm/imx/imx-ldb.c @@ -422,7 +422,7 @@ static int imx_ldb_register(struct drm_device *drm, drm_encoder_helper_add(&imx_ldb_ch->encoder, &imx_ldb_encoder_helper_funcs); drm_encoder_init(drm, &imx_ldb_ch->encoder, &imx_ldb_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); drm_connector_helper_add(&imx_ldb_ch->connector, &imx_ldb_connector_helper_funcs); diff --git a/drivers/gpu/drm/imx/imx-tve.c b/drivers/gpu/drm/imx/imx-tve.c index e671ad369416..e61a8fca77cd 100644 --- a/drivers/gpu/drm/imx/imx-tve.c +++ b/drivers/gpu/drm/imx/imx-tve.c @@ -508,7 +508,7 @@ static int imx_tve_register(struct drm_device *drm, struct imx_tve *tve) drm_encoder_helper_add(&tve->encoder, &imx_tve_encoder_helper_funcs); drm_encoder_init(drm, &tve->encoder, &imx_tve_encoder_funcs, - encoder_type); + encoder_type, NULL); drm_connector_helper_add(&tve->connector, &imx_tve_connector_helper_funcs); @@ -721,6 +721,7 @@ static const struct of_device_id imx_tve_dt_ids[] = { { .compatible = "fsl,imx53-tve", }, { /* sentinel */ } }; +MODULE_DEVICE_TABLE(of, imx_tve_dt_ids); static struct platform_driver imx_tve_driver = { .probe = imx_tve_probe, diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c index 7bc8301fafff..4ab841eebee1 100644 --- a/drivers/gpu/drm/imx/ipuv3-crtc.c +++ b/drivers/gpu/drm/imx/ipuv3-crtc.c @@ -212,7 +212,8 @@ static void ipu_crtc_handle_pageflip(struct ipu_crtc *ipu_crtc) spin_lock_irqsave(&drm->event_lock, flags); if (ipu_crtc->page_flip_event) - drm_send_vblank_event(drm, -1, ipu_crtc->page_flip_event); + drm_crtc_send_vblank_event(&ipu_crtc->base, + ipu_crtc->page_flip_event); ipu_crtc->page_flip_event = NULL; imx_drm_crtc_vblank_put(ipu_crtc->imx_crtc); spin_unlock_irqrestore(&drm->event_lock, flags); @@ -349,7 +350,6 @@ static int ipu_crtc_init(struct ipu_crtc *ipu_crtc, struct ipu_soc *ipu = dev_get_drvdata(ipu_crtc->dev->parent); int dp = -EINVAL; int ret; - int id; ret = ipu_get_resources(ipu_crtc, pdata); if (ret) { @@ -358,18 +358,23 @@ static int ipu_crtc_init(struct ipu_crtc *ipu_crtc, return ret; } + if (pdata->dp >= 0) + dp = IPU_DP_FLOW_SYNC_BG; + ipu_crtc->plane[0] = ipu_plane_init(drm, ipu, pdata->dma[0], dp, 0, + DRM_PLANE_TYPE_PRIMARY); + if (IS_ERR(ipu_crtc->plane[0])) { + ret = PTR_ERR(ipu_crtc->plane[0]); + goto err_put_resources; + } + ret = imx_drm_add_crtc(drm, &ipu_crtc->base, &ipu_crtc->imx_crtc, - &ipu_crtc_helper_funcs, ipu_crtc->dev->of_node); + &ipu_crtc->plane[0]->base, &ipu_crtc_helper_funcs, + ipu_crtc->dev->of_node); if (ret) { dev_err(ipu_crtc->dev, "adding crtc failed with %d.\n", ret); goto err_put_resources; } - if (pdata->dp >= 0) - dp = IPU_DP_FLOW_SYNC_BG; - id = imx_drm_crtc_id(ipu_crtc->imx_crtc); - ipu_crtc->plane[0] = ipu_plane_init(ipu_crtc->base.dev, ipu, - pdata->dma[0], dp, BIT(id), true); ret = ipu_plane_get_resources(ipu_crtc->plane[0]); if (ret) { dev_err(ipu_crtc->dev, "getting plane 0 resources failed with %d.\n", @@ -379,10 +384,10 @@ static int ipu_crtc_init(struct ipu_crtc *ipu_crtc, /* If this crtc is using the DP, add an overlay plane */ if (pdata->dp >= 0 && pdata->dma[1] > 0) { - ipu_crtc->plane[1] = ipu_plane_init(ipu_crtc->base.dev, ipu, - pdata->dma[1], - IPU_DP_FLOW_SYNC_FG, - BIT(id), false); + ipu_crtc->plane[1] = ipu_plane_init(drm, ipu, pdata->dma[1], + IPU_DP_FLOW_SYNC_FG, + drm_crtc_mask(&ipu_crtc->base), + DRM_PLANE_TYPE_OVERLAY); if (IS_ERR(ipu_crtc->plane[1])) ipu_crtc->plane[1] = NULL; } @@ -407,28 +412,6 @@ err_put_resources: return ret; } -static struct device_node *ipu_drm_get_port_by_id(struct device_node *parent, - int port_id) -{ - struct device_node *port; - int id, ret; - - port = of_get_child_by_name(parent, "port"); - while (port) { - ret = of_property_read_u32(port, "reg", &id); - if (!ret && id == port_id) - return port; - - do { - port = of_get_next_child(parent, port); - if (!port) - return NULL; - } while (of_node_cmp(port->name, "port")); - } - - return NULL; -} - static int ipu_drm_bind(struct device *dev, struct device *master, void *data) { struct ipu_client_platformdata *pdata = dev->platform_data; @@ -470,23 +453,11 @@ static const struct component_ops ipu_crtc_ops = { static int ipu_drm_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; - struct ipu_client_platformdata *pdata = dev->platform_data; int ret; if (!dev->platform_data) return -EINVAL; - if (!dev->of_node) { - /* Associate crtc device with the corresponding DI port node */ - dev->of_node = ipu_drm_get_port_by_id(dev->parent->of_node, - pdata->di + 2); - if (!dev->of_node) { - dev_err(dev, "missing port@%d node in %s\n", - pdata->di + 2, dev->parent->of_node->full_name); - return -ENODEV; - } - } - ret = dma_set_coherent_mask(dev, DMA_BIT_MASK(32)); if (ret) return ret; diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c index 575f4c84388f..591ba2f1ae03 100644 --- a/drivers/gpu/drm/imx/ipuv3-plane.c +++ b/drivers/gpu/drm/imx/ipuv3-plane.c @@ -381,7 +381,7 @@ static struct drm_plane_funcs ipu_plane_funcs = { struct ipu_plane *ipu_plane_init(struct drm_device *dev, struct ipu_soc *ipu, int dma, int dp, unsigned int possible_crtcs, - bool priv) + enum drm_plane_type type) { struct ipu_plane *ipu_plane; int ret; @@ -399,10 +399,10 @@ struct ipu_plane *ipu_plane_init(struct drm_device *dev, struct ipu_soc *ipu, ipu_plane->dma = dma; ipu_plane->dp_flow = dp; - ret = drm_plane_init(dev, &ipu_plane->base, possible_crtcs, - &ipu_plane_funcs, ipu_plane_formats, - ARRAY_SIZE(ipu_plane_formats), - priv); + ret = drm_universal_plane_init(dev, &ipu_plane->base, possible_crtcs, + &ipu_plane_funcs, ipu_plane_formats, + ARRAY_SIZE(ipu_plane_formats), type, + NULL); if (ret) { DRM_ERROR("failed to initialize plane\n"); kfree(ipu_plane); diff --git a/drivers/gpu/drm/imx/ipuv3-plane.h b/drivers/gpu/drm/imx/ipuv3-plane.h index 9b5eff18f5b8..3a443b413c60 100644 --- a/drivers/gpu/drm/imx/ipuv3-plane.h +++ b/drivers/gpu/drm/imx/ipuv3-plane.h @@ -34,7 +34,7 @@ struct ipu_plane { struct ipu_plane *ipu_plane_init(struct drm_device *dev, struct ipu_soc *ipu, int dma, int dp, unsigned int possible_crtcs, - bool priv); + enum drm_plane_type type); /* Init IDMAC, DMFC, DP */ int ipu_plane_mode_set(struct ipu_plane *plane, struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c index b4deb9cf9d71..fcbe4d2eeabf 100644 --- a/drivers/gpu/drm/imx/parallel-display.c +++ b/drivers/gpu/drm/imx/parallel-display.c @@ -54,7 +54,11 @@ static int imx_pd_connector_get_modes(struct drm_connector *connector) if (imxpd->panel && imxpd->panel->funcs && imxpd->panel->funcs->get_modes) { + struct drm_display_info *di = &connector->display_info; + num_modes = imxpd->panel->funcs->get_modes(imxpd->panel); + if (!imxpd->bus_format && di->num_bus_formats) + imxpd->bus_format = di->bus_formats[0]; if (num_modes > 0) return num_modes; } @@ -188,7 +192,7 @@ static int imx_pd_register(struct drm_device *drm, drm_encoder_helper_add(&imxpd->encoder, &imx_pd_encoder_helper_funcs); drm_encoder_init(drm, &imxpd->encoder, &imx_pd_encoder_funcs, - DRM_MODE_ENCODER_NONE); + DRM_MODE_ENCODER_NONE, NULL); drm_connector_helper_add(&imxpd->connector, &imx_pd_connector_helper_funcs); diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h index 912151c36d59..205b2801d3b8 100644 --- a/drivers/gpu/drm/mgag200/mgag200_drv.h +++ b/drivers/gpu/drm/mgag200/mgag200_drv.h @@ -252,7 +252,7 @@ void mgag200_fbdev_fini(struct mga_device *mdev); /* mgag200_main.c */ int mgag200_framebuffer_init(struct drm_device *dev, struct mga_framebuffer *mfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c index b35b5b2db4ec..d9b04b008feb 100644 --- a/drivers/gpu/drm/mgag200/mgag200_fb.c +++ b/drivers/gpu/drm/mgag200/mgag200_fb.c @@ -138,7 +138,7 @@ static struct fb_ops mgag200fb_ops = { }; static int mgag200fb_create_object(struct mga_fbdev *afbdev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **gobj_p) { struct drm_device *dev = afbdev->helper.dev; diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c index b1a0f5656175..9147444d5bf2 100644 --- a/drivers/gpu/drm/mgag200/mgag200_main.c +++ b/drivers/gpu/drm/mgag200/mgag200_main.c @@ -29,7 +29,7 @@ static const struct drm_framebuffer_funcs mga_fb_funcs = { int mgag200_framebuffer_init(struct drm_device *dev, struct mga_framebuffer *gfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -47,7 +47,7 @@ int mgag200_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * mgag200_user_framebuffer_create(struct drm_device *dev, struct drm_file *filp, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct mga_framebuffer *mga_fb; diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index c99d3fe12881..31802128dfbb 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -1538,7 +1538,7 @@ static struct drm_encoder *mga_encoder_init(struct drm_device *dev) encoder->possible_crtcs = 0x1; drm_encoder_init(dev, encoder, &mga_encoder_encoder_funcs, - DRM_MODE_ENCODER_DAC); + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &mga_encoder_helper_funcs); return encoder; diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig index 84d3ec98e6b9..215495c2780c 100644 --- a/drivers/gpu/drm/msm/Kconfig +++ b/drivers/gpu/drm/msm/Kconfig @@ -54,3 +54,11 @@ config DRM_MSM_DSI_20NM_PHY default y help Choose this option if the 20nm DSI PHY is used on the platform. + +config DRM_MSM_DSI_28NM_8960_PHY + bool "Enable DSI 28nm 8960 PHY driver in MSM DRM" + depends on DRM_MSM_DSI + default y + help + Choose this option if the 28nm DSI PHY 8960 variant is used on the + platform. diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index 1c90290be716..065ad4138799 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -54,6 +54,7 @@ msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o msm-$(CONFIG_COMMON_CLK) += mdp/mdp4/mdp4_lvds_pll.o msm-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \ + mdp/mdp4/mdp4_dsi_encoder.o \ dsi/dsi_cfg.o \ dsi/dsi_host.o \ dsi/dsi_manager.o \ @@ -62,10 +63,12 @@ msm-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \ msm-$(CONFIG_DRM_MSM_DSI_28NM_PHY) += dsi/phy/dsi_phy_28nm.o msm-$(CONFIG_DRM_MSM_DSI_20NM_PHY) += dsi/phy/dsi_phy_20nm.o +msm-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/phy/dsi_phy_28nm_8960.o ifeq ($(CONFIG_DRM_MSM_DSI_PLL),y) msm-y += dsi/pll/dsi_pll.o msm-$(CONFIG_DRM_MSM_DSI_28NM_PHY) += dsi/pll/dsi_pll_28nm.o +msm-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/pll/dsi_pll_28nm_8960.o endif obj-$(CONFIG_DRM_MSM) += msm.o diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 1ea2df524fac..950d27d26b30 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -19,10 +19,6 @@ #include "adreno_gpu.h" -#if defined(DOWNSTREAM_CONFIG_MSM_BUS_SCALING) && !defined(CONFIG_OF) -# include <mach/kgsl.h> -#endif - #define ANY_ID 0xff bool hang_debug = false; @@ -168,7 +164,6 @@ static void set_gpu_pdev(struct drm_device *dev, static int adreno_bind(struct device *dev, struct device *master, void *data) { static struct adreno_platform_config config = {}; -#ifdef CONFIG_OF struct device_node *child, *node = dev->of_node; u32 val; int ret; @@ -205,53 +200,6 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) return -ENXIO; } -#else - struct kgsl_device_platform_data *pdata = dev->platform_data; - uint32_t version = socinfo_get_version(); - if (cpu_is_apq8064ab()) { - config.fast_rate = 450000000; - config.slow_rate = 27000000; - config.bus_freq = 4; - config.rev = ADRENO_REV(3, 2, 1, 0); - } else if (cpu_is_apq8064()) { - config.fast_rate = 400000000; - config.slow_rate = 27000000; - config.bus_freq = 4; - - if (SOCINFO_VERSION_MAJOR(version) == 2) - config.rev = ADRENO_REV(3, 2, 0, 2); - else if ((SOCINFO_VERSION_MAJOR(version) == 1) && - (SOCINFO_VERSION_MINOR(version) == 1)) - config.rev = ADRENO_REV(3, 2, 0, 1); - else - config.rev = ADRENO_REV(3, 2, 0, 0); - - } else if (cpu_is_msm8960ab()) { - config.fast_rate = 400000000; - config.slow_rate = 320000000; - config.bus_freq = 4; - - if (SOCINFO_VERSION_MINOR(version) == 0) - config.rev = ADRENO_REV(3, 2, 1, 0); - else - config.rev = ADRENO_REV(3, 2, 1, 1); - - } else if (cpu_is_msm8930()) { - config.fast_rate = 400000000; - config.slow_rate = 27000000; - config.bus_freq = 3; - - if ((SOCINFO_VERSION_MAJOR(version) == 1) && - (SOCINFO_VERSION_MINOR(version) == 2)) - config.rev = ADRENO_REV(3, 0, 5, 2); - else - config.rev = ADRENO_REV(3, 0, 5, 0); - - } -# ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING - config.bus_scale_table = pdata->bus_scale_table; -# endif -#endif dev->platform_data = &config; set_gpu_pdev(dev_get_drvdata(master), to_platform_device(dev)); return 0; diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h index 5f5a3732cdf6..749fbb28ec3d 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.h +++ b/drivers/gpu/drm/msm/dsi/dsi.h @@ -31,10 +31,12 @@ enum msm_dsi_phy_type { MSM_DSI_PHY_28NM_HPM, MSM_DSI_PHY_28NM_LP, MSM_DSI_PHY_20NM, + MSM_DSI_PHY_28NM_8960, MSM_DSI_PHY_MAX }; #define DSI_DEV_REGULATOR_MAX 8 +#define DSI_BUS_CLK_MAX 4 /* Regulators for DSI devices */ struct dsi_reg_entry { @@ -89,7 +91,7 @@ int msm_dsi_manager_phy_enable(int id, u32 *clk_pre, u32 *clk_post); void msm_dsi_manager_phy_disable(int id); int msm_dsi_manager_cmd_xfer(int id, const struct mipi_dsi_msg *msg); -bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 iova, u32 len); +bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 dma_base, u32 len); int msm_dsi_manager_register(struct msm_dsi *msm_dsi); void msm_dsi_manager_unregister(struct msm_dsi *msm_dsi); @@ -143,7 +145,7 @@ int msm_dsi_host_cmd_tx(struct mipi_dsi_host *host, int msm_dsi_host_cmd_rx(struct mipi_dsi_host *host, const struct mipi_dsi_msg *msg); void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host, - u32 iova, u32 len); + u32 dma_base, u32 len); int msm_dsi_host_enable(struct mipi_dsi_host *host); int msm_dsi_host_disable(struct mipi_dsi_host *host); int msm_dsi_host_power_on(struct mipi_dsi_host *host); diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c index 5872d5e5934f..2a827d8093a2 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -13,9 +13,26 @@ #include "dsi_cfg.h" -/* DSI v2 has not been supported by now */ -static const struct msm_dsi_config dsi_v2_cfg = { +static const char * const dsi_v2_bus_clk_names[] = { + "core_mmss_clk", "iface_clk", "bus_clk", +}; + +static const struct msm_dsi_config apq8064_dsi_cfg = { .io_offset = 0, + .reg_cfg = { + .num = 3, + .regs = { + {"vdda", 1200000, 1200000, 100000, 100}, + {"avdd", 3000000, 3000000, 110000, 100}, + {"vddio", 1800000, 1800000, 100000, 100}, + }, + }, + .bus_clk_names = dsi_v2_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_bus_clk_names), +}; + +static const char * const dsi_6g_bus_clk_names[] = { + "mdp_core_clk", "iface_clk", "bus_clk", "core_mmss_clk", }; static const struct msm_dsi_config msm8974_apq8084_dsi_cfg = { @@ -29,6 +46,12 @@ static const struct msm_dsi_config msm8974_apq8084_dsi_cfg = { {"vddio", 1800000, 1800000, 100000, 100}, }, }, + .bus_clk_names = dsi_6g_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names), +}; + +static const char * const dsi_8916_bus_clk_names[] = { + "mdp_core_clk", "iface_clk", "bus_clk", }; static const struct msm_dsi_config msm8916_dsi_cfg = { @@ -42,6 +65,8 @@ static const struct msm_dsi_config msm8916_dsi_cfg = { {"vddio", 1800000, 1800000, 100000, 100}, }, }, + .bus_clk_names = dsi_8916_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_8916_bus_clk_names), }; static const struct msm_dsi_config msm8994_dsi_cfg = { @@ -57,11 +82,13 @@ static const struct msm_dsi_config msm8994_dsi_cfg = { {"lab_reg", -1, -1, -1, -1}, {"ibb_reg", -1, -1, -1, -1}, }, - } + }, + .bus_clk_names = dsi_6g_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names), }; static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = { - {MSM_DSI_VER_MAJOR_V2, U32_MAX, &dsi_v2_cfg}, + {MSM_DSI_VER_MAJOR_V2, MSM_DSI_V2_VER_MINOR_8064, &apq8064_dsi_cfg}, {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0, &msm8974_apq8084_dsi_cfg}, {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1, diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h index 4cf887240177..a68c836744a3 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h @@ -25,11 +25,15 @@ #define MSM_DSI_6G_VER_MINOR_V1_3 0x10030000 #define MSM_DSI_6G_VER_MINOR_V1_3_1 0x10030001 +#define MSM_DSI_V2_VER_MINOR_8064 0x0 + #define DSI_6G_REG_SHIFT 4 struct msm_dsi_config { u32 io_offset; struct dsi_reg_config reg_cfg; + const char * const *bus_clk_names; + const int num_bus_clks; }; struct msm_dsi_cfg_handler { diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 4c49868efcda..48f9967b4a1b 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -24,26 +24,36 @@ #include <linux/of_graph.h> #include <linux/regulator/consumer.h> #include <linux/spinlock.h> +#include <linux/mfd/syscon.h> +#include <linux/regmap.h> #include <video/mipi_display.h> #include "dsi.h" #include "dsi.xml.h" +#include "sfpb.xml.h" #include "dsi_cfg.h" static int dsi_get_version(const void __iomem *base, u32 *major, u32 *minor) { u32 ver; - u32 ver_6g; if (!major || !minor) return -EINVAL; - /* From DSI6G(v3), addition of a 6G_HW_VERSION register at offset 0 + /* + * From DSI6G(v3), addition of a 6G_HW_VERSION register at offset 0 * makes all other registers 4-byte shifted down. + * + * In order to identify between DSI6G(v3) and beyond, and DSIv2 and + * older, we read the DSI_VERSION register without any shift(offset + * 0x1f0). In the case of DSIv2, this hast to be a non-zero value. In + * the case of DSI6G, this has to be zero (the offset points to a + * scratch register which we never touch) */ - ver_6g = msm_readl(base + REG_DSI_6G_HW_VERSION); - if (ver_6g == 0) { - ver = msm_readl(base + REG_DSI_VERSION); + + ver = msm_readl(base + REG_DSI_VERSION); + if (ver) { + /* older dsi host, there is no register shift */ ver = FIELD(ver, DSI_VERSION_MAJOR); if (ver <= MSM_DSI_VER_MAJOR_V2) { /* old versions */ @@ -54,12 +64,17 @@ static int dsi_get_version(const void __iomem *base, u32 *major, u32 *minor) return -EINVAL; } } else { + /* + * newer host, offset 0 has 6G_HW_VERSION, the rest of the + * registers are shifted down, read DSI_VERSION again with + * the shifted offset + */ ver = msm_readl(base + DSI_6G_REG_SHIFT + REG_DSI_VERSION); ver = FIELD(ver, DSI_VERSION_MAJOR); if (ver == MSM_DSI_VER_MAJOR_6G) { /* 6G version */ *major = ver; - *minor = ver_6g; + *minor = msm_readl(base + REG_DSI_6G_HW_VERSION); return 0; } else { return -EINVAL; @@ -91,10 +106,9 @@ struct msm_dsi_host { void __iomem *ctrl_base; struct regulator_bulk_data supplies[DSI_DEV_REGULATOR_MAX]; - struct clk *mdp_core_clk; - struct clk *ahb_clk; - struct clk *axi_clk; - struct clk *mmss_misc_ahb_clk; + + struct clk *bus_clks[DSI_BUS_CLK_MAX]; + struct clk *byte_clk; struct clk *esc_clk; struct clk *pixel_clk; @@ -102,6 +116,14 @@ struct msm_dsi_host { struct clk *pixel_clk_src; u32 byte_clk_rate; + u32 esc_clk_rate; + + /* DSI v2 specific clocks */ + struct clk *src_clk; + struct clk *esc_clk_src; + struct clk *dsi_clk_src; + + u32 src_clk_rate; struct gpio_desc *disp_en_gpio; struct gpio_desc *te_gpio; @@ -119,9 +141,19 @@ struct msm_dsi_host { struct work_struct err_work; struct workqueue_struct *workqueue; + /* DSI 6G TX buffer*/ struct drm_gem_object *tx_gem_obj; + + /* DSI v2 TX buffer */ + void *tx_buf; + dma_addr_t tx_buf_paddr; + + int tx_size; + u8 *rx_buf; + struct regmap *sfpb; + struct drm_display_mode *mode; /* connected device info */ @@ -165,21 +197,31 @@ static const struct msm_dsi_cfg_handler *dsi_get_config( struct msm_dsi_host *msm_host) { const struct msm_dsi_cfg_handler *cfg_hnd = NULL; + struct device *dev = &msm_host->pdev->dev; struct regulator *gdsc_reg; + struct clk *ahb_clk; int ret; u32 major = 0, minor = 0; - gdsc_reg = regulator_get(&msm_host->pdev->dev, "gdsc"); + gdsc_reg = regulator_get(dev, "gdsc"); if (IS_ERR(gdsc_reg)) { pr_err("%s: cannot get gdsc\n", __func__); goto exit; } + + ahb_clk = clk_get(dev, "iface_clk"); + if (IS_ERR(ahb_clk)) { + pr_err("%s: cannot get interface clock\n", __func__); + goto put_gdsc; + } + ret = regulator_enable(gdsc_reg); if (ret) { pr_err("%s: unable to enable gdsc\n", __func__); - goto put_gdsc; + goto put_clk; } - ret = clk_prepare_enable(msm_host->ahb_clk); + + ret = clk_prepare_enable(ahb_clk); if (ret) { pr_err("%s: unable to enable ahb_clk\n", __func__); goto disable_gdsc; @@ -196,9 +238,11 @@ static const struct msm_dsi_cfg_handler *dsi_get_config( DBG("%s: Version %x:%x\n", __func__, major, minor); disable_clks: - clk_disable_unprepare(msm_host->ahb_clk); + clk_disable_unprepare(ahb_clk); disable_gdsc: regulator_disable(gdsc_reg); +put_clk: + clk_put(ahb_clk); put_gdsc: regulator_put(gdsc_reg); exit: @@ -295,40 +339,23 @@ static int dsi_regulator_init(struct msm_dsi_host *msm_host) static int dsi_clk_init(struct msm_dsi_host *msm_host) { struct device *dev = &msm_host->pdev->dev; - int ret = 0; - - msm_host->mdp_core_clk = devm_clk_get(dev, "mdp_core_clk"); - if (IS_ERR(msm_host->mdp_core_clk)) { - ret = PTR_ERR(msm_host->mdp_core_clk); - pr_err("%s: Unable to get mdp core clk. ret=%d\n", - __func__, ret); - goto exit; - } - - msm_host->ahb_clk = devm_clk_get(dev, "iface_clk"); - if (IS_ERR(msm_host->ahb_clk)) { - ret = PTR_ERR(msm_host->ahb_clk); - pr_err("%s: Unable to get mdss ahb clk. ret=%d\n", - __func__, ret); - goto exit; - } - - msm_host->axi_clk = devm_clk_get(dev, "bus_clk"); - if (IS_ERR(msm_host->axi_clk)) { - ret = PTR_ERR(msm_host->axi_clk); - pr_err("%s: Unable to get axi bus clk. ret=%d\n", - __func__, ret); - goto exit; - } - - msm_host->mmss_misc_ahb_clk = devm_clk_get(dev, "core_mmss_clk"); - if (IS_ERR(msm_host->mmss_misc_ahb_clk)) { - ret = PTR_ERR(msm_host->mmss_misc_ahb_clk); - pr_err("%s: Unable to get mmss misc ahb clk. ret=%d\n", - __func__, ret); - goto exit; + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; + const struct msm_dsi_config *cfg = cfg_hnd->cfg; + int i, ret = 0; + + /* get bus clocks */ + for (i = 0; i < cfg->num_bus_clks; i++) { + msm_host->bus_clks[i] = devm_clk_get(dev, + cfg->bus_clk_names[i]); + if (IS_ERR(msm_host->bus_clks[i])) { + ret = PTR_ERR(msm_host->bus_clks[i]); + pr_err("%s: Unable to get %s, ret = %d\n", + __func__, cfg->bus_clk_names[i], ret); + goto exit; + } } + /* get link and source clocks */ msm_host->byte_clk = devm_clk_get(dev, "byte_clk"); if (IS_ERR(msm_host->byte_clk)) { ret = PTR_ERR(msm_host->byte_clk); @@ -356,80 +383,85 @@ static int dsi_clk_init(struct msm_dsi_host *msm_host) goto exit; } - msm_host->byte_clk_src = devm_clk_get(dev, "byte_clk_src"); - if (IS_ERR(msm_host->byte_clk_src)) { - ret = PTR_ERR(msm_host->byte_clk_src); + msm_host->byte_clk_src = clk_get_parent(msm_host->byte_clk); + if (!msm_host->byte_clk_src) { + ret = -ENODEV; pr_err("%s: can't find byte_clk_src. ret=%d\n", __func__, ret); - msm_host->byte_clk_src = NULL; goto exit; } - msm_host->pixel_clk_src = devm_clk_get(dev, "pixel_clk_src"); - if (IS_ERR(msm_host->pixel_clk_src)) { - ret = PTR_ERR(msm_host->pixel_clk_src); + msm_host->pixel_clk_src = clk_get_parent(msm_host->pixel_clk); + if (!msm_host->pixel_clk_src) { + ret = -ENODEV; pr_err("%s: can't find pixel_clk_src. ret=%d\n", __func__, ret); - msm_host->pixel_clk_src = NULL; goto exit; } + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_V2) { + msm_host->src_clk = devm_clk_get(dev, "src_clk"); + if (IS_ERR(msm_host->src_clk)) { + ret = PTR_ERR(msm_host->src_clk); + pr_err("%s: can't find dsi_src_clk. ret=%d\n", + __func__, ret); + msm_host->src_clk = NULL; + goto exit; + } + + msm_host->esc_clk_src = clk_get_parent(msm_host->esc_clk); + if (!msm_host->esc_clk_src) { + ret = -ENODEV; + pr_err("%s: can't get esc_clk_src. ret=%d\n", + __func__, ret); + goto exit; + } + + msm_host->dsi_clk_src = clk_get_parent(msm_host->src_clk); + if (!msm_host->dsi_clk_src) { + ret = -ENODEV; + pr_err("%s: can't get dsi_clk_src. ret=%d\n", + __func__, ret); + } + } exit: return ret; } static int dsi_bus_clk_enable(struct msm_dsi_host *msm_host) { - int ret; + const struct msm_dsi_config *cfg = msm_host->cfg_hnd->cfg; + int i, ret; DBG("id=%d", msm_host->id); - ret = clk_prepare_enable(msm_host->mdp_core_clk); - if (ret) { - pr_err("%s: failed to enable mdp_core_clock, %d\n", - __func__, ret); - goto core_clk_err; - } - - ret = clk_prepare_enable(msm_host->ahb_clk); - if (ret) { - pr_err("%s: failed to enable ahb clock, %d\n", __func__, ret); - goto ahb_clk_err; - } - - ret = clk_prepare_enable(msm_host->axi_clk); - if (ret) { - pr_err("%s: failed to enable ahb clock, %d\n", __func__, ret); - goto axi_clk_err; - } - - ret = clk_prepare_enable(msm_host->mmss_misc_ahb_clk); - if (ret) { - pr_err("%s: failed to enable mmss misc ahb clk, %d\n", - __func__, ret); - goto misc_ahb_clk_err; + for (i = 0; i < cfg->num_bus_clks; i++) { + ret = clk_prepare_enable(msm_host->bus_clks[i]); + if (ret) { + pr_err("%s: failed to enable bus clock %d ret %d\n", + __func__, i, ret); + goto err; + } } return 0; +err: + for (; i > 0; i--) + clk_disable_unprepare(msm_host->bus_clks[i]); -misc_ahb_clk_err: - clk_disable_unprepare(msm_host->axi_clk); -axi_clk_err: - clk_disable_unprepare(msm_host->ahb_clk); -ahb_clk_err: - clk_disable_unprepare(msm_host->mdp_core_clk); -core_clk_err: return ret; } static void dsi_bus_clk_disable(struct msm_dsi_host *msm_host) { + const struct msm_dsi_config *cfg = msm_host->cfg_hnd->cfg; + int i; + DBG(""); - clk_disable_unprepare(msm_host->mmss_misc_ahb_clk); - clk_disable_unprepare(msm_host->axi_clk); - clk_disable_unprepare(msm_host->ahb_clk); - clk_disable_unprepare(msm_host->mdp_core_clk); + + for (i = cfg->num_bus_clks - 1; i >= 0; i--) + clk_disable_unprepare(msm_host->bus_clks[i]); } -static int dsi_link_clk_enable(struct msm_dsi_host *msm_host) +static int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host) { int ret; @@ -476,11 +508,98 @@ error: return ret; } -static void dsi_link_clk_disable(struct msm_dsi_host *msm_host) +static int dsi_link_clk_enable_v2(struct msm_dsi_host *msm_host) { + int ret; + + DBG("Set clk rates: pclk=%d, byteclk=%d, esc_clk=%d, dsi_src_clk=%d", + msm_host->mode->clock, msm_host->byte_clk_rate, + msm_host->esc_clk_rate, msm_host->src_clk_rate); + + ret = clk_set_rate(msm_host->byte_clk, msm_host->byte_clk_rate); + if (ret) { + pr_err("%s: Failed to set rate byte clk, %d\n", __func__, ret); + goto error; + } + + ret = clk_set_rate(msm_host->esc_clk, msm_host->esc_clk_rate); + if (ret) { + pr_err("%s: Failed to set rate esc clk, %d\n", __func__, ret); + goto error; + } + + ret = clk_set_rate(msm_host->src_clk, msm_host->src_clk_rate); + if (ret) { + pr_err("%s: Failed to set rate src clk, %d\n", __func__, ret); + goto error; + } + + ret = clk_set_rate(msm_host->pixel_clk, msm_host->mode->clock * 1000); + if (ret) { + pr_err("%s: Failed to set rate pixel clk, %d\n", __func__, ret); + goto error; + } + + ret = clk_prepare_enable(msm_host->byte_clk); + if (ret) { + pr_err("%s: Failed to enable dsi byte clk\n", __func__); + goto error; + } + + ret = clk_prepare_enable(msm_host->esc_clk); + if (ret) { + pr_err("%s: Failed to enable dsi esc clk\n", __func__); + goto esc_clk_err; + } + + ret = clk_prepare_enable(msm_host->src_clk); + if (ret) { + pr_err("%s: Failed to enable dsi src clk\n", __func__); + goto src_clk_err; + } + + ret = clk_prepare_enable(msm_host->pixel_clk); + if (ret) { + pr_err("%s: Failed to enable dsi pixel clk\n", __func__); + goto pixel_clk_err; + } + + return 0; + +pixel_clk_err: + clk_disable_unprepare(msm_host->src_clk); +src_clk_err: clk_disable_unprepare(msm_host->esc_clk); - clk_disable_unprepare(msm_host->pixel_clk); +esc_clk_err: clk_disable_unprepare(msm_host->byte_clk); +error: + return ret; +} + +static int dsi_link_clk_enable(struct msm_dsi_host *msm_host) +{ + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; + + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) + return dsi_link_clk_enable_6g(msm_host); + else + return dsi_link_clk_enable_v2(msm_host); +} + +static void dsi_link_clk_disable(struct msm_dsi_host *msm_host) +{ + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; + + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) { + clk_disable_unprepare(msm_host->esc_clk); + clk_disable_unprepare(msm_host->pixel_clk); + clk_disable_unprepare(msm_host->byte_clk); + } else { + clk_disable_unprepare(msm_host->pixel_clk); + clk_disable_unprepare(msm_host->src_clk); + clk_disable_unprepare(msm_host->esc_clk); + clk_disable_unprepare(msm_host->byte_clk); + } } static int dsi_clk_ctrl(struct msm_dsi_host *msm_host, bool enable) @@ -515,6 +634,7 @@ unlock_ret: static int dsi_calc_clk_rate(struct msm_dsi_host *msm_host) { struct drm_display_mode *mode = msm_host->mode; + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; u8 lanes = msm_host->lanes; u32 bpp = dsi_get_bpp(msm_host->format); u32 pclk_rate; @@ -534,6 +654,47 @@ static int dsi_calc_clk_rate(struct msm_dsi_host *msm_host) DBG("pclk=%d, bclk=%d", pclk_rate, msm_host->byte_clk_rate); + msm_host->esc_clk_rate = clk_get_rate(msm_host->esc_clk); + + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_V2) { + unsigned int esc_mhz, esc_div; + unsigned long byte_mhz; + + msm_host->src_clk_rate = (pclk_rate * bpp) / 8; + + /* + * esc clock is byte clock followed by a 4 bit divider, + * we need to find an escape clock frequency within the + * mipi DSI spec range within the maximum divider limit + * We iterate here between an escape clock frequencey + * between 20 Mhz to 5 Mhz and pick up the first one + * that can be supported by our divider + */ + + byte_mhz = msm_host->byte_clk_rate / 1000000; + + for (esc_mhz = 20; esc_mhz >= 5; esc_mhz--) { + esc_div = DIV_ROUND_UP(byte_mhz, esc_mhz); + + /* + * TODO: Ideally, we shouldn't know what sort of divider + * is available in mmss_cc, we're just assuming that + * it'll always be a 4 bit divider. Need to come up with + * a better way here. + */ + if (esc_div >= 1 && esc_div <= 16) + break; + } + + if (esc_mhz < 5) + return -EINVAL; + + msm_host->esc_clk_rate = msm_host->byte_clk_rate / esc_div; + + DBG("esc=%d, src=%d", msm_host->esc_clk_rate, + msm_host->src_clk_rate); + } + return 0; } @@ -835,29 +996,46 @@ static void dsi_wait4video_eng_busy(struct msm_dsi_host *msm_host) static int dsi_tx_buf_alloc(struct msm_dsi_host *msm_host, int size) { struct drm_device *dev = msm_host->dev; + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; int ret; u32 iova; - mutex_lock(&dev->struct_mutex); - msm_host->tx_gem_obj = msm_gem_new(dev, size, MSM_BO_UNCACHED); - if (IS_ERR(msm_host->tx_gem_obj)) { - ret = PTR_ERR(msm_host->tx_gem_obj); - pr_err("%s: failed to allocate gem, %d\n", __func__, ret); - msm_host->tx_gem_obj = NULL; + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) { + mutex_lock(&dev->struct_mutex); + msm_host->tx_gem_obj = msm_gem_new(dev, size, MSM_BO_UNCACHED); + if (IS_ERR(msm_host->tx_gem_obj)) { + ret = PTR_ERR(msm_host->tx_gem_obj); + pr_err("%s: failed to allocate gem, %d\n", + __func__, ret); + msm_host->tx_gem_obj = NULL; + mutex_unlock(&dev->struct_mutex); + return ret; + } + + ret = msm_gem_get_iova_locked(msm_host->tx_gem_obj, 0, &iova); mutex_unlock(&dev->struct_mutex); - return ret; - } + if (ret) { + pr_err("%s: failed to get iova, %d\n", __func__, ret); + return ret; + } - ret = msm_gem_get_iova_locked(msm_host->tx_gem_obj, 0, &iova); - if (ret) { - pr_err("%s: failed to get iova, %d\n", __func__, ret); - return ret; - } - mutex_unlock(&dev->struct_mutex); + if (iova & 0x07) { + pr_err("%s: buf NOT 8 bytes aligned\n", __func__); + return -EINVAL; + } - if (iova & 0x07) { - pr_err("%s: buf NOT 8 bytes aligned\n", __func__); - return -EINVAL; + msm_host->tx_size = msm_host->tx_gem_obj->size; + } else { + msm_host->tx_buf = dma_alloc_coherent(dev->dev, size, + &msm_host->tx_buf_paddr, GFP_KERNEL); + if (!msm_host->tx_buf) { + ret = -ENOMEM; + pr_err("%s: failed to allocate tx buf, %d\n", + __func__, ret); + return ret; + } + + msm_host->tx_size = size; } return 0; @@ -874,14 +1052,19 @@ static void dsi_tx_buf_free(struct msm_dsi_host *msm_host) msm_host->tx_gem_obj = NULL; mutex_unlock(&dev->struct_mutex); } + + if (msm_host->tx_buf) + dma_free_coherent(dev->dev, msm_host->tx_size, msm_host->tx_buf, + msm_host->tx_buf_paddr); } /* * prepare cmd buffer to be txed */ -static int dsi_cmd_dma_add(struct drm_gem_object *tx_gem, - const struct mipi_dsi_msg *msg) +static int dsi_cmd_dma_add(struct msm_dsi_host *msm_host, + const struct mipi_dsi_msg *msg) { + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; struct mipi_dsi_packet packet; int len; int ret; @@ -894,17 +1077,20 @@ static int dsi_cmd_dma_add(struct drm_gem_object *tx_gem, } len = (packet.size + 3) & (~0x3); - if (len > tx_gem->size) { + if (len > msm_host->tx_size) { pr_err("%s: packet size is too big\n", __func__); return -EINVAL; } - data = msm_gem_vaddr(tx_gem); - - if (IS_ERR(data)) { - ret = PTR_ERR(data); - pr_err("%s: get vaddr failed, %d\n", __func__, ret); - return ret; + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) { + data = msm_gem_vaddr(msm_host->tx_gem_obj); + if (IS_ERR(data)) { + ret = PTR_ERR(data); + pr_err("%s: get vaddr failed, %d\n", __func__, ret); + return ret; + } + } else { + data = msm_host->tx_buf; } /* MSM specific command format in memory */ @@ -970,17 +1156,21 @@ static int dsi_long_read_resp(u8 *buf, const struct mipi_dsi_msg *msg) return msg->rx_len; } - static int dsi_cmd_dma_tx(struct msm_dsi_host *msm_host, int len) { + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; int ret; - u32 iova; + u32 dma_base; bool triggered; - ret = msm_gem_get_iova(msm_host->tx_gem_obj, 0, &iova); - if (ret) { - pr_err("%s: failed to get iova: %d\n", __func__, ret); - return ret; + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) { + ret = msm_gem_get_iova(msm_host->tx_gem_obj, 0, &dma_base); + if (ret) { + pr_err("%s: failed to get iova: %d\n", __func__, ret); + return ret; + } + } else { + dma_base = msm_host->tx_buf_paddr; } reinit_completion(&msm_host->dma_comp); @@ -988,7 +1178,7 @@ static int dsi_cmd_dma_tx(struct msm_dsi_host *msm_host, int len) dsi_wait4video_eng_busy(msm_host); triggered = msm_dsi_manager_cmd_xfer_trigger( - msm_host->id, iova, len); + msm_host->id, dma_base, len); if (triggered) { ret = wait_for_completion_timeout(&msm_host->dma_comp, msecs_to_jiffies(200)); @@ -1060,7 +1250,7 @@ static int dsi_cmds2buf_tx(struct msm_dsi_host *msm_host, int bllp_len = msm_host->mode->hdisplay * dsi_get_bpp(msm_host->format) / 8; - len = dsi_cmd_dma_add(msm_host->tx_gem_obj, msg); + len = dsi_cmd_dma_add(msm_host, msg); if (!len) { pr_err("%s: failed to add cmd type = 0x%x\n", __func__, msg->type); @@ -1383,6 +1573,16 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host) msm_host->device_node = device_node; + if (of_property_read_bool(np, "syscon-sfpb")) { + msm_host->sfpb = syscon_regmap_lookup_by_phandle(np, + "syscon-sfpb"); + if (IS_ERR(msm_host->sfpb)) { + dev_err(dev, "%s: failed to get sfpb regmap\n", + __func__); + return PTR_ERR(msm_host->sfpb); + } + } + return 0; } @@ -1408,12 +1608,6 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi) goto fail; } - ret = dsi_clk_init(msm_host); - if (ret) { - pr_err("%s: unable to initialize dsi clks\n", __func__); - goto fail; - } - msm_host->ctrl_base = msm_ioremap(pdev, "dsi_ctrl", "DSI CTRL"); if (IS_ERR(msm_host->ctrl_base)) { pr_err("%s: unable to map Dsi ctrl base\n", __func__); @@ -1437,6 +1631,12 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi) goto fail; } + ret = dsi_clk_init(msm_host); + if (ret) { + pr_err("%s: unable to initialize dsi clks\n", __func__); + goto fail; + } + msm_host->rx_buf = devm_kzalloc(&pdev->dev, SZ_4K, GFP_KERNEL); if (!msm_host->rx_buf) { pr_err("%s: alloc rx temp buf failed\n", __func__); @@ -1750,11 +1950,12 @@ int msm_dsi_host_cmd_rx(struct mipi_dsi_host *host, return ret; } -void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host, u32 iova, u32 len) +void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host, u32 dma_base, + u32 len) { struct msm_dsi_host *msm_host = to_msm_dsi_host(host); - dsi_write(msm_host, REG_DSI_DMA_BASE, iova); + dsi_write(msm_host, REG_DSI_DMA_BASE, dma_base); dsi_write(msm_host, REG_DSI_DMA_LEN, len); dsi_write(msm_host, REG_DSI_TRIG_DMA, 1); @@ -1766,6 +1967,7 @@ int msm_dsi_host_set_src_pll(struct mipi_dsi_host *host, struct msm_dsi_pll *src_pll) { struct msm_dsi_host *msm_host = to_msm_dsi_host(host); + const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; struct clk *byte_clk_provider, *pixel_clk_provider; int ret; @@ -1791,6 +1993,22 @@ int msm_dsi_host_set_src_pll(struct mipi_dsi_host *host, goto exit; } + if (cfg_hnd->major == MSM_DSI_VER_MAJOR_V2) { + ret = clk_set_parent(msm_host->dsi_clk_src, pixel_clk_provider); + if (ret) { + pr_err("%s: can't set parent to dsi_clk_src. ret=%d\n", + __func__, ret); + goto exit; + } + + ret = clk_set_parent(msm_host->esc_clk_src, byte_clk_provider); + if (ret) { + pr_err("%s: can't set parent to esc_clk_src. ret=%d\n", + __func__, ret); + goto exit; + } + } + exit: return ret; } @@ -1828,6 +2046,20 @@ int msm_dsi_host_disable(struct mipi_dsi_host *host) return 0; } +static void msm_dsi_sfpb_config(struct msm_dsi_host *msm_host, bool enable) +{ + enum sfpb_ahb_arb_master_port_en en; + + if (!msm_host->sfpb) + return; + + en = enable ? SFPB_MASTER_PORT_ENABLE : SFPB_MASTER_PORT_DISABLE; + + regmap_update_bits(msm_host->sfpb, REG_SFPB_GPREG, + SFPB_GPREG_MASTER_PORT_EN__MASK, + SFPB_GPREG_MASTER_PORT_EN(en)); +} + int msm_dsi_host_power_on(struct mipi_dsi_host *host) { struct msm_dsi_host *msm_host = to_msm_dsi_host(host); @@ -1840,6 +2072,8 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host) goto unlock_ret; } + msm_dsi_sfpb_config(msm_host, true); + ret = dsi_calc_clk_rate(msm_host); if (ret) { pr_err("%s: unable to calc clk rate, %d\n", __func__, ret); @@ -1862,7 +2096,7 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host) dsi_phy_sw_reset(msm_host); ret = msm_dsi_manager_phy_enable(msm_host->id, msm_host->byte_clk_rate * 8, - clk_get_rate(msm_host->esc_clk), + msm_host->esc_clk_rate, &clk_pre, &clk_post); dsi_bus_clk_disable(msm_host); if (ret) { @@ -1927,6 +2161,8 @@ int msm_dsi_host_power_off(struct mipi_dsi_host *host) dsi_host_regulator_disable(msm_host); + msm_dsi_sfpb_config(msm_host, false); + DBG("-"); msm_host->power_on = false; diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c index 0455ff75074a..58ba7ec17f51 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_manager.c +++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c @@ -774,7 +774,7 @@ restore_host0: return ret; } -bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 iova, u32 len) +bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 dma_base, u32 len) { struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id); struct msm_dsi *msm_dsi0 = dsi_mgr_get_dsi(DSI_0); @@ -784,9 +784,9 @@ bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 iova, u32 len) return false; if (IS_SYNC_NEEDED() && msm_dsi0) - msm_dsi_host_cmd_xfer_commit(msm_dsi0->host, iova, len); + msm_dsi_host_cmd_xfer_commit(msm_dsi0->host, dma_base, len); - msm_dsi_host_cmd_xfer_commit(host, iova, len); + msm_dsi_host_cmd_xfer_commit(host, dma_base, len); return true; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index f1f955f571fa..91a95fb04a4a 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -277,6 +277,10 @@ static const struct of_device_id dsi_phy_dt_match[] = { { .compatible = "qcom,dsi-phy-20nm", .data = &dsi_phy_20nm_cfgs }, #endif +#ifdef CONFIG_DRM_MSM_DSI_28NM_8960_PHY + { .compatible = "qcom,dsi-phy-28nm-8960", + .data = &dsi_phy_28nm_8960_cfgs }, +#endif {} }; diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h index 0456b253239f..0d54ed00386d 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h @@ -43,6 +43,7 @@ struct msm_dsi_phy_cfg { extern const struct msm_dsi_phy_cfg dsi_phy_28nm_hpm_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_28nm_lp_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_20nm_cfgs; +extern const struct msm_dsi_phy_cfg dsi_phy_28nm_8960_cfgs; struct msm_dsi_dphy_timing { u32 clk_pre; diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c new file mode 100644 index 000000000000..197b039ca1f1 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c @@ -0,0 +1,195 @@ +/* + * Copyright (c) 2012-2015, The Linux Foundation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 and + * only version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include "dsi_phy.h" +#include "dsi.xml.h" + +static void dsi_28nm_dphy_set_timing(struct msm_dsi_phy *phy, + struct msm_dsi_dphy_timing *timing) +{ + void __iomem *base = phy->base; + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_0, + DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO(timing->clk_zero)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_1, + DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL(timing->clk_trail)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_2, + DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE(timing->clk_prepare)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_3, 0x0); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_4, + DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT(timing->hs_exit)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_5, + DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO(timing->hs_zero)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_6, + DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE(timing->hs_prepare)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_7, + DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL(timing->hs_trail)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_8, + DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST(timing->hs_rqst)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_9, + DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO(timing->ta_go) | + DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE(timing->ta_sure)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_10, + DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET(timing->ta_get)); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_11, + DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD(0)); +} + +static void dsi_28nm_phy_regulator_init(struct msm_dsi_phy *phy) +{ + void __iomem *base = phy->reg_base; + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_0, 0x3); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_1, 1); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_2, 1); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_3, 0); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_4, + 0x100); +} + +static void dsi_28nm_phy_regulator_ctrl(struct msm_dsi_phy *phy) +{ + void __iomem *base = phy->reg_base; + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_0, 0x3); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_1, 0xa); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_2, 0x4); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_3, 0x0); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_4, 0x20); +} + +static void dsi_28nm_phy_calibration(struct msm_dsi_phy *phy) +{ + void __iomem *base = phy->reg_base; + u32 status; + int i = 5000; + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CAL_PWR_CFG, + 0x3); + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_SW_CFG_2, 0x0); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_1, 0x5a); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_3, 0x10); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_4, 0x1); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_0, 0x1); + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_TRIGGER, 0x1); + usleep_range(5000, 6000); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_TRIGGER, 0x0); + + do { + status = dsi_phy_read(base + + REG_DSI_28nm_8960_PHY_MISC_CAL_STATUS); + + if (!(status & DSI_28nm_8960_PHY_MISC_CAL_STATUS_CAL_BUSY)) + break; + + udelay(1); + } while (--i > 0); +} + +static void dsi_28nm_phy_lane_config(struct msm_dsi_phy *phy) +{ + void __iomem *base = phy->base; + int i; + + for (i = 0; i < 4; i++) { + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_CFG_0(i), 0x80); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_CFG_1(i), 0x45); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_CFG_2(i), 0x00); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_TEST_DATAPATH(i), + 0x00); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_TEST_STR_0(i), + 0x01); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_TEST_STR_1(i), + 0x66); + } + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_CFG_0, 0x40); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_CFG_1, 0x67); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_CFG_2, 0x0); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_TEST_DATAPATH, 0x0); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_TEST_STR0, 0x1); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_TEST_STR1, 0x88); +} + +static int dsi_28nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, + const unsigned long bit_rate, const unsigned long esc_rate) +{ + struct msm_dsi_dphy_timing *timing = &phy->timing; + void __iomem *base = phy->base; + + DBG(""); + + if (msm_dsi_dphy_timing_calc(timing, bit_rate, esc_rate)) { + dev_err(&phy->pdev->dev, + "%s: D-PHY timing calculation failed\n", __func__); + return -EINVAL; + } + + dsi_28nm_phy_regulator_init(phy); + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LDO_CTRL, 0x04); + + /* strength control */ + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_STRENGTH_0, 0xff); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_STRENGTH_1, 0x00); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_STRENGTH_2, 0x06); + + /* phy ctrl */ + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_0, 0x5f); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_1, 0x00); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_2, 0x00); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_3, 0x10); + + dsi_28nm_phy_regulator_ctrl(phy); + + dsi_28nm_phy_calibration(phy); + + dsi_28nm_phy_lane_config(phy); + + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_4, 0x0f); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_1, 0x03); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_0, 0x03); + dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_4, 0x0); + + dsi_28nm_dphy_set_timing(phy, timing); + + return 0; +} + +static void dsi_28nm_phy_disable(struct msm_dsi_phy *phy) +{ + dsi_phy_write(phy->base + REG_DSI_28nm_8960_PHY_CTRL_0, 0x0); + + /* + * Wait for the registers writes to complete in order to + * ensure that the phy is completely disabled + */ + wmb(); +} + +const struct msm_dsi_phy_cfg dsi_phy_28nm_8960_cfgs = { + .type = MSM_DSI_PHY_28NM_8960, + .src_pll_truthtable = { {true, true}, {false, true} }, + .reg_cfg = { + .num = 1, + .regs = { + {"vddio", 1800000, 1800000, 100000, 100}, + }, + }, + .ops = { + .enable = dsi_28nm_phy_enable, + .disable = dsi_28nm_phy_disable, + }, +}; diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c index 5104fc9f9a53..5cd438f91afe 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c @@ -151,6 +151,9 @@ struct msm_dsi_pll *msm_dsi_pll_init(struct platform_device *pdev, case MSM_DSI_PHY_28NM_LP: pll = msm_dsi_pll_28nm_init(pdev, type, id); break; + case MSM_DSI_PHY_28NM_8960: + pll = msm_dsi_pll_28nm_8960_init(pdev, id); + break; default: pll = ERR_PTR(-ENXIO); break; diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h index 063caa2c5740..80b6038334a6 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h @@ -93,6 +93,16 @@ static inline struct msm_dsi_pll *msm_dsi_pll_28nm_init( return ERR_PTR(-ENODEV); } #endif +#ifdef CONFIG_DRM_MSM_DSI_28NM_8960_PHY +struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev, + int id); +#else +struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev, + int id) +{ + return ERR_PTR(-ENODEV); +} +#endif #endif /* __DSI_PLL_H__ */ diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c new file mode 100644 index 000000000000..38c90e1eb002 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c @@ -0,0 +1,533 @@ +/* + * Copyright (c) 2012-2015, The Linux Foundation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 and + * only version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/clk-provider.h> + +#include "dsi_pll.h" +#include "dsi.xml.h" + +/* + * DSI PLL 28nm (8960/A family) - clock diagram (eg: DSI1): + * + * + * +------+ + * dsi1vco_clk ----o-----| DIV1 |---dsi1pllbit (not exposed as clock) + * F * byte_clk | +------+ + * | bit clock divider (F / 8) + * | + * | +------+ + * o-----| DIV2 |---dsi0pllbyte---o---> To byte RCG + * | +------+ | (sets parent rate) + * | byte clock divider (F) | + * | | + * | o---> To esc RCG + * | (doesn't set parent rate) + * | + * | +------+ + * o-----| DIV3 |----dsi0pll------o---> To dsi RCG + * +------+ | (sets parent rate) + * dsi clock divider (F * magic) | + * | + * o---> To pixel rcg + * (doesn't set parent rate) + */ + +#define POLL_MAX_READS 8000 +#define POLL_TIMEOUT_US 1 + +#define NUM_PROVIDED_CLKS 2 + +#define VCO_REF_CLK_RATE 27000000 +#define VCO_MIN_RATE 600000000 +#define VCO_MAX_RATE 1200000000 + +#define DSI_BYTE_PLL_CLK 0 +#define DSI_PIXEL_PLL_CLK 1 + +#define VCO_PREF_DIV_RATIO 27 + +struct pll_28nm_cached_state { + unsigned long vco_rate; + u8 postdiv3; + u8 postdiv2; + u8 postdiv1; +}; + +struct clk_bytediv { + struct clk_hw hw; + void __iomem *reg; +}; + +struct dsi_pll_28nm { + struct msm_dsi_pll base; + + int id; + struct platform_device *pdev; + void __iomem *mmio; + + /* custom byte clock divider */ + struct clk_bytediv *bytediv; + + /* private clocks: */ + struct clk *clks[NUM_DSI_CLOCKS_MAX]; + u32 num_clks; + + /* clock-provider: */ + struct clk *provided_clks[NUM_PROVIDED_CLKS]; + struct clk_onecell_data clk_data; + + struct pll_28nm_cached_state cached_state; +}; + +#define to_pll_28nm(x) container_of(x, struct dsi_pll_28nm, base) + +static bool pll_28nm_poll_for_ready(struct dsi_pll_28nm *pll_28nm, + int nb_tries, int timeout_us) +{ + bool pll_locked = false; + u32 val; + + while (nb_tries--) { + val = pll_read(pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_RDY); + pll_locked = !!(val & DSI_28nm_8960_PHY_PLL_RDY_PLL_RDY); + + if (pll_locked) + break; + + udelay(timeout_us); + } + DBG("DSI PLL is %slocked", pll_locked ? "" : "*not* "); + + return pll_locked; +} + +/* + * Clock Callbacks + */ +static int dsi_pll_28nm_clk_set_rate(struct clk_hw *hw, unsigned long rate, + unsigned long parent_rate) +{ + struct msm_dsi_pll *pll = hw_clk_to_pll(hw); + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + void __iomem *base = pll_28nm->mmio; + u32 val, temp, fb_divider; + + DBG("rate=%lu, parent's=%lu", rate, parent_rate); + + temp = rate / 10; + val = VCO_REF_CLK_RATE / 10; + fb_divider = (temp * VCO_PREF_DIV_RATIO) / val; + fb_divider = fb_divider / 2 - 1; + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_1, + fb_divider & 0xff); + + val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_2); + + val |= (fb_divider >> 8) & 0x07; + + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_2, + val); + + val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_3); + + val |= (VCO_PREF_DIV_RATIO - 1) & 0x3f; + + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_3, + val); + + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_6, + 0xf); + + val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8); + val |= 0x7 << 4; + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8, + val); + + return 0; +} + +static int dsi_pll_28nm_clk_is_enabled(struct clk_hw *hw) +{ + struct msm_dsi_pll *pll = hw_clk_to_pll(hw); + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + + return pll_28nm_poll_for_ready(pll_28nm, POLL_MAX_READS, + POLL_TIMEOUT_US); +} + +static unsigned long dsi_pll_28nm_clk_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + struct msm_dsi_pll *pll = hw_clk_to_pll(hw); + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + void __iomem *base = pll_28nm->mmio; + unsigned long vco_rate; + u32 status, fb_divider, temp, ref_divider; + + VERB("parent_rate=%lu", parent_rate); + + status = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_0); + + if (status & DSI_28nm_8960_PHY_PLL_CTRL_0_ENABLE) { + fb_divider = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_1); + fb_divider &= 0xff; + temp = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_2) & 0x07; + fb_divider = (temp << 8) | fb_divider; + fb_divider += 1; + + ref_divider = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_3); + ref_divider &= 0x3f; + ref_divider += 1; + + /* multiply by 2 */ + vco_rate = (parent_rate / ref_divider) * fb_divider * 2; + } else { + vco_rate = 0; + } + + DBG("returning vco rate = %lu", vco_rate); + + return vco_rate; +} + +static const struct clk_ops clk_ops_dsi_pll_28nm_vco = { + .round_rate = msm_dsi_pll_helper_clk_round_rate, + .set_rate = dsi_pll_28nm_clk_set_rate, + .recalc_rate = dsi_pll_28nm_clk_recalc_rate, + .prepare = msm_dsi_pll_helper_clk_prepare, + .unprepare = msm_dsi_pll_helper_clk_unprepare, + .is_enabled = dsi_pll_28nm_clk_is_enabled, +}; + +/* + * Custom byte clock divier clk_ops + * + * This clock is the entry point to configuring the PLL. The user (dsi host) + * will set this clock's rate to the desired byte clock rate. The VCO lock + * frequency is a multiple of the byte clock rate. The multiplication factor + * (shown as F in the diagram above) is a function of the byte clock rate. + * + * This custom divider clock ensures that its parent (VCO) is set to the + * desired rate, and that the byte clock postdivider (POSTDIV2) is configured + * accordingly + */ +#define to_clk_bytediv(_hw) container_of(_hw, struct clk_bytediv, hw) + +static unsigned long clk_bytediv_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + struct clk_bytediv *bytediv = to_clk_bytediv(hw); + unsigned int div; + + div = pll_read(bytediv->reg) & 0xff; + + return parent_rate / (div + 1); +} + +/* find multiplication factor(wrt byte clock) at which the VCO should be set */ +static unsigned int get_vco_mul_factor(unsigned long byte_clk_rate) +{ + unsigned long bit_mhz; + + /* convert to bit clock in Mhz */ + bit_mhz = (byte_clk_rate * 8) / 1000000; + + if (bit_mhz < 125) + return 64; + else if (bit_mhz < 250) + return 32; + else if (bit_mhz < 600) + return 16; + else + return 8; +} + +static long clk_bytediv_round_rate(struct clk_hw *hw, unsigned long rate, + unsigned long *prate) +{ + unsigned long best_parent; + unsigned int factor; + + factor = get_vco_mul_factor(rate); + + best_parent = rate * factor; + *prate = clk_hw_round_rate(clk_hw_get_parent(hw), best_parent); + + return *prate / factor; +} + +static int clk_bytediv_set_rate(struct clk_hw *hw, unsigned long rate, + unsigned long parent_rate) +{ + struct clk_bytediv *bytediv = to_clk_bytediv(hw); + u32 val; + unsigned int factor; + + factor = get_vco_mul_factor(rate); + + val = pll_read(bytediv->reg); + val |= (factor - 1) & 0xff; + pll_write(bytediv->reg, val); + + return 0; +} + +/* Our special byte clock divider ops */ +static const struct clk_ops clk_bytediv_ops = { + .round_rate = clk_bytediv_round_rate, + .set_rate = clk_bytediv_set_rate, + .recalc_rate = clk_bytediv_recalc_rate, +}; + +/* + * PLL Callbacks + */ +static int dsi_pll_28nm_enable_seq(struct msm_dsi_pll *pll) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + struct device *dev = &pll_28nm->pdev->dev; + void __iomem *base = pll_28nm->mmio; + bool locked; + unsigned int bit_div, byte_div; + int max_reads = 1000, timeout_us = 100; + u32 val; + + DBG("id=%d", pll_28nm->id); + + /* + * before enabling the PLL, configure the bit clock divider since we + * don't expose it as a clock to the outside world + * 1: read back the byte clock divider that should already be set + * 2: divide by 8 to get bit clock divider + * 3: write it to POSTDIV1 + */ + val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_9); + byte_div = val + 1; + bit_div = byte_div / 8; + + val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8); + val &= ~0xf; + val |= (bit_div - 1); + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8, val); + + /* enable the PLL */ + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_0, + DSI_28nm_8960_PHY_PLL_CTRL_0_ENABLE); + + locked = pll_28nm_poll_for_ready(pll_28nm, max_reads, timeout_us); + + if (unlikely(!locked)) + dev_err(dev, "DSI PLL lock failed\n"); + else + DBG("DSI PLL lock success"); + + return locked ? 0 : -EINVAL; +} + +static void dsi_pll_28nm_disable_seq(struct msm_dsi_pll *pll) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + + DBG("id=%d", pll_28nm->id); + pll_write(pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_CTRL_0, 0x00); +} + +static void dsi_pll_28nm_save_state(struct msm_dsi_pll *pll) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + struct pll_28nm_cached_state *cached_state = &pll_28nm->cached_state; + void __iomem *base = pll_28nm->mmio; + + cached_state->postdiv3 = + pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_10); + cached_state->postdiv2 = + pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_9); + cached_state->postdiv1 = + pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8); + + cached_state->vco_rate = clk_hw_get_rate(&pll->clk_hw); +} + +static int dsi_pll_28nm_restore_state(struct msm_dsi_pll *pll) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + struct pll_28nm_cached_state *cached_state = &pll_28nm->cached_state; + void __iomem *base = pll_28nm->mmio; + int ret; + + ret = dsi_pll_28nm_clk_set_rate(&pll->clk_hw, + cached_state->vco_rate, 0); + if (ret) { + dev_err(&pll_28nm->pdev->dev, + "restore vco rate failed. ret=%d\n", ret); + return ret; + } + + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_10, + cached_state->postdiv3); + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_9, + cached_state->postdiv2); + pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8, + cached_state->postdiv1); + + return 0; +} + +static int dsi_pll_28nm_get_provider(struct msm_dsi_pll *pll, + struct clk **byte_clk_provider, + struct clk **pixel_clk_provider) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + + if (byte_clk_provider) + *byte_clk_provider = pll_28nm->provided_clks[DSI_BYTE_PLL_CLK]; + if (pixel_clk_provider) + *pixel_clk_provider = + pll_28nm->provided_clks[DSI_PIXEL_PLL_CLK]; + + return 0; +} + +static void dsi_pll_28nm_destroy(struct msm_dsi_pll *pll) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll); + + msm_dsi_pll_helper_unregister_clks(pll_28nm->pdev, + pll_28nm->clks, pll_28nm->num_clks); +} + +static int pll_28nm_register(struct dsi_pll_28nm *pll_28nm) +{ + char *clk_name, *parent_name, *vco_name; + struct clk_init_data vco_init = { + .parent_names = (const char *[]){ "pxo" }, + .num_parents = 1, + .ops = &clk_ops_dsi_pll_28nm_vco, + }; + struct device *dev = &pll_28nm->pdev->dev; + struct clk **clks = pll_28nm->clks; + struct clk **provided_clks = pll_28nm->provided_clks; + struct clk_bytediv *bytediv; + struct clk_init_data bytediv_init = { }; + int ret, num = 0; + + DBG("%d", pll_28nm->id); + + bytediv = devm_kzalloc(dev, sizeof(*bytediv), GFP_KERNEL); + if (!bytediv) + return -ENOMEM; + + vco_name = devm_kzalloc(dev, 32, GFP_KERNEL); + if (!vco_name) + return -ENOMEM; + + parent_name = devm_kzalloc(dev, 32, GFP_KERNEL); + if (!parent_name) + return -ENOMEM; + + clk_name = devm_kzalloc(dev, 32, GFP_KERNEL); + if (!clk_name) + return -ENOMEM; + + pll_28nm->bytediv = bytediv; + + snprintf(vco_name, 32, "dsi%dvco_clk", pll_28nm->id); + vco_init.name = vco_name; + + pll_28nm->base.clk_hw.init = &vco_init; + + clks[num++] = clk_register(dev, &pll_28nm->base.clk_hw); + + /* prepare and register bytediv */ + bytediv->hw.init = &bytediv_init; + bytediv->reg = pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_CTRL_9; + + snprintf(parent_name, 32, "dsi%dvco_clk", pll_28nm->id); + snprintf(clk_name, 32, "dsi%dpllbyte", pll_28nm->id); + + bytediv_init.name = clk_name; + bytediv_init.ops = &clk_bytediv_ops; + bytediv_init.flags = CLK_SET_RATE_PARENT; + bytediv_init.parent_names = (const char * const *) &parent_name; + bytediv_init.num_parents = 1; + + /* DIV2 */ + clks[num++] = provided_clks[DSI_BYTE_PLL_CLK] = + clk_register(dev, &bytediv->hw); + + snprintf(clk_name, 32, "dsi%dpll", pll_28nm->id); + /* DIV3 */ + clks[num++] = provided_clks[DSI_PIXEL_PLL_CLK] = + clk_register_divider(dev, clk_name, + parent_name, 0, pll_28nm->mmio + + REG_DSI_28nm_8960_PHY_PLL_CTRL_10, + 0, 8, 0, NULL); + + pll_28nm->num_clks = num; + + pll_28nm->clk_data.clk_num = NUM_PROVIDED_CLKS; + pll_28nm->clk_data.clks = provided_clks; + + ret = of_clk_add_provider(dev->of_node, + of_clk_src_onecell_get, &pll_28nm->clk_data); + if (ret) { + dev_err(dev, "failed to register clk provider: %d\n", ret); + return ret; + } + + return 0; +} + +struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev, + int id) +{ + struct dsi_pll_28nm *pll_28nm; + struct msm_dsi_pll *pll; + int ret; + + if (!pdev) + return ERR_PTR(-ENODEV); + + pll_28nm = devm_kzalloc(&pdev->dev, sizeof(*pll_28nm), GFP_KERNEL); + if (!pll_28nm) + return ERR_PTR(-ENOMEM); + + pll_28nm->pdev = pdev; + pll_28nm->id = id + 1; + + pll_28nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL"); + if (IS_ERR_OR_NULL(pll_28nm->mmio)) { + dev_err(&pdev->dev, "%s: failed to map pll base\n", __func__); + return ERR_PTR(-ENOMEM); + } + + pll = &pll_28nm->base; + pll->min_rate = VCO_MIN_RATE; + pll->max_rate = VCO_MAX_RATE; + pll->get_provider = dsi_pll_28nm_get_provider; + pll->destroy = dsi_pll_28nm_destroy; + pll->disable_seq = dsi_pll_28nm_disable_seq; + pll->save_state = dsi_pll_28nm_save_state; + pll->restore_state = dsi_pll_28nm_restore_state; + + pll->en_seq_cnt = 1; + pll->enable_seqs[0] = dsi_pll_28nm_enable_seq; + + ret = pll_28nm_register(pll_28nm); + if (ret) { + dev_err(&pdev->dev, "failed to register PLL: %d\n", ret); + return ERR_PTR(ret); + } + + return pll; +} diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index 1f4a95eeb348..9a0989c0b4de 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -17,6 +17,8 @@ */ #include <linux/of_irq.h> +#include <linux/of_gpio.h> + #include "hdmi.h" void hdmi_set_mode(struct hdmi *hdmi, bool power_on) @@ -322,8 +324,6 @@ fail: * The hdmi device: */ -#include <linux/of_gpio.h> - #define HDMI_CFG(item, entry) \ .item ## _names = item ##_names_ ## entry, \ .item ## _cnt = ARRAY_SIZE(item ## _names_ ## entry) @@ -388,17 +388,6 @@ static struct hdmi_platform_config hdmi_tx_8996_config = { .hpd_freq = hpd_clk_freq_8x74, }; -static const struct of_device_id dt_match[] = { - { .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8996_config }, - { .compatible = "qcom,hdmi-tx-8994", .data = &hdmi_tx_8994_config }, - { .compatible = "qcom,hdmi-tx-8084", .data = &hdmi_tx_8084_config }, - { .compatible = "qcom,hdmi-tx-8974", .data = &hdmi_tx_8974_config }, - { .compatible = "qcom,hdmi-tx-8960", .data = &hdmi_tx_8960_config }, - { .compatible = "qcom,hdmi-tx-8660", .data = &hdmi_tx_8660_config }, - {} -}; - -#ifdef CONFIG_OF static int get_gpio(struct device *dev, struct device_node *of_node, const char *name) { int gpio = of_get_named_gpio(of_node, name, 0); @@ -413,7 +402,6 @@ static int get_gpio(struct device *dev, struct device_node *of_node, const char } return gpio; } -#endif static int hdmi_bind(struct device *dev, struct device *master, void *data) { @@ -421,16 +409,12 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data) struct msm_drm_private *priv = drm->dev_private; static struct hdmi_platform_config *hdmi_cfg; struct hdmi *hdmi; -#ifdef CONFIG_OF struct device_node *of_node = dev->of_node; - const struct of_device_id *match; - match = of_match_node(dt_match, of_node); - if (match && match->data) { - hdmi_cfg = (struct hdmi_platform_config *)match->data; - DBG("hdmi phy: %s", match->compatible); - } else { - dev_err(dev, "unknown phy: %s\n", of_node->name); + hdmi_cfg = (struct hdmi_platform_config *) + of_device_get_match_data(dev); + if (!hdmi_cfg) { + dev_err(dev, "unknown hdmi_cfg: %s\n", of_node->name); return -ENXIO; } @@ -443,55 +427,6 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data) hdmi_cfg->mux_sel_gpio = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-sel"); hdmi_cfg->mux_lpm_gpio = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-lpm"); -#else - static struct hdmi_platform_config config = {}; - static const char *hpd_clk_names[] = { - "core_clk", "master_iface_clk", "slave_iface_clk", - }; - if (cpu_is_apq8064()) { - static const char *hpd_reg_names[] = {"8921_hdmi_mvs"}; - config.phy_init = hdmi_phy_8960_init; - config.hpd_reg_names = hpd_reg_names; - config.hpd_reg_cnt = ARRAY_SIZE(hpd_reg_names); - config.hpd_clk_names = hpd_clk_names; - config.hpd_clk_cnt = ARRAY_SIZE(hpd_clk_names); - config.ddc_clk_gpio = 70; - config.ddc_data_gpio = 71; - config.hpd_gpio = 72; - config.mux_en_gpio = -1; - config.mux_sel_gpio = -1; - } else if (cpu_is_msm8960() || cpu_is_msm8960ab()) { - static const char *hpd_reg_names[] = {"8921_hdmi_mvs"}; - config.phy_init = hdmi_phy_8960_init; - config.hpd_reg_names = hpd_reg_names; - config.hpd_reg_cnt = ARRAY_SIZE(hpd_reg_names); - config.hpd_clk_names = hpd_clk_names; - config.hpd_clk_cnt = ARRAY_SIZE(hpd_clk_names); - config.ddc_clk_gpio = 100; - config.ddc_data_gpio = 101; - config.hpd_gpio = 102; - config.mux_en_gpio = -1; - config.mux_sel_gpio = -1; - } else if (cpu_is_msm8x60()) { - static const char *hpd_reg_names[] = { - "8901_hdmi_mvs", "8901_mpp0" - }; - config.phy_init = hdmi_phy_8x60_init; - config.hpd_reg_names = hpd_reg_names; - config.hpd_reg_cnt = ARRAY_SIZE(hpd_reg_names); - config.hpd_clk_names = hpd_clk_names; - config.hpd_clk_cnt = ARRAY_SIZE(hpd_clk_names); - config.ddc_clk_gpio = 170; - config.ddc_data_gpio = 171; - config.hpd_gpio = 172; - config.mux_en_gpio = -1; - config.mux_sel_gpio = -1; - } - config.mmio_name = "hdmi_msm_hdmi_addr"; - config.qfprom_mmio_name = "hdmi_msm_qfprom_addr"; - - hdmi_cfg = &config; -#endif dev->platform_data = hdmi_cfg; hdmi = hdmi_init(to_platform_device(dev)); @@ -529,6 +464,16 @@ static int hdmi_dev_remove(struct platform_device *pdev) return 0; } +static const struct of_device_id dt_match[] = { + { .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8996_config }, + { .compatible = "qcom,hdmi-tx-8994", .data = &hdmi_tx_8994_config }, + { .compatible = "qcom,hdmi-tx-8084", .data = &hdmi_tx_8084_config }, + { .compatible = "qcom,hdmi-tx-8974", .data = &hdmi_tx_8974_config }, + { .compatible = "qcom,hdmi-tx-8960", .data = &hdmi_tx_8960_config }, + { .compatible = "qcom,hdmi-tx-8660", .data = &hdmi_tx_8660_config }, + {} +}; + static struct platform_driver hdmi_driver = { .probe = hdmi_dev_probe, .remove = hdmi_dev_remove, diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c index 6ac9aa165768..28df397c3b04 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c @@ -678,7 +678,8 @@ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, drm_flip_work_init(&mdp4_crtc->unref_cursor_work, "unref cursor", unref_cursor_worker); - drm_crtc_init_with_planes(dev, crtc, plane, NULL, &mdp4_crtc_funcs); + drm_crtc_init_with_planes(dev, crtc, plane, NULL, &mdp4_crtc_funcs, + NULL); drm_crtc_helper_add(crtc, &mdp4_crtc_helper_funcs); plane->crtc = crtc; diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dsi_encoder.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dsi_encoder.c new file mode 100644 index 000000000000..2f57e9453b67 --- /dev/null +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dsi_encoder.c @@ -0,0 +1,198 @@ +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + * Copyright (c) 2014, Inforce Computing. All rights reserved. + * + * Author: Vinay Simha <vinaysimha@inforcecomputing.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "mdp4_kms.h" + +#include "drm_crtc.h" +#include "drm_crtc_helper.h" + +struct mdp4_dsi_encoder { + struct drm_encoder base; + struct drm_panel *panel; + bool enabled; +}; +#define to_mdp4_dsi_encoder(x) container_of(x, struct mdp4_dsi_encoder, base) + +static struct mdp4_kms *get_kms(struct drm_encoder *encoder) +{ + struct msm_drm_private *priv = encoder->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +static void mdp4_dsi_encoder_destroy(struct drm_encoder *encoder) +{ + struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder); + + drm_encoder_cleanup(encoder); + kfree(mdp4_dsi_encoder); +} + +static const struct drm_encoder_funcs mdp4_dsi_encoder_funcs = { + .destroy = mdp4_dsi_encoder_destroy, +}; + +static bool mdp4_dsi_encoder_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + return true; +} + +static void mdp4_dsi_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct mdp4_kms *mdp4_kms = get_kms(encoder); + uint32_t dsi_hsync_skew, vsync_period, vsync_len, ctrl_pol; + uint32_t display_v_start, display_v_end; + uint32_t hsync_start_x, hsync_end_x; + + mode = adjusted_mode; + + DBG("set mode: %d:\"%s\" %d %d %d %d %d %d %d %d %d %d 0x%x 0x%x", + mode->base.id, mode->name, + mode->vrefresh, mode->clock, + mode->hdisplay, mode->hsync_start, + mode->hsync_end, mode->htotal, + mode->vdisplay, mode->vsync_start, + mode->vsync_end, mode->vtotal, + mode->type, mode->flags); + + ctrl_pol = 0; + if (mode->flags & DRM_MODE_FLAG_NHSYNC) + ctrl_pol |= MDP4_DSI_CTRL_POLARITY_HSYNC_LOW; + if (mode->flags & DRM_MODE_FLAG_NVSYNC) + ctrl_pol |= MDP4_DSI_CTRL_POLARITY_VSYNC_LOW; + /* probably need to get DATA_EN polarity from panel.. */ + + dsi_hsync_skew = 0; /* get this from panel? */ + + hsync_start_x = (mode->htotal - mode->hsync_start); + hsync_end_x = mode->htotal - (mode->hsync_start - mode->hdisplay) - 1; + + vsync_period = mode->vtotal * mode->htotal; + vsync_len = (mode->vsync_end - mode->vsync_start) * mode->htotal; + display_v_start = (mode->vtotal - mode->vsync_start) * mode->htotal + dsi_hsync_skew; + display_v_end = vsync_period - ((mode->vsync_start - mode->vdisplay) * mode->htotal) + dsi_hsync_skew - 1; + + mdp4_write(mdp4_kms, REG_MDP4_DSI_HSYNC_CTRL, + MDP4_DSI_HSYNC_CTRL_PULSEW(mode->hsync_end - mode->hsync_start) | + MDP4_DSI_HSYNC_CTRL_PERIOD(mode->htotal)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_VSYNC_PERIOD, vsync_period); + mdp4_write(mdp4_kms, REG_MDP4_DSI_VSYNC_LEN, vsync_len); + mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_HCTRL, + MDP4_DSI_DISPLAY_HCTRL_START(hsync_start_x) | + MDP4_DSI_DISPLAY_HCTRL_END(hsync_end_x)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_VSTART, display_v_start); + mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_VEND, display_v_end); + + mdp4_write(mdp4_kms, REG_MDP4_DSI_CTRL_POLARITY, ctrl_pol); + mdp4_write(mdp4_kms, REG_MDP4_DSI_UNDERFLOW_CLR, + MDP4_DSI_UNDERFLOW_CLR_ENABLE_RECOVERY | + MDP4_DSI_UNDERFLOW_CLR_COLOR(0xff)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_HCTL, + MDP4_DSI_ACTIVE_HCTL_START(0) | + MDP4_DSI_ACTIVE_HCTL_END(0)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_HSYNC_SKEW, dsi_hsync_skew); + mdp4_write(mdp4_kms, REG_MDP4_DSI_BORDER_CLR, 0); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_VSTART, 0); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_VEND, 0); +} + +static void mdp4_dsi_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + + if (!mdp4_dsi_encoder->enabled) + return; + + mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 0); + + /* + * Wait for a vsync so we know the ENABLE=0 latched before + * the (connector) source of the vsync's gets disabled, + * otherwise we end up in a funny state if we re-enable + * before the disable latches, which results that some of + * the settings changes for the new modeset (like new + * scanout buffer) don't latch properly.. + */ + mdp_irq_wait(&mdp4_kms->base, MDP4_IRQ_PRIMARY_VSYNC); + + mdp4_dsi_encoder->enabled = false; +} + +static void mdp4_dsi_encoder_enable(struct drm_encoder *encoder) +{ + struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + + if (mdp4_dsi_encoder->enabled) + return; + + mdp4_crtc_set_config(encoder->crtc, + MDP4_DMA_CONFIG_PACK_ALIGN_MSB | + MDP4_DMA_CONFIG_DEFLKR_EN | + MDP4_DMA_CONFIG_DITHER_EN | + MDP4_DMA_CONFIG_R_BPC(BPC8) | + MDP4_DMA_CONFIG_G_BPC(BPC8) | + MDP4_DMA_CONFIG_B_BPC(BPC8) | + MDP4_DMA_CONFIG_PACK(0x21)); + + mdp4_crtc_set_intf(encoder->crtc, INTF_DSI_VIDEO, 0); + + mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 1); + + mdp4_dsi_encoder->enabled = true; +} + +static const struct drm_encoder_helper_funcs mdp4_dsi_encoder_helper_funcs = { + .mode_fixup = mdp4_dsi_encoder_mode_fixup, + .mode_set = mdp4_dsi_encoder_mode_set, + .disable = mdp4_dsi_encoder_disable, + .enable = mdp4_dsi_encoder_enable, +}; + +/* initialize encoder */ +struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev) +{ + struct drm_encoder *encoder = NULL; + struct mdp4_dsi_encoder *mdp4_dsi_encoder; + int ret; + + mdp4_dsi_encoder = kzalloc(sizeof(*mdp4_dsi_encoder), GFP_KERNEL); + if (!mdp4_dsi_encoder) { + ret = -ENOMEM; + goto fail; + } + + encoder = &mdp4_dsi_encoder->base; + + drm_encoder_init(dev, encoder, &mdp4_dsi_encoder_funcs, + DRM_MODE_ENCODER_DSI, NULL); + drm_encoder_helper_add(encoder, &mdp4_dsi_encoder_helper_funcs); + + return encoder; + +fail: + if (encoder) + mdp4_dsi_encoder_destroy(encoder); + + return ERR_PTR(ret); +} diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dtv_encoder.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dtv_encoder.c index 89614c6a6c1b..a21df54cb50f 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dtv_encoder.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dtv_encoder.c @@ -262,7 +262,7 @@ struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev) encoder = &mdp4_dtv_encoder->base; drm_encoder_init(dev, encoder, &mdp4_dtv_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &mdp4_dtv_encoder_helper_funcs); mdp4_dtv_encoder->src_clk = devm_clk_get(dev->dev, "src_clk"); diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c index 5ed38cf548a1..a521207db8a1 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c @@ -29,7 +29,7 @@ void mdp4_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask, static void mdp4_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus) { - DRM_ERROR("errors: %08x\n", irqstatus); + DRM_ERROR_RATELIMITED("errors: %08x\n", irqstatus); } void mdp4_irq_preinstall(struct msm_kms *kms) diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c index 077f7521a971..5a8e3d6bcbff 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c @@ -169,7 +169,14 @@ static long mdp4_round_pixclk(struct msm_kms *kms, unsigned long rate, struct drm_encoder *encoder) { /* if we had >1 encoder, we'd need something more clever: */ - return mdp4_dtv_round_pixclk(encoder, rate); + switch (encoder->encoder_type) { + case DRM_MODE_ENCODER_TMDS: + return mdp4_dtv_round_pixclk(encoder, rate); + case DRM_MODE_ENCODER_LVDS: + case DRM_MODE_ENCODER_DSI: + default: + return rate; + } } static void mdp4_preclose(struct msm_kms *kms, struct drm_file *file) @@ -240,19 +247,18 @@ int mdp4_enable(struct mdp4_kms *mdp4_kms) return 0; } -#ifdef CONFIG_OF -static struct drm_panel *detect_panel(struct drm_device *dev) +static struct device_node *mdp4_detect_lcdc_panel(struct drm_device *dev) { struct device_node *endpoint, *panel_node; struct device_node *np = dev->dev->of_node; - struct drm_panel *panel = NULL; endpoint = of_graph_get_next_endpoint(np, NULL); if (!endpoint) { - dev_err(dev->dev, "no valid endpoint\n"); - return ERR_PTR(-ENODEV); + DBG("no endpoint in MDP4 to fetch LVDS panel\n"); + return NULL; } + /* don't proceed if we have an endpoint but no panel_node tied to it */ panel_node = of_graph_get_remote_port_parent(endpoint); if (!panel_node) { dev_err(dev->dev, "no valid panel node\n"); @@ -262,132 +268,185 @@ static struct drm_panel *detect_panel(struct drm_device *dev) of_node_put(endpoint); - panel = of_drm_find_panel(panel_node); - if (!panel) { - of_node_put(panel_node); - return ERR_PTR(-EPROBE_DEFER); - } - - return panel; + return panel_node; } -#else -static struct drm_panel *detect_panel(struct drm_device *dev) -{ - // ??? maybe use a module param to specify which panel is attached? -} -#endif -static int modeset_init(struct mdp4_kms *mdp4_kms) +static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, + int intf_type) { struct drm_device *dev = mdp4_kms->dev; struct msm_drm_private *priv = dev->dev_private; - struct drm_plane *plane; - struct drm_crtc *crtc; struct drm_encoder *encoder; struct drm_connector *connector; - struct drm_panel *panel; + struct device_node *panel_node; + struct drm_encoder *dsi_encs[MSM_DSI_ENCODER_NUM]; + int i, dsi_id; int ret; - /* construct non-private planes: */ - plane = mdp4_plane_init(dev, VG1, false); - if (IS_ERR(plane)) { - dev_err(dev->dev, "failed to construct plane for VG1\n"); - ret = PTR_ERR(plane); - goto fail; - } - priv->planes[priv->num_planes++] = plane; + switch (intf_type) { + case DRM_MODE_ENCODER_LVDS: + /* + * bail out early if: + * - there is no panel node (no need to initialize lcdc + * encoder and lvds connector), or + * - panel node is a bad pointer + */ + panel_node = mdp4_detect_lcdc_panel(dev); + if (IS_ERR_OR_NULL(panel_node)) + return PTR_ERR(panel_node); + + encoder = mdp4_lcdc_encoder_init(dev, panel_node); + if (IS_ERR(encoder)) { + dev_err(dev->dev, "failed to construct LCDC encoder\n"); + return PTR_ERR(encoder); + } - plane = mdp4_plane_init(dev, VG2, false); - if (IS_ERR(plane)) { - dev_err(dev->dev, "failed to construct plane for VG2\n"); - ret = PTR_ERR(plane); - goto fail; - } - priv->planes[priv->num_planes++] = plane; + /* LCDC can be hooked to DMA_P (TODO: Add DMA_S later?) */ + encoder->possible_crtcs = 1 << DMA_P; - /* - * Setup the LCDC/LVDS path: RGB2 -> DMA_P -> LCDC -> LVDS: - */ + connector = mdp4_lvds_connector_init(dev, panel_node, encoder); + if (IS_ERR(connector)) { + dev_err(dev->dev, "failed to initialize LVDS connector\n"); + return PTR_ERR(connector); + } - panel = detect_panel(dev); - if (IS_ERR(panel)) { - ret = PTR_ERR(panel); - dev_err(dev->dev, "failed to detect LVDS panel: %d\n", ret); - goto fail; - } + priv->encoders[priv->num_encoders++] = encoder; + priv->connectors[priv->num_connectors++] = connector; - plane = mdp4_plane_init(dev, RGB2, true); - if (IS_ERR(plane)) { - dev_err(dev->dev, "failed to construct plane for RGB2\n"); - ret = PTR_ERR(plane); - goto fail; - } + break; + case DRM_MODE_ENCODER_TMDS: + encoder = mdp4_dtv_encoder_init(dev); + if (IS_ERR(encoder)) { + dev_err(dev->dev, "failed to construct DTV encoder\n"); + return PTR_ERR(encoder); + } - crtc = mdp4_crtc_init(dev, plane, priv->num_crtcs, 0, DMA_P); - if (IS_ERR(crtc)) { - dev_err(dev->dev, "failed to construct crtc for DMA_P\n"); - ret = PTR_ERR(crtc); - goto fail; - } + /* DTV can be hooked to DMA_E: */ + encoder->possible_crtcs = 1 << 1; - encoder = mdp4_lcdc_encoder_init(dev, panel); - if (IS_ERR(encoder)) { - dev_err(dev->dev, "failed to construct LCDC encoder\n"); - ret = PTR_ERR(encoder); - goto fail; - } + if (priv->hdmi) { + /* Construct bridge/connector for HDMI: */ + ret = hdmi_modeset_init(priv->hdmi, dev, encoder); + if (ret) { + dev_err(dev->dev, "failed to initialize HDMI: %d\n", ret); + return ret; + } + } - /* LCDC can be hooked to DMA_P: */ - encoder->possible_crtcs = 1 << priv->num_crtcs; + priv->encoders[priv->num_encoders++] = encoder; - priv->crtcs[priv->num_crtcs++] = crtc; - priv->encoders[priv->num_encoders++] = encoder; + break; + case DRM_MODE_ENCODER_DSI: + /* only DSI1 supported for now */ + dsi_id = 0; - connector = mdp4_lvds_connector_init(dev, panel, encoder); - if (IS_ERR(connector)) { - ret = PTR_ERR(connector); - dev_err(dev->dev, "failed to initialize LVDS connector: %d\n", ret); - goto fail; - } + if (!priv->dsi[dsi_id]) + break; - priv->connectors[priv->num_connectors++] = connector; + for (i = 0; i < MSM_DSI_ENCODER_NUM; i++) { + dsi_encs[i] = mdp4_dsi_encoder_init(dev); + if (IS_ERR(dsi_encs[i])) { + ret = PTR_ERR(dsi_encs[i]); + dev_err(dev->dev, + "failed to construct DSI encoder: %d\n", + ret); + return ret; + } - /* - * Setup DTV/HDMI path: RGB1 -> DMA_E -> DTV -> HDMI: - */ + /* TODO: Add DMA_S later? */ + dsi_encs[i]->possible_crtcs = 1 << DMA_P; + priv->encoders[priv->num_encoders++] = dsi_encs[i]; + } - plane = mdp4_plane_init(dev, RGB1, true); - if (IS_ERR(plane)) { - dev_err(dev->dev, "failed to construct plane for RGB1\n"); - ret = PTR_ERR(plane); - goto fail; - } + ret = msm_dsi_modeset_init(priv->dsi[dsi_id], dev, dsi_encs); + if (ret) { + dev_err(dev->dev, "failed to initialize DSI: %d\n", + ret); + return ret; + } - crtc = mdp4_crtc_init(dev, plane, priv->num_crtcs, 1, DMA_E); - if (IS_ERR(crtc)) { - dev_err(dev->dev, "failed to construct crtc for DMA_E\n"); - ret = PTR_ERR(crtc); - goto fail; + break; + default: + dev_err(dev->dev, "Invalid or unsupported interface\n"); + return -EINVAL; } - encoder = mdp4_dtv_encoder_init(dev); - if (IS_ERR(encoder)) { - dev_err(dev->dev, "failed to construct DTV encoder\n"); - ret = PTR_ERR(encoder); - goto fail; + return 0; +} + +static int modeset_init(struct mdp4_kms *mdp4_kms) +{ + struct drm_device *dev = mdp4_kms->dev; + struct msm_drm_private *priv = dev->dev_private; + struct drm_plane *plane; + struct drm_crtc *crtc; + int i, ret; + static const enum mdp4_pipe rgb_planes[] = { + RGB1, RGB2, + }; + static const enum mdp4_pipe vg_planes[] = { + VG1, VG2, + }; + static const enum mdp4_dma mdp4_crtcs[] = { + DMA_P, DMA_E, + }; + static const char * const mdp4_crtc_names[] = { + "DMA_P", "DMA_E", + }; + static const int mdp4_intfs[] = { + DRM_MODE_ENCODER_LVDS, + DRM_MODE_ENCODER_DSI, + DRM_MODE_ENCODER_TMDS, + }; + + /* construct non-private planes: */ + for (i = 0; i < ARRAY_SIZE(vg_planes); i++) { + plane = mdp4_plane_init(dev, vg_planes[i], false); + if (IS_ERR(plane)) { + dev_err(dev->dev, + "failed to construct plane for VG%d\n", i + 1); + ret = PTR_ERR(plane); + goto fail; + } + priv->planes[priv->num_planes++] = plane; } - /* DTV can be hooked to DMA_E: */ - encoder->possible_crtcs = 1 << priv->num_crtcs; + for (i = 0; i < ARRAY_SIZE(mdp4_crtcs); i++) { + plane = mdp4_plane_init(dev, rgb_planes[i], true); + if (IS_ERR(plane)) { + dev_err(dev->dev, + "failed to construct plane for RGB%d\n", i + 1); + ret = PTR_ERR(plane); + goto fail; + } + + crtc = mdp4_crtc_init(dev, plane, priv->num_crtcs, i, + mdp4_crtcs[i]); + if (IS_ERR(crtc)) { + dev_err(dev->dev, "failed to construct crtc for %s\n", + mdp4_crtc_names[i]); + ret = PTR_ERR(crtc); + goto fail; + } + + priv->crtcs[priv->num_crtcs++] = crtc; + } - priv->crtcs[priv->num_crtcs++] = crtc; - priv->encoders[priv->num_encoders++] = encoder; + /* + * we currently set up two relatively fixed paths: + * + * LCDC/LVDS path: RGB1 -> DMA_P -> LCDC -> LVDS + * or + * DSI path: RGB1 -> DMA_P -> DSI1 -> DSI Panel + * + * DTV/HDMI path: RGB2 -> DMA_E -> DTV -> HDMI + */ - if (priv->hdmi) { - /* Construct bridge/connector for HDMI: */ - ret = hdmi_modeset_init(priv->hdmi, dev, encoder); + for (i = 0; i < ARRAY_SIZE(mdp4_intfs); i++) { + ret = mdp4_modeset_init_intf(mdp4_kms, mdp4_intfs[i]); if (ret) { - dev_err(dev->dev, "failed to initialize HDMI: %d\n", ret); + dev_err(dev->dev, "failed to initialize intf: %d, %d\n", + i, ret); goto fail; } } @@ -558,17 +617,10 @@ fail: static struct mdp4_platform_config *mdp4_get_config(struct platform_device *dev) { static struct mdp4_platform_config config = {}; -#ifdef CONFIG_OF - /* TODO */ + + /* TODO: Chips that aren't apq8064 have a 200 Mhz max_clk */ config.max_clk = 266667000; config.iommu = iommu_domain_alloc(&platform_bus_type); -#else - if (cpu_is_apq8064()) - config.max_clk = 266667000; - else - config.max_clk = 200000000; - - config.iommu = msm_get_iommu_domain(DISPLAY_READ_DOMAIN); -#endif + return &config; } diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h index 8a7f6e1e2bca..d2c96ef431f4 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h @@ -157,7 +157,7 @@ static inline uint32_t mixercfg(uint32_t mixer_cfg, int mixer, COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1); break; default: - WARN_ON("invalid pipe"); + WARN(1, "invalid pipe"); break; } @@ -212,10 +212,19 @@ struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev); long mdp4_lcdc_round_pixclk(struct drm_encoder *encoder, unsigned long rate); struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, - struct drm_panel *panel); + struct device_node *panel_node); struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev, - struct drm_panel *panel, struct drm_encoder *encoder); + struct device_node *panel_node, struct drm_encoder *encoder); + +#ifdef CONFIG_DRM_MSM_DSI +struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev); +#else +static inline struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev) +{ + return ERR_PTR(-ENODEV); +} +#endif #ifdef CONFIG_COMMON_CLK struct clk *mpd4_lvds_pll_init(struct drm_device *dev); diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c index 4cd6e721aa0a..cd63fedb67cc 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c @@ -23,6 +23,7 @@ struct mdp4_lcdc_encoder { struct drm_encoder base; + struct device_node *panel_node; struct drm_panel *panel; struct clk *lcdc_clk; unsigned long int pixclock; @@ -338,7 +339,7 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder) struct mdp4_lcdc_encoder *mdp4_lcdc_encoder = to_mdp4_lcdc_encoder(encoder); struct mdp4_kms *mdp4_kms = get_kms(encoder); - struct drm_panel *panel = mdp4_lcdc_encoder->panel; + struct drm_panel *panel; int i, ret; if (WARN_ON(!mdp4_lcdc_encoder->enabled)) @@ -346,6 +347,7 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder) mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 0); + panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node); if (panel) { drm_panel_disable(panel); drm_panel_unprepare(panel); @@ -381,7 +383,7 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder) to_mdp4_lcdc_encoder(encoder); unsigned long pc = mdp4_lcdc_encoder->pixclock; struct mdp4_kms *mdp4_kms = get_kms(encoder); - struct drm_panel *panel = mdp4_lcdc_encoder->panel; + struct drm_panel *panel; int i, ret; if (WARN_ON(mdp4_lcdc_encoder->enabled)) @@ -414,6 +416,7 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder) if (ret) dev_err(dev->dev, "failed to enable lcdc_clk: %d\n", ret); + panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node); if (panel) { drm_panel_prepare(panel); drm_panel_enable(panel); @@ -442,7 +445,7 @@ long mdp4_lcdc_round_pixclk(struct drm_encoder *encoder, unsigned long rate) /* initialize encoder */ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, - struct drm_panel *panel) + struct device_node *panel_node) { struct drm_encoder *encoder = NULL; struct mdp4_lcdc_encoder *mdp4_lcdc_encoder; @@ -455,12 +458,12 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, goto fail; } - mdp4_lcdc_encoder->panel = panel; + mdp4_lcdc_encoder->panel_node = panel_node; encoder = &mdp4_lcdc_encoder->base; drm_encoder_init(dev, encoder, &mdp4_lcdc_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); drm_encoder_helper_add(encoder, &mdp4_lcdc_encoder_helper_funcs); /* TODO: do we need different pll in other cases? */ diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c index 921185133d38..e73e1742b250 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c @@ -23,6 +23,7 @@ struct mdp4_lvds_connector { struct drm_connector base; struct drm_encoder *encoder; + struct device_node *panel_node; struct drm_panel *panel; }; #define to_mdp4_lvds_connector(x) container_of(x, struct mdp4_lvds_connector, base) @@ -33,6 +34,10 @@ static enum drm_connector_status mdp4_lvds_connector_detect( struct mdp4_lvds_connector *mdp4_lvds_connector = to_mdp4_lvds_connector(connector); + if (!mdp4_lvds_connector->panel) + mdp4_lvds_connector->panel = + of_drm_find_panel(mdp4_lvds_connector->panel_node); + return mdp4_lvds_connector->panel ? connector_status_connected : connector_status_disconnected; @@ -42,10 +47,6 @@ static void mdp4_lvds_connector_destroy(struct drm_connector *connector) { struct mdp4_lvds_connector *mdp4_lvds_connector = to_mdp4_lvds_connector(connector); - struct drm_panel *panel = mdp4_lvds_connector->panel; - - if (panel) - drm_panel_detach(panel); drm_connector_unregister(connector); drm_connector_cleanup(connector); @@ -60,9 +61,14 @@ static int mdp4_lvds_connector_get_modes(struct drm_connector *connector) struct drm_panel *panel = mdp4_lvds_connector->panel; int ret = 0; - if (panel) + if (panel) { + drm_panel_attach(panel, connector); + ret = panel->funcs->get_modes(panel); + drm_panel_detach(panel); + } + return ret; } @@ -111,7 +117,7 @@ static const struct drm_connector_helper_funcs mdp4_lvds_connector_helper_funcs /* initialize connector */ struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev, - struct drm_panel *panel, struct drm_encoder *encoder) + struct device_node *panel_node, struct drm_encoder *encoder) { struct drm_connector *connector = NULL; struct mdp4_lvds_connector *mdp4_lvds_connector; @@ -124,7 +130,7 @@ struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev, } mdp4_lvds_connector->encoder = encoder; - mdp4_lvds_connector->panel = panel; + mdp4_lvds_connector->panel_node = panel_node; connector = &mdp4_lvds_connector->base; @@ -141,9 +147,6 @@ struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev, drm_mode_connector_attach_encoder(connector, encoder); - if (panel) - drm_panel_attach(panel, connector); - return connector; fail: diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c index 30d57e74c42f..9f96dfe67769 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c @@ -397,7 +397,8 @@ struct drm_plane *mdp4_plane_init(struct drm_device *dev, type = private_plane ? DRM_PLANE_TYPE_PRIMARY : DRM_PLANE_TYPE_OVERLAY; ret = drm_universal_plane_init(dev, plane, 0xff, &mdp4_plane_funcs, - mdp4_plane->formats, mdp4_plane->nformats, type); + mdp4_plane->formats, mdp4_plane->nformats, + type, NULL); if (ret) goto fail; diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c index bb1225aa2f75..57f73f0c120d 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c @@ -553,9 +553,7 @@ fail: static struct mdp5_cfg_platform *mdp5_get_config(struct platform_device *dev) { static struct mdp5_cfg_platform config = {}; -#ifdef CONFIG_OF - /* TODO */ -#endif + config.iommu = iommu_domain_alloc(&platform_bus_type); return &config; diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c index 8e6c9b598a57..1aa21dba663d 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cmd_encoder.c @@ -326,7 +326,7 @@ struct drm_encoder *mdp5_cmd_encoder_init(struct drm_device *dev, mdp5_cmd_enc->ctl = ctl; drm_encoder_init(dev, encoder, &mdp5_cmd_encoder_funcs, - DRM_MODE_ENCODER_DSI); + DRM_MODE_ENCODER_DSI, NULL); drm_encoder_helper_add(encoder, &mdp5_cmd_encoder_helper_funcs); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c index 7f9f4ac88029..20cee5ce4071 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c @@ -797,7 +797,8 @@ struct drm_crtc *mdp5_crtc_init(struct drm_device *dev, snprintf(mdp5_crtc->name, sizeof(mdp5_crtc->name), "%s:%d", pipe2name(mdp5_plane_pipe(plane)), id); - drm_crtc_init_with_planes(dev, crtc, plane, NULL, &mdp5_crtc_funcs); + drm_crtc_init_with_planes(dev, crtc, plane, NULL, &mdp5_crtc_funcs, + NULL); drm_flip_work_init(&mdp5_crtc->unref_cursor_work, "unref cursor", unref_cursor_worker); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c index c9e32b08a7a0..0d737cad03a6 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c @@ -293,6 +293,24 @@ static const struct drm_encoder_helper_funcs mdp5_encoder_helper_funcs = { .enable = mdp5_encoder_enable, }; +int mdp5_encoder_get_linecount(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + int intf = mdp5_encoder->intf.num; + + return mdp5_read(mdp5_kms, REG_MDP5_INTF_LINE_COUNT(intf)); +} + +u32 mdp5_encoder_get_framecount(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + int intf = mdp5_encoder->intf.num; + + return mdp5_read(mdp5_kms, REG_MDP5_INTF_FRAME_COUNT(intf)); +} + int mdp5_encoder_set_split_display(struct drm_encoder *encoder, struct drm_encoder *slave_encoder) { @@ -354,7 +372,7 @@ struct drm_encoder *mdp5_encoder_init(struct drm_device *dev, spin_lock_init(&mdp5_encoder->intf_lock); - drm_encoder_init(dev, encoder, &mdp5_encoder_funcs, enc_type); + drm_encoder_init(dev, encoder, &mdp5_encoder_funcs, enc_type, NULL); drm_encoder_helper_add(encoder, &mdp5_encoder_helper_funcs); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c index b0d4b53b97f4..73bc3e312fd4 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c @@ -31,7 +31,7 @@ void mdp5_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask, static void mdp5_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus) { - DRM_ERROR("errors: %08x\n", irqstatus); + DRM_ERROR_RATELIMITED("errors: %08x\n", irqstatus); } void mdp5_irq_preinstall(struct msm_kms *kms) diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c index b532faa8026d..e115318402bd 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c @@ -468,6 +468,127 @@ static int get_clk(struct platform_device *pdev, struct clk **clkp, return 0; } +static struct drm_encoder *get_encoder_from_crtc(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_encoder *encoder; + + drm_for_each_encoder(encoder, dev) + if (encoder->crtc == crtc) + return encoder; + + return NULL; +} + +static int mdp5_get_scanoutpos(struct drm_device *dev, unsigned int pipe, + unsigned int flags, int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) +{ + struct msm_drm_private *priv = dev->dev_private; + struct drm_crtc *crtc; + struct drm_encoder *encoder; + int line, vsw, vbp, vactive_start, vactive_end, vfp_end; + int ret = 0; + + crtc = priv->crtcs[pipe]; + if (!crtc) { + DRM_ERROR("Invalid crtc %d\n", pipe); + return 0; + } + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) { + DRM_ERROR("no encoder found for crtc %d\n", pipe); + return 0; + } + + ret |= DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE; + + vsw = mode->crtc_vsync_end - mode->crtc_vsync_start; + vbp = mode->crtc_vtotal - mode->crtc_vsync_end; + + /* + * the line counter is 1 at the start of the VSYNC pulse and VTOTAL at + * the end of VFP. Translate the porch values relative to the line + * counter positions. + */ + + vactive_start = vsw + vbp + 1; + + vactive_end = vactive_start + mode->crtc_vdisplay; + + /* last scan line before VSYNC */ + vfp_end = mode->crtc_vtotal; + + if (stime) + *stime = ktime_get(); + + line = mdp5_encoder_get_linecount(encoder); + + if (line < vactive_start) { + line -= vactive_start; + ret |= DRM_SCANOUTPOS_IN_VBLANK; + } else if (line > vactive_end) { + line = line - vfp_end - vactive_start; + ret |= DRM_SCANOUTPOS_IN_VBLANK; + } else { + line -= vactive_start; + } + + *vpos = line; + *hpos = 0; + + if (etime) + *etime = ktime_get(); + + return ret; +} + +static int mdp5_get_vblank_timestamp(struct drm_device *dev, unsigned int pipe, + int *max_error, + struct timeval *vblank_time, + unsigned flags) +{ + struct msm_drm_private *priv = dev->dev_private; + struct drm_crtc *crtc; + + if (pipe < 0 || pipe >= priv->num_crtcs) { + DRM_ERROR("Invalid crtc %d\n", pipe); + return -EINVAL; + } + + crtc = priv->crtcs[pipe]; + if (!crtc) { + DRM_ERROR("Invalid crtc %d\n", pipe); + return -EINVAL; + } + + return drm_calc_vbltimestamp_from_scanoutpos(dev, pipe, max_error, + vblank_time, flags, + &crtc->mode); +} + +static u32 mdp5_get_vblank_counter(struct drm_device *dev, unsigned int pipe) +{ + struct msm_drm_private *priv = dev->dev_private; + struct drm_crtc *crtc; + struct drm_encoder *encoder; + + if (pipe < 0 || pipe >= priv->num_crtcs) + return 0; + + crtc = priv->crtcs[pipe]; + if (!crtc) + return 0; + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) + return 0; + + return mdp5_encoder_get_framecount(encoder); +} + struct msm_kms *mdp5_kms_init(struct drm_device *dev) { struct platform_device *pdev = dev->platformdev; @@ -590,6 +711,8 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) !config->hw->intf.base[i]) continue; mdp5_write(mdp5_kms, REG_MDP5_INTF_TIMING_ENGINE_EN(i), 0); + + mdp5_write(mdp5_kms, REG_MDP5_INTF_FRAME_LINE_COUNT_EN(i), 0x3); } mdp5_disable(mdp5_kms); mdelay(16); @@ -635,6 +758,12 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) dev->mode_config.max_width = config->hw->lm.max_width; dev->mode_config.max_height = config->hw->lm.max_height; + dev->driver->get_vblank_timestamp = mdp5_get_vblank_timestamp; + dev->driver->get_scanout_position = mdp5_get_scanoutpos; + dev->driver->get_vblank_counter = mdp5_get_vblank_counter; + dev->max_vblank_count = 0xffffffff; + dev->vblank_disable_immediate = true; + return kms; fail: diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h index 84f65d415598..00730ba08a60 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h @@ -222,6 +222,8 @@ struct drm_encoder *mdp5_encoder_init(struct drm_device *dev, struct mdp5_interface *intf, struct mdp5_ctl *ctl); int mdp5_encoder_set_split_display(struct drm_encoder *encoder, struct drm_encoder *slave_encoder); +int mdp5_encoder_get_linecount(struct drm_encoder *encoder); +u32 mdp5_encoder_get_framecount(struct drm_encoder *encoder); #ifdef CONFIG_DRM_MSM_DSI struct drm_encoder *mdp5_cmd_encoder_init(struct drm_device *dev, diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c index 81cd49045ffc..432c09836b0e 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c @@ -904,7 +904,7 @@ struct drm_plane *mdp5_plane_init(struct drm_device *dev, type = private_plane ? DRM_PLANE_TYPE_PRIMARY : DRM_PLANE_TYPE_OVERLAY; ret = drm_universal_plane_init(dev, plane, 0xff, &mdp5_plane_funcs, mdp5_plane->formats, mdp5_plane->nformats, - type); + type, NULL); if (ret) goto fail; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index b88ce514eb8e..9a30807b900b 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -237,20 +237,9 @@ static int msm_unload(struct drm_device *dev) static int get_mdp_ver(struct platform_device *pdev) { -#ifdef CONFIG_OF - static const struct of_device_id match_types[] = { { - .compatible = "qcom,mdss_mdp", - .data = (void *)5, - }, { - /* end node */ - } }; struct device *dev = &pdev->dev; - const struct of_device_id *match; - match = of_match_node(match_types, dev->of_node); - if (match) - return (int)(unsigned long)match->data; -#endif - return 4; + + return (int) (unsigned long) of_device_get_match_data(dev); } #include <linux/of_address.h> @@ -258,10 +247,10 @@ static int get_mdp_ver(struct platform_device *pdev) static int msm_init_vram(struct drm_device *dev) { struct msm_drm_private *priv = dev->dev_private; + struct device_node *node; unsigned long size = 0; int ret = 0; -#ifdef CONFIG_OF /* In the device-tree world, we could have a 'memory-region' * phandle, which gives us a link to our "vram". Allocating * is all nicely abstracted behind the dma api, but we need @@ -278,7 +267,6 @@ static int msm_init_vram(struct drm_device *dev) * as corruption on screen before we have a chance to * load and do initial modeset) */ - struct device_node *node; node = of_parse_phandle(dev->dev->of_node, "memory-region", 0); if (node) { @@ -288,14 +276,12 @@ static int msm_init_vram(struct drm_device *dev) return ret; size = r.end - r.start; DRM_INFO("using VRAM carveout: %lx@%pa\n", size, &r.start); - } else -#endif - /* if we have no IOMMU, then we need to use carveout allocator. - * Grab the entire CMA chunk carved out in early startup in - * mach-msm: - */ - if (!iommu_present(&platform_bus_type)) { + /* if we have no IOMMU, then we need to use carveout allocator. + * Grab the entire CMA chunk carved out in early startup in + * mach-msm: + */ + } else if (!iommu_present(&platform_bus_type)) { DRM_INFO("using %s VRAM carveout\n", vram); size = memparse(vram, NULL); } @@ -1035,9 +1021,9 @@ static const struct dev_pm_ops msm_pm_ops = { * Componentized driver support: */ -#ifdef CONFIG_OF -/* NOTE: the CONFIG_OF case duplicates the same code as exynos or imx - * (or probably any other).. so probably some room for some helpers +/* + * NOTE: duplication of the same code as exynos or imx (or probably any other). + * so probably some room for some helpers */ static int compare_of(struct device *dev, void *data) { @@ -1062,12 +1048,6 @@ static int add_components(struct device *dev, struct component_match **matchptr, return 0; } -#else -static int compare_dev(struct device *dev, void *data) -{ - return dev == data; -} -#endif static int msm_drm_bind(struct device *dev) { @@ -1091,35 +1071,9 @@ static const struct component_master_ops msm_drm_ops = { static int msm_pdev_probe(struct platform_device *pdev) { struct component_match *match = NULL; -#ifdef CONFIG_OF + add_components(&pdev->dev, &match, "connectors"); add_components(&pdev->dev, &match, "gpus"); -#else - /* For non-DT case, it kinda sucks. We don't actually have a way - * to know whether or not we are waiting for certain devices (or if - * they are simply not present). But for non-DT we only need to - * care about apq8064/apq8060/etc (all mdp4/a3xx): - */ - static const char *devnames[] = { - "hdmi_msm.0", "kgsl-3d0.0", - }; - int i; - - DBG("Adding components.."); - - for (i = 0; i < ARRAY_SIZE(devnames); i++) { - struct device *dev; - - dev = bus_find_device_by_name(&platform_bus_type, - NULL, devnames[i]); - if (!dev) { - dev_info(&pdev->dev, "still waiting for %s\n", devnames[i]); - return -EPROBE_DEFER; - } - - component_match_add(&pdev->dev, &match, compare_dev, dev); - } -#endif pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32); return component_master_add_with_match(&pdev->dev, &msm_drm_ops, match); @@ -1138,8 +1092,10 @@ static const struct platform_device_id msm_id[] = { }; static const struct of_device_id dt_match[] = { - { .compatible = "qcom,mdp" }, /* mdp4 */ - { .compatible = "qcom,mdss_mdp" }, /* mdp5 */ + { .compatible = "qcom,mdp4", .data = (void *) 4 }, /* mdp4 */ + { .compatible = "qcom,mdp5", .data = (void *) 5 }, /* mdp5 */ + /* to support downstream DT files */ + { .compatible = "qcom,mdss_mdp", .data = (void *) 5 }, /* mdp5 */ {} }; MODULE_DEVICE_TABLE(of, dt_match); diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 3be7a56b14f1..c1e7bba2fdb7 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -31,14 +31,9 @@ #include <linux/iommu.h> #include <linux/types.h> #include <linux/of_graph.h> +#include <linux/of_device.h> #include <asm/sizes.h> -#ifndef CONFIG_OF -#include <mach/board.h> -#include <mach/socinfo.h> -#include <mach/iommu_domains.h> -#endif - #include <drm/drmP.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> @@ -240,9 +235,9 @@ uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, int id, int plane); struct drm_gem_object *msm_framebuffer_bo(struct drm_framebuffer *fb, int plane); const struct msm_format *msm_framebuffer_format(struct drm_framebuffer *fb); struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos); + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos); struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev, - struct drm_file *file, struct drm_mode_fb_cmd2 *mode_cmd); + struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd); struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev); diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c index 121713281417..a474d6cf5d9f 100644 --- a/drivers/gpu/drm/msm/msm_fb.c +++ b/drivers/gpu/drm/msm/msm_fb.c @@ -138,7 +138,7 @@ const struct msm_format *msm_framebuffer_format(struct drm_framebuffer *fb) } struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev, - struct drm_file *file, struct drm_mode_fb_cmd2 *mode_cmd) + struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *bos[4] = {0}; struct drm_framebuffer *fb; @@ -168,7 +168,7 @@ out_unref: } struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos) + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos) { struct msm_drm_private *priv = dev->dev_private; struct msm_kms *kms = priv->kms; diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index 3f6ec077b51d..d95af6eba602 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -121,7 +121,7 @@ static int msm_fbdev_create(struct drm_fb_helper *helper, /* note: if fb creation failed, we can't rely on fb destroy * to unref the bo: */ - drm_gem_object_unreference(fbdev->bo); + drm_gem_object_unreference_unlocked(fbdev->bo); ret = PTR_ERR(fb); goto fail; } diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c index 3d96b49fe662..6f04397d43a7 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c +++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c @@ -1081,8 +1081,6 @@ nouveau_crtc_set_config(struct drm_mode_set *set) } static const struct drm_crtc_funcs nv04_crtc_funcs = { - .save = nv_crtc_save, - .restore = nv_crtc_restore, .cursor_set = nv04_crtc_cursor_set, .cursor_move = nv04_crtc_cursor_move, .gamma_set = nv_crtc_gamma_set, @@ -1123,6 +1121,9 @@ nv04_crtc_create(struct drm_device *dev, int crtc_num) nv_crtc->index = crtc_num; nv_crtc->last_dpms = NV_DPMS_CLEARED; + nv_crtc->save = nv_crtc_save; + nv_crtc->restore = nv_crtc_restore; + drm_crtc_init(dev, &nv_crtc->base, &nv04_crtc_funcs); drm_crtc_helper_add(&nv_crtc->base, &nv04_crtc_helper_funcs); drm_mode_crtc_set_gamma_size(&nv_crtc->base, 256); diff --git a/drivers/gpu/drm/nouveau/dispnv04/dac.c b/drivers/gpu/drm/nouveau/dispnv04/dac.c index 78cb033bc015..b48eec395f07 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/dac.c +++ b/drivers/gpu/drm/nouveau/dispnv04/dac.c @@ -504,8 +504,6 @@ static void nv04_dac_destroy(struct drm_encoder *encoder) static const struct drm_encoder_helper_funcs nv04_dac_helper_funcs = { .dpms = nv04_dac_dpms, - .save = nv04_dac_save, - .restore = nv04_dac_restore, .mode_fixup = nv04_dac_mode_fixup, .prepare = nv04_dac_prepare, .commit = nv04_dac_commit, @@ -515,8 +513,6 @@ static const struct drm_encoder_helper_funcs nv04_dac_helper_funcs = { static const struct drm_encoder_helper_funcs nv17_dac_helper_funcs = { .dpms = nv04_dac_dpms, - .save = nv04_dac_save, - .restore = nv04_dac_restore, .mode_fixup = nv04_dac_mode_fixup, .prepare = nv04_dac_prepare, .commit = nv04_dac_commit, @@ -545,12 +541,16 @@ nv04_dac_create(struct drm_connector *connector, struct dcb_output *entry) nv_encoder->dcb = entry; nv_encoder->or = ffs(entry->or) - 1; + nv_encoder->enc_save = nv04_dac_save; + nv_encoder->enc_restore = nv04_dac_restore; + if (nv_gf4_disp_arch(dev)) helper = &nv17_dac_helper_funcs; else helper = &nv04_dac_helper_funcs; - drm_encoder_init(dev, encoder, &nv04_dac_funcs, DRM_MODE_ENCODER_DAC); + drm_encoder_init(dev, encoder, &nv04_dac_funcs, DRM_MODE_ENCODER_DAC, + NULL); drm_encoder_helper_add(encoder, helper); encoder->possible_crtcs = entry->heads; diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c index 429ab5e3025a..05bfd151d1d8 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c +++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c @@ -652,8 +652,6 @@ static void nv04_tmds_slave_init(struct drm_encoder *encoder) static const struct drm_encoder_helper_funcs nv04_lvds_helper_funcs = { .dpms = nv04_lvds_dpms, - .save = nv04_dfp_save, - .restore = nv04_dfp_restore, .mode_fixup = nv04_dfp_mode_fixup, .prepare = nv04_dfp_prepare, .commit = nv04_dfp_commit, @@ -663,8 +661,6 @@ static const struct drm_encoder_helper_funcs nv04_lvds_helper_funcs = { static const struct drm_encoder_helper_funcs nv04_tmds_helper_funcs = { .dpms = nv04_tmds_dpms, - .save = nv04_dfp_save, - .restore = nv04_dfp_restore, .mode_fixup = nv04_dfp_mode_fixup, .prepare = nv04_dfp_prepare, .commit = nv04_dfp_commit, @@ -701,12 +697,15 @@ nv04_dfp_create(struct drm_connector *connector, struct dcb_output *entry) if (!nv_encoder) return -ENOMEM; + nv_encoder->enc_save = nv04_dfp_save; + nv_encoder->enc_restore = nv04_dfp_restore; + encoder = to_drm_encoder(nv_encoder); nv_encoder->dcb = entry; nv_encoder->or = ffs(entry->or) - 1; - drm_encoder_init(connector->dev, encoder, &nv04_dfp_funcs, type); + drm_encoder_init(connector->dev, encoder, &nv04_dfp_funcs, type, NULL); drm_encoder_helper_add(encoder, helper); encoder->possible_crtcs = entry->heads; diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c index 9e650081c357..b4a6bc433ef5 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c @@ -39,7 +39,8 @@ nv04_display_create(struct drm_device *dev) struct dcb_table *dcb = &drm->vbios.dcb; struct drm_connector *connector, *ct; struct drm_encoder *encoder; - struct drm_crtc *crtc; + struct nouveau_encoder *nv_encoder; + struct nouveau_crtc *crtc; struct nv04_display *disp; int i, ret; @@ -107,14 +108,11 @@ nv04_display_create(struct drm_device *dev) } /* Save previous state */ - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) - crtc->funcs->save(crtc); - - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - const struct drm_encoder_helper_funcs *func = encoder->helper_private; + list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) + crtc->save(&crtc->base); - func->save(encoder); - } + list_for_each_entry(nv_encoder, &dev->mode_config.encoder_list, base.base.head) + nv_encoder->enc_save(&nv_encoder->base.base); nouveau_overlay_init(dev); @@ -126,8 +124,9 @@ nv04_display_destroy(struct drm_device *dev) { struct nv04_display *disp = nv04_display(dev); struct nouveau_drm *drm = nouveau_drm(dev); - struct drm_encoder *encoder; + struct nouveau_encoder *encoder; struct drm_crtc *crtc; + struct nouveau_crtc *nv_crtc; /* Turn every CRTC off. */ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { @@ -139,14 +138,11 @@ nv04_display_destroy(struct drm_device *dev) } /* Restore state */ - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - const struct drm_encoder_helper_funcs *func = encoder->helper_private; + list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.base.head) + encoder->enc_restore(&encoder->base.base); - func->restore(encoder); - } - - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) - crtc->funcs->restore(crtc); + list_for_each_entry(nv_crtc, &dev->mode_config.crtc_list, base.head) + nv_crtc->restore(&nv_crtc->base); nouveau_hw_save_vga_fonts(dev, 0); @@ -159,8 +155,8 @@ nv04_display_destroy(struct drm_device *dev) int nv04_display_init(struct drm_device *dev) { - struct drm_encoder *encoder; - struct drm_crtc *crtc; + struct nouveau_encoder *encoder; + struct nouveau_crtc *crtc; /* meh.. modeset apparently doesn't setup all the regs and depends * on pre-existing state, for now load the state of the card *before* @@ -170,14 +166,11 @@ nv04_display_init(struct drm_device *dev) * save/restore "pre-load" state, but more general so we can save * on suspend too. */ - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - const struct drm_encoder_helper_funcs *func = encoder->helper_private; - - func->restore(encoder); - } + list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) + crtc->save(&crtc->base); - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) - crtc->funcs->restore(crtc); + list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.base.head) + encoder->enc_save(&encoder->base.base); return 0; } diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c index 5345eb5378a8..54e9fb9eb5c0 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c +++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c @@ -192,8 +192,6 @@ static const struct drm_encoder_funcs nv04_tv_funcs = { static const struct drm_encoder_helper_funcs nv04_tv_helper_funcs = { .dpms = nv04_tv_dpms, - .save = drm_i2c_encoder_save, - .restore = drm_i2c_encoder_restore, .mode_fixup = drm_i2c_encoder_mode_fixup, .prepare = nv04_tv_prepare, .commit = nv04_tv_commit, @@ -225,9 +223,13 @@ nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry) /* Initialize the common members */ encoder = to_drm_encoder(nv_encoder); - drm_encoder_init(dev, encoder, &nv04_tv_funcs, DRM_MODE_ENCODER_TVDAC); + drm_encoder_init(dev, encoder, &nv04_tv_funcs, DRM_MODE_ENCODER_TVDAC, + NULL); drm_encoder_helper_add(encoder, &nv04_tv_helper_funcs); + nv_encoder->enc_save = drm_i2c_encoder_save; + nv_encoder->enc_restore = drm_i2c_encoder_restore; + encoder->possible_crtcs = entry->heads; encoder->possible_clones = 0; nv_encoder->dcb = entry; diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c index b734195d80a0..d9644c0c5a83 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c +++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c @@ -771,8 +771,6 @@ static void nv17_tv_destroy(struct drm_encoder *encoder) static struct drm_encoder_helper_funcs nv17_tv_helper_funcs = { .dpms = nv17_tv_dpms, - .save = nv17_tv_save, - .restore = nv17_tv_restore, .mode_fixup = nv17_tv_mode_fixup, .prepare = nv17_tv_prepare, .commit = nv17_tv_commit, @@ -816,10 +814,14 @@ nv17_tv_create(struct drm_connector *connector, struct dcb_output *entry) tv_enc->base.dcb = entry; tv_enc->base.or = ffs(entry->or) - 1; - drm_encoder_init(dev, encoder, &nv17_tv_funcs, DRM_MODE_ENCODER_TVDAC); + drm_encoder_init(dev, encoder, &nv17_tv_funcs, DRM_MODE_ENCODER_TVDAC, + NULL); drm_encoder_helper_add(encoder, &nv17_tv_helper_funcs); to_encoder_slave(encoder)->slave_funcs = &nv17_tv_slave_funcs; + tv_enc->base.enc_save = nv17_tv_save; + tv_enc->base.enc_restore = nv17_tv_restore; + encoder->possible_crtcs = entry->heads; encoder->possible_clones = 0; diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h index 28bc202f9753..40f845e31272 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h @@ -7,6 +7,7 @@ struct nvkm_instmem { const struct nvkm_instmem_func *func; struct nvkm_subdev subdev; + spinlock_t lock; struct list_head list; u32 reserved; diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c index 8b8332e46f24..d5e6938cc6bc 100644 --- a/drivers/gpu/drm/nouveau/nouveau_acpi.c +++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c @@ -367,6 +367,7 @@ static int nouveau_rom_call(acpi_handle rom_handle, uint8_t *bios, return -ENODEV; } obj = (union acpi_object *)buffer.pointer; + len = min(len, (int)obj->buffer.length); memcpy(bios+offset, obj->buffer.pointer, len); kfree(buffer.pointer); return len; diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 2e7cbe933533..5dd1d0111cac 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -898,8 +898,6 @@ nouveau_connector_helper_funcs = { static const struct drm_connector_funcs nouveau_connector_funcs = { .dpms = drm_helper_connector_dpms, - .save = NULL, - .restore = NULL, .detect = nouveau_connector_detect, .destroy = nouveau_connector_destroy, .fill_modes = drm_helper_probe_single_connector_modes, @@ -910,8 +908,6 @@ nouveau_connector_funcs = { static const struct drm_connector_funcs nouveau_connector_funcs_lvds = { .dpms = drm_helper_connector_dpms, - .save = NULL, - .restore = NULL, .detect = nouveau_connector_detect_lvds, .destroy = nouveau_connector_destroy, .fill_modes = drm_helper_probe_single_connector_modes, @@ -944,8 +940,6 @@ nouveau_connector_dp_dpms(struct drm_connector *connector, int mode) static const struct drm_connector_funcs nouveau_connector_funcs_dp = { .dpms = nouveau_connector_dp_dpms, - .save = NULL, - .restore = NULL, .detect = nouveau_connector_detect, .destroy = nouveau_connector_destroy, .fill_modes = drm_helper_probe_single_connector_modes, diff --git a/drivers/gpu/drm/nouveau/nouveau_crtc.h b/drivers/gpu/drm/nouveau/nouveau_crtc.h index f19cb1c5fc5a..863f10b8d818 100644 --- a/drivers/gpu/drm/nouveau/nouveau_crtc.h +++ b/drivers/gpu/drm/nouveau/nouveau_crtc.h @@ -73,6 +73,9 @@ struct nouveau_crtc { int (*set_dither)(struct nouveau_crtc *crtc, bool update); int (*set_scale)(struct nouveau_crtc *crtc, bool update); int (*set_color_vibrance)(struct nouveau_crtc *crtc, bool update); + + void (*save)(struct drm_crtc *crtc); + void (*restore)(struct drm_crtc *crtc); }; static inline struct nouveau_crtc *nouveau_crtc(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index db6bc6760545..18676b8c1721 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -246,7 +246,7 @@ static const struct drm_framebuffer_funcs nouveau_framebuffer_funcs = { int nouveau_framebuffer_init(struct drm_device *dev, struct nouveau_framebuffer *nv_fb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct nouveau_bo *nvbo) { struct nouveau_display *disp = nouveau_display(dev); @@ -272,7 +272,7 @@ nouveau_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * nouveau_user_framebuffer_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct nouveau_framebuffer *nouveau_fb; struct drm_gem_object *gem; @@ -829,7 +829,6 @@ nouveau_finish_page_flip(struct nouveau_channel *chan, struct drm_device *dev = drm->dev; struct nouveau_page_flip_state *s; unsigned long flags; - int crtcid = -1; spin_lock_irqsave(&dev->event_lock, flags); @@ -841,15 +840,19 @@ nouveau_finish_page_flip(struct nouveau_channel *chan, s = list_first_entry(&fctx->flip, struct nouveau_page_flip_state, head); if (s->event) { - /* Vblank timestamps/counts are only correct on >= NV-50 */ - if (drm->device.info.family >= NV_DEVICE_INFO_V0_TESLA) - crtcid = s->crtc; + if (drm->device.info.family < NV_DEVICE_INFO_V0_TESLA) { + drm_arm_vblank_event(dev, s->crtc, s->event); + } else { + drm_send_vblank_event(dev, s->crtc, s->event); - drm_send_vblank_event(dev, crtcid, s->event); + /* Give up ownership of vblank for page-flipped crtc */ + drm_vblank_put(dev, s->crtc); + } + } + else { + /* Give up ownership of vblank for page-flipped crtc */ + drm_vblank_put(dev, s->crtc); } - - /* Give up ownership of vblank for page-flipped crtc */ - drm_vblank_put(dev, s->crtc); list_del(&s->head); if (ps) diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h index 856abe0f070d..5a57d8b472c4 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.h +++ b/drivers/gpu/drm/nouveau/nouveau_display.h @@ -23,7 +23,7 @@ nouveau_framebuffer(struct drm_framebuffer *fb) } int nouveau_framebuffer_init(struct drm_device *, struct nouveau_framebuffer *, - struct drm_mode_fb_cmd2 *, struct nouveau_bo *); + const struct drm_mode_fb_cmd2 *, struct nouveau_bo *); struct nouveau_page_flip_state { struct list_head head; diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.h b/drivers/gpu/drm/nouveau/nouveau_drm.h index 3050042e6c6d..a02813e994ec 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.h +++ b/drivers/gpu/drm/nouveau/nouveau_drm.h @@ -39,6 +39,7 @@ #include <nvif/client.h> #include <nvif/device.h> +#include <nvif/ioctl.h> #include <drmP.h> @@ -65,9 +66,10 @@ struct nouveau_drm_tile { }; enum nouveau_drm_object_route { - NVDRM_OBJECT_NVIF = 0, + NVDRM_OBJECT_NVIF = NVIF_IOCTL_V0_OWNER_NVIF, NVDRM_OBJECT_USIF, NVDRM_OBJECT_ABI16, + NVDRM_OBJECT_ANY = NVIF_IOCTL_V0_OWNER_ANY, }; enum nouveau_drm_notify_route { diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h index b37da95105b0..c38a86408363 100644 --- a/drivers/gpu/drm/nouveau/nouveau_encoder.h +++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h @@ -63,6 +63,9 @@ struct nouveau_encoder { u32 datarate; } dp; }; + + void (*enc_save)(struct drm_encoder *encoder); + void (*enc_restore)(struct drm_encoder *encoder); }; struct nouveau_encoder * diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.h b/drivers/gpu/drm/nouveau/nouveau_fbcon.h index 1e2e9e27a03b..ca77ad001978 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.h +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.h @@ -34,7 +34,6 @@ struct nouveau_fbdev { struct drm_fb_helper helper; struct nouveau_framebuffer nouveau_fb; - struct list_head fbdev_list; struct drm_device *dev; unsigned int saved_flags; struct nvif_object surf2d; diff --git a/drivers/gpu/drm/nouveau/nouveau_usif.c b/drivers/gpu/drm/nouveau/nouveau_usif.c index 89dc4ce63490..6ae1b3494bcd 100644 --- a/drivers/gpu/drm/nouveau/nouveau_usif.c +++ b/drivers/gpu/drm/nouveau/nouveau_usif.c @@ -313,7 +313,10 @@ usif_ioctl(struct drm_file *filp, void __user *user, u32 argc) if (nvif_unpack(argv->v0, 0, 0, true)) { /* block access to objects not created via this interface */ owner = argv->v0.owner; - argv->v0.owner = NVDRM_OBJECT_USIF; + if (argv->v0.object == 0ULL) + argv->v0.owner = NVDRM_OBJECT_ANY; /* except client */ + else + argv->v0.owner = NVDRM_OBJECT_USIF; } else goto done; diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c index c053c50b346a..44e1952582aa 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ b/drivers/gpu/drm/nouveau/nv50_display.c @@ -28,6 +28,7 @@ #include <drm/drm_crtc_helper.h> #include <drm/drm_plane_helper.h> #include <drm/drm_dp_helper.h> +#include <drm/drm_fb_helper.h> #include <nvif/class.h> @@ -1717,7 +1718,7 @@ nv50_dac_create(struct drm_connector *connector, struct dcb_output *dcbe) encoder = to_drm_encoder(nv_encoder); encoder->possible_crtcs = dcbe->heads; encoder->possible_clones = 0; - drm_encoder_init(connector->dev, encoder, &nv50_dac_func, type); + drm_encoder_init(connector->dev, encoder, &nv50_dac_func, type, NULL); drm_encoder_helper_add(encoder, &nv50_dac_hfunc); drm_mode_connector_attach_encoder(connector, encoder); @@ -2125,7 +2126,7 @@ nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe) encoder = to_drm_encoder(nv_encoder); encoder->possible_crtcs = dcbe->heads; encoder->possible_clones = 0; - drm_encoder_init(connector->dev, encoder, &nv50_sor_func, type); + drm_encoder_init(connector->dev, encoder, &nv50_sor_func, type, NULL); drm_encoder_helper_add(encoder, &nv50_sor_hfunc); drm_mode_connector_attach_encoder(connector, encoder); @@ -2305,7 +2306,7 @@ nv50_pior_create(struct drm_connector *connector, struct dcb_output *dcbe) encoder = to_drm_encoder(nv_encoder); encoder->possible_crtcs = dcbe->heads; encoder->possible_clones = 0; - drm_encoder_init(connector->dev, encoder, &nv50_pior_func, type); + drm_encoder_init(connector->dev, encoder, &nv50_pior_func, type, NULL); drm_encoder_helper_add(encoder, &nv50_pior_hfunc); drm_mode_connector_attach_encoder(connector, encoder); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c index e3c783d0e2ab..caf22b589edc 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c @@ -279,6 +279,12 @@ nvkm_device_pci_10de_0fe3[] = { }; static const struct nvkm_device_pci_vendor +nvkm_device_pci_10de_0fe4[] = { + { 0x144d, 0xc740, NULL, { .War00C800_0 = true } }, + {} +}; + +static const struct nvkm_device_pci_vendor nvkm_device_pci_10de_104b[] = { { 0x1043, 0x844c, "GeForce GT 625" }, { 0x1043, 0x846b, "GeForce GT 625" }, @@ -689,6 +695,12 @@ nvkm_device_pci_10de_1199[] = { }; static const struct nvkm_device_pci_vendor +nvkm_device_pci_10de_11e0[] = { + { 0x1558, 0x5106, NULL, { .War00C800_0 = true } }, + {} +}; + +static const struct nvkm_device_pci_vendor nvkm_device_pci_10de_11e3[] = { { 0x17aa, 0x3683, "GeForce GTX 760A" }, {} @@ -1370,7 +1382,7 @@ nvkm_device_pci_10de[] = { { 0x0fe1, "GeForce GT 730M" }, { 0x0fe2, "GeForce GT 745M" }, { 0x0fe3, "GeForce GT 745M", nvkm_device_pci_10de_0fe3 }, - { 0x0fe4, "GeForce GT 750M" }, + { 0x0fe4, "GeForce GT 750M", nvkm_device_pci_10de_0fe4 }, { 0x0fe9, "GeForce GT 750M" }, { 0x0fea, "GeForce GT 755M" }, { 0x0fec, "GeForce 710A" }, @@ -1485,7 +1497,7 @@ nvkm_device_pci_10de[] = { { 0x11c6, "GeForce GTX 650 Ti" }, { 0x11c8, "GeForce GTX 650" }, { 0x11cb, "GeForce GT 740" }, - { 0x11e0, "GeForce GTX 770M" }, + { 0x11e0, "GeForce GTX 770M", nvkm_device_pci_10de_11e0 }, { 0x11e1, "GeForce GTX 765M" }, { 0x11e2, "GeForce GTX 765M" }, { 0x11e3, "GeForce GTX 760M", nvkm_device_pci_10de_11e3 }, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c index b5b875928aba..74de7a96c22a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf117.c @@ -207,6 +207,8 @@ gf117_grctx_generate_attrib(struct gf100_grctx *info) const u32 b = beta * gr->ppc_tpc_nr[gpc][ppc]; const u32 t = timeslice_mode; const u32 o = PPC_UNIT(gpc, ppc, 0); + if (!(gr->ppc_mask[gpc] & (1 << ppc))) + continue; mmio_skip(info, o + 0xc0, (t << 28) | (b << 16) | ++bo); mmio_wr32(info, o + 0xc0, (t << 28) | (b << 16) | --bo); bo += grctx->attrib_nr_max * gr->ppc_tpc_nr[gpc][ppc]; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc index 194afe910d21..7dacb3cc0668 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc @@ -52,10 +52,12 @@ mmio_list_base: #endif #ifdef INCLUDE_CODE +#define gpc_addr(reg,addr) /* +*/ imm32(reg,addr) /* +*/ or reg NV_PGRAPH_GPCX_GPCCS_MMIO_CTRL_BASE_ENABLE #define gpc_wr32(addr,reg) /* +*/ gpc_addr($r14,addr) /* */ mov b32 $r15 reg /* -*/ imm32($r14, addr) /* -*/ or $r14 NV_PGRAPH_GPCX_GPCCS_MMIO_CTRL_BASE_ENABLE /* */ call(nv_wr32) // reports an exception to the host @@ -161,7 +163,7 @@ init: #if NV_PGRAPH_GPCX_UNK__SIZE > 0 // figure out which, and how many, UNKs are actually present - imm32($r14, 0x500c30) + gpc_addr($r14, 0x500c30) clear b32 $r2 clear b32 $r3 clear b32 $r4 diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h index 64d07df4b8b1..bb820ff28621 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h @@ -314,7 +314,7 @@ uint32_t gf117_grgpc_code[] = { 0x03f01200, 0x0002d000, 0x17f104bd, - 0x10fe0542, + 0x10fe0545, 0x0007f100, 0x0003f007, 0xbd0000d0, @@ -338,184 +338,184 @@ uint32_t gf117_grgpc_code[] = { 0x02d00103, 0xf104bd00, 0xf00c30e7, - 0x24bd50e3, - 0x44bd34bd, -/* 0x0430: init_unk_loop */ - 0xb06821f4, - 0x0bf400f6, - 0x01f7f00f, - 0xfd04f2bb, - 0x30b6054f, -/* 0x0445: init_unk_next */ - 0x0120b601, - 0xb004e0b6, - 0x1bf40126, -/* 0x0451: init_unk_done */ - 0x070380e2, - 0xf1080480, - 0xf0010027, - 0x22cf0223, - 0x9534bd00, - 0x07f10825, - 0x03f0c000, - 0x0005d001, - 0x07f104bd, - 0x03f0c100, - 0x0005d001, - 0x0e9804bd, - 0x010f9800, - 0x015021f5, - 0xbb002fbb, - 0x0e98003f, - 0x020f9801, - 0x015021f5, - 0xfd050e98, - 0x2ebb00ef, - 0x003ebb00, - 0x98020e98, - 0x21f5030f, - 0x0e980150, - 0x00effd07, - 0xbb002ebb, - 0x35b6003e, - 0x0007f102, - 0x0103f0d3, - 0xbd0003d0, - 0x0825b604, - 0xb60635b6, - 0x30b60120, - 0x0824b601, - 0xb90834b6, - 0x21f5022f, - 0x2fbb02d3, - 0x003fbb00, - 0x010007f1, - 0xd00203f0, + 0xe5f050e3, + 0xbd24bd01, +/* 0x0433: init_unk_loop */ + 0xf444bd34, + 0xf6b06821, + 0x0f0bf400, + 0xbb01f7f0, + 0x4ffd04f2, + 0x0130b605, +/* 0x0448: init_unk_next */ + 0xb60120b6, + 0x26b004e0, + 0xe21bf401, +/* 0x0454: init_unk_done */ + 0x80070380, + 0x27f10804, + 0x23f00100, + 0x0022cf02, + 0x259534bd, + 0x0007f108, + 0x0103f0c0, + 0xbd0005d0, + 0x0007f104, + 0x0103f0c1, + 0xbd0005d0, + 0x000e9804, + 0xf5010f98, + 0xbb015021, + 0x3fbb002f, + 0x010e9800, + 0xf5020f98, + 0x98015021, + 0xeffd050e, + 0x002ebb00, + 0x98003ebb, + 0x0f98020e, + 0x5021f503, + 0x070e9801, + 0xbb00effd, + 0x3ebb002e, + 0x0235b600, + 0xd30007f1, + 0xd00103f0, 0x04bd0003, - 0x29f024bd, - 0x0007f11f, - 0x0203f008, - 0xbd0002d0, -/* 0x0505: main */ - 0x0031f404, - 0xf00028f4, - 0x21f424d7, - 0xf401f439, - 0xf404e4b0, - 0x81fe1e18, - 0x0627f001, - 0x12fd20bd, - 0x01e4b604, - 0xfe051efd, - 0x21f50018, - 0x0ef405fa, -/* 0x0535: main_not_ctx_xfer */ - 0x10ef94d3, - 0xf501f5f0, - 0xf4037e21, -/* 0x0542: ih */ - 0x80f9c60e, - 0xf90188fe, - 0xf990f980, - 0xf9b0f9a0, - 0xf9e0f9d0, - 0xf104bdf0, - 0xf00200a7, - 0xaacf00a3, - 0x04abc400, - 0xf02c0bf4, - 0xe7f124d7, - 0xe3f01a00, - 0x00eecf00, - 0x1900f7f1, - 0xcf00f3f0, - 0x21f400ff, - 0x01e7f004, - 0x1d0007f1, - 0xd00003f0, - 0x04bd000e, -/* 0x0590: ih_no_fifo */ - 0x010007f1, - 0xd00003f0, - 0x04bd000a, - 0xe0fcf0fc, - 0xb0fcd0fc, - 0x90fca0fc, - 0x88fe80fc, - 0xf480fc00, - 0x01f80032, -/* 0x05b4: hub_barrier_done */ - 0x9801f7f0, - 0xfebb040e, - 0x02ffb904, - 0x9418e7f1, - 0xf440e3f0, - 0x00f89d21, -/* 0x05cc: ctx_redswitch */ - 0xf120f7f0, + 0xb60825b6, + 0x20b60635, + 0x0130b601, + 0xb60824b6, + 0x2fb90834, + 0xd321f502, + 0x002fbb02, + 0xf1003fbb, + 0xf0010007, + 0x03d00203, + 0xbd04bd00, + 0x1f29f024, + 0x080007f1, + 0xd00203f0, + 0x04bd0002, +/* 0x0508: main */ + 0xf40031f4, + 0xd7f00028, + 0x3921f424, + 0xb0f401f4, + 0x18f404e4, + 0x0181fe1e, + 0xbd0627f0, + 0x0412fd20, + 0xfd01e4b6, + 0x18fe051e, + 0xfd21f500, + 0xd30ef405, +/* 0x0538: main_not_ctx_xfer */ + 0xf010ef94, + 0x21f501f5, + 0x0ef4037e, +/* 0x0545: ih */ + 0xfe80f9c6, + 0x80f90188, + 0xa0f990f9, + 0xd0f9b0f9, + 0xf0f9e0f9, + 0xa7f104bd, + 0xa3f00200, + 0x00aacf00, + 0xf404abc4, + 0xd7f02c0b, + 0x00e7f124, + 0x00e3f01a, + 0xf100eecf, + 0xf01900f7, + 0xffcf00f3, + 0x0421f400, + 0xf101e7f0, + 0xf01d0007, + 0x0ed00003, +/* 0x0593: ih_no_fifo */ + 0xf104bd00, + 0xf0010007, + 0x0ad00003, + 0xfc04bd00, + 0xfce0fcf0, + 0xfcb0fcd0, + 0xfc90fca0, + 0x0088fe80, + 0x32f480fc, +/* 0x05b7: hub_barrier_done */ + 0xf001f800, + 0x0e9801f7, + 0x04febb04, + 0xf102ffb9, + 0xf09418e7, + 0x21f440e3, +/* 0x05cf: ctx_redswitch */ + 0xf000f89d, + 0x07f120f7, + 0x03f08500, + 0x000fd001, + 0xe7f004bd, +/* 0x05e1: ctx_redswitch_delay */ + 0x01e2b608, + 0xf1fd1bf4, + 0xf10800f5, + 0xf10200f5, 0xf0850007, 0x0fd00103, - 0xf004bd00, -/* 0x05de: ctx_redswitch_delay */ - 0xe2b608e7, - 0xfd1bf401, - 0x0800f5f1, - 0x0200f5f1, - 0x850007f1, - 0xd00103f0, - 0x04bd000f, -/* 0x05fa: ctx_xfer */ - 0x07f100f8, - 0x03f08100, - 0x000fd002, - 0x11f404bd, - 0xcc21f507, -/* 0x060d: ctx_xfer_not_load */ - 0x6a21f505, - 0xf124bd02, - 0xf047fc07, - 0x02d00203, - 0xf004bd00, - 0x20b6012c, - 0xfc07f103, - 0x0203f04a, - 0xbd0002d0, - 0x01acf004, - 0xf102a5f0, - 0xf00000b7, - 0x0c9850b3, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98000c, - 0x00e7f001, - 0x016f21f5, - 0xf101acf0, - 0xf04000b7, - 0x0c9850b3, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98010c, - 0x060f9802, - 0x0800e7f1, - 0x016f21f5, + 0xf804bd00, +/* 0x05fd: ctx_xfer */ + 0x0007f100, + 0x0203f081, + 0xbd000fd0, + 0x0711f404, + 0x05cf21f5, +/* 0x0610: ctx_xfer_not_load */ + 0x026a21f5, + 0x07f124bd, + 0x03f047fc, + 0x0002d002, + 0x2cf004bd, + 0x0320b601, + 0x4afc07f1, + 0xd00203f0, + 0x04bd0002, 0xf001acf0, - 0xb7f104a5, - 0xb3f03000, + 0xb7f102a5, + 0xb3f00000, 0x040c9850, 0xbb0fc4b6, 0x0c9800bc, - 0x030d9802, - 0xf1080f98, - 0xf50200e7, - 0xf5016f21, - 0xf4025e21, - 0x12f40601, -/* 0x06a9: ctx_xfer_post */ - 0x7f21f507, -/* 0x06ad: ctx_xfer_done */ - 0xb421f502, - 0x0000f805, - 0x00000000, + 0x010d9800, + 0xf500e7f0, + 0xf0016f21, + 0xb7f101ac, + 0xb3f04000, + 0x040c9850, + 0xbb0fc4b6, + 0x0c9800bc, + 0x020d9801, + 0xf1060f98, + 0xf50800e7, + 0xf0016f21, + 0xa5f001ac, + 0x00b7f104, + 0x50b3f030, + 0xb6040c98, + 0xbcbb0fc4, + 0x020c9800, + 0x98030d98, + 0xe7f1080f, + 0x21f50200, + 0x21f5016f, + 0x01f4025e, + 0x0712f406, +/* 0x06ac: ctx_xfer_post */ + 0x027f21f5, +/* 0x06b0: ctx_xfer_done */ + 0x05b721f5, + 0x000000f8, 0x00000000, 0x00000000, 0x00000000, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h index 2f596433c222..911976d20940 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h @@ -314,7 +314,7 @@ uint32_t gk104_grgpc_code[] = { 0x03f01200, 0x0002d000, 0x17f104bd, - 0x10fe0542, + 0x10fe0545, 0x0007f100, 0x0003f007, 0xbd0000d0, @@ -338,184 +338,184 @@ uint32_t gk104_grgpc_code[] = { 0x02d00103, 0xf104bd00, 0xf00c30e7, - 0x24bd50e3, - 0x44bd34bd, -/* 0x0430: init_unk_loop */ - 0xb06821f4, - 0x0bf400f6, - 0x01f7f00f, - 0xfd04f2bb, - 0x30b6054f, -/* 0x0445: init_unk_next */ - 0x0120b601, - 0xb004e0b6, - 0x1bf40126, -/* 0x0451: init_unk_done */ - 0x070380e2, - 0xf1080480, - 0xf0010027, - 0x22cf0223, - 0x9534bd00, - 0x07f10825, - 0x03f0c000, - 0x0005d001, - 0x07f104bd, - 0x03f0c100, - 0x0005d001, - 0x0e9804bd, - 0x010f9800, - 0x015021f5, - 0xbb002fbb, - 0x0e98003f, - 0x020f9801, - 0x015021f5, - 0xfd050e98, - 0x2ebb00ef, - 0x003ebb00, - 0x98020e98, - 0x21f5030f, - 0x0e980150, - 0x00effd07, - 0xbb002ebb, - 0x35b6003e, - 0x0007f102, - 0x0103f0d3, - 0xbd0003d0, - 0x0825b604, - 0xb60635b6, - 0x30b60120, - 0x0824b601, - 0xb90834b6, - 0x21f5022f, - 0x2fbb02d3, - 0x003fbb00, - 0x010007f1, - 0xd00203f0, + 0xe5f050e3, + 0xbd24bd01, +/* 0x0433: init_unk_loop */ + 0xf444bd34, + 0xf6b06821, + 0x0f0bf400, + 0xbb01f7f0, + 0x4ffd04f2, + 0x0130b605, +/* 0x0448: init_unk_next */ + 0xb60120b6, + 0x26b004e0, + 0xe21bf401, +/* 0x0454: init_unk_done */ + 0x80070380, + 0x27f10804, + 0x23f00100, + 0x0022cf02, + 0x259534bd, + 0x0007f108, + 0x0103f0c0, + 0xbd0005d0, + 0x0007f104, + 0x0103f0c1, + 0xbd0005d0, + 0x000e9804, + 0xf5010f98, + 0xbb015021, + 0x3fbb002f, + 0x010e9800, + 0xf5020f98, + 0x98015021, + 0xeffd050e, + 0x002ebb00, + 0x98003ebb, + 0x0f98020e, + 0x5021f503, + 0x070e9801, + 0xbb00effd, + 0x3ebb002e, + 0x0235b600, + 0xd30007f1, + 0xd00103f0, 0x04bd0003, - 0x29f024bd, - 0x0007f11f, - 0x0203f008, - 0xbd0002d0, -/* 0x0505: main */ - 0x0031f404, - 0xf00028f4, - 0x21f424d7, - 0xf401f439, - 0xf404e4b0, - 0x81fe1e18, - 0x0627f001, - 0x12fd20bd, - 0x01e4b604, - 0xfe051efd, - 0x21f50018, - 0x0ef405fa, -/* 0x0535: main_not_ctx_xfer */ - 0x10ef94d3, - 0xf501f5f0, - 0xf4037e21, -/* 0x0542: ih */ - 0x80f9c60e, - 0xf90188fe, - 0xf990f980, - 0xf9b0f9a0, - 0xf9e0f9d0, - 0xf104bdf0, - 0xf00200a7, - 0xaacf00a3, - 0x04abc400, - 0xf02c0bf4, - 0xe7f124d7, - 0xe3f01a00, - 0x00eecf00, - 0x1900f7f1, - 0xcf00f3f0, - 0x21f400ff, - 0x01e7f004, - 0x1d0007f1, - 0xd00003f0, - 0x04bd000e, -/* 0x0590: ih_no_fifo */ - 0x010007f1, - 0xd00003f0, - 0x04bd000a, - 0xe0fcf0fc, - 0xb0fcd0fc, - 0x90fca0fc, - 0x88fe80fc, - 0xf480fc00, - 0x01f80032, -/* 0x05b4: hub_barrier_done */ - 0x9801f7f0, - 0xfebb040e, - 0x02ffb904, - 0x9418e7f1, - 0xf440e3f0, - 0x00f89d21, -/* 0x05cc: ctx_redswitch */ - 0xf120f7f0, + 0xb60825b6, + 0x20b60635, + 0x0130b601, + 0xb60824b6, + 0x2fb90834, + 0xd321f502, + 0x002fbb02, + 0xf1003fbb, + 0xf0010007, + 0x03d00203, + 0xbd04bd00, + 0x1f29f024, + 0x080007f1, + 0xd00203f0, + 0x04bd0002, +/* 0x0508: main */ + 0xf40031f4, + 0xd7f00028, + 0x3921f424, + 0xb0f401f4, + 0x18f404e4, + 0x0181fe1e, + 0xbd0627f0, + 0x0412fd20, + 0xfd01e4b6, + 0x18fe051e, + 0xfd21f500, + 0xd30ef405, +/* 0x0538: main_not_ctx_xfer */ + 0xf010ef94, + 0x21f501f5, + 0x0ef4037e, +/* 0x0545: ih */ + 0xfe80f9c6, + 0x80f90188, + 0xa0f990f9, + 0xd0f9b0f9, + 0xf0f9e0f9, + 0xa7f104bd, + 0xa3f00200, + 0x00aacf00, + 0xf404abc4, + 0xd7f02c0b, + 0x00e7f124, + 0x00e3f01a, + 0xf100eecf, + 0xf01900f7, + 0xffcf00f3, + 0x0421f400, + 0xf101e7f0, + 0xf01d0007, + 0x0ed00003, +/* 0x0593: ih_no_fifo */ + 0xf104bd00, + 0xf0010007, + 0x0ad00003, + 0xfc04bd00, + 0xfce0fcf0, + 0xfcb0fcd0, + 0xfc90fca0, + 0x0088fe80, + 0x32f480fc, +/* 0x05b7: hub_barrier_done */ + 0xf001f800, + 0x0e9801f7, + 0x04febb04, + 0xf102ffb9, + 0xf09418e7, + 0x21f440e3, +/* 0x05cf: ctx_redswitch */ + 0xf000f89d, + 0x07f120f7, + 0x03f08500, + 0x000fd001, + 0xe7f004bd, +/* 0x05e1: ctx_redswitch_delay */ + 0x01e2b608, + 0xf1fd1bf4, + 0xf10800f5, + 0xf10200f5, 0xf0850007, 0x0fd00103, - 0xf004bd00, -/* 0x05de: ctx_redswitch_delay */ - 0xe2b608e7, - 0xfd1bf401, - 0x0800f5f1, - 0x0200f5f1, - 0x850007f1, - 0xd00103f0, - 0x04bd000f, -/* 0x05fa: ctx_xfer */ - 0x07f100f8, - 0x03f08100, - 0x000fd002, - 0x11f404bd, - 0xcc21f507, -/* 0x060d: ctx_xfer_not_load */ - 0x6a21f505, - 0xf124bd02, - 0xf047fc07, - 0x02d00203, - 0xf004bd00, - 0x20b6012c, - 0xfc07f103, - 0x0203f04a, - 0xbd0002d0, - 0x01acf004, - 0xf102a5f0, - 0xf00000b7, - 0x0c9850b3, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98000c, - 0x00e7f001, - 0x016f21f5, - 0xf101acf0, - 0xf04000b7, - 0x0c9850b3, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98010c, - 0x060f9802, - 0x0800e7f1, - 0x016f21f5, + 0xf804bd00, +/* 0x05fd: ctx_xfer */ + 0x0007f100, + 0x0203f081, + 0xbd000fd0, + 0x0711f404, + 0x05cf21f5, +/* 0x0610: ctx_xfer_not_load */ + 0x026a21f5, + 0x07f124bd, + 0x03f047fc, + 0x0002d002, + 0x2cf004bd, + 0x0320b601, + 0x4afc07f1, + 0xd00203f0, + 0x04bd0002, 0xf001acf0, - 0xb7f104a5, - 0xb3f03000, + 0xb7f102a5, + 0xb3f00000, 0x040c9850, 0xbb0fc4b6, 0x0c9800bc, - 0x030d9802, - 0xf1080f98, - 0xf50200e7, - 0xf5016f21, - 0xf4025e21, - 0x12f40601, -/* 0x06a9: ctx_xfer_post */ - 0x7f21f507, -/* 0x06ad: ctx_xfer_done */ - 0xb421f502, - 0x0000f805, - 0x00000000, + 0x010d9800, + 0xf500e7f0, + 0xf0016f21, + 0xb7f101ac, + 0xb3f04000, + 0x040c9850, + 0xbb0fc4b6, + 0x0c9800bc, + 0x020d9801, + 0xf1060f98, + 0xf50800e7, + 0xf0016f21, + 0xa5f001ac, + 0x00b7f104, + 0x50b3f030, + 0xb6040c98, + 0xbcbb0fc4, + 0x020c9800, + 0x98030d98, + 0xe7f1080f, + 0x21f50200, + 0x21f5016f, + 0x01f4025e, + 0x0712f406, +/* 0x06ac: ctx_xfer_post */ + 0x027f21f5, +/* 0x06b0: ctx_xfer_done */ + 0x05b721f5, + 0x000000f8, 0x00000000, 0x00000000, 0x00000000, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h index ee8e54db8fc9..1c6e11b05df2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h @@ -314,7 +314,7 @@ uint32_t gk110_grgpc_code[] = { 0x03f01200, 0x0002d000, 0x17f104bd, - 0x10fe0542, + 0x10fe0545, 0x0007f100, 0x0003f007, 0xbd0000d0, @@ -338,184 +338,184 @@ uint32_t gk110_grgpc_code[] = { 0x02d00103, 0xf104bd00, 0xf00c30e7, - 0x24bd50e3, - 0x44bd34bd, -/* 0x0430: init_unk_loop */ - 0xb06821f4, - 0x0bf400f6, - 0x01f7f00f, - 0xfd04f2bb, - 0x30b6054f, -/* 0x0445: init_unk_next */ - 0x0120b601, - 0xb004e0b6, - 0x1bf40226, -/* 0x0451: init_unk_done */ - 0x070380e2, - 0xf1080480, - 0xf0010027, - 0x22cf0223, - 0x9534bd00, - 0x07f10825, - 0x03f0c000, - 0x0005d001, - 0x07f104bd, - 0x03f0c100, - 0x0005d001, - 0x0e9804bd, - 0x010f9800, - 0x015021f5, - 0xbb002fbb, - 0x0e98003f, - 0x020f9801, - 0x015021f5, - 0xfd050e98, - 0x2ebb00ef, - 0x003ebb00, - 0x98020e98, - 0x21f5030f, - 0x0e980150, - 0x00effd07, - 0xbb002ebb, - 0x35b6003e, - 0x0007f102, - 0x0103f0d3, - 0xbd0003d0, - 0x0825b604, - 0xb60635b6, - 0x30b60120, - 0x0824b601, - 0xb90834b6, - 0x21f5022f, - 0x2fbb02d3, - 0x003fbb00, - 0x010007f1, - 0xd00203f0, + 0xe5f050e3, + 0xbd24bd01, +/* 0x0433: init_unk_loop */ + 0xf444bd34, + 0xf6b06821, + 0x0f0bf400, + 0xbb01f7f0, + 0x4ffd04f2, + 0x0130b605, +/* 0x0448: init_unk_next */ + 0xb60120b6, + 0x26b004e0, + 0xe21bf402, +/* 0x0454: init_unk_done */ + 0x80070380, + 0x27f10804, + 0x23f00100, + 0x0022cf02, + 0x259534bd, + 0x0007f108, + 0x0103f0c0, + 0xbd0005d0, + 0x0007f104, + 0x0103f0c1, + 0xbd0005d0, + 0x000e9804, + 0xf5010f98, + 0xbb015021, + 0x3fbb002f, + 0x010e9800, + 0xf5020f98, + 0x98015021, + 0xeffd050e, + 0x002ebb00, + 0x98003ebb, + 0x0f98020e, + 0x5021f503, + 0x070e9801, + 0xbb00effd, + 0x3ebb002e, + 0x0235b600, + 0xd30007f1, + 0xd00103f0, 0x04bd0003, - 0x29f024bd, - 0x0007f11f, - 0x0203f030, - 0xbd0002d0, -/* 0x0505: main */ - 0x0031f404, - 0xf00028f4, - 0x21f424d7, - 0xf401f439, - 0xf404e4b0, - 0x81fe1e18, - 0x0627f001, - 0x12fd20bd, - 0x01e4b604, - 0xfe051efd, - 0x21f50018, - 0x0ef405fa, -/* 0x0535: main_not_ctx_xfer */ - 0x10ef94d3, - 0xf501f5f0, - 0xf4037e21, -/* 0x0542: ih */ - 0x80f9c60e, - 0xf90188fe, - 0xf990f980, - 0xf9b0f9a0, - 0xf9e0f9d0, - 0xf104bdf0, - 0xf00200a7, - 0xaacf00a3, - 0x04abc400, - 0xf02c0bf4, - 0xe7f124d7, - 0xe3f01a00, - 0x00eecf00, - 0x1900f7f1, - 0xcf00f3f0, - 0x21f400ff, - 0x01e7f004, - 0x1d0007f1, - 0xd00003f0, - 0x04bd000e, -/* 0x0590: ih_no_fifo */ - 0x010007f1, - 0xd00003f0, - 0x04bd000a, - 0xe0fcf0fc, - 0xb0fcd0fc, - 0x90fca0fc, - 0x88fe80fc, - 0xf480fc00, - 0x01f80032, -/* 0x05b4: hub_barrier_done */ - 0x9801f7f0, - 0xfebb040e, - 0x02ffb904, - 0x9418e7f1, - 0xf440e3f0, - 0x00f89d21, -/* 0x05cc: ctx_redswitch */ - 0xf120f7f0, + 0xb60825b6, + 0x20b60635, + 0x0130b601, + 0xb60824b6, + 0x2fb90834, + 0xd321f502, + 0x002fbb02, + 0xf1003fbb, + 0xf0010007, + 0x03d00203, + 0xbd04bd00, + 0x1f29f024, + 0x300007f1, + 0xd00203f0, + 0x04bd0002, +/* 0x0508: main */ + 0xf40031f4, + 0xd7f00028, + 0x3921f424, + 0xb0f401f4, + 0x18f404e4, + 0x0181fe1e, + 0xbd0627f0, + 0x0412fd20, + 0xfd01e4b6, + 0x18fe051e, + 0xfd21f500, + 0xd30ef405, +/* 0x0538: main_not_ctx_xfer */ + 0xf010ef94, + 0x21f501f5, + 0x0ef4037e, +/* 0x0545: ih */ + 0xfe80f9c6, + 0x80f90188, + 0xa0f990f9, + 0xd0f9b0f9, + 0xf0f9e0f9, + 0xa7f104bd, + 0xa3f00200, + 0x00aacf00, + 0xf404abc4, + 0xd7f02c0b, + 0x00e7f124, + 0x00e3f01a, + 0xf100eecf, + 0xf01900f7, + 0xffcf00f3, + 0x0421f400, + 0xf101e7f0, + 0xf01d0007, + 0x0ed00003, +/* 0x0593: ih_no_fifo */ + 0xf104bd00, + 0xf0010007, + 0x0ad00003, + 0xfc04bd00, + 0xfce0fcf0, + 0xfcb0fcd0, + 0xfc90fca0, + 0x0088fe80, + 0x32f480fc, +/* 0x05b7: hub_barrier_done */ + 0xf001f800, + 0x0e9801f7, + 0x04febb04, + 0xf102ffb9, + 0xf09418e7, + 0x21f440e3, +/* 0x05cf: ctx_redswitch */ + 0xf000f89d, + 0x07f120f7, + 0x03f08500, + 0x000fd001, + 0xe7f004bd, +/* 0x05e1: ctx_redswitch_delay */ + 0x01e2b608, + 0xf1fd1bf4, + 0xf10800f5, + 0xf10200f5, 0xf0850007, 0x0fd00103, - 0xf004bd00, -/* 0x05de: ctx_redswitch_delay */ - 0xe2b608e7, - 0xfd1bf401, - 0x0800f5f1, - 0x0200f5f1, - 0x850007f1, - 0xd00103f0, - 0x04bd000f, -/* 0x05fa: ctx_xfer */ - 0x07f100f8, - 0x03f08100, - 0x000fd002, - 0x11f404bd, - 0xcc21f507, -/* 0x060d: ctx_xfer_not_load */ - 0x6a21f505, - 0xf124bd02, - 0xf047fc07, - 0x02d00203, - 0xf004bd00, - 0x20b6012c, - 0xfc07f103, - 0x0203f04a, - 0xbd0002d0, - 0x01acf004, - 0xf102a5f0, - 0xf00000b7, - 0x0c9850b3, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98000c, - 0x00e7f001, - 0x016f21f5, - 0xf101acf0, - 0xf04000b7, - 0x0c9850b3, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98010c, - 0x060f9802, - 0x0800e7f1, - 0x016f21f5, + 0xf804bd00, +/* 0x05fd: ctx_xfer */ + 0x0007f100, + 0x0203f081, + 0xbd000fd0, + 0x0711f404, + 0x05cf21f5, +/* 0x0610: ctx_xfer_not_load */ + 0x026a21f5, + 0x07f124bd, + 0x03f047fc, + 0x0002d002, + 0x2cf004bd, + 0x0320b601, + 0x4afc07f1, + 0xd00203f0, + 0x04bd0002, 0xf001acf0, - 0xb7f104a5, - 0xb3f03000, + 0xb7f102a5, + 0xb3f00000, 0x040c9850, 0xbb0fc4b6, 0x0c9800bc, - 0x030d9802, - 0xf1080f98, - 0xf50200e7, - 0xf5016f21, - 0xf4025e21, - 0x12f40601, -/* 0x06a9: ctx_xfer_post */ - 0x7f21f507, -/* 0x06ad: ctx_xfer_done */ - 0xb421f502, - 0x0000f805, - 0x00000000, + 0x010d9800, + 0xf500e7f0, + 0xf0016f21, + 0xb7f101ac, + 0xb3f04000, + 0x040c9850, + 0xbb0fc4b6, + 0x0c9800bc, + 0x020d9801, + 0xf1060f98, + 0xf50800e7, + 0xf0016f21, + 0xa5f001ac, + 0x00b7f104, + 0x50b3f030, + 0xb6040c98, + 0xbcbb0fc4, + 0x020c9800, + 0x98030d98, + 0xe7f1080f, + 0x21f50200, + 0x21f5016f, + 0x01f4025e, + 0x0712f406, +/* 0x06ac: ctx_xfer_post */ + 0x027f21f5, +/* 0x06b0: ctx_xfer_done */ + 0x05b721f5, + 0x000000f8, 0x00000000, 0x00000000, 0x00000000, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h index fbcc342f896f..84af7ec6a78e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h @@ -276,7 +276,7 @@ uint32_t gk208_grgpc_code[] = { 0x02020014, 0xf6120040, 0x04bd0002, - 0xfe048141, + 0xfe048441, 0x00400010, 0x0000f607, 0x040204bd, @@ -295,165 +295,165 @@ uint32_t gk208_grgpc_code[] = { 0x01c90080, 0xbd0002f6, 0x0c308e04, - 0xbd24bd50, -/* 0x0383: init_unk_loop */ - 0x7e44bd34, - 0xb0000065, - 0x0bf400f6, - 0xbb010f0e, - 0x4ffd04f2, - 0x0130b605, -/* 0x0398: init_unk_next */ - 0xb60120b6, - 0x26b004e0, - 0xe21bf401, -/* 0x03a4: init_unk_done */ - 0xb50703b5, - 0x00820804, - 0x22cf0201, - 0x9534bd00, - 0x00800825, - 0x05f601c0, - 0x8004bd00, - 0xf601c100, - 0x04bd0005, - 0x98000e98, - 0x207e010f, - 0x2fbb0001, - 0x003fbb00, - 0x98010e98, - 0x207e020f, - 0x0e980001, - 0x00effd05, - 0xbb002ebb, - 0x0e98003e, - 0x030f9802, - 0x0001207e, - 0xfd070e98, - 0x2ebb00ef, - 0x003ebb00, - 0x800235b6, - 0xf601d300, - 0x04bd0003, - 0xb60825b6, - 0x20b60635, - 0x0130b601, - 0xb60824b6, - 0x2fb20834, - 0x0002687e, - 0xbb002fbb, - 0x0080003f, - 0x03f60201, - 0xbd04bd00, - 0x1f29f024, - 0x02300080, - 0xbd0002f6, -/* 0x0445: main */ - 0x0031f404, - 0x0d0028f4, - 0x00377e24, - 0xf401f400, - 0xf404e4b0, - 0x81fe1d18, - 0xbd060201, - 0x0412fd20, - 0xfd01e4b6, - 0x18fe051e, - 0x05187e00, - 0xd40ef400, -/* 0x0474: main_not_ctx_xfer */ - 0xf010ef94, - 0xf87e01f5, - 0x0ef40002, -/* 0x0481: ih */ - 0xfe80f9c7, - 0x80f90188, - 0xa0f990f9, - 0xd0f9b0f9, - 0xf0f9e0f9, - 0x004a04bd, - 0x00aacf02, - 0xf404abc4, - 0x240d1f0b, - 0xcf1a004e, - 0x004f00ee, - 0x00ffcf19, - 0x0000047e, - 0x0040010e, - 0x000ef61d, -/* 0x04be: ih_no_fifo */ - 0x004004bd, - 0x000af601, - 0xf0fc04bd, - 0xd0fce0fc, - 0xa0fcb0fc, - 0x80fc90fc, - 0xfc0088fe, - 0x0032f480, -/* 0x04de: hub_barrier_done */ - 0x010f01f8, - 0xbb040e98, - 0xffb204fe, - 0x4094188e, - 0x00008f7e, -/* 0x04f2: ctx_redswitch */ - 0x200f00f8, + 0x01e5f050, + 0x34bd24bd, +/* 0x0386: init_unk_loop */ + 0x657e44bd, + 0xf6b00000, + 0x0e0bf400, + 0xf2bb010f, + 0x054ffd04, +/* 0x039b: init_unk_next */ + 0xb60130b6, + 0xe0b60120, + 0x0126b004, +/* 0x03a7: init_unk_done */ + 0xb5e21bf4, + 0x04b50703, + 0x01008208, + 0x0022cf02, + 0x259534bd, + 0xc0008008, + 0x0005f601, + 0x008004bd, + 0x05f601c1, + 0x9804bd00, + 0x0f98000e, + 0x01207e01, + 0x002fbb00, + 0x98003fbb, + 0x0f98010e, + 0x01207e02, + 0x050e9800, + 0xbb00effd, + 0x3ebb002e, + 0x020e9800, + 0x7e030f98, + 0x98000120, + 0xeffd070e, + 0x002ebb00, + 0xb6003ebb, + 0x00800235, + 0x03f601d3, + 0xb604bd00, + 0x35b60825, + 0x0120b606, + 0xb60130b6, + 0x34b60824, + 0x7e2fb208, + 0xbb000268, + 0x3fbb002f, + 0x01008000, + 0x0003f602, + 0x24bd04bd, + 0x801f29f0, + 0xf6023000, + 0x04bd0002, +/* 0x0448: main */ + 0xf40031f4, + 0x240d0028, + 0x0000377e, + 0xb0f401f4, + 0x18f404e4, + 0x0181fe1d, + 0x20bd0602, + 0xb60412fd, + 0x1efd01e4, + 0x0018fe05, + 0x00051b7e, +/* 0x0477: main_not_ctx_xfer */ + 0x94d40ef4, + 0xf5f010ef, + 0x02f87e01, + 0xc70ef400, +/* 0x0484: ih */ + 0x88fe80f9, + 0xf980f901, + 0xf9a0f990, + 0xf9d0f9b0, + 0xbdf0f9e0, + 0x02004a04, + 0xc400aacf, + 0x0bf404ab, + 0x4e240d1f, + 0xeecf1a00, + 0x19004f00, + 0x7e00ffcf, + 0x0e000004, + 0x1d004001, + 0xbd000ef6, +/* 0x04c1: ih_no_fifo */ + 0x01004004, + 0xbd000af6, + 0xfcf0fc04, + 0xfcd0fce0, + 0xfca0fcb0, + 0xfe80fc90, + 0x80fc0088, + 0xf80032f4, +/* 0x04e1: hub_barrier_done */ + 0x98010f01, + 0xfebb040e, + 0x8effb204, + 0x7e409418, + 0xf800008f, +/* 0x04f5: ctx_redswitch */ + 0x80200f00, + 0xf6018500, + 0x04bd000f, +/* 0x0502: ctx_redswitch_delay */ + 0xe2b6080e, + 0xfd1bf401, + 0x0800f5f1, + 0x0200f5f1, 0x01850080, 0xbd000ff6, -/* 0x04ff: ctx_redswitch_delay */ - 0xb6080e04, - 0x1bf401e2, - 0x00f5f1fd, - 0x00f5f108, - 0x85008002, - 0x000ff601, - 0x00f804bd, -/* 0x0518: ctx_xfer */ - 0x02810080, - 0xbd000ff6, - 0x0711f404, - 0x0004f27e, -/* 0x0528: ctx_xfer_not_load */ - 0x0002167e, - 0xfc8024bd, - 0x02f60247, - 0xf004bd00, - 0x20b6012c, - 0x4afc8003, +/* 0x051b: ctx_xfer */ + 0x8000f804, + 0xf6028100, + 0x04bd000f, + 0x7e0711f4, +/* 0x052b: ctx_xfer_not_load */ + 0x7e0004f5, + 0xbd000216, + 0x47fc8024, 0x0002f602, - 0xacf004bd, - 0x02a5f001, - 0x5000008b, - 0xb6040c98, - 0xbcbb0fc4, - 0x000c9800, - 0x0e010d98, - 0x013d7e00, - 0x01acf000, - 0x5040008b, - 0xb6040c98, - 0xbcbb0fc4, - 0x010c9800, - 0x98020d98, - 0x004e060f, - 0x013d7e08, - 0x01acf000, - 0x8b04a5f0, - 0x98503000, + 0x2cf004bd, + 0x0320b601, + 0x024afc80, + 0xbd0002f6, + 0x01acf004, + 0x8b02a5f0, + 0x98500000, 0xc4b6040c, 0x00bcbb0f, - 0x98020c98, - 0x0f98030d, - 0x02004e08, + 0x98000c98, + 0x000e010d, 0x00013d7e, - 0x00020a7e, - 0xf40601f4, -/* 0x05b2: ctx_xfer_post */ - 0x277e0712, -/* 0x05b6: ctx_xfer_done */ - 0xde7e0002, - 0x00f80004, - 0x00000000, + 0x8b01acf0, + 0x98504000, + 0xc4b6040c, + 0x00bcbb0f, + 0x98010c98, + 0x0f98020d, + 0x08004e06, + 0x00013d7e, + 0xf001acf0, + 0x008b04a5, + 0x0c985030, + 0x0fc4b604, + 0x9800bcbb, + 0x0d98020c, + 0x080f9803, + 0x7e02004e, + 0x7e00013d, + 0xf400020a, + 0x12f40601, +/* 0x05b5: ctx_xfer_post */ + 0x02277e07, +/* 0x05b9: ctx_xfer_done */ + 0x04e17e00, + 0x0000f800, 0x00000000, 0x00000000, 0x00000000, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h index 51f5c3c6e966..11bf363a6ae9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h @@ -289,7 +289,7 @@ uint32_t gm107_grgpc_code[] = { 0x020014fe, 0x12004002, 0xbd0002f6, - 0x05b04104, + 0x05b34104, 0x400010fe, 0x00f60700, 0x0204bd00, @@ -308,259 +308,259 @@ uint32_t gm107_grgpc_code[] = { 0xc900800f, 0x0002f601, 0x308e04bd, - 0x24bd500c, - 0x44bd34bd, -/* 0x03b0: init_unk_loop */ - 0x0000657e, - 0xf400f6b0, - 0x010f0e0b, - 0xfd04f2bb, - 0x30b6054f, -/* 0x03c5: init_unk_next */ - 0x0120b601, - 0xb004e0b6, - 0x1bf40226, -/* 0x03d1: init_unk_done */ - 0x0703b5e2, - 0x820804b5, - 0xcf020100, - 0x34bd0022, - 0x80082595, - 0xf601c000, + 0xe5f0500c, + 0xbd24bd01, +/* 0x03b3: init_unk_loop */ + 0x7e44bd34, + 0xb0000065, + 0x0bf400f6, + 0xbb010f0e, + 0x4ffd04f2, + 0x0130b605, +/* 0x03c8: init_unk_next */ + 0xb60120b6, + 0x26b004e0, + 0xe21bf402, +/* 0x03d4: init_unk_done */ + 0xb50703b5, + 0x00820804, + 0x22cf0201, + 0x9534bd00, + 0x00800825, + 0x05f601c0, + 0x8004bd00, + 0xf601c100, 0x04bd0005, - 0x01c10080, - 0xbd0005f6, - 0x000e9804, - 0x7e010f98, - 0xbb000120, - 0x3fbb002f, - 0x010e9800, - 0x7e020f98, - 0x98000120, - 0xeffd050e, - 0x002ebb00, - 0x98003ebb, - 0x0f98020e, - 0x01207e03, - 0x070e9800, - 0xbb00effd, - 0x3ebb002e, - 0x0235b600, - 0x01d30080, - 0xbd0003f6, - 0x0825b604, - 0xb60635b6, - 0x30b60120, - 0x0824b601, - 0xb20834b6, - 0x02687e2f, - 0x002fbb00, - 0x0f003fbb, - 0x8effb23f, - 0xf0501d60, - 0x8f7e01e5, - 0x0c0f0000, - 0xa88effb2, - 0xe5f0501d, - 0x008f7e01, - 0x03147e00, - 0xb23f0f00, - 0x1d608eff, - 0x01e5f050, - 0x00008f7e, - 0xffb2000f, - 0x501d9c8e, - 0x7e01e5f0, - 0x0f00008f, - 0x03147e01, - 0x8effb200, + 0x98000e98, + 0x207e010f, + 0x2fbb0001, + 0x003fbb00, + 0x98010e98, + 0x207e020f, + 0x0e980001, + 0x00effd05, + 0xbb002ebb, + 0x0e98003e, + 0x030f9802, + 0x0001207e, + 0xfd070e98, + 0x2ebb00ef, + 0x003ebb00, + 0x800235b6, + 0xf601d300, + 0x04bd0003, + 0xb60825b6, + 0x20b60635, + 0x0130b601, + 0xb60824b6, + 0x2fb20834, + 0x0002687e, + 0xbb002fbb, + 0x3f0f003f, + 0x501d608e, + 0xb201e5f0, + 0x008f7eff, + 0x8e0c0f00, 0xf0501da8, - 0x8f7e01e5, - 0xff0f0000, - 0x988effb2, + 0xffb201e5, + 0x00008f7e, + 0x0003147e, + 0x608e3f0f, 0xe5f0501d, - 0x008f7e01, - 0xb2020f00, - 0x1da88eff, + 0x7effb201, + 0x0f00008f, + 0x1d9c8e00, 0x01e5f050, - 0x00008f7e, + 0x8f7effb2, + 0x010f0000, 0x0003147e, - 0x85050498, - 0x98504000, - 0x64b60406, - 0x0056bb0f, -/* 0x04e0: tpc_strand_init_tpc_loop */ - 0x05705eb8, - 0x00657e00, - 0xbdf6b200, -/* 0x04ed: tpc_strand_init_idx_loop */ - 0x605eb874, - 0x7fb20005, - 0x00008f7e, - 0x05885eb8, - 0x082f9500, - 0x00008f7e, - 0x058c5eb8, - 0x082f9500, + 0x501da88e, + 0xb201e5f0, + 0x008f7eff, + 0x8eff0f00, + 0xf0501d98, + 0xffb201e5, 0x00008f7e, - 0x05905eb8, - 0x00657e00, - 0x06f5b600, - 0xb601f0b6, - 0x2fbb08f4, - 0x003fbb00, - 0xb60170b6, - 0x1bf40162, - 0x0050b7bf, - 0x0142b608, - 0x0fa81bf4, - 0x8effb23f, - 0xf0501d60, - 0x8f7e01e5, - 0x0d0f0000, - 0xa88effb2, + 0xa88e020f, 0xe5f0501d, - 0x008f7e01, - 0x03147e00, - 0x01008000, - 0x0003f602, - 0x24bd04bd, - 0x801f29f0, - 0xf6023000, - 0x04bd0002, -/* 0x0574: main */ - 0xf40031f4, - 0x240d0028, - 0x0000377e, - 0xb0f401f4, - 0x18f404e4, - 0x0181fe1d, - 0x20bd0602, - 0xb60412fd, - 0x1efd01e4, - 0x0018fe05, - 0x0006477e, -/* 0x05a3: main_not_ctx_xfer */ - 0x94d40ef4, - 0xf5f010ef, - 0x02f87e01, - 0xc70ef400, -/* 0x05b0: ih */ - 0x88fe80f9, - 0xf980f901, - 0xf9a0f990, - 0xf9d0f9b0, - 0xbdf0f9e0, - 0x02004a04, - 0xc400aacf, - 0x0bf404ab, - 0x4e240d1f, - 0xeecf1a00, - 0x19004f00, - 0x7e00ffcf, - 0x0e000004, - 0x1d004001, - 0xbd000ef6, -/* 0x05ed: ih_no_fifo */ - 0x01004004, - 0xbd000af6, - 0xfcf0fc04, - 0xfcd0fce0, - 0xfca0fcb0, - 0xfe80fc90, - 0x80fc0088, - 0xf80032f4, -/* 0x060d: hub_barrier_done */ - 0x98010f01, - 0xfebb040e, - 0x8effb204, - 0x7e409418, - 0xf800008f, -/* 0x0621: ctx_redswitch */ - 0x80200f00, + 0x7effb201, + 0x7e00008f, + 0x98000314, + 0x00850504, + 0x06985040, + 0x0f64b604, +/* 0x04e3: tpc_strand_init_tpc_loop */ + 0xb80056bb, + 0x0005705e, + 0x0000657e, + 0x74bdf6b2, +/* 0x04f0: tpc_strand_init_idx_loop */ + 0x05605eb8, + 0x7e7fb200, + 0xb800008f, + 0x0005885e, + 0x7e082f95, + 0xb800008f, + 0x00058c5e, + 0x7e082f95, + 0xb800008f, + 0x0005905e, + 0x0000657e, + 0xb606f5b6, + 0xf4b601f0, + 0x002fbb08, + 0xb6003fbb, + 0x62b60170, + 0xbf1bf401, + 0x080050b7, + 0xf40142b6, + 0x3f0fa81b, + 0x501d608e, + 0xb201e5f0, + 0x008f7eff, + 0x8e0d0f00, + 0xf0501da8, + 0xffb201e5, + 0x00008f7e, + 0x0003147e, + 0x02010080, + 0xbd0003f6, + 0xf024bd04, + 0x00801f29, + 0x02f60230, +/* 0x0577: main */ + 0xf404bd00, + 0x28f40031, + 0x7e240d00, + 0xf4000037, + 0xe4b0f401, + 0x1d18f404, + 0x020181fe, + 0xfd20bd06, + 0xe4b60412, + 0x051efd01, + 0x7e0018fe, + 0xf400064a, +/* 0x05a6: main_not_ctx_xfer */ + 0xef94d40e, + 0x01f5f010, + 0x0002f87e, +/* 0x05b3: ih */ + 0xf9c70ef4, + 0x0188fe80, + 0x90f980f9, + 0xb0f9a0f9, + 0xe0f9d0f9, + 0x04bdf0f9, + 0xcf02004a, + 0xabc400aa, + 0x1f0bf404, + 0x004e240d, + 0x00eecf1a, + 0xcf19004f, + 0x047e00ff, + 0x010e0000, + 0xf61d0040, + 0x04bd000e, +/* 0x05f0: ih_no_fifo */ + 0xf6010040, + 0x04bd000a, + 0xe0fcf0fc, + 0xb0fcd0fc, + 0x90fca0fc, + 0x88fe80fc, + 0xf480fc00, + 0x01f80032, +/* 0x0610: hub_barrier_done */ + 0x0e98010f, + 0x04febb04, + 0x188effb2, + 0x8f7e4094, + 0x00f80000, +/* 0x0624: ctx_redswitch */ + 0x0080200f, + 0x0ff60185, + 0x0e04bd00, +/* 0x0631: ctx_redswitch_delay */ + 0x01e2b608, + 0xf1fd1bf4, + 0xf10800f5, + 0x800200f5, 0xf6018500, 0x04bd000f, -/* 0x062e: ctx_redswitch_delay */ - 0xe2b6080e, - 0xfd1bf401, - 0x0800f5f1, - 0x0200f5f1, - 0x01850080, - 0xbd000ff6, -/* 0x0647: ctx_xfer */ - 0x8000f804, - 0xf6028100, - 0x04bd000f, - 0xc48effb2, - 0xe5f0501d, - 0x008f7e01, - 0x0711f400, - 0x0006217e, -/* 0x0664: ctx_xfer_not_load */ - 0x0002167e, - 0xfc8024bd, - 0x02f60247, - 0xf004bd00, - 0x20b6012c, - 0x4afc8003, +/* 0x064a: ctx_xfer */ + 0x008000f8, + 0x0ff60281, + 0x8e04bd00, + 0xf0501dc4, + 0xffb201e5, + 0x00008f7e, + 0x7e0711f4, +/* 0x0667: ctx_xfer_not_load */ + 0x7e000624, + 0xbd000216, + 0x47fc8024, 0x0002f602, - 0x0c0f04bd, - 0xa88effb2, - 0xe5f0501d, - 0x008f7e01, - 0x03147e00, - 0xb23f0f00, - 0x1d608eff, - 0x01e5f050, + 0x2cf004bd, + 0x0320b601, + 0x024afc80, + 0xbd0002f6, + 0x8e0c0f04, + 0xf0501da8, + 0xffb201e5, 0x00008f7e, - 0xffb2000f, - 0x501d9c8e, - 0x7e01e5f0, + 0x0003147e, + 0x608e3f0f, + 0xe5f0501d, + 0x7effb201, 0x0f00008f, - 0x03147e01, - 0x01fcf000, - 0xb203f0b6, - 0x1da88eff, + 0x1d9c8e00, 0x01e5f050, - 0x00008f7e, - 0xf001acf0, - 0x008b02a5, - 0x0c985000, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98000c, - 0x7e000e01, - 0xf000013d, - 0x008b01ac, - 0x0c985040, - 0x0fc4b604, - 0x9800bcbb, - 0x0d98010c, - 0x060f9802, - 0x7e08004e, - 0xf000013d, + 0x8f7effb2, + 0x010f0000, + 0x0003147e, + 0xb601fcf0, + 0xa88e03f0, + 0xe5f0501d, + 0x7effb201, + 0xf000008f, 0xa5f001ac, - 0x30008b04, + 0x00008b02, 0x040c9850, 0xbb0fc4b6, 0x0c9800bc, - 0x030d9802, - 0x4e080f98, - 0x3d7e0200, - 0x0a7e0001, - 0x147e0002, - 0x01f40003, - 0x1a12f406, -/* 0x073c: ctx_xfer_post */ - 0x0002277e, - 0xffb20d0f, - 0x501da88e, - 0x7e01e5f0, - 0x7e00008f, -/* 0x0753: ctx_xfer_done */ - 0x7e000314, - 0xf800060d, - 0x00000000, + 0x010d9800, + 0x3d7e000e, + 0xacf00001, + 0x40008b01, + 0x040c9850, + 0xbb0fc4b6, + 0x0c9800bc, + 0x020d9801, + 0x4e060f98, + 0x3d7e0800, + 0xacf00001, + 0x04a5f001, + 0x5030008b, + 0xb6040c98, + 0xbcbb0fc4, + 0x020c9800, + 0x98030d98, + 0x004e080f, + 0x013d7e02, + 0x020a7e00, + 0x03147e00, + 0x0601f400, +/* 0x073f: ctx_xfer_post */ + 0x7e1a12f4, + 0x0f000227, + 0x1da88e0d, + 0x01e5f050, + 0x8f7effb2, + 0x147e0000, +/* 0x0756: ctx_xfer_done */ + 0x107e0003, + 0x00f80006, 0x00000000, 0x00000000, 0x00000000, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index dda7a7d224c9..9f5dfc85147a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -143,7 +143,7 @@ gf100_gr_zbc_depth_get(struct gf100_gr *gr, int format, static int gf100_fermi_mthd_zbc_color(struct nvkm_object *object, void *data, u32 size) { - struct gf100_gr *gr = (void *)object->engine; + struct gf100_gr *gr = gf100_gr(nvkm_gr(object->engine)); union { struct fermi_a_zbc_color_v0 v0; } *args = data; @@ -189,7 +189,7 @@ gf100_fermi_mthd_zbc_color(struct nvkm_object *object, void *data, u32 size) static int gf100_fermi_mthd_zbc_depth(struct nvkm_object *object, void *data, u32 size) { - struct gf100_gr *gr = (void *)object->engine; + struct gf100_gr *gr = gf100_gr(nvkm_gr(object->engine)); union { struct fermi_a_zbc_depth_v0 v0; } *args = data; @@ -1530,6 +1530,8 @@ gf100_gr_oneinit(struct nvkm_gr *base) gr->ppc_nr[i] = gr->func->ppc_nr; for (j = 0; j < gr->ppc_nr[i]; j++) { u8 mask = nvkm_rd32(device, GPC_UNIT(i, 0x0c30 + (j * 4))); + if (mask) + gr->ppc_mask[i] |= (1 << j); gr->ppc_tpc_nr[i][j] = hweight8(mask); } } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h index 4611961b1187..02e78b8d93f6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h @@ -97,6 +97,7 @@ struct gf100_gr { u8 tpc_nr[GPC_MAX]; u8 tpc_total; u8 ppc_nr[GPC_MAX]; + u8 ppc_mask[GPC_MAX]; u8 ppc_tpc_nr[GPC_MAX][4]; struct nvkm_memory *unk4188b4; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c index 895ba74057d4..1d7dd38292b3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c @@ -97,7 +97,9 @@ static void * nvkm_instobj_dtor(struct nvkm_memory *memory) { struct nvkm_instobj *iobj = nvkm_instobj(memory); + spin_lock(&iobj->imem->lock); list_del(&iobj->head); + spin_unlock(&iobj->imem->lock); nvkm_memory_del(&iobj->parent); return iobj; } @@ -190,7 +192,9 @@ nvkm_instobj_new(struct nvkm_instmem *imem, u32 size, u32 align, bool zero, nvkm_memory_ctor(&nvkm_instobj_func_slow, &iobj->memory); iobj->parent = memory; iobj->imem = imem; + spin_lock(&iobj->imem->lock); list_add_tail(&iobj->head, &imem->list); + spin_unlock(&iobj->imem->lock); memory = &iobj->memory; } @@ -309,5 +313,6 @@ nvkm_instmem_ctor(const struct nvkm_instmem_func *func, { nvkm_subdev_ctor(&nvkm_instmem, device, index, 0, &imem->subdev); imem->func = func; + spin_lock_init(&imem->lock); INIT_LIST_HEAD(&imem->list); } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c index b61509e26ec9..b735173a18ff 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c @@ -59,7 +59,7 @@ gk104_volt_set(struct nvkm_volt *base, u32 uv) duty = (uv - bios->base) * div / bios->pwm_range; nvkm_wr32(device, 0x20340, div); - nvkm_wr32(device, 0x20344, 0x8000000 | duty); + nvkm_wr32(device, 0x20344, 0x80000000 | duty); return 0; } diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c index ad09590e8a46..2ed0754ed19e 100644 --- a/drivers/gpu/drm/omapdrm/omap_crtc.c +++ b/drivers/gpu/drm/omapdrm/omap_crtc.c @@ -524,7 +524,7 @@ struct drm_crtc *omap_crtc_init(struct drm_device *dev, omap_crtc->mgr = omap_dss_get_overlay_manager(channel); ret = drm_crtc_init_with_planes(dev, crtc, plane, NULL, - &omap_crtc_funcs); + &omap_crtc_funcs, NULL); if (ret < 0) { kfree(omap_crtc); return NULL; diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h index 5c367aad8a6e..130fca70bfd7 100644 --- a/drivers/gpu/drm/omapdrm/omap_drv.h +++ b/drivers/gpu/drm/omapdrm/omap_drv.h @@ -172,9 +172,9 @@ void copy_timings_drm_to_omap(struct omap_video_timings *timings, uint32_t omap_framebuffer_get_formats(uint32_t *pixel_formats, uint32_t max_formats, enum omap_color_mode supported_modes); struct drm_framebuffer *omap_framebuffer_create(struct drm_device *dev, - struct drm_file *file, struct drm_mode_fb_cmd2 *mode_cmd); + struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd); struct drm_framebuffer *omap_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos); + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos); struct drm_gem_object *omap_framebuffer_bo(struct drm_framebuffer *fb, int p); int omap_framebuffer_pin(struct drm_framebuffer *fb); void omap_framebuffer_unpin(struct drm_framebuffer *fb); @@ -248,7 +248,7 @@ struct omap_dss_device *omap_encoder_get_dssdev(struct drm_encoder *encoder); static inline int objects_lookup(struct drm_device *dev, struct drm_file *filp, uint32_t pixel_format, - struct drm_gem_object **bos, uint32_t *handles) + struct drm_gem_object **bos, const uint32_t *handles) { int i, n = drm_format_num_planes(pixel_format); diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c index 7d9b32a0eb43..0c104ad7ef66 100644 --- a/drivers/gpu/drm/omapdrm/omap_encoder.c +++ b/drivers/gpu/drm/omapdrm/omap_encoder.c @@ -178,7 +178,7 @@ struct drm_encoder *omap_encoder_init(struct drm_device *dev, encoder = &omap_encoder->base; drm_encoder_init(dev, encoder, &omap_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &omap_encoder_helper_funcs); return encoder; diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c index 636a1f921569..ad202dfc1a49 100644 --- a/drivers/gpu/drm/omapdrm/omap_fb.c +++ b/drivers/gpu/drm/omapdrm/omap_fb.c @@ -364,7 +364,7 @@ void omap_framebuffer_describe(struct drm_framebuffer *fb, struct seq_file *m) #endif struct drm_framebuffer *omap_framebuffer_create(struct drm_device *dev, - struct drm_file *file, struct drm_mode_fb_cmd2 *mode_cmd) + struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *bos[4]; struct drm_framebuffer *fb; @@ -386,7 +386,7 @@ struct drm_framebuffer *omap_framebuffer_create(struct drm_device *dev, } struct drm_framebuffer *omap_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos) + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos) { struct omap_framebuffer *omap_fb = NULL; struct drm_framebuffer *fb = NULL; diff --git a/drivers/gpu/drm/omapdrm/omap_plane.c b/drivers/gpu/drm/omapdrm/omap_plane.c index 3054bda72688..d5ecabd6c14c 100644 --- a/drivers/gpu/drm/omapdrm/omap_plane.c +++ b/drivers/gpu/drm/omapdrm/omap_plane.c @@ -366,7 +366,7 @@ struct drm_plane *omap_plane_init(struct drm_device *dev, ret = drm_universal_plane_init(dev, plane, (1 << priv->num_crtcs) - 1, &omap_plane_funcs, omap_plane->formats, - omap_plane->nformats, type); + omap_plane->nformats, type, NULL); if (ret < 0) goto error; diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index 183aea1abebc..86276519b2ef 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -521,7 +521,7 @@ static const struct drm_framebuffer_funcs qxl_fb_funcs = { int qxl_framebuffer_init(struct drm_device *dev, struct qxl_framebuffer *qfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -876,16 +876,6 @@ static const struct drm_connector_helper_funcs qxl_connector_helper_funcs = { .best_encoder = qxl_best_encoder, }; -static void qxl_conn_save(struct drm_connector *connector) -{ - DRM_DEBUG("\n"); -} - -static void qxl_conn_restore(struct drm_connector *connector) -{ - DRM_DEBUG("\n"); -} - static enum drm_connector_status qxl_conn_detect( struct drm_connector *connector, bool force) @@ -932,10 +922,8 @@ static void qxl_conn_destroy(struct drm_connector *connector) static const struct drm_connector_funcs qxl_connector_funcs = { .dpms = drm_helper_connector_dpms, - .save = qxl_conn_save, - .restore = qxl_conn_restore, .detect = qxl_conn_detect, - .fill_modes = drm_helper_probe_single_connector_modes_nomerge, + .fill_modes = drm_helper_probe_single_connector_modes, .set_property = qxl_conn_set_property, .destroy = qxl_conn_destroy, }; @@ -980,7 +968,7 @@ static int qdev_output_init(struct drm_device *dev, int num_output) &qxl_connector_funcs, DRM_MODE_CONNECTOR_VIRTUAL); drm_encoder_init(dev, &qxl_output->enc, &qxl_enc_funcs, - DRM_MODE_ENCODER_VIRTUAL); + DRM_MODE_ENCODER_VIRTUAL, NULL); /* we get HPD via client monitors config */ connector->polled = DRM_CONNECTOR_POLL_HPD; @@ -1003,7 +991,7 @@ static int qdev_output_init(struct drm_device *dev, int num_output) static struct drm_framebuffer * qxl_user_framebuffer_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct qxl_framebuffer *qxl_fb; diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h index 01a86948eb8c..6e6b9b1519b8 100644 --- a/drivers/gpu/drm/qxl/qxl_drv.h +++ b/drivers/gpu/drm/qxl/qxl_drv.h @@ -390,7 +390,7 @@ void qxl_fbdev_set_suspend(struct qxl_device *qdev, int state); int qxl_framebuffer_init(struct drm_device *dev, struct qxl_framebuffer *rfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); void qxl_display_read_client_monitors_config(struct qxl_device *qdev); void qxl_send_monitors_config(struct qxl_device *qdev); diff --git a/drivers/gpu/drm/qxl/qxl_fb.c b/drivers/gpu/drm/qxl/qxl_fb.c index c4a552637c93..7136e521e6db 100644 --- a/drivers/gpu/drm/qxl/qxl_fb.c +++ b/drivers/gpu/drm/qxl/qxl_fb.c @@ -40,7 +40,6 @@ struct qxl_fbdev { struct drm_fb_helper helper; struct qxl_framebuffer qfb; - struct list_head fbdev_list; struct qxl_device *qdev; spinlock_t delayed_ops_lock; @@ -283,7 +282,7 @@ int qxl_get_handle_for_primary_fb(struct qxl_device *qdev, } static int qxlfb_create_pinned_object(struct qxl_fbdev *qfbdev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **gobj_p) { struct qxl_device *qdev = qfbdev->qdev; diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index dac78ad24b31..801dd60ac192 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -25,6 +25,7 @@ */ #include <drm/drmP.h> #include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_helper.h> #include <drm/radeon_drm.h> #include <drm/drm_fixed.h> #include "radeon.h" diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index bb292143997e..01b20e14a247 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -2767,23 +2767,27 @@ radeon_add_atom_encoder(struct drm_device *dev, case ENCODER_OBJECT_ID_INTERNAL_LVTM1: if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { radeon_encoder->rmx_type = RMX_FULL; - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_LVDS); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_LVDS, NULL); radeon_encoder->enc_priv = radeon_atombios_get_lvds_info(radeon_encoder); } else { - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_TMDS, NULL); radeon_encoder->enc_priv = radeon_atombios_set_dig_info(radeon_encoder); } drm_encoder_helper_add(encoder, &radeon_atom_dig_helper_funcs); break; case ENCODER_OBJECT_ID_INTERNAL_DAC1: - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_DAC); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_DAC, NULL); radeon_encoder->enc_priv = radeon_atombios_set_dac_info(radeon_encoder); drm_encoder_helper_add(encoder, &radeon_atom_dac_helper_funcs); break; case ENCODER_OBJECT_ID_INTERNAL_DAC2: case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1: case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2: - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_TVDAC); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_TVDAC, NULL); radeon_encoder->enc_priv = radeon_atombios_set_dac_info(radeon_encoder); drm_encoder_helper_add(encoder, &radeon_atom_dac_helper_funcs); break; @@ -2797,13 +2801,16 @@ radeon_add_atom_encoder(struct drm_device *dev, case ENCODER_OBJECT_ID_INTERNAL_UNIPHY3: if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { radeon_encoder->rmx_type = RMX_FULL; - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_LVDS); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_LVDS, NULL); radeon_encoder->enc_priv = radeon_atombios_get_lvds_info(radeon_encoder); } else if (radeon_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) { - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_DAC); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_DAC, NULL); radeon_encoder->enc_priv = radeon_atombios_set_dig_info(radeon_encoder); } else { - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_TMDS, NULL); radeon_encoder->enc_priv = radeon_atombios_set_dig_info(radeon_encoder); } drm_encoder_helper_add(encoder, &radeon_atom_dig_helper_funcs); @@ -2820,11 +2827,14 @@ radeon_add_atom_encoder(struct drm_device *dev, /* these are handled by the primary encoders */ radeon_encoder->is_ext_encoder = true; if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_LVDS); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_LVDS, NULL); else if (radeon_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_DAC); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_DAC, NULL); else - drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &radeon_atom_enc_funcs, + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &radeon_atom_ext_helper_funcs); break; } diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 248953d2fdb7..0154db43860c 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -8472,7 +8472,7 @@ restart_ih: if (queue_dp) schedule_work(&rdev->dp_work); if (queue_hotplug) - schedule_work(&rdev->hotplug_work); + schedule_delayed_work(&rdev->hotplug_work, 0); if (queue_reset) { rdev->needs_reset = true; wake_up_all(&rdev->fence_queue); @@ -9630,6 +9630,9 @@ static void dce8_program_watermarks(struct radeon_device *rdev, (rdev->disp_priority == 2)) { DRM_DEBUG_KMS("force priority to high\n"); } + + /* Save number of lines the linebuffer leads before the scanout */ + radeon_crtc->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay); } /* select wm A */ diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 7f33767d7ed6..2ad462896896 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2372,6 +2372,9 @@ static void evergreen_program_watermarks(struct radeon_device *rdev, c.full = dfixed_div(c, a); priority_b_mark = dfixed_trunc(c); priority_b_cnt |= priority_b_mark & PRIORITY_MARK_MASK; + + /* Save number of lines the linebuffer leads before the scanout */ + radeon_crtc->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay); } /* select wm A */ @@ -5344,7 +5347,7 @@ restart_ih: if (queue_dp) schedule_work(&rdev->dp_work); if (queue_hotplug) - schedule_work(&rdev->hotplug_work); + schedule_delayed_work(&rdev->hotplug_work, 0); if (queue_hdmi) schedule_work(&rdev->audio_work); if (queue_thermal && rdev->pm.dpm_enabled) diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 238b13f045c1..9e7e2bf03b81 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -806,7 +806,7 @@ int r100_irq_process(struct radeon_device *rdev) status = r100_irq_ack(rdev); } if (queue_hotplug) - schedule_work(&rdev->hotplug_work); + schedule_delayed_work(&rdev->hotplug_work, 0); if (rdev->msi_enabled) { switch (rdev->family) { case CHIP_RS400: @@ -3217,6 +3217,9 @@ void r100_bandwidth_update(struct radeon_device *rdev) uint32_t pixel_bytes1 = 0; uint32_t pixel_bytes2 = 0; + /* Guess line buffer size to be 8192 pixels */ + u32 lb_size = 8192; + if (!rdev->mode_info.mode_config_initialized) return; @@ -3631,6 +3634,13 @@ void r100_bandwidth_update(struct radeon_device *rdev) DRM_DEBUG_KMS("GRPH2_BUFFER_CNTL from to %x\n", (unsigned int)RREG32(RADEON_GRPH2_BUFFER_CNTL)); } + + /* Save number of lines the linebuffer leads before the scanout */ + if (mode1) + rdev->mode_info.crtcs[0]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode1->crtc_hdisplay); + + if (mode2) + rdev->mode_info.crtcs[1]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode2->crtc_hdisplay); } int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *ring) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 4ea5b10ff5f4..cc2fdf0be37a 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -4276,7 +4276,7 @@ restart_ih: WREG32(IH_RB_RPTR, rptr); } if (queue_hotplug) - schedule_work(&rdev->hotplug_work); + schedule_delayed_work(&rdev->hotplug_work, 0); if (queue_hdmi) schedule_work(&rdev->audio_work); if (queue_thermal && rdev->pm.dpm_enabled) diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index b6cbd816537e..87db64983ea8 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -2414,7 +2414,7 @@ struct radeon_device { struct r600_ih ih; /* r6/700 interrupt ring */ struct radeon_rlc rlc; struct radeon_mec mec; - struct work_struct hotplug_work; + struct delayed_work hotplug_work; struct work_struct dp_work; struct work_struct audio_work; int num_crtc; /* number of crtcs */ diff --git a/drivers/gpu/drm/radeon/radeon_agp.c b/drivers/gpu/drm/radeon/radeon_agp.c index fe994aac3b04..c77d349c561c 100644 --- a/drivers/gpu/drm/radeon/radeon_agp.c +++ b/drivers/gpu/drm/radeon/radeon_agp.c @@ -54,6 +54,9 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = { /* Intel 82855PM host bridge / Mobility 9600 M10 RV350 Needs AGPMode 1 (lp #195051) */ { PCI_VENDOR_ID_INTEL, 0x3340, PCI_VENDOR_ID_ATI, 0x4e50, PCI_VENDOR_ID_IBM, 0x0550, 1}, + /* Intel 82855PM host bridge / RV250/M9 GL [Mobility FireGL 9000/Radeon 9000] needs AGPMode 1 (Thinkpad T40p) */ + { PCI_VENDOR_ID_INTEL, 0x3340, PCI_VENDOR_ID_ATI, 0x4c66, + PCI_VENDOR_ID_IBM, 0x054d, 1}, /* Intel 82855PM host bridge / Mobility M7 needs AGPMode 1 */ { PCI_VENDOR_ID_INTEL, 0x3340, PCI_VENDOR_ID_ATI, 0x4c57, PCI_VENDOR_ID_IBM, 0x0530, 1}, diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 5a2cafb4f1bc..340f3f549f29 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1234,13 +1234,32 @@ radeon_dvi_detect(struct drm_connector *connector, bool force) if (r < 0) return connector_status_disconnected; + if (radeon_connector->detected_hpd_without_ddc) { + force = true; + radeon_connector->detected_hpd_without_ddc = false; + } + if (!force && radeon_check_hpd_status_unchanged(connector)) { ret = connector->status; goto exit; } - if (radeon_connector->ddc_bus) + if (radeon_connector->ddc_bus) { dret = radeon_ddc_probe(radeon_connector, false); + + /* Sometimes the pins required for the DDC probe on DVI + * connectors don't make contact at the same time that the ones + * for HPD do. If the DDC probe fails even though we had an HPD + * signal, try again later */ + if (!dret && !force && + connector->status != connector_status_connected) { + DRM_DEBUG_KMS("hpd detected without ddc, retrying in 1 second\n"); + radeon_connector->detected_hpd_without_ddc = true; + schedule_delayed_work(&rdev->hotplug_work, + msecs_to_jiffies(1000)); + goto exit; + } + } if (dret) { radeon_connector->detected_by_load = false; radeon_connector_free_edid(connector); diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index a8d9927ed9eb..b3bb92368ae0 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -322,7 +322,9 @@ void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id) * to complete in this vblank? */ if (update_pending && - (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id, 0, + (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, + crtc_id, + USE_REAL_VBLANKSTART, &vpos, &hpos, NULL, NULL, &rdev->mode_info.crtcs[crtc_id]->base.hwmode)) && ((vpos >= (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100) || @@ -401,6 +403,8 @@ static void radeon_flip_work_func(struct work_struct *__work) struct drm_crtc *crtc = &radeon_crtc->base; unsigned long flags; int r; + int vpos, hpos, stat, min_udelay; + struct drm_vblank_crtc *vblank = &crtc->dev->vblank[work->crtc_id]; down_read(&rdev->exclusive_lock); if (work->fence) { @@ -437,6 +441,41 @@ static void radeon_flip_work_func(struct work_struct *__work) /* set the proper interrupt */ radeon_irq_kms_pflip_irq_get(rdev, radeon_crtc->crtc_id); + /* If this happens to execute within the "virtually extended" vblank + * interval before the start of the real vblank interval then it needs + * to delay programming the mmio flip until the real vblank is entered. + * This prevents completing a flip too early due to the way we fudge + * our vblank counter and vblank timestamps in order to work around the + * problem that the hw fires vblank interrupts before actual start of + * vblank (when line buffer refilling is done for a frame). It + * complements the fudging logic in radeon_get_crtc_scanoutpos() for + * timestamping and radeon_get_vblank_counter_kms() for vblank counts. + * + * In practice this won't execute very often unless on very fast + * machines because the time window for this to happen is very small. + */ + for (;;) { + /* GET_DISTANCE_TO_VBLANKSTART returns distance to real vblank + * start in hpos, and to the "fudged earlier" vblank start in + * vpos. + */ + stat = radeon_get_crtc_scanoutpos(rdev->ddev, work->crtc_id, + GET_DISTANCE_TO_VBLANKSTART, + &vpos, &hpos, NULL, NULL, + &crtc->hwmode); + + if ((stat & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE)) != + (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE) || + !(vpos >= 0 && hpos <= 0)) + break; + + /* Sleep at least until estimated real start of hw vblank */ + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + min_udelay = (-hpos + 1) * max(vblank->linedur_ns / 1000, 5); + usleep_range(min_udelay, 2 * min_udelay); + spin_lock_irqsave(&crtc->dev->event_lock, flags); + }; + /* do the flip (mmio) */ radeon_page_flip(rdev, radeon_crtc->crtc_id, work->base); @@ -1292,7 +1331,7 @@ static const struct drm_framebuffer_funcs radeon_fb_funcs = { int radeon_framebuffer_init(struct drm_device *dev, struct radeon_framebuffer *rfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -1309,7 +1348,7 @@ radeon_framebuffer_init(struct drm_device *dev, static struct drm_framebuffer * radeon_user_framebuffer_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct radeon_framebuffer *radeon_fb; @@ -1768,6 +1807,15 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc, * \param dev Device to query. * \param crtc Crtc to query. * \param flags Flags from caller (DRM_CALLED_FROM_VBLIRQ or 0). + * For driver internal use only also supports these flags: + * + * USE_REAL_VBLANKSTART to use the real start of vblank instead + * of a fudged earlier start of vblank. + * + * GET_DISTANCE_TO_VBLANKSTART to return distance to the + * fudged earlier start of vblank in *vpos and the distance + * to true start of vblank in *hpos. + * * \param *vpos Location where vertical scanout position should be stored. * \param *hpos Location where horizontal scanout position should go. * \param *stime Target location for timestamp taken immediately before @@ -1911,10 +1959,40 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, vbl_end = 0; } + /* Called from driver internal vblank counter query code? */ + if (flags & GET_DISTANCE_TO_VBLANKSTART) { + /* Caller wants distance from real vbl_start in *hpos */ + *hpos = *vpos - vbl_start; + } + + /* Fudge vblank to start a few scanlines earlier to handle the + * problem that vblank irqs fire a few scanlines before start + * of vblank. Some driver internal callers need the true vblank + * start to be used and signal this via the USE_REAL_VBLANKSTART flag. + * + * The cause of the "early" vblank irq is that the irq is triggered + * by the line buffer logic when the line buffer read position enters + * the vblank, whereas our crtc scanout position naturally lags the + * line buffer read position. + */ + if (!(flags & USE_REAL_VBLANKSTART)) + vbl_start -= rdev->mode_info.crtcs[pipe]->lb_vblank_lead_lines; + /* Test scanout position against vblank region. */ if ((*vpos < vbl_start) && (*vpos >= vbl_end)) in_vbl = false; + /* In vblank? */ + if (in_vbl) + ret |= DRM_SCANOUTPOS_IN_VBLANK; + + /* Called from driver internal vblank counter query code? */ + if (flags & GET_DISTANCE_TO_VBLANKSTART) { + /* Caller wants distance from fudged earlier vbl_start */ + *vpos -= vbl_start; + return ret; + } + /* Check if inside vblank area and apply corrective offsets: * vpos will then be >=0 in video scanout area, but negative * within vblank area, counting down the number of lines until @@ -1930,31 +2008,5 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, /* Correct for shifted end of vbl at vbl_end. */ *vpos = *vpos - vbl_end; - /* In vblank? */ - if (in_vbl) - ret |= DRM_SCANOUTPOS_IN_VBLANK; - - /* Is vpos outside nominal vblank area, but less than - * 1/100 of a frame height away from start of vblank? - * If so, assume this isn't a massively delayed vblank - * interrupt, but a vblank interrupt that fired a few - * microseconds before true start of vblank. Compensate - * by adding a full frame duration to the final timestamp. - * Happens, e.g., on ATI R500, R600. - * - * We only do this if DRM_CALLED_FROM_VBLIRQ. - */ - if ((flags & DRM_CALLED_FROM_VBLIRQ) && !in_vbl) { - vbl_start = mode->crtc_vdisplay; - vtotal = mode->crtc_vtotal; - - if (vbl_start - *vpos < vtotal / 100) { - *vpos -= vtotal; - - /* Signal this correction as "applied". */ - ret |= 0x8; - } - } - return ret; } diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c index 744f5c49c664..94323f51ffcf 100644 --- a/drivers/gpu/drm/radeon/radeon_dp_mst.c +++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c @@ -641,7 +641,7 @@ radeon_dp_create_fake_mst_encoder(struct radeon_connector *connector) } drm_encoder_init(dev, &radeon_encoder->base, &radeon_dp_mst_enc_funcs, - DRM_MODE_ENCODER_DPMST); + DRM_MODE_ENCODER_DPMST, NULL); drm_encoder_helper_add(encoder, &radeon_mst_helper_funcs); mst_enc = radeon_encoder->enc_priv; diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 26da2f4d7b4f..adc44bbc81a9 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -44,7 +44,6 @@ struct radeon_fbdev { struct drm_fb_helper helper; struct radeon_framebuffer rfb; - struct list_head fbdev_list; struct radeon_device *rdev; }; diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c index 171d3e43c30c..979f3bf65f2c 100644 --- a/drivers/gpu/drm/radeon/radeon_irq_kms.c +++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c @@ -74,7 +74,7 @@ irqreturn_t radeon_driver_irq_handler_kms(int irq, void *arg) static void radeon_hotplug_work_func(struct work_struct *work) { struct radeon_device *rdev = container_of(work, struct radeon_device, - hotplug_work); + hotplug_work.work); struct drm_device *dev = rdev->ddev; struct drm_mode_config *mode_config = &dev->mode_config; struct drm_connector *connector; @@ -302,7 +302,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev) } } - INIT_WORK(&rdev->hotplug_work, radeon_hotplug_work_func); + INIT_DELAYED_WORK(&rdev->hotplug_work, radeon_hotplug_work_func); INIT_WORK(&rdev->dp_work, radeon_dp_work_func); INIT_WORK(&rdev->audio_work, r600_audio_update_hdmi); @@ -310,7 +310,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev) r = drm_irq_install(rdev->ddev, rdev->ddev->pdev->irq); if (r) { rdev->irq.installed = false; - flush_work(&rdev->hotplug_work); + flush_delayed_work(&rdev->hotplug_work); return r; } @@ -333,7 +333,7 @@ void radeon_irq_kms_fini(struct radeon_device *rdev) rdev->irq.installed = false; if (rdev->msi_enabled) pci_disable_msi(rdev->pdev); - flush_work(&rdev->hotplug_work); + flush_delayed_work(&rdev->hotplug_work); } } diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 0ec6fcca16d3..d290a8a09036 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -755,6 +755,8 @@ void radeon_driver_preclose_kms(struct drm_device *dev, */ u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc) { + int vpos, hpos, stat; + u32 count; struct radeon_device *rdev = dev->dev_private; if (crtc < 0 || crtc >= rdev->num_crtc) { @@ -762,7 +764,53 @@ u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc) return -EINVAL; } - return radeon_get_vblank_counter(rdev, crtc); + /* The hw increments its frame counter at start of vsync, not at start + * of vblank, as is required by DRM core vblank counter handling. + * Cook the hw count here to make it appear to the caller as if it + * incremented at start of vblank. We measure distance to start of + * vblank in vpos. vpos therefore will be >= 0 between start of vblank + * and start of vsync, so vpos >= 0 means to bump the hw frame counter + * result by 1 to give the proper appearance to caller. + */ + if (rdev->mode_info.crtcs[crtc]) { + /* Repeat readout if needed to provide stable result if + * we cross start of vsync during the queries. + */ + do { + count = radeon_get_vblank_counter(rdev, crtc); + /* Ask radeon_get_crtc_scanoutpos to return vpos as + * distance to start of vblank, instead of regular + * vertical scanout pos. + */ + stat = radeon_get_crtc_scanoutpos( + dev, crtc, GET_DISTANCE_TO_VBLANKSTART, + &vpos, &hpos, NULL, NULL, + &rdev->mode_info.crtcs[crtc]->base.hwmode); + } while (count != radeon_get_vblank_counter(rdev, crtc)); + + if (((stat & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE)) != + (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE))) { + DRM_DEBUG_VBL("Query failed! stat %d\n", stat); + } + else { + DRM_DEBUG_VBL("crtc %d: dist from vblank start %d\n", + crtc, vpos); + + /* Bump counter if we are at >= leading edge of vblank, + * but before vsync where vpos would turn negative and + * the hw counter really increments. + */ + if (vpos >= 0) + count++; + } + } + else { + /* Fallback to use value as is. */ + count = radeon_get_vblank_counter(rdev, crtc); + DRM_DEBUG_VBL("NULL mode info! Returned count may be wrong.\n"); + } + + return count; } /** diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index 678b4386540d..32b338ff436b 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -25,6 +25,7 @@ */ #include <drm/drmP.h> #include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_helper.h> #include <drm/radeon_drm.h> #include <drm/drm_fixed.h> #include "radeon.h" diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c index 30de43366eae..88dc973fb209 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c @@ -1772,7 +1772,8 @@ radeon_add_legacy_encoder(struct drm_device *dev, uint32_t encoder_enum, uint32_ switch (radeon_encoder->encoder_id) { case ENCODER_OBJECT_ID_INTERNAL_LVDS: encoder->possible_crtcs = 0x1; - drm_encoder_init(dev, encoder, &radeon_legacy_lvds_enc_funcs, DRM_MODE_ENCODER_LVDS); + drm_encoder_init(dev, encoder, &radeon_legacy_lvds_enc_funcs, + DRM_MODE_ENCODER_LVDS, NULL); drm_encoder_helper_add(encoder, &radeon_legacy_lvds_helper_funcs); if (rdev->is_atom_bios) radeon_encoder->enc_priv = radeon_atombios_get_lvds_info(radeon_encoder); @@ -1781,12 +1782,14 @@ radeon_add_legacy_encoder(struct drm_device *dev, uint32_t encoder_enum, uint32_ radeon_encoder->rmx_type = RMX_FULL; break; case ENCODER_OBJECT_ID_INTERNAL_TMDS1: - drm_encoder_init(dev, encoder, &radeon_legacy_tmds_int_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &radeon_legacy_tmds_int_enc_funcs, + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &radeon_legacy_tmds_int_helper_funcs); radeon_encoder->enc_priv = radeon_legacy_get_tmds_info(radeon_encoder); break; case ENCODER_OBJECT_ID_INTERNAL_DAC1: - drm_encoder_init(dev, encoder, &radeon_legacy_primary_dac_enc_funcs, DRM_MODE_ENCODER_DAC); + drm_encoder_init(dev, encoder, &radeon_legacy_primary_dac_enc_funcs, + DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(encoder, &radeon_legacy_primary_dac_helper_funcs); if (rdev->is_atom_bios) radeon_encoder->enc_priv = radeon_atombios_get_primary_dac_info(radeon_encoder); @@ -1794,7 +1797,8 @@ radeon_add_legacy_encoder(struct drm_device *dev, uint32_t encoder_enum, uint32_ radeon_encoder->enc_priv = radeon_combios_get_primary_dac_info(radeon_encoder); break; case ENCODER_OBJECT_ID_INTERNAL_DAC2: - drm_encoder_init(dev, encoder, &radeon_legacy_tv_dac_enc_funcs, DRM_MODE_ENCODER_TVDAC); + drm_encoder_init(dev, encoder, &radeon_legacy_tv_dac_enc_funcs, + DRM_MODE_ENCODER_TVDAC, NULL); drm_encoder_helper_add(encoder, &radeon_legacy_tv_dac_helper_funcs); if (rdev->is_atom_bios) radeon_encoder->enc_priv = radeon_atombios_get_tv_dac_info(radeon_encoder); @@ -1802,7 +1806,8 @@ radeon_add_legacy_encoder(struct drm_device *dev, uint32_t encoder_enum, uint32_ radeon_encoder->enc_priv = radeon_combios_get_tv_dac_info(radeon_encoder); break; case ENCODER_OBJECT_ID_INTERNAL_DVO1: - drm_encoder_init(dev, encoder, &radeon_legacy_tmds_ext_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &radeon_legacy_tmds_ext_enc_funcs, + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(encoder, &radeon_legacy_tmds_ext_helper_funcs); if (!rdev->is_atom_bios) radeon_encoder->enc_priv = radeon_legacy_get_ext_tmds_info(radeon_encoder); diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 830e171c3a9e..cddd41b32eda 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -367,6 +367,7 @@ struct radeon_crtc { u32 line_time; u32 wm_low; u32 wm_high; + u32 lb_vblank_lead_lines; struct drm_display_mode hw_mode; enum radeon_output_csc output_csc; }; @@ -553,6 +554,7 @@ struct radeon_connector { void *con_priv; bool dac_load_detect; bool detected_by_load; /* if the connection status was determined by load */ + bool detected_hpd_without_ddc; /* if an HPD signal was detected on DVI, but ddc probing failed */ uint16_t connector_object_id; struct radeon_hpd hpd; struct radeon_router router; @@ -686,6 +688,9 @@ struct atom_voltage_table struct atom_voltage_table_entry entries[MAX_VOLTAGE_ENTRIES]; }; +/* Driver internal use only flags of radeon_get_crtc_scanoutpos() */ +#define USE_REAL_VBLANKSTART (1 << 30) +#define GET_DISTANCE_TO_VBLANKSTART (1 << 31) extern void radeon_add_atom_connector(struct drm_device *dev, @@ -929,7 +934,7 @@ extern void radeon_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green u16 *blue, int regno); int radeon_framebuffer_init(struct drm_device *dev, struct radeon_framebuffer *rfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); int radeonfb_remove(struct drm_device *dev, struct drm_framebuffer *fb); diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index f4f03dcc1530..59abebd6b5dc 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -1756,7 +1756,9 @@ static bool radeon_pm_in_vbl(struct radeon_device *rdev) */ for (crtc = 0; (crtc < rdev->num_crtc) && in_vbl; crtc++) { if (rdev->pm.active_crtcs & (1 << crtc)) { - vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, 0, + vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, + crtc, + USE_REAL_VBLANKSTART, &vpos, &hpos, NULL, NULL, &rdev->mode_info.crtcs[crtc]->base.hwmode); if ((vbl_status & DRM_SCANOUTPOS_VALID) && diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c index 97a904835759..6244f4e44e9a 100644 --- a/drivers/gpu/drm/radeon/rs600.c +++ b/drivers/gpu/drm/radeon/rs600.c @@ -813,7 +813,7 @@ int rs600_irq_process(struct radeon_device *rdev) status = rs600_irq_ack(rdev); } if (queue_hotplug) - schedule_work(&rdev->hotplug_work); + schedule_delayed_work(&rdev->hotplug_work, 0); if (queue_hdmi) schedule_work(&rdev->audio_work); if (rdev->msi_enabled) { diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c index 516ca27cfa12..6bc44c24e837 100644 --- a/drivers/gpu/drm/radeon/rs690.c +++ b/drivers/gpu/drm/radeon/rs690.c @@ -207,6 +207,9 @@ void rs690_line_buffer_adjust(struct radeon_device *rdev, { u32 tmp; + /* Guess line buffer size to be 8192 pixels */ + u32 lb_size = 8192; + /* * Line Buffer Setup * There is a single line buffer shared by both display controllers. @@ -243,6 +246,13 @@ void rs690_line_buffer_adjust(struct radeon_device *rdev, tmp |= V_006520_DC_LB_MEMORY_SPLIT_D1_1Q_D2_3Q; } WREG32(R_006520_DC_LB_MEMORY_SPLIT, tmp); + + /* Save number of lines the linebuffer leads before the scanout */ + if (mode1) + rdev->mode_info.crtcs[0]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode1->crtc_hdisplay); + + if (mode2) + rdev->mode_info.crtcs[1]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode2->crtc_hdisplay); } struct rs690_watermark { diff --git a/drivers/gpu/drm/radeon/rv730_dpm.c b/drivers/gpu/drm/radeon/rv730_dpm.c index 3f5e1cf138ba..d37ba2cb886e 100644 --- a/drivers/gpu/drm/radeon/rv730_dpm.c +++ b/drivers/gpu/drm/radeon/rv730_dpm.c @@ -464,7 +464,7 @@ void rv730_stop_dpm(struct radeon_device *rdev) result = rv770_send_msg_to_smc(rdev, PPSMC_MSG_TwoLevelsDisabled); if (result != PPSMC_Result_OK) - DRM_ERROR("Could not force DPM to low\n"); + DRM_DEBUG("Could not force DPM to low\n"); WREG32_P(GENERAL_PWRMGT, 0, ~GLOBAL_PWRMGT_EN); diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c index b9c770745a7a..e830c8935db0 100644 --- a/drivers/gpu/drm/radeon/rv770_dpm.c +++ b/drivers/gpu/drm/radeon/rv770_dpm.c @@ -193,7 +193,7 @@ void rv770_stop_dpm(struct radeon_device *rdev) result = rv770_send_msg_to_smc(rdev, PPSMC_MSG_TwoLevelsDisabled); if (result != PPSMC_Result_OK) - DRM_ERROR("Could not force DPM to low.\n"); + DRM_DEBUG("Could not force DPM to low.\n"); WREG32_P(GENERAL_PWRMGT, 0, ~GLOBAL_PWRMGT_EN); @@ -1418,7 +1418,7 @@ int rv770_resume_smc(struct radeon_device *rdev) int rv770_set_sw_state(struct radeon_device *rdev) { if (rv770_send_msg_to_smc(rdev, PPSMC_MSG_SwitchToSwState) != PPSMC_Result_OK) - return -EINVAL; + DRM_DEBUG("rv770_set_sw_state failed\n"); return 0; } diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 07037e32dea3..f878d6962da5 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -2376,6 +2376,9 @@ static void dce6_program_watermarks(struct radeon_device *rdev, c.full = dfixed_div(c, a); priority_b_mark = dfixed_trunc(c); priority_b_cnt |= priority_b_mark & PRIORITY_MARK_MASK; + + /* Save number of lines the linebuffer leads before the scanout */ + radeon_crtc->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay); } /* select wm A */ @@ -6848,7 +6851,7 @@ restart_ih: if (queue_dp) schedule_work(&rdev->dp_work); if (queue_hotplug) - schedule_work(&rdev->hotplug_work); + schedule_delayed_work(&rdev->hotplug_work, 0); if (queue_thermal && rdev->pm.dpm_enabled) schedule_work(&rdev->pm.dpm.thermal.work); rdev->ih.rptr = rptr; diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c index 48cb19949ca3..88a4b706be16 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c @@ -613,7 +613,7 @@ int rcar_du_crtc_create(struct rcar_du_group *rgrp, unsigned int index) ret = drm_crtc_init_with_planes(rcdu->ddev, crtc, &rgrp->planes[index % 2].plane, - NULL, &crtc_funcs); + NULL, &crtc_funcs, NULL); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c index d0ae1e8009c6..c08700757feb 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c @@ -173,7 +173,7 @@ int rcar_du_encoder_init(struct rcar_du_device *rcdu, goto done; } else { ret = drm_encoder_init(rcdu->ddev, encoder, &encoder_funcs, - encoder_type); + encoder_type, NULL); if (ret < 0) goto done; diff --git a/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c b/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c index 81da8419282b..11267de26a51 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_hdmienc.c @@ -151,7 +151,7 @@ int rcar_du_hdmienc_init(struct rcar_du_device *rcdu, goto error; ret = drm_encoder_init(rcdu->ddev, encoder, &encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); if (ret < 0) goto error; diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/rcar-du/rcar_du_kms.c index ca12e8ca5552..43bce69d8560 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_kms.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.c @@ -136,7 +136,7 @@ int rcar_du_dumb_create(struct drm_file *file, struct drm_device *dev, static struct drm_framebuffer * rcar_du_fb_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct rcar_du_device *rcdu = dev->dev_private; const struct rcar_du_format_info *format; diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.c b/drivers/gpu/drm/rcar-du/rcar_du_plane.c index ffa583712cd9..c3ed9522c0e1 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_plane.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_plane.c @@ -410,7 +410,8 @@ int rcar_du_planes_init(struct rcar_du_group *rgrp) ret = drm_universal_plane_init(rcdu->ddev, &plane->plane, crtcs, &rcar_du_plane_funcs, formats, - ARRAY_SIZE(formats), type); + ARRAY_SIZE(formats), type, + NULL); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c index 80d6fc8a5cee..525b5a81e96e 100644 --- a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c @@ -295,7 +295,7 @@ static int dw_hdmi_rockchip_bind(struct device *dev, struct device *master, drm_encoder_helper_add(encoder, &dw_hdmi_rockchip_encoder_helper_funcs); drm_encoder_init(drm, encoder, &dw_hdmi_rockchip_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); return dw_hdmi_bind(dev, master, data, encoder, iores, irq, plat_data); } diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c index 002645bb5bbf..b8ac5911c102 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c @@ -72,7 +72,7 @@ static struct drm_framebuffer_funcs rockchip_drm_fb_funcs = { }; static struct rockchip_drm_fb * -rockchip_fb_alloc(struct drm_device *dev, struct drm_mode_fb_cmd2 *mode_cmd, +rockchip_fb_alloc(struct drm_device *dev, const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **obj, unsigned int num_planes) { struct rockchip_drm_fb *rockchip_fb; @@ -102,7 +102,7 @@ rockchip_fb_alloc(struct drm_device *dev, struct drm_mode_fb_cmd2 *mode_cmd, static struct drm_framebuffer * rockchip_user_fb_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct rockchip_drm_fb *rockchip_fb; struct drm_gem_object *objs[ROCKCHIP_MAX_FB_BUFFER]; @@ -173,7 +173,7 @@ static const struct drm_mode_config_funcs rockchip_drm_mode_config_funcs = { struct drm_framebuffer * rockchip_drm_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { struct rockchip_drm_fb *rockchip_fb; diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.h b/drivers/gpu/drm/rockchip/rockchip_drm_fb.h index 09574d48226f..2fe47f1ee98f 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_fb.h +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.h @@ -17,7 +17,7 @@ struct drm_framebuffer * rockchip_drm_framebuffer_init(struct drm_device *dev, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); void rockchip_drm_framebuffer_fini(struct drm_framebuffer *fb); diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c index 8caea0a33dd8..d908321b94ce 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c @@ -67,6 +67,7 @@ static int rockchip_drm_gem_object_mmap(struct drm_gem_object *obj, * VM_PFNMAP flag that was set by drm_gem_mmap_obj()/drm_gem_mmap(). */ vma->vm_flags &= ~VM_PFNMAP; + vma->vm_pgoff = 0; ret = dma_mmap_attrs(drm->dev, vma, rk_obj->kvaddr, rk_obj->dma_addr, obj->size, &rk_obj->dma_attrs); diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c index 5d8ae5e49c44..dd8e0860ad4e 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -374,6 +374,7 @@ static const struct of_device_id vop_driver_dt_match[] = { .data = &rk3288_vop }, {}, }; +MODULE_DEVICE_TABLE(of, vop_driver_dt_match); static inline void vop_writel(struct vop *vop, uint32_t offset, uint32_t v) { @@ -959,8 +960,8 @@ static int vop_update_plane_event(struct drm_plane *plane, val = (dest.y2 - dest.y1 - 1) << 16; val |= (dest.x2 - dest.x1 - 1) & 0xffff; VOP_WIN_SET(vop, win, dsp_info, val); - val = (dsp_sty - 1) << 16; - val |= (dsp_stx - 1) & 0xffff; + val = dsp_sty << 16; + val |= dsp_stx & 0xffff; VOP_WIN_SET(vop, win, dsp_st, val); VOP_WIN_SET(vop, win, rb_swap, rb_swap); @@ -1289,7 +1290,7 @@ static void vop_win_state_complete(struct vop_win *vop_win, if (state->event) { spin_lock_irqsave(&drm->event_lock, flags); - drm_send_vblank_event(drm, -1, state->event); + drm_crtc_send_vblank_event(crtc, state->event); spin_unlock_irqrestore(&drm->event_lock, flags); } @@ -1477,7 +1478,7 @@ static int vop_create_crtc(struct vop *vop) 0, &vop_plane_funcs, win_data->phy->data_formats, win_data->phy->nformats, - win_data->type); + win_data->type, NULL); if (ret) { DRM_ERROR("failed to initialize plane\n"); goto err_cleanup_planes; @@ -1491,7 +1492,7 @@ static int vop_create_crtc(struct vop *vop) } ret = drm_crtc_init_with_planes(drm_dev, crtc, primary, cursor, - &vop_crtc_funcs); + &vop_crtc_funcs, NULL); if (ret) return ret; @@ -1514,7 +1515,7 @@ static int vop_create_crtc(struct vop *vop) &vop_plane_funcs, win_data->phy->data_formats, win_data->phy->nformats, - win_data->type); + win_data->type, NULL); if (ret) { DRM_ERROR("failed to initialize overlay plane\n"); goto err_cleanup_crtc; @@ -1575,32 +1576,25 @@ static int vop_initial(struct vop *vop) return PTR_ERR(vop->dclk); } - ret = clk_prepare(vop->hclk); - if (ret < 0) { - dev_err(vop->dev, "failed to prepare hclk\n"); - return ret; - } - ret = clk_prepare(vop->dclk); if (ret < 0) { dev_err(vop->dev, "failed to prepare dclk\n"); - goto err_unprepare_hclk; + return ret; } - ret = clk_prepare(vop->aclk); + /* Enable both the hclk and aclk to setup the vop */ + ret = clk_prepare_enable(vop->hclk); if (ret < 0) { - dev_err(vop->dev, "failed to prepare aclk\n"); + dev_err(vop->dev, "failed to prepare/enable hclk\n"); goto err_unprepare_dclk; } - /* - * enable hclk, so that we can config vop register. - */ - ret = clk_enable(vop->hclk); + ret = clk_prepare_enable(vop->aclk); if (ret < 0) { - dev_err(vop->dev, "failed to prepare aclk\n"); - goto err_unprepare_aclk; + dev_err(vop->dev, "failed to prepare/enable aclk\n"); + goto err_disable_hclk; } + /* * do hclk_reset, reset all vop registers. */ @@ -1608,7 +1602,7 @@ static int vop_initial(struct vop *vop) if (IS_ERR(ahb_rst)) { dev_err(vop->dev, "failed to get ahb reset\n"); ret = PTR_ERR(ahb_rst); - goto err_disable_hclk; + goto err_disable_aclk; } reset_control_assert(ahb_rst); usleep_range(10, 20); @@ -1634,26 +1628,25 @@ static int vop_initial(struct vop *vop) if (IS_ERR(vop->dclk_rst)) { dev_err(vop->dev, "failed to get dclk reset\n"); ret = PTR_ERR(vop->dclk_rst); - goto err_unprepare_aclk; + goto err_disable_aclk; } reset_control_assert(vop->dclk_rst); usleep_range(10, 20); reset_control_deassert(vop->dclk_rst); clk_disable(vop->hclk); + clk_disable(vop->aclk); vop->is_enabled = false; return 0; +err_disable_aclk: + clk_disable_unprepare(vop->aclk); err_disable_hclk: - clk_disable(vop->hclk); -err_unprepare_aclk: - clk_unprepare(vop->aclk); + clk_disable_unprepare(vop->hclk); err_unprepare_dclk: clk_unprepare(vop->dclk); -err_unprepare_hclk: - clk_unprepare(vop->hclk); return ret; } diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c index e9272b0a8592..b80802f55143 100644 --- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c +++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c @@ -613,7 +613,7 @@ int shmob_drm_encoder_create(struct shmob_drm_device *sdev) encoder->possible_crtcs = 1; ret = drm_encoder_init(sdev->ddev, encoder, &encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/shmobile/shmob_drm_kms.c index aaf98ace4a90..388a0fc13564 100644 --- a/drivers/gpu/drm/shmobile/shmob_drm_kms.c +++ b/drivers/gpu/drm/shmobile/shmob_drm_kms.c @@ -104,7 +104,7 @@ const struct shmob_drm_format_info *shmob_drm_format_info(u32 fourcc) static struct drm_framebuffer * shmob_drm_fb_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { const struct shmob_drm_format_info *format; diff --git a/drivers/gpu/drm/sti/sti_crtc.c b/drivers/gpu/drm/sti/sti_crtc.c index 3ae09dcd4fd8..de11c7cfb02f 100644 --- a/drivers/gpu/drm/sti/sti_crtc.c +++ b/drivers/gpu/drm/sti/sti_crtc.c @@ -367,7 +367,7 @@ int sti_crtc_init(struct drm_device *drm_dev, struct sti_mixer *mixer, int res; res = drm_crtc_init_with_planes(drm_dev, crtc, primary, cursor, - &sti_crtc_funcs); + &sti_crtc_funcs, NULL); if (res) { DRM_ERROR("Can't initialze CRTC\n"); return -EINVAL; diff --git a/drivers/gpu/drm/sti/sti_cursor.c b/drivers/gpu/drm/sti/sti_cursor.c index dd1032195051..807863106b8d 100644 --- a/drivers/gpu/drm/sti/sti_cursor.c +++ b/drivers/gpu/drm/sti/sti_cursor.c @@ -272,7 +272,7 @@ struct drm_plane *sti_cursor_create(struct drm_device *drm_dev, &sti_plane_helpers_funcs, cursor_supported_formats, ARRAY_SIZE(cursor_supported_formats), - DRM_PLANE_TYPE_CURSOR); + DRM_PLANE_TYPE_CURSOR, NULL); if (res) { DRM_ERROR("Failed to initialize universal plane\n"); goto err_plane; diff --git a/drivers/gpu/drm/sti/sti_gdp.c b/drivers/gpu/drm/sti/sti_gdp.c index c85dc7d6b005..f9a1d92c9d95 100644 --- a/drivers/gpu/drm/sti/sti_gdp.c +++ b/drivers/gpu/drm/sti/sti_gdp.c @@ -630,7 +630,7 @@ struct drm_plane *sti_gdp_create(struct drm_device *drm_dev, &sti_plane_helpers_funcs, gdp_supported_formats, ARRAY_SIZE(gdp_supported_formats), - type); + type, NULL); if (res) { DRM_ERROR("Failed to initialize universal plane\n"); goto err; diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c index d735daccd458..49cce833f2c8 100644 --- a/drivers/gpu/drm/sti/sti_hda.c +++ b/drivers/gpu/drm/sti/sti_hda.c @@ -543,8 +543,6 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector) count++; } - drm_mode_sort(&connector->modes); - return count; } diff --git a/drivers/gpu/drm/sti/sti_hqvdp.c b/drivers/gpu/drm/sti/sti_hqvdp.c index ea0690bc77d5..43861b52261d 100644 --- a/drivers/gpu/drm/sti/sti_hqvdp.c +++ b/drivers/gpu/drm/sti/sti_hqvdp.c @@ -973,7 +973,7 @@ static struct drm_plane *sti_hqvdp_create(struct drm_device *drm_dev, &sti_plane_helpers_funcs, hqvdp_supported_formats, ARRAY_SIZE(hqvdp_supported_formats), - DRM_PLANE_TYPE_OVERLAY); + DRM_PLANE_TYPE_OVERLAY, NULL); if (res) { DRM_ERROR("Failed to initialize universal plane\n"); return NULL; diff --git a/drivers/gpu/drm/sti/sti_tvout.c b/drivers/gpu/drm/sti/sti_tvout.c index c8a4c5dae2b6..f2afcf5438b8 100644 --- a/drivers/gpu/drm/sti/sti_tvout.c +++ b/drivers/gpu/drm/sti/sti_tvout.c @@ -512,7 +512,8 @@ sti_tvout_create_dvo_encoder(struct drm_device *dev, drm_encoder->possible_clones = 1 << 0; drm_encoder_init(dev, drm_encoder, - &sti_tvout_encoder_funcs, DRM_MODE_ENCODER_LVDS); + &sti_tvout_encoder_funcs, DRM_MODE_ENCODER_LVDS, + NULL); drm_encoder_helper_add(drm_encoder, &sti_dvo_encoder_helper_funcs); @@ -564,7 +565,7 @@ static struct drm_encoder *sti_tvout_create_hda_encoder(struct drm_device *dev, drm_encoder->possible_clones = 1 << 0; drm_encoder_init(dev, drm_encoder, - &sti_tvout_encoder_funcs, DRM_MODE_ENCODER_DAC); + &sti_tvout_encoder_funcs, DRM_MODE_ENCODER_DAC, NULL); drm_encoder_helper_add(drm_encoder, &sti_hda_encoder_helper_funcs); @@ -613,7 +614,7 @@ static struct drm_encoder *sti_tvout_create_hdmi_encoder(struct drm_device *dev, drm_encoder->possible_clones = 1 << 1; drm_encoder_init(dev, drm_encoder, - &sti_tvout_encoder_funcs, DRM_MODE_ENCODER_TMDS); + &sti_tvout_encoder_funcs, DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(drm_encoder, &sti_hdmi_encoder_helper_funcs); diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig index 74d9d621453d..63ebb154b9b5 100644 --- a/drivers/gpu/drm/tegra/Kconfig +++ b/drivers/gpu/drm/tegra/Kconfig @@ -16,18 +16,6 @@ config DRM_TEGRA if DRM_TEGRA -config DRM_TEGRA_FBDEV - bool "Enable legacy fbdev support" - select DRM_KMS_FB_HELPER - select FB_SYS_FILLRECT - select FB_SYS_COPYAREA - select FB_SYS_IMAGEBLIT - default y - help - Choose this option if you have a need for the legacy fbdev support. - Note that this support also provides the Linux console on top of - the Tegra modesetting driver. - config DRM_TEGRA_DEBUG bool "NVIDIA Tegra DRM debug support" help diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index e9f24a85a103..1f5cb68357c7 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -660,7 +660,8 @@ static struct drm_plane *tegra_dc_primary_plane_create(struct drm_device *drm, err = drm_universal_plane_init(drm, &plane->base, possible_crtcs, &tegra_primary_plane_funcs, formats, - num_formats, DRM_PLANE_TYPE_PRIMARY); + num_formats, DRM_PLANE_TYPE_PRIMARY, + NULL); if (err < 0) { kfree(plane); return ERR_PTR(err); @@ -827,7 +828,8 @@ static struct drm_plane *tegra_dc_cursor_plane_create(struct drm_device *drm, err = drm_universal_plane_init(drm, &plane->base, 1 << dc->pipe, &tegra_cursor_plane_funcs, formats, - num_formats, DRM_PLANE_TYPE_CURSOR); + num_formats, DRM_PLANE_TYPE_CURSOR, + NULL); if (err < 0) { kfree(plane); return ERR_PTR(err); @@ -890,7 +892,8 @@ static struct drm_plane *tegra_dc_overlay_plane_create(struct drm_device *drm, err = drm_universal_plane_init(drm, &plane->base, 1 << dc->pipe, &tegra_overlay_plane_funcs, formats, - num_formats, DRM_PLANE_TYPE_OVERLAY); + num_formats, DRM_PLANE_TYPE_OVERLAY, + NULL); if (err < 0) { kfree(plane); return ERR_PTR(err); @@ -1732,7 +1735,7 @@ static int tegra_dc_init(struct host1x_client *client) } err = drm_crtc_init_with_planes(drm, &dc->base, primary, cursor, - &tegra_crtc_funcs); + &tegra_crtc_funcs, NULL); if (err < 0) goto cleanup; diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index 159ef515cab1..e0f827790a5e 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -106,7 +106,7 @@ static int tegra_atomic_commit(struct drm_device *drm, static const struct drm_mode_config_funcs tegra_drm_mode_funcs = { .fb_create = tegra_fb_create, -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION .output_poll_changed = tegra_fb_output_poll_changed, #endif .atomic_check = drm_atomic_helper_check, @@ -260,7 +260,7 @@ static void tegra_drm_context_free(struct tegra_drm_context *context) static void tegra_drm_lastclose(struct drm_device *drm) { -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_drm *tegra = drm->dev_private; tegra_fbdev_restore_mode(tegra->fbdev); diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h index ec49275ffb24..d88a2d18c1a4 100644 --- a/drivers/gpu/drm/tegra/drm.h +++ b/drivers/gpu/drm/tegra/drm.h @@ -30,7 +30,7 @@ struct tegra_fb { unsigned int num_planes; }; -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_fbdev { struct drm_fb_helper base; struct tegra_fb *fb; @@ -46,7 +46,7 @@ struct tegra_drm { struct mutex clients_lock; struct list_head clients; -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_fbdev *fbdev; #endif @@ -268,12 +268,12 @@ int tegra_fb_get_tiling(struct drm_framebuffer *framebuffer, struct tegra_bo_tiling *tiling); struct drm_framebuffer *tegra_fb_create(struct drm_device *drm, struct drm_file *file, - struct drm_mode_fb_cmd2 *cmd); + const struct drm_mode_fb_cmd2 *cmd); int tegra_drm_fb_prepare(struct drm_device *drm); void tegra_drm_fb_free(struct drm_device *drm); int tegra_drm_fb_init(struct drm_device *drm); void tegra_drm_fb_exit(struct drm_device *drm); -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev); void tegra_fb_output_poll_changed(struct drm_device *drm); #endif diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c index f0a138ef68ce..50d46ae3786b 100644 --- a/drivers/gpu/drm/tegra/dsi.c +++ b/drivers/gpu/drm/tegra/dsi.c @@ -1023,7 +1023,7 @@ static int tegra_dsi_init(struct host1x_client *client) drm_encoder_init(drm, &dsi->output.encoder, &tegra_dsi_encoder_funcs, - DRM_MODE_ENCODER_DSI); + DRM_MODE_ENCODER_DSI, NULL); drm_encoder_helper_add(&dsi->output.encoder, &tegra_dsi_encoder_helper_funcs); diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c index 1004075fd088..ede9e94f3312 100644 --- a/drivers/gpu/drm/tegra/fb.c +++ b/drivers/gpu/drm/tegra/fb.c @@ -18,7 +18,7 @@ static inline struct tegra_fb *to_tegra_fb(struct drm_framebuffer *fb) return container_of(fb, struct tegra_fb, base); } -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION static inline struct tegra_fbdev *to_tegra_fbdev(struct drm_fb_helper *helper) { return container_of(helper, struct tegra_fbdev, base); @@ -92,7 +92,7 @@ static struct drm_framebuffer_funcs tegra_fb_funcs = { }; static struct tegra_fb *tegra_fb_alloc(struct drm_device *drm, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct tegra_bo **planes, unsigned int num_planes) { @@ -131,7 +131,7 @@ static struct tegra_fb *tegra_fb_alloc(struct drm_device *drm, struct drm_framebuffer *tegra_fb_create(struct drm_device *drm, struct drm_file *file, - struct drm_mode_fb_cmd2 *cmd) + const struct drm_mode_fb_cmd2 *cmd) { unsigned int hsub, vsub, i; struct tegra_bo *planes[4]; @@ -181,7 +181,7 @@ unreference: return ERR_PTR(err); } -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION static struct fb_ops tegra_fb_ops = { .owner = THIS_MODULE, .fb_fillrect = drm_fb_helper_sys_fillrect, @@ -370,7 +370,7 @@ void tegra_fb_output_poll_changed(struct drm_device *drm) int tegra_drm_fb_prepare(struct drm_device *drm) { -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_drm *tegra = drm->dev_private; tegra->fbdev = tegra_fbdev_create(drm); @@ -383,7 +383,7 @@ int tegra_drm_fb_prepare(struct drm_device *drm) void tegra_drm_fb_free(struct drm_device *drm) { -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_drm *tegra = drm->dev_private; tegra_fbdev_free(tegra->fbdev); @@ -392,7 +392,7 @@ void tegra_drm_fb_free(struct drm_device *drm) int tegra_drm_fb_init(struct drm_device *drm) { -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_drm *tegra = drm->dev_private; int err; @@ -407,7 +407,7 @@ int tegra_drm_fb_init(struct drm_device *drm) void tegra_drm_fb_exit(struct drm_device *drm) { -#ifdef CONFIG_DRM_TEGRA_FBDEV +#ifdef CONFIG_DRM_FBDEV_EMULATION struct tegra_drm *tegra = drm->dev_private; tegra_fbdev_exit(tegra->fbdev); diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c index 52b32cbd9de6..b7ef4929e347 100644 --- a/drivers/gpu/drm/tegra/hdmi.c +++ b/drivers/gpu/drm/tegra/hdmi.c @@ -1320,7 +1320,7 @@ static int tegra_hdmi_init(struct host1x_client *client) hdmi->output.connector.dpms = DRM_MODE_DPMS_OFF; drm_encoder_init(drm, &hdmi->output.encoder, &tegra_hdmi_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(&hdmi->output.encoder, &tegra_hdmi_encoder_helper_funcs); diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c index bc9735b4ad60..e246334e0252 100644 --- a/drivers/gpu/drm/tegra/rgb.c +++ b/drivers/gpu/drm/tegra/rgb.c @@ -287,7 +287,7 @@ int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc) output->connector.dpms = DRM_MODE_DPMS_OFF; drm_encoder_init(drm, &output->encoder, &tegra_rgb_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); drm_encoder_helper_add(&output->encoder, &tegra_rgb_encoder_helper_funcs); diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c index 3eff7cf75d25..3e012ee25242 100644 --- a/drivers/gpu/drm/tegra/sor.c +++ b/drivers/gpu/drm/tegra/sor.c @@ -2178,7 +2178,7 @@ static int tegra_sor_init(struct host1x_client *client) sor->output.connector.dpms = DRM_MODE_DPMS_OFF; drm_encoder_init(drm, &sor->output.encoder, &tegra_sor_encoder_funcs, - encoder); + encoder, NULL); drm_encoder_helper_add(&sor->output.encoder, helpers); drm_mode_connector_attach_encoder(&sor->output.connector, diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c index 876cad58b1f9..4ddb21e7f52f 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c @@ -46,7 +46,7 @@ void tilcdc_module_cleanup(struct tilcdc_module *mod) static struct of_device_id tilcdc_of_match[]; static struct drm_framebuffer *tilcdc_fb_create(struct drm_device *dev, - struct drm_file *file_priv, struct drm_mode_fb_cmd2 *mode_cmd) + struct drm_file *file_priv, const struct drm_mode_fb_cmd2 *mode_cmd) { return drm_fb_cma_create(dev, file_priv, mode_cmd); } diff --git a/drivers/gpu/drm/tilcdc/tilcdc_panel.c b/drivers/gpu/drm/tilcdc/tilcdc_panel.c index 0af8bed7ce1e..4dda6e2f464b 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_panel.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_panel.c @@ -128,7 +128,7 @@ static struct drm_encoder *panel_encoder_create(struct drm_device *dev, encoder->possible_crtcs = 1; ret = drm_encoder_init(dev, encoder, &panel_encoder_funcs, - DRM_MODE_ENCODER_LVDS); + DRM_MODE_ENCODER_LVDS, NULL); if (ret < 0) goto fail; diff --git a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c index 354c47ca6374..5052a8af7ecb 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c @@ -138,7 +138,7 @@ static struct drm_encoder *tfp410_encoder_create(struct drm_device *dev, encoder->possible_crtcs = 1; ret = drm_encoder_init(dev, encoder, &tfp410_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); if (ret < 0) goto fail; diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h index 80adbac82bde..4a064efcea58 100644 --- a/drivers/gpu/drm/udl/udl_drv.h +++ b/drivers/gpu/drm/udl/udl_drv.h @@ -108,7 +108,7 @@ void udl_fbdev_unplug(struct drm_device *dev); struct drm_framebuffer * udl_fb_user_fb_create(struct drm_device *dev, struct drm_file *file, - struct drm_mode_fb_cmd2 *mode_cmd); + const struct drm_mode_fb_cmd2 *mode_cmd); int udl_render_hline(struct drm_device *dev, int bpp, struct urb **urb_ptr, const char *front, char **urb_buf_ptr, diff --git a/drivers/gpu/drm/udl/udl_encoder.c b/drivers/gpu/drm/udl/udl_encoder.c index 4052c4656498..a181a647fcf9 100644 --- a/drivers/gpu/drm/udl/udl_encoder.c +++ b/drivers/gpu/drm/udl/udl_encoder.c @@ -73,7 +73,8 @@ struct drm_encoder *udl_encoder_init(struct drm_device *dev) if (!encoder) return NULL; - drm_encoder_init(dev, encoder, &udl_enc_funcs, DRM_MODE_ENCODER_TMDS); + drm_encoder_init(dev, encoder, &udl_enc_funcs, DRM_MODE_ENCODER_TMDS, + NULL); drm_encoder_helper_add(encoder, &udl_helper_funcs); encoder->possible_crtcs = 1; return encoder; diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c index 62c7b1dafaa4..200419d4d43c 100644 --- a/drivers/gpu/drm/udl/udl_fb.c +++ b/drivers/gpu/drm/udl/udl_fb.c @@ -33,7 +33,6 @@ module_param(fb_defio, int, S_IWUSR | S_IRUSR | S_IWGRP | S_IRGRP); struct udl_fbdev { struct drm_fb_helper helper; struct udl_framebuffer ufb; - struct list_head fbdev_list; int fb_count; }; @@ -456,7 +455,7 @@ static const struct drm_framebuffer_funcs udlfb_funcs = { static int udl_framebuffer_init(struct drm_device *dev, struct udl_framebuffer *ufb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct udl_gem_object *obj) { int ret; @@ -624,7 +623,7 @@ void udl_fbdev_unplug(struct drm_device *dev) struct drm_framebuffer * udl_fb_user_fb_create(struct drm_device *dev, struct drm_file *file, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj; struct udl_framebuffer *ufb; diff --git a/drivers/gpu/drm/vc4/Makefile b/drivers/gpu/drm/vc4/Makefile index 32b4f9cd8f52..4c6a99f0398c 100644 --- a/drivers/gpu/drm/vc4/Makefile +++ b/drivers/gpu/drm/vc4/Makefile @@ -8,10 +8,19 @@ vc4-y := \ vc4_crtc.o \ vc4_drv.o \ vc4_kms.o \ + vc4_gem.o \ vc4_hdmi.o \ vc4_hvs.o \ - vc4_plane.o + vc4_irq.o \ + vc4_plane.o \ + vc4_render_cl.o \ + vc4_trace_points.o \ + vc4_v3d.o \ + vc4_validate.o \ + vc4_validate_shaders.o vc4-$(CONFIG_DEBUG_FS) += vc4_debugfs.o obj-$(CONFIG_DRM_VC4) += vc4.o + +CFLAGS_vc4_trace_points.o := -I$(src) diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c index ab9f5108ae1a..18dfe3ec9a62 100644 --- a/drivers/gpu/drm/vc4/vc4_bo.c +++ b/drivers/gpu/drm/vc4/vc4_bo.c @@ -12,19 +12,236 @@ * access to system memory with no MMU in between. To support it, we * use the GEM CMA helper functions to allocate contiguous ranges of * physical memory for our BOs. + * + * Since the CMA allocator is very slow, we keep a cache of recently + * freed BOs around so that the kernel's allocation of objects for 3D + * rendering can return quickly. */ #include "vc4_drv.h" +#include "uapi/drm/vc4_drm.h" + +static void vc4_bo_stats_dump(struct vc4_dev *vc4) +{ + DRM_INFO("num bos allocated: %d\n", + vc4->bo_stats.num_allocated); + DRM_INFO("size bos allocated: %dkb\n", + vc4->bo_stats.size_allocated / 1024); + DRM_INFO("num bos used: %d\n", + vc4->bo_stats.num_allocated - vc4->bo_stats.num_cached); + DRM_INFO("size bos used: %dkb\n", + (vc4->bo_stats.size_allocated - + vc4->bo_stats.size_cached) / 1024); + DRM_INFO("num bos cached: %d\n", + vc4->bo_stats.num_cached); + DRM_INFO("size bos cached: %dkb\n", + vc4->bo_stats.size_cached / 1024); +} + +#ifdef CONFIG_DEBUG_FS +int vc4_bo_stats_debugfs(struct seq_file *m, void *unused) +{ + struct drm_info_node *node = (struct drm_info_node *)m->private; + struct drm_device *dev = node->minor->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct vc4_bo_stats stats; + + /* Take a snapshot of the current stats with the lock held. */ + mutex_lock(&vc4->bo_lock); + stats = vc4->bo_stats; + mutex_unlock(&vc4->bo_lock); + + seq_printf(m, "num bos allocated: %d\n", + stats.num_allocated); + seq_printf(m, "size bos allocated: %dkb\n", + stats.size_allocated / 1024); + seq_printf(m, "num bos used: %d\n", + stats.num_allocated - stats.num_cached); + seq_printf(m, "size bos used: %dkb\n", + (stats.size_allocated - stats.size_cached) / 1024); + seq_printf(m, "num bos cached: %d\n", + stats.num_cached); + seq_printf(m, "size bos cached: %dkb\n", + stats.size_cached / 1024); + + return 0; +} +#endif + +static uint32_t bo_page_index(size_t size) +{ + return (size / PAGE_SIZE) - 1; +} + +/* Must be called with bo_lock held. */ +static void vc4_bo_destroy(struct vc4_bo *bo) +{ + struct drm_gem_object *obj = &bo->base.base; + struct vc4_dev *vc4 = to_vc4_dev(obj->dev); + + if (bo->validated_shader) { + kfree(bo->validated_shader->texture_samples); + kfree(bo->validated_shader); + bo->validated_shader = NULL; + } + + vc4->bo_stats.num_allocated--; + vc4->bo_stats.size_allocated -= obj->size; + drm_gem_cma_free_object(obj); +} + +/* Must be called with bo_lock held. */ +static void vc4_bo_remove_from_cache(struct vc4_bo *bo) +{ + struct drm_gem_object *obj = &bo->base.base; + struct vc4_dev *vc4 = to_vc4_dev(obj->dev); + + vc4->bo_stats.num_cached--; + vc4->bo_stats.size_cached -= obj->size; + + list_del(&bo->unref_head); + list_del(&bo->size_head); +} + +static struct list_head *vc4_get_cache_list_for_size(struct drm_device *dev, + size_t size) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + uint32_t page_index = bo_page_index(size); + + if (vc4->bo_cache.size_list_size <= page_index) { + uint32_t new_size = max(vc4->bo_cache.size_list_size * 2, + page_index + 1); + struct list_head *new_list; + uint32_t i; + + new_list = kmalloc_array(new_size, sizeof(struct list_head), + GFP_KERNEL); + if (!new_list) + return NULL; + + /* Rebase the old cached BO lists to their new list + * head locations. + */ + for (i = 0; i < vc4->bo_cache.size_list_size; i++) { + struct list_head *old_list = + &vc4->bo_cache.size_list[i]; + + if (list_empty(old_list)) + INIT_LIST_HEAD(&new_list[i]); + else + list_replace(old_list, &new_list[i]); + } + /* And initialize the brand new BO list heads. */ + for (i = vc4->bo_cache.size_list_size; i < new_size; i++) + INIT_LIST_HEAD(&new_list[i]); + + kfree(vc4->bo_cache.size_list); + vc4->bo_cache.size_list = new_list; + vc4->bo_cache.size_list_size = new_size; + } + + return &vc4->bo_cache.size_list[page_index]; +} + +void vc4_bo_cache_purge(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + mutex_lock(&vc4->bo_lock); + while (!list_empty(&vc4->bo_cache.time_list)) { + struct vc4_bo *bo = list_last_entry(&vc4->bo_cache.time_list, + struct vc4_bo, unref_head); + vc4_bo_remove_from_cache(bo); + vc4_bo_destroy(bo); + } + mutex_unlock(&vc4->bo_lock); +} + +static struct vc4_bo *vc4_bo_get_from_cache(struct drm_device *dev, + uint32_t size) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + uint32_t page_index = bo_page_index(size); + struct vc4_bo *bo = NULL; + + size = roundup(size, PAGE_SIZE); + + mutex_lock(&vc4->bo_lock); + if (page_index >= vc4->bo_cache.size_list_size) + goto out; -struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t size) + if (list_empty(&vc4->bo_cache.size_list[page_index])) + goto out; + + bo = list_first_entry(&vc4->bo_cache.size_list[page_index], + struct vc4_bo, size_head); + vc4_bo_remove_from_cache(bo); + kref_init(&bo->base.base.refcount); + +out: + mutex_unlock(&vc4->bo_lock); + return bo; +} + +/** + * vc4_gem_create_object - Implementation of driver->gem_create_object. + * + * This lets the CMA helpers allocate object structs for us, and keep + * our BO stats correct. + */ +struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size) { + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct vc4_bo *bo; + + bo = kzalloc(sizeof(*bo), GFP_KERNEL); + if (!bo) + return ERR_PTR(-ENOMEM); + + mutex_lock(&vc4->bo_lock); + vc4->bo_stats.num_allocated++; + vc4->bo_stats.size_allocated += size; + mutex_unlock(&vc4->bo_lock); + + return &bo->base.base; +} + +struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t unaligned_size, + bool from_cache) +{ + size_t size = roundup(unaligned_size, PAGE_SIZE); + struct vc4_dev *vc4 = to_vc4_dev(dev); struct drm_gem_cma_object *cma_obj; - cma_obj = drm_gem_cma_create(dev, size); - if (IS_ERR(cma_obj)) + if (size == 0) return NULL; - else - return to_vc4_bo(&cma_obj->base); + + /* First, try to get a vc4_bo from the kernel BO cache. */ + if (from_cache) { + struct vc4_bo *bo = vc4_bo_get_from_cache(dev, size); + + if (bo) + return bo; + } + + cma_obj = drm_gem_cma_create(dev, size); + if (IS_ERR(cma_obj)) { + /* + * If we've run out of CMA memory, kill the cache of + * CMA allocations we've got laying around and try again. + */ + vc4_bo_cache_purge(dev); + + cma_obj = drm_gem_cma_create(dev, size); + if (IS_ERR(cma_obj)) { + DRM_ERROR("Failed to allocate from CMA:\n"); + vc4_bo_stats_dump(vc4); + return NULL; + } + } + + return to_vc4_bo(&cma_obj->base); } int vc4_dumb_create(struct drm_file *file_priv, @@ -41,7 +258,191 @@ int vc4_dumb_create(struct drm_file *file_priv, if (args->size < args->pitch * args->height) args->size = args->pitch * args->height; - bo = vc4_bo_create(dev, roundup(args->size, PAGE_SIZE)); + bo = vc4_bo_create(dev, args->size, false); + if (!bo) + return -ENOMEM; + + ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle); + drm_gem_object_unreference_unlocked(&bo->base.base); + + return ret; +} + +/* Must be called with bo_lock held. */ +static void vc4_bo_cache_free_old(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + unsigned long expire_time = jiffies - msecs_to_jiffies(1000); + + while (!list_empty(&vc4->bo_cache.time_list)) { + struct vc4_bo *bo = list_last_entry(&vc4->bo_cache.time_list, + struct vc4_bo, unref_head); + if (time_before(expire_time, bo->free_time)) { + mod_timer(&vc4->bo_cache.time_timer, + round_jiffies_up(jiffies + + msecs_to_jiffies(1000))); + return; + } + + vc4_bo_remove_from_cache(bo); + vc4_bo_destroy(bo); + } +} + +/* Called on the last userspace/kernel unreference of the BO. Returns + * it to the BO cache if possible, otherwise frees it. + * + * Note that this is called with the struct_mutex held. + */ +void vc4_free_object(struct drm_gem_object *gem_bo) +{ + struct drm_device *dev = gem_bo->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct vc4_bo *bo = to_vc4_bo(gem_bo); + struct list_head *cache_list; + + mutex_lock(&vc4->bo_lock); + /* If the object references someone else's memory, we can't cache it. + */ + if (gem_bo->import_attach) { + vc4_bo_destroy(bo); + goto out; + } + + /* Don't cache if it was publicly named. */ + if (gem_bo->name) { + vc4_bo_destroy(bo); + goto out; + } + + cache_list = vc4_get_cache_list_for_size(dev, gem_bo->size); + if (!cache_list) { + vc4_bo_destroy(bo); + goto out; + } + + if (bo->validated_shader) { + kfree(bo->validated_shader->texture_samples); + kfree(bo->validated_shader); + bo->validated_shader = NULL; + } + + bo->free_time = jiffies; + list_add(&bo->size_head, cache_list); + list_add(&bo->unref_head, &vc4->bo_cache.time_list); + + vc4->bo_stats.num_cached++; + vc4->bo_stats.size_cached += gem_bo->size; + + vc4_bo_cache_free_old(dev); + +out: + mutex_unlock(&vc4->bo_lock); +} + +static void vc4_bo_cache_time_work(struct work_struct *work) +{ + struct vc4_dev *vc4 = + container_of(work, struct vc4_dev, bo_cache.time_work); + struct drm_device *dev = vc4->dev; + + mutex_lock(&vc4->bo_lock); + vc4_bo_cache_free_old(dev); + mutex_unlock(&vc4->bo_lock); +} + +static void vc4_bo_cache_time_timer(unsigned long data) +{ + struct drm_device *dev = (struct drm_device *)data; + struct vc4_dev *vc4 = to_vc4_dev(dev); + + schedule_work(&vc4->bo_cache.time_work); +} + +struct dma_buf * +vc4_prime_export(struct drm_device *dev, struct drm_gem_object *obj, int flags) +{ + struct vc4_bo *bo = to_vc4_bo(obj); + + if (bo->validated_shader) { + DRM_ERROR("Attempting to export shader BO\n"); + return ERR_PTR(-EINVAL); + } + + return drm_gem_prime_export(dev, obj, flags); +} + +int vc4_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct drm_gem_object *gem_obj; + struct vc4_bo *bo; + int ret; + + ret = drm_gem_mmap(filp, vma); + if (ret) + return ret; + + gem_obj = vma->vm_private_data; + bo = to_vc4_bo(gem_obj); + + if (bo->validated_shader && (vma->vm_flags & VM_WRITE)) { + DRM_ERROR("mmaping of shader BOs for writing not allowed.\n"); + return -EINVAL; + } + + /* + * Clear the VM_PFNMAP flag that was set by drm_gem_mmap(), and set the + * vm_pgoff (used as a fake buffer offset by DRM) to 0 as we want to map + * the whole buffer. + */ + vma->vm_flags &= ~VM_PFNMAP; + vma->vm_pgoff = 0; + + ret = dma_mmap_writecombine(bo->base.base.dev->dev, vma, + bo->base.vaddr, bo->base.paddr, + vma->vm_end - vma->vm_start); + if (ret) + drm_gem_vm_close(vma); + + return ret; +} + +int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma) +{ + struct vc4_bo *bo = to_vc4_bo(obj); + + if (bo->validated_shader && (vma->vm_flags & VM_WRITE)) { + DRM_ERROR("mmaping of shader BOs for writing not allowed.\n"); + return -EINVAL; + } + + return drm_gem_cma_prime_mmap(obj, vma); +} + +void *vc4_prime_vmap(struct drm_gem_object *obj) +{ + struct vc4_bo *bo = to_vc4_bo(obj); + + if (bo->validated_shader) { + DRM_ERROR("mmaping of shader BOs not allowed.\n"); + return ERR_PTR(-EINVAL); + } + + return drm_gem_cma_prime_vmap(obj); +} + +int vc4_create_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_vc4_create_bo *args = data; + struct vc4_bo *bo = NULL; + int ret; + + /* + * We can't allocate from the BO cache, because the BOs don't + * get zeroed, and that might leak data between users. + */ + bo = vc4_bo_create(dev, args->size, false); if (!bo) return -ENOMEM; @@ -50,3 +451,107 @@ int vc4_dumb_create(struct drm_file *file_priv, return ret; } + +int vc4_mmap_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_vc4_mmap_bo *args = data; + struct drm_gem_object *gem_obj; + + gem_obj = drm_gem_object_lookup(dev, file_priv, args->handle); + if (!gem_obj) { + DRM_ERROR("Failed to look up GEM BO %d\n", args->handle); + return -EINVAL; + } + + /* The mmap offset was set up at BO allocation time. */ + args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node); + + drm_gem_object_unreference_unlocked(gem_obj); + return 0; +} + +int +vc4_create_shader_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_vc4_create_shader_bo *args = data; + struct vc4_bo *bo = NULL; + int ret; + + if (args->size == 0) + return -EINVAL; + + if (args->size % sizeof(u64) != 0) + return -EINVAL; + + if (args->flags != 0) { + DRM_INFO("Unknown flags set: 0x%08x\n", args->flags); + return -EINVAL; + } + + if (args->pad != 0) { + DRM_INFO("Pad set: 0x%08x\n", args->pad); + return -EINVAL; + } + + bo = vc4_bo_create(dev, args->size, true); + if (!bo) + return -ENOMEM; + + ret = copy_from_user(bo->base.vaddr, + (void __user *)(uintptr_t)args->data, + args->size); + if (ret != 0) + goto fail; + /* Clear the rest of the memory from allocating from the BO + * cache. + */ + memset(bo->base.vaddr + args->size, 0, + bo->base.base.size - args->size); + + bo->validated_shader = vc4_validate_shader(&bo->base); + if (!bo->validated_shader) { + ret = -EINVAL; + goto fail; + } + + /* We have to create the handle after validation, to avoid + * races for users to do doing things like mmap the shader BO. + */ + ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle); + + fail: + drm_gem_object_unreference_unlocked(&bo->base.base); + + return ret; +} + +void vc4_bo_cache_init(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + mutex_init(&vc4->bo_lock); + + INIT_LIST_HEAD(&vc4->bo_cache.time_list); + + INIT_WORK(&vc4->bo_cache.time_work, vc4_bo_cache_time_work); + setup_timer(&vc4->bo_cache.time_timer, + vc4_bo_cache_time_timer, + (unsigned long)dev); +} + +void vc4_bo_cache_destroy(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + del_timer(&vc4->bo_cache.time_timer); + cancel_work_sync(&vc4->bo_cache.time_work); + + vc4_bo_cache_purge(dev); + + if (vc4->bo_stats.num_allocated) { + DRM_ERROR("Destroying BO cache while BOs still allocated:\n"); + vc4_bo_stats_dump(vc4); + } +} diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 265064c62d49..8d0d70e51ef2 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -35,6 +35,7 @@ #include "drm_atomic_helper.h" #include "drm_crtc_helper.h" #include "linux/clk.h" +#include "drm_fb_cma_helper.h" #include "linux/component.h" #include "linux/of_device.h" #include "vc4_drv.h" @@ -476,10 +477,106 @@ static irqreturn_t vc4_crtc_irq_handler(int irq, void *data) return ret; } +struct vc4_async_flip_state { + struct drm_crtc *crtc; + struct drm_framebuffer *fb; + struct drm_pending_vblank_event *event; + + struct vc4_seqno_cb cb; +}; + +/* Called when the V3D execution for the BO being flipped to is done, so that + * we can actually update the plane's address to point to it. + */ +static void +vc4_async_page_flip_complete(struct vc4_seqno_cb *cb) +{ + struct vc4_async_flip_state *flip_state = + container_of(cb, struct vc4_async_flip_state, cb); + struct drm_crtc *crtc = flip_state->crtc; + struct drm_device *dev = crtc->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct drm_plane *plane = crtc->primary; + + vc4_plane_async_set_fb(plane, flip_state->fb); + if (flip_state->event) { + unsigned long flags; + + spin_lock_irqsave(&dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, flip_state->event); + spin_unlock_irqrestore(&dev->event_lock, flags); + } + + drm_framebuffer_unreference(flip_state->fb); + kfree(flip_state); + + up(&vc4->async_modeset); +} + +/* Implements async (non-vblank-synced) page flips. + * + * The page flip ioctl needs to return immediately, so we grab the + * modeset semaphore on the pipe, and queue the address update for + * when V3D is done with the BO being flipped to. + */ +static int vc4_async_page_flip(struct drm_crtc *crtc, + struct drm_framebuffer *fb, + struct drm_pending_vblank_event *event, + uint32_t flags) +{ + struct drm_device *dev = crtc->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct drm_plane *plane = crtc->primary; + int ret = 0; + struct vc4_async_flip_state *flip_state; + struct drm_gem_cma_object *cma_bo = drm_fb_cma_get_gem_obj(fb, 0); + struct vc4_bo *bo = to_vc4_bo(&cma_bo->base); + + flip_state = kzalloc(sizeof(*flip_state), GFP_KERNEL); + if (!flip_state) + return -ENOMEM; + + drm_framebuffer_reference(fb); + flip_state->fb = fb; + flip_state->crtc = crtc; + flip_state->event = event; + + /* Make sure all other async modesetes have landed. */ + ret = down_interruptible(&vc4->async_modeset); + if (ret) { + kfree(flip_state); + return ret; + } + + /* Immediately update the plane's legacy fb pointer, so that later + * modeset prep sees the state that will be present when the semaphore + * is released. + */ + drm_atomic_set_fb_for_plane(plane->state, fb); + plane->fb = fb; + + vc4_queue_seqno_cb(dev, &flip_state->cb, bo->seqno, + vc4_async_page_flip_complete); + + /* Driver takes ownership of state on successful async commit. */ + return 0; +} + +static int vc4_page_flip(struct drm_crtc *crtc, + struct drm_framebuffer *fb, + struct drm_pending_vblank_event *event, + uint32_t flags) +{ + if (flags & DRM_MODE_PAGE_FLIP_ASYNC) + return vc4_async_page_flip(crtc, fb, event, flags); + else + return drm_atomic_helper_page_flip(crtc, fb, event, flags); +} + static const struct drm_crtc_funcs vc4_crtc_funcs = { .set_config = drm_atomic_helper_set_config, .destroy = vc4_crtc_destroy, - .page_flip = drm_atomic_helper_page_flip, + .page_flip = vc4_page_flip, .set_property = NULL, .cursor_set = NULL, /* handled by drm_mode_cursor_universal */ .cursor_move = NULL, /* handled by drm_mode_cursor_universal */ @@ -606,7 +703,7 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data) } drm_crtc_init_with_planes(drm, crtc, primary_plane, cursor_plane, - &vc4_crtc_funcs); + &vc4_crtc_funcs, NULL); drm_crtc_helper_add(crtc, &vc4_crtc_helper_funcs); primary_plane->crtc = crtc; cursor_plane->crtc = crtc; diff --git a/drivers/gpu/drm/vc4/vc4_debugfs.c b/drivers/gpu/drm/vc4/vc4_debugfs.c index 4297b0a5b74e..d76ad10b07fd 100644 --- a/drivers/gpu/drm/vc4/vc4_debugfs.c +++ b/drivers/gpu/drm/vc4/vc4_debugfs.c @@ -16,11 +16,14 @@ #include "vc4_regs.h" static const struct drm_info_list vc4_debugfs_list[] = { + {"bo_stats", vc4_bo_stats_debugfs, 0}, {"hdmi_regs", vc4_hdmi_debugfs_regs, 0}, {"hvs_regs", vc4_hvs_debugfs_regs, 0}, {"crtc0_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)0}, {"crtc1_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)1}, {"crtc2_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)2}, + {"v3d_ident", vc4_v3d_debugfs_ident, 0}, + {"v3d_regs", vc4_v3d_debugfs_regs, 0}, }; #define VC4_DEBUGFS_ENTRIES ARRAY_SIZE(vc4_debugfs_list) diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index d5db9e0f3b73..cbcbbb83500e 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -16,6 +16,7 @@ #include <linux/platform_device.h> #include "drm_fb_cma_helper.h" +#include "uapi/drm/vc4_drm.h" #include "vc4_drv.h" #include "vc4_regs.h" @@ -63,7 +64,7 @@ static const struct file_operations vc4_drm_fops = { .open = drm_open, .release = drm_release, .unlocked_ioctl = drm_ioctl, - .mmap = drm_gem_cma_mmap, + .mmap = vc4_mmap, .poll = drm_poll, .read = drm_read, #ifdef CONFIG_COMPAT @@ -73,16 +74,30 @@ static const struct file_operations vc4_drm_fops = { }; static const struct drm_ioctl_desc vc4_drm_ioctls[] = { + DRM_IOCTL_DEF_DRV(VC4_SUBMIT_CL, vc4_submit_cl_ioctl, 0), + DRM_IOCTL_DEF_DRV(VC4_WAIT_SEQNO, vc4_wait_seqno_ioctl, 0), + DRM_IOCTL_DEF_DRV(VC4_WAIT_BO, vc4_wait_bo_ioctl, 0), + DRM_IOCTL_DEF_DRV(VC4_CREATE_BO, vc4_create_bo_ioctl, 0), + DRM_IOCTL_DEF_DRV(VC4_MMAP_BO, vc4_mmap_bo_ioctl, 0), + DRM_IOCTL_DEF_DRV(VC4_CREATE_SHADER_BO, vc4_create_shader_bo_ioctl, 0), + DRM_IOCTL_DEF_DRV(VC4_GET_HANG_STATE, vc4_get_hang_state_ioctl, + DRM_ROOT_ONLY), }; static struct drm_driver vc4_drm_driver = { .driver_features = (DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_GEM | + DRIVER_HAVE_IRQ | DRIVER_PRIME), .lastclose = vc4_lastclose, .preclose = vc4_drm_preclose, + .irq_handler = vc4_irq, + .irq_preinstall = vc4_irq_preinstall, + .irq_postinstall = vc4_irq_postinstall, + .irq_uninstall = vc4_irq_uninstall, + .enable_vblank = vc4_enable_vblank, .disable_vblank = vc4_disable_vblank, .get_vblank_counter = drm_vblank_count, @@ -92,18 +107,19 @@ static struct drm_driver vc4_drm_driver = { .debugfs_cleanup = vc4_debugfs_cleanup, #endif - .gem_free_object = drm_gem_cma_free_object, + .gem_create_object = vc4_create_object, + .gem_free_object = vc4_free_object, .gem_vm_ops = &drm_gem_cma_vm_ops, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_prime_import = drm_gem_prime_import, - .gem_prime_export = drm_gem_prime_export, + .gem_prime_export = vc4_prime_export, .gem_prime_get_sg_table = drm_gem_cma_prime_get_sg_table, .gem_prime_import_sg_table = drm_gem_cma_prime_import_sg_table, - .gem_prime_vmap = drm_gem_cma_prime_vmap, + .gem_prime_vmap = vc4_prime_vmap, .gem_prime_vunmap = drm_gem_cma_prime_vunmap, - .gem_prime_mmap = drm_gem_cma_prime_mmap, + .gem_prime_mmap = vc4_prime_mmap, .dumb_create = vc4_dumb_create, .dumb_map_offset = drm_gem_cma_dumb_map_offset, @@ -170,13 +186,17 @@ static int vc4_drm_bind(struct device *dev) drm_dev_set_unique(drm, dev_name(dev)); + vc4_bo_cache_init(drm); + drm_mode_config_init(drm); if (ret) goto unref; + vc4_gem_init(drm); + ret = component_bind_all(dev, drm); if (ret) - goto unref; + goto gem_destroy; ret = drm_dev_register(drm, 0); if (ret < 0) @@ -200,8 +220,11 @@ unregister: drm_dev_unregister(drm); unbind_all: component_unbind_all(dev, drm); +gem_destroy: + vc4_gem_destroy(drm); unref: drm_dev_unref(drm); + vc4_bo_cache_destroy(drm); return ret; } @@ -228,6 +251,7 @@ static struct platform_driver *const component_drivers[] = { &vc4_hdmi_driver, &vc4_crtc_driver, &vc4_hvs_driver, + &vc4_v3d_driver, }; static int vc4_platform_drm_probe(struct platform_device *pdev) diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index fd8319fa682e..080865ec2bae 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -15,8 +15,89 @@ struct vc4_dev { struct vc4_hdmi *hdmi; struct vc4_hvs *hvs; struct vc4_crtc *crtc[3]; + struct vc4_v3d *v3d; struct drm_fbdev_cma *fbdev; + + struct vc4_hang_state *hang_state; + + /* The kernel-space BO cache. Tracks buffers that have been + * unreferenced by all other users (refcounts of 0!) but not + * yet freed, so we can do cheap allocations. + */ + struct vc4_bo_cache { + /* Array of list heads for entries in the BO cache, + * based on number of pages, so we can do O(1) lookups + * in the cache when allocating. + */ + struct list_head *size_list; + uint32_t size_list_size; + + /* List of all BOs in the cache, ordered by age, so we + * can do O(1) lookups when trying to free old + * buffers. + */ + struct list_head time_list; + struct work_struct time_work; + struct timer_list time_timer; + } bo_cache; + + struct vc4_bo_stats { + u32 num_allocated; + u32 size_allocated; + u32 num_cached; + u32 size_cached; + } bo_stats; + + /* Protects bo_cache and the BO stats. */ + struct mutex bo_lock; + + /* Sequence number for the last job queued in job_list. + * Starts at 0 (no jobs emitted). + */ + uint64_t emit_seqno; + + /* Sequence number for the last completed job on the GPU. + * Starts at 0 (no jobs completed). + */ + uint64_t finished_seqno; + + /* List of all struct vc4_exec_info for jobs to be executed. + * The first job in the list is the one currently programmed + * into ct0ca/ct1ca for execution. + */ + struct list_head job_list; + /* List of the finished vc4_exec_infos waiting to be freed by + * job_done_work. + */ + struct list_head job_done_list; + /* Spinlock used to synchronize the job_list and seqno + * accesses between the IRQ handler and GEM ioctls. + */ + spinlock_t job_lock; + wait_queue_head_t job_wait_queue; + struct work_struct job_done_work; + + /* List of struct vc4_seqno_cb for callbacks to be made from a + * workqueue when the given seqno is passed. + */ + struct list_head seqno_cb_list; + + /* The binner overflow memory that's currently set up in + * BPOA/BPOS registers. When overflow occurs and a new one is + * allocated, the previous one will be moved to + * vc4->current_exec's free list. + */ + struct vc4_bo *overflow_mem; + struct work_struct overflow_mem_work; + + struct { + uint32_t last_ct0ca, last_ct1ca; + struct timer_list timer; + struct work_struct reset_work; + } hangcheck; + + struct semaphore async_modeset; }; static inline struct vc4_dev * @@ -27,6 +108,25 @@ to_vc4_dev(struct drm_device *dev) struct vc4_bo { struct drm_gem_cma_object base; + + /* seqno of the last job to render to this BO. */ + uint64_t seqno; + + /* List entry for the BO's position in either + * vc4_exec_info->unref_list or vc4_dev->bo_cache.time_list + */ + struct list_head unref_head; + + /* Time in jiffies when the BO was put in vc4->bo_cache. */ + unsigned long free_time; + + /* List entry for the BO's position in vc4_dev->bo_cache.size_list */ + struct list_head size_head; + + /* Struct for shader validation state, if created by + * DRM_IOCTL_VC4_CREATE_SHADER_BO. + */ + struct vc4_validated_shader_info *validated_shader; }; static inline struct vc4_bo * @@ -35,6 +135,17 @@ to_vc4_bo(struct drm_gem_object *bo) return (struct vc4_bo *)bo; } +struct vc4_seqno_cb { + struct work_struct work; + uint64_t seqno; + void (*func)(struct vc4_seqno_cb *cb); +}; + +struct vc4_v3d { + struct platform_device *pdev; + void __iomem *regs; +}; + struct vc4_hvs { struct platform_device *pdev; void __iomem *regs; @@ -72,9 +183,142 @@ to_vc4_encoder(struct drm_encoder *encoder) return container_of(encoder, struct vc4_encoder, base); } +#define V3D_READ(offset) readl(vc4->v3d->regs + offset) +#define V3D_WRITE(offset, val) writel(val, vc4->v3d->regs + offset) #define HVS_READ(offset) readl(vc4->hvs->regs + offset) #define HVS_WRITE(offset, val) writel(val, vc4->hvs->regs + offset) +struct vc4_exec_info { + /* Sequence number for this bin/render job. */ + uint64_t seqno; + + /* Kernel-space copy of the ioctl arguments */ + struct drm_vc4_submit_cl *args; + + /* This is the array of BOs that were looked up at the start of exec. + * Command validation will use indices into this array. + */ + struct drm_gem_cma_object **bo; + uint32_t bo_count; + + /* Pointers for our position in vc4->job_list */ + struct list_head head; + + /* List of other BOs used in the job that need to be released + * once the job is complete. + */ + struct list_head unref_list; + + /* Current unvalidated indices into @bo loaded by the non-hardware + * VC4_PACKET_GEM_HANDLES. + */ + uint32_t bo_index[2]; + + /* This is the BO where we store the validated command lists, shader + * records, and uniforms. + */ + struct drm_gem_cma_object *exec_bo; + + /** + * This tracks the per-shader-record state (packet 64) that + * determines the length of the shader record and the offset + * it's expected to be found at. It gets read in from the + * command lists. + */ + struct vc4_shader_state { + uint32_t addr; + /* Maximum vertex index referenced by any primitive using this + * shader state. + */ + uint32_t max_index; + } *shader_state; + + /** How many shader states the user declared they were using. */ + uint32_t shader_state_size; + /** How many shader state records the validator has seen. */ + uint32_t shader_state_count; + + bool found_tile_binning_mode_config_packet; + bool found_start_tile_binning_packet; + bool found_increment_semaphore_packet; + bool found_flush; + uint8_t bin_tiles_x, bin_tiles_y; + struct drm_gem_cma_object *tile_bo; + uint32_t tile_alloc_offset; + + /** + * Computed addresses pointing into exec_bo where we start the + * bin thread (ct0) and render thread (ct1). + */ + uint32_t ct0ca, ct0ea; + uint32_t ct1ca, ct1ea; + + /* Pointer to the unvalidated bin CL (if present). */ + void *bin_u; + + /* Pointers to the shader recs. These paddr gets incremented as CL + * packets are relocated in validate_gl_shader_state, and the vaddrs + * (u and v) get incremented and size decremented as the shader recs + * themselves are validated. + */ + void *shader_rec_u; + void *shader_rec_v; + uint32_t shader_rec_p; + uint32_t shader_rec_size; + + /* Pointers to the uniform data. These pointers are incremented, and + * size decremented, as each batch of uniforms is uploaded. + */ + void *uniforms_u; + void *uniforms_v; + uint32_t uniforms_p; + uint32_t uniforms_size; +}; + +static inline struct vc4_exec_info * +vc4_first_job(struct vc4_dev *vc4) +{ + if (list_empty(&vc4->job_list)) + return NULL; + return list_first_entry(&vc4->job_list, struct vc4_exec_info, head); +} + +/** + * struct vc4_texture_sample_info - saves the offsets into the UBO for texture + * setup parameters. + * + * This will be used at draw time to relocate the reference to the texture + * contents in p0, and validate that the offset combined with + * width/height/stride/etc. from p1 and p2/p3 doesn't sample outside the BO. + * Note that the hardware treats unprovided config parameters as 0, so not all + * of them need to be set up for every texure sample, and we'll store ~0 as + * the offset to mark the unused ones. + * + * See the VC4 3D architecture guide page 41 ("Texture and Memory Lookup Unit + * Setup") for definitions of the texture parameters. + */ +struct vc4_texture_sample_info { + bool is_direct; + uint32_t p_offset[4]; +}; + +/** + * struct vc4_validated_shader_info - information about validated shaders that + * needs to be used from command list validation. + * + * For a given shader, each time a shader state record references it, we need + * to verify that the shader doesn't read more uniforms than the shader state + * record's uniform BO pointer can provide, and we need to apply relocations + * and validate the shader state record's uniforms that define the texture + * samples. + */ +struct vc4_validated_shader_info { + uint32_t uniforms_size; + uint32_t uniforms_src_size; + uint32_t num_texture_samples; + struct vc4_texture_sample_info *texture_samples; +}; + /** * _wait_for - magic (register) wait macro * @@ -104,13 +348,29 @@ to_vc4_encoder(struct drm_encoder *encoder) #define wait_for(COND, MS) _wait_for(COND, MS, 1) /* vc4_bo.c */ +struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size); void vc4_free_object(struct drm_gem_object *gem_obj); -struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t size); +struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t size, + bool from_cache); int vc4_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); struct dma_buf *vc4_prime_export(struct drm_device *dev, struct drm_gem_object *obj, int flags); +int vc4_create_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +int vc4_create_shader_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +int vc4_mmap_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +int vc4_get_hang_state_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +int vc4_mmap(struct file *filp, struct vm_area_struct *vma); +int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma); +void *vc4_prime_vmap(struct drm_gem_object *obj); +void vc4_bo_cache_init(struct drm_device *dev); +void vc4_bo_cache_destroy(struct drm_device *dev); +int vc4_bo_stats_debugfs(struct seq_file *m, void *arg); /* vc4_crtc.c */ extern struct platform_driver vc4_crtc_driver; @@ -126,10 +386,34 @@ void vc4_debugfs_cleanup(struct drm_minor *minor); /* vc4_drv.c */ void __iomem *vc4_ioremap_regs(struct platform_device *dev, int index); +/* vc4_gem.c */ +void vc4_gem_init(struct drm_device *dev); +void vc4_gem_destroy(struct drm_device *dev); +int vc4_submit_cl_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +int vc4_wait_seqno_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +int vc4_wait_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +void vc4_submit_next_job(struct drm_device *dev); +int vc4_wait_for_seqno(struct drm_device *dev, uint64_t seqno, + uint64_t timeout_ns, bool interruptible); +void vc4_job_handle_completed(struct vc4_dev *vc4); +int vc4_queue_seqno_cb(struct drm_device *dev, + struct vc4_seqno_cb *cb, uint64_t seqno, + void (*func)(struct vc4_seqno_cb *cb)); + /* vc4_hdmi.c */ extern struct platform_driver vc4_hdmi_driver; int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused); +/* vc4_irq.c */ +irqreturn_t vc4_irq(int irq, void *arg); +void vc4_irq_preinstall(struct drm_device *dev); +int vc4_irq_postinstall(struct drm_device *dev); +void vc4_irq_uninstall(struct drm_device *dev); +void vc4_irq_reset(struct drm_device *dev); + /* vc4_hvs.c */ extern struct platform_driver vc4_hvs_driver; void vc4_hvs_dump_state(struct drm_device *dev); @@ -143,3 +427,35 @@ struct drm_plane *vc4_plane_init(struct drm_device *dev, enum drm_plane_type type); u32 vc4_plane_write_dlist(struct drm_plane *plane, u32 __iomem *dlist); u32 vc4_plane_dlist_size(struct drm_plane_state *state); +void vc4_plane_async_set_fb(struct drm_plane *plane, + struct drm_framebuffer *fb); + +/* vc4_v3d.c */ +extern struct platform_driver vc4_v3d_driver; +int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused); +int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused); +int vc4_v3d_set_power(struct vc4_dev *vc4, bool on); + +/* vc4_validate.c */ +int +vc4_validate_bin_cl(struct drm_device *dev, + void *validated, + void *unvalidated, + struct vc4_exec_info *exec); + +int +vc4_validate_shader_recs(struct drm_device *dev, struct vc4_exec_info *exec); + +struct drm_gem_cma_object *vc4_use_bo(struct vc4_exec_info *exec, + uint32_t hindex); + +int vc4_get_rcl(struct drm_device *dev, struct vc4_exec_info *exec); + +bool vc4_check_tex_size(struct vc4_exec_info *exec, + struct drm_gem_cma_object *fbo, + uint32_t offset, uint8_t tiling_format, + uint32_t width, uint32_t height, uint8_t cpp); + +/* vc4_validate_shader.c */ +struct vc4_validated_shader_info * +vc4_validate_shader(struct drm_gem_cma_object *shader_obj); diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c new file mode 100644 index 000000000000..39f29e759334 --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -0,0 +1,867 @@ +/* + * Copyright © 2014 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/device.h> +#include <linux/io.h> + +#include "uapi/drm/vc4_drm.h" +#include "vc4_drv.h" +#include "vc4_regs.h" +#include "vc4_trace.h" + +static void +vc4_queue_hangcheck(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + mod_timer(&vc4->hangcheck.timer, + round_jiffies_up(jiffies + msecs_to_jiffies(100))); +} + +struct vc4_hang_state { + struct drm_vc4_get_hang_state user_state; + + u32 bo_count; + struct drm_gem_object **bo; +}; + +static void +vc4_free_hang_state(struct drm_device *dev, struct vc4_hang_state *state) +{ + unsigned int i; + + mutex_lock(&dev->struct_mutex); + for (i = 0; i < state->user_state.bo_count; i++) + drm_gem_object_unreference(state->bo[i]); + mutex_unlock(&dev->struct_mutex); + + kfree(state); +} + +int +vc4_get_hang_state_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_vc4_get_hang_state *get_state = data; + struct drm_vc4_get_hang_state_bo *bo_state; + struct vc4_hang_state *kernel_state; + struct drm_vc4_get_hang_state *state; + struct vc4_dev *vc4 = to_vc4_dev(dev); + unsigned long irqflags; + u32 i; + int ret; + + spin_lock_irqsave(&vc4->job_lock, irqflags); + kernel_state = vc4->hang_state; + if (!kernel_state) { + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + return -ENOENT; + } + state = &kernel_state->user_state; + + /* If the user's array isn't big enough, just return the + * required array size. + */ + if (get_state->bo_count < state->bo_count) { + get_state->bo_count = state->bo_count; + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + return 0; + } + + vc4->hang_state = NULL; + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + + /* Save the user's BO pointer, so we don't stomp it with the memcpy. */ + state->bo = get_state->bo; + memcpy(get_state, state, sizeof(*state)); + + bo_state = kcalloc(state->bo_count, sizeof(*bo_state), GFP_KERNEL); + if (!bo_state) { + ret = -ENOMEM; + goto err_free; + } + + for (i = 0; i < state->bo_count; i++) { + struct vc4_bo *vc4_bo = to_vc4_bo(kernel_state->bo[i]); + u32 handle; + + ret = drm_gem_handle_create(file_priv, kernel_state->bo[i], + &handle); + + if (ret) { + state->bo_count = i - 1; + goto err; + } + bo_state[i].handle = handle; + bo_state[i].paddr = vc4_bo->base.paddr; + bo_state[i].size = vc4_bo->base.base.size; + } + + ret = copy_to_user((void __user *)(uintptr_t)get_state->bo, + bo_state, + state->bo_count * sizeof(*bo_state)); + kfree(bo_state); + +err_free: + + vc4_free_hang_state(dev, kernel_state); + +err: + return ret; +} + +static void +vc4_save_hang_state(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct drm_vc4_get_hang_state *state; + struct vc4_hang_state *kernel_state; + struct vc4_exec_info *exec; + struct vc4_bo *bo; + unsigned long irqflags; + unsigned int i, unref_list_count; + + kernel_state = kcalloc(1, sizeof(*state), GFP_KERNEL); + if (!kernel_state) + return; + + state = &kernel_state->user_state; + + spin_lock_irqsave(&vc4->job_lock, irqflags); + exec = vc4_first_job(vc4); + if (!exec) { + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + return; + } + + unref_list_count = 0; + list_for_each_entry(bo, &exec->unref_list, unref_head) + unref_list_count++; + + state->bo_count = exec->bo_count + unref_list_count; + kernel_state->bo = kcalloc(state->bo_count, sizeof(*kernel_state->bo), + GFP_ATOMIC); + if (!kernel_state->bo) { + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + return; + } + + for (i = 0; i < exec->bo_count; i++) { + drm_gem_object_reference(&exec->bo[i]->base); + kernel_state->bo[i] = &exec->bo[i]->base; + } + + list_for_each_entry(bo, &exec->unref_list, unref_head) { + drm_gem_object_reference(&bo->base.base); + kernel_state->bo[i] = &bo->base.base; + i++; + } + + state->start_bin = exec->ct0ca; + state->start_render = exec->ct1ca; + + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + + state->ct0ca = V3D_READ(V3D_CTNCA(0)); + state->ct0ea = V3D_READ(V3D_CTNEA(0)); + + state->ct1ca = V3D_READ(V3D_CTNCA(1)); + state->ct1ea = V3D_READ(V3D_CTNEA(1)); + + state->ct0cs = V3D_READ(V3D_CTNCS(0)); + state->ct1cs = V3D_READ(V3D_CTNCS(1)); + + state->ct0ra0 = V3D_READ(V3D_CT00RA0); + state->ct1ra0 = V3D_READ(V3D_CT01RA0); + + state->bpca = V3D_READ(V3D_BPCA); + state->bpcs = V3D_READ(V3D_BPCS); + state->bpoa = V3D_READ(V3D_BPOA); + state->bpos = V3D_READ(V3D_BPOS); + + state->vpmbase = V3D_READ(V3D_VPMBASE); + + state->dbge = V3D_READ(V3D_DBGE); + state->fdbgo = V3D_READ(V3D_FDBGO); + state->fdbgb = V3D_READ(V3D_FDBGB); + state->fdbgr = V3D_READ(V3D_FDBGR); + state->fdbgs = V3D_READ(V3D_FDBGS); + state->errstat = V3D_READ(V3D_ERRSTAT); + + spin_lock_irqsave(&vc4->job_lock, irqflags); + if (vc4->hang_state) { + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + vc4_free_hang_state(dev, kernel_state); + } else { + vc4->hang_state = kernel_state; + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + } +} + +static void +vc4_reset(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + DRM_INFO("Resetting GPU.\n"); + vc4_v3d_set_power(vc4, false); + vc4_v3d_set_power(vc4, true); + + vc4_irq_reset(dev); + + /* Rearm the hangcheck -- another job might have been waiting + * for our hung one to get kicked off, and vc4_irq_reset() + * would have started it. + */ + vc4_queue_hangcheck(dev); +} + +static void +vc4_reset_work(struct work_struct *work) +{ + struct vc4_dev *vc4 = + container_of(work, struct vc4_dev, hangcheck.reset_work); + + vc4_save_hang_state(vc4->dev); + + vc4_reset(vc4->dev); +} + +static void +vc4_hangcheck_elapsed(unsigned long data) +{ + struct drm_device *dev = (struct drm_device *)data; + struct vc4_dev *vc4 = to_vc4_dev(dev); + uint32_t ct0ca, ct1ca; + + /* If idle, we can stop watching for hangs. */ + if (list_empty(&vc4->job_list)) + return; + + ct0ca = V3D_READ(V3D_CTNCA(0)); + ct1ca = V3D_READ(V3D_CTNCA(1)); + + /* If we've made any progress in execution, rearm the timer + * and wait. + */ + if (ct0ca != vc4->hangcheck.last_ct0ca || + ct1ca != vc4->hangcheck.last_ct1ca) { + vc4->hangcheck.last_ct0ca = ct0ca; + vc4->hangcheck.last_ct1ca = ct1ca; + vc4_queue_hangcheck(dev); + return; + } + + /* We've gone too long with no progress, reset. This has to + * be done from a work struct, since resetting can sleep and + * this timer hook isn't allowed to. + */ + schedule_work(&vc4->hangcheck.reset_work); +} + +static void +submit_cl(struct drm_device *dev, uint32_t thread, uint32_t start, uint32_t end) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + /* Set the current and end address of the control list. + * Writing the end register is what starts the job. + */ + V3D_WRITE(V3D_CTNCA(thread), start); + V3D_WRITE(V3D_CTNEA(thread), end); +} + +int +vc4_wait_for_seqno(struct drm_device *dev, uint64_t seqno, uint64_t timeout_ns, + bool interruptible) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + int ret = 0; + unsigned long timeout_expire; + DEFINE_WAIT(wait); + + if (vc4->finished_seqno >= seqno) + return 0; + + if (timeout_ns == 0) + return -ETIME; + + timeout_expire = jiffies + nsecs_to_jiffies(timeout_ns); + + trace_vc4_wait_for_seqno_begin(dev, seqno, timeout_ns); + for (;;) { + prepare_to_wait(&vc4->job_wait_queue, &wait, + interruptible ? TASK_INTERRUPTIBLE : + TASK_UNINTERRUPTIBLE); + + if (interruptible && signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + + if (vc4->finished_seqno >= seqno) + break; + + if (timeout_ns != ~0ull) { + if (time_after_eq(jiffies, timeout_expire)) { + ret = -ETIME; + break; + } + schedule_timeout(timeout_expire - jiffies); + } else { + schedule(); + } + } + + finish_wait(&vc4->job_wait_queue, &wait); + trace_vc4_wait_for_seqno_end(dev, seqno); + + if (ret && ret != -ERESTARTSYS) { + DRM_ERROR("timeout waiting for render thread idle\n"); + return ret; + } + + return 0; +} + +static void +vc4_flush_caches(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + /* Flush the GPU L2 caches. These caches sit on top of system + * L3 (the 128kb or so shared with the CPU), and are + * non-allocating in the L3. + */ + V3D_WRITE(V3D_L2CACTL, + V3D_L2CACTL_L2CCLR); + + V3D_WRITE(V3D_SLCACTL, + VC4_SET_FIELD(0xf, V3D_SLCACTL_T1CC) | + VC4_SET_FIELD(0xf, V3D_SLCACTL_T0CC) | + VC4_SET_FIELD(0xf, V3D_SLCACTL_UCC) | + VC4_SET_FIELD(0xf, V3D_SLCACTL_ICC)); +} + +/* Sets the registers for the next job to be actually be executed in + * the hardware. + * + * The job_lock should be held during this. + */ +void +vc4_submit_next_job(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct vc4_exec_info *exec = vc4_first_job(vc4); + + if (!exec) + return; + + vc4_flush_caches(dev); + + /* Disable the binner's pre-loaded overflow memory address */ + V3D_WRITE(V3D_BPOA, 0); + V3D_WRITE(V3D_BPOS, 0); + + if (exec->ct0ca != exec->ct0ea) + submit_cl(dev, 0, exec->ct0ca, exec->ct0ea); + submit_cl(dev, 1, exec->ct1ca, exec->ct1ea); +} + +static void +vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno) +{ + struct vc4_bo *bo; + unsigned i; + + for (i = 0; i < exec->bo_count; i++) { + bo = to_vc4_bo(&exec->bo[i]->base); + bo->seqno = seqno; + } + + list_for_each_entry(bo, &exec->unref_list, unref_head) { + bo->seqno = seqno; + } +} + +/* Queues a struct vc4_exec_info for execution. If no job is + * currently executing, then submits it. + * + * Unlike most GPUs, our hardware only handles one command list at a + * time. To queue multiple jobs at once, we'd need to edit the + * previous command list to have a jump to the new one at the end, and + * then bump the end address. That's a change for a later date, + * though. + */ +static void +vc4_queue_submit(struct drm_device *dev, struct vc4_exec_info *exec) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + uint64_t seqno; + unsigned long irqflags; + + spin_lock_irqsave(&vc4->job_lock, irqflags); + + seqno = ++vc4->emit_seqno; + exec->seqno = seqno; + vc4_update_bo_seqnos(exec, seqno); + + list_add_tail(&exec->head, &vc4->job_list); + + /* If no job was executing, kick ours off. Otherwise, it'll + * get started when the previous job's frame done interrupt + * occurs. + */ + if (vc4_first_job(vc4) == exec) { + vc4_submit_next_job(dev); + vc4_queue_hangcheck(dev); + } + + spin_unlock_irqrestore(&vc4->job_lock, irqflags); +} + +/** + * Looks up a bunch of GEM handles for BOs and stores the array for + * use in the command validator that actually writes relocated + * addresses pointing to them. + */ +static int +vc4_cl_lookup_bos(struct drm_device *dev, + struct drm_file *file_priv, + struct vc4_exec_info *exec) +{ + struct drm_vc4_submit_cl *args = exec->args; + uint32_t *handles; + int ret = 0; + int i; + + exec->bo_count = args->bo_handle_count; + + if (!exec->bo_count) { + /* See comment on bo_index for why we have to check + * this. + */ + DRM_ERROR("Rendering requires BOs to validate\n"); + return -EINVAL; + } + + exec->bo = kcalloc(exec->bo_count, sizeof(struct drm_gem_cma_object *), + GFP_KERNEL); + if (!exec->bo) { + DRM_ERROR("Failed to allocate validated BO pointers\n"); + return -ENOMEM; + } + + handles = drm_malloc_ab(exec->bo_count, sizeof(uint32_t)); + if (!handles) { + DRM_ERROR("Failed to allocate incoming GEM handles\n"); + goto fail; + } + + ret = copy_from_user(handles, + (void __user *)(uintptr_t)args->bo_handles, + exec->bo_count * sizeof(uint32_t)); + if (ret) { + DRM_ERROR("Failed to copy in GEM handles\n"); + goto fail; + } + + spin_lock(&file_priv->table_lock); + for (i = 0; i < exec->bo_count; i++) { + struct drm_gem_object *bo = idr_find(&file_priv->object_idr, + handles[i]); + if (!bo) { + DRM_ERROR("Failed to look up GEM BO %d: %d\n", + i, handles[i]); + ret = -EINVAL; + spin_unlock(&file_priv->table_lock); + goto fail; + } + drm_gem_object_reference(bo); + exec->bo[i] = (struct drm_gem_cma_object *)bo; + } + spin_unlock(&file_priv->table_lock); + +fail: + kfree(handles); + return 0; +} + +static int +vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec) +{ + struct drm_vc4_submit_cl *args = exec->args; + void *temp = NULL; + void *bin; + int ret = 0; + uint32_t bin_offset = 0; + uint32_t shader_rec_offset = roundup(bin_offset + args->bin_cl_size, + 16); + uint32_t uniforms_offset = shader_rec_offset + args->shader_rec_size; + uint32_t exec_size = uniforms_offset + args->uniforms_size; + uint32_t temp_size = exec_size + (sizeof(struct vc4_shader_state) * + args->shader_rec_count); + struct vc4_bo *bo; + + if (uniforms_offset < shader_rec_offset || + exec_size < uniforms_offset || + args->shader_rec_count >= (UINT_MAX / + sizeof(struct vc4_shader_state)) || + temp_size < exec_size) { + DRM_ERROR("overflow in exec arguments\n"); + goto fail; + } + + /* Allocate space where we'll store the copied in user command lists + * and shader records. + * + * We don't just copy directly into the BOs because we need to + * read the contents back for validation, and I think the + * bo->vaddr is uncached access. + */ + temp = kmalloc(temp_size, GFP_KERNEL); + if (!temp) { + DRM_ERROR("Failed to allocate storage for copying " + "in bin/render CLs.\n"); + ret = -ENOMEM; + goto fail; + } + bin = temp + bin_offset; + exec->shader_rec_u = temp + shader_rec_offset; + exec->uniforms_u = temp + uniforms_offset; + exec->shader_state = temp + exec_size; + exec->shader_state_size = args->shader_rec_count; + + ret = copy_from_user(bin, + (void __user *)(uintptr_t)args->bin_cl, + args->bin_cl_size); + if (ret) { + DRM_ERROR("Failed to copy in bin cl\n"); + goto fail; + } + + ret = copy_from_user(exec->shader_rec_u, + (void __user *)(uintptr_t)args->shader_rec, + args->shader_rec_size); + if (ret) { + DRM_ERROR("Failed to copy in shader recs\n"); + goto fail; + } + + ret = copy_from_user(exec->uniforms_u, + (void __user *)(uintptr_t)args->uniforms, + args->uniforms_size); + if (ret) { + DRM_ERROR("Failed to copy in uniforms cl\n"); + goto fail; + } + + bo = vc4_bo_create(dev, exec_size, true); + if (!bo) { + DRM_ERROR("Couldn't allocate BO for binning\n"); + ret = PTR_ERR(exec->exec_bo); + goto fail; + } + exec->exec_bo = &bo->base; + + list_add_tail(&to_vc4_bo(&exec->exec_bo->base)->unref_head, + &exec->unref_list); + + exec->ct0ca = exec->exec_bo->paddr + bin_offset; + + exec->bin_u = bin; + + exec->shader_rec_v = exec->exec_bo->vaddr + shader_rec_offset; + exec->shader_rec_p = exec->exec_bo->paddr + shader_rec_offset; + exec->shader_rec_size = args->shader_rec_size; + + exec->uniforms_v = exec->exec_bo->vaddr + uniforms_offset; + exec->uniforms_p = exec->exec_bo->paddr + uniforms_offset; + exec->uniforms_size = args->uniforms_size; + + ret = vc4_validate_bin_cl(dev, + exec->exec_bo->vaddr + bin_offset, + bin, + exec); + if (ret) + goto fail; + + ret = vc4_validate_shader_recs(dev, exec); + +fail: + kfree(temp); + return ret; +} + +static void +vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec) +{ + unsigned i; + + /* Need the struct lock for drm_gem_object_unreference(). */ + mutex_lock(&dev->struct_mutex); + if (exec->bo) { + for (i = 0; i < exec->bo_count; i++) + drm_gem_object_unreference(&exec->bo[i]->base); + kfree(exec->bo); + } + + while (!list_empty(&exec->unref_list)) { + struct vc4_bo *bo = list_first_entry(&exec->unref_list, + struct vc4_bo, unref_head); + list_del(&bo->unref_head); + drm_gem_object_unreference(&bo->base.base); + } + mutex_unlock(&dev->struct_mutex); + + kfree(exec); +} + +void +vc4_job_handle_completed(struct vc4_dev *vc4) +{ + unsigned long irqflags; + struct vc4_seqno_cb *cb, *cb_temp; + + spin_lock_irqsave(&vc4->job_lock, irqflags); + while (!list_empty(&vc4->job_done_list)) { + struct vc4_exec_info *exec = + list_first_entry(&vc4->job_done_list, + struct vc4_exec_info, head); + list_del(&exec->head); + + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + vc4_complete_exec(vc4->dev, exec); + spin_lock_irqsave(&vc4->job_lock, irqflags); + } + + list_for_each_entry_safe(cb, cb_temp, &vc4->seqno_cb_list, work.entry) { + if (cb->seqno <= vc4->finished_seqno) { + list_del_init(&cb->work.entry); + schedule_work(&cb->work); + } + } + + spin_unlock_irqrestore(&vc4->job_lock, irqflags); +} + +static void vc4_seqno_cb_work(struct work_struct *work) +{ + struct vc4_seqno_cb *cb = container_of(work, struct vc4_seqno_cb, work); + + cb->func(cb); +} + +int vc4_queue_seqno_cb(struct drm_device *dev, + struct vc4_seqno_cb *cb, uint64_t seqno, + void (*func)(struct vc4_seqno_cb *cb)) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + int ret = 0; + unsigned long irqflags; + + cb->func = func; + INIT_WORK(&cb->work, vc4_seqno_cb_work); + + spin_lock_irqsave(&vc4->job_lock, irqflags); + if (seqno > vc4->finished_seqno) { + cb->seqno = seqno; + list_add_tail(&cb->work.entry, &vc4->seqno_cb_list); + } else { + schedule_work(&cb->work); + } + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + + return ret; +} + +/* Scheduled when any job has been completed, this walks the list of + * jobs that had completed and unrefs their BOs and frees their exec + * structs. + */ +static void +vc4_job_done_work(struct work_struct *work) +{ + struct vc4_dev *vc4 = + container_of(work, struct vc4_dev, job_done_work); + + vc4_job_handle_completed(vc4); +} + +static int +vc4_wait_for_seqno_ioctl_helper(struct drm_device *dev, + uint64_t seqno, + uint64_t *timeout_ns) +{ + unsigned long start = jiffies; + int ret = vc4_wait_for_seqno(dev, seqno, *timeout_ns, true); + + if ((ret == -EINTR || ret == -ERESTARTSYS) && *timeout_ns != ~0ull) { + uint64_t delta = jiffies_to_nsecs(jiffies - start); + + if (*timeout_ns >= delta) + *timeout_ns -= delta; + } + + return ret; +} + +int +vc4_wait_seqno_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_vc4_wait_seqno *args = data; + + return vc4_wait_for_seqno_ioctl_helper(dev, args->seqno, + &args->timeout_ns); +} + +int +vc4_wait_bo_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + int ret; + struct drm_vc4_wait_bo *args = data; + struct drm_gem_object *gem_obj; + struct vc4_bo *bo; + + gem_obj = drm_gem_object_lookup(dev, file_priv, args->handle); + if (!gem_obj) { + DRM_ERROR("Failed to look up GEM BO %d\n", args->handle); + return -EINVAL; + } + bo = to_vc4_bo(gem_obj); + + ret = vc4_wait_for_seqno_ioctl_helper(dev, bo->seqno, + &args->timeout_ns); + + drm_gem_object_unreference_unlocked(gem_obj); + return ret; +} + +/** + * Submits a command list to the VC4. + * + * This is what is called batchbuffer emitting on other hardware. + */ +int +vc4_submit_cl_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct drm_vc4_submit_cl *args = data; + struct vc4_exec_info *exec; + int ret; + + if ((args->flags & ~VC4_SUBMIT_CL_USE_CLEAR_COLOR) != 0) { + DRM_ERROR("Unknown flags: 0x%02x\n", args->flags); + return -EINVAL; + } + + exec = kcalloc(1, sizeof(*exec), GFP_KERNEL); + if (!exec) { + DRM_ERROR("malloc failure on exec struct\n"); + return -ENOMEM; + } + + exec->args = args; + INIT_LIST_HEAD(&exec->unref_list); + + ret = vc4_cl_lookup_bos(dev, file_priv, exec); + if (ret) + goto fail; + + if (exec->args->bin_cl_size != 0) { + ret = vc4_get_bcl(dev, exec); + if (ret) + goto fail; + } else { + exec->ct0ca = 0; + exec->ct0ea = 0; + } + + ret = vc4_get_rcl(dev, exec); + if (ret) + goto fail; + + /* Clear this out of the struct we'll be putting in the queue, + * since it's part of our stack. + */ + exec->args = NULL; + + vc4_queue_submit(dev, exec); + + /* Return the seqno for our job. */ + args->seqno = vc4->emit_seqno; + + return 0; + +fail: + vc4_complete_exec(vc4->dev, exec); + + return ret; +} + +void +vc4_gem_init(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + INIT_LIST_HEAD(&vc4->job_list); + INIT_LIST_HEAD(&vc4->job_done_list); + INIT_LIST_HEAD(&vc4->seqno_cb_list); + spin_lock_init(&vc4->job_lock); + + INIT_WORK(&vc4->hangcheck.reset_work, vc4_reset_work); + setup_timer(&vc4->hangcheck.timer, + vc4_hangcheck_elapsed, + (unsigned long)dev); + + INIT_WORK(&vc4->job_done_work, vc4_job_done_work); +} + +void +vc4_gem_destroy(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + /* Waiting for exec to finish would need to be done before + * unregistering V3D. + */ + WARN_ON(vc4->emit_seqno != vc4->finished_seqno); + + /* V3D should already have disabled its interrupt and cleared + * the overflow allocation registers. Now free the object. + */ + if (vc4->overflow_mem) { + drm_gem_object_unreference_unlocked(&vc4->overflow_mem->base.base); + vc4->overflow_mem = NULL; + } + + vc4_bo_cache_destroy(dev); + + if (vc4->hang_state) + vc4_free_hang_state(dev, vc4->hang_state); +} diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index da9a36d6e1d1..c69c0460196b 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -519,7 +519,7 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data) WARN_ON_ONCE((HD_READ(VC4_HD_M_CTL) & VC4_HD_M_ENABLE) == 0); drm_encoder_init(drm, hdmi->encoder, &vc4_hdmi_encoder_funcs, - DRM_MODE_ENCODER_TMDS); + DRM_MODE_ENCODER_TMDS, NULL); drm_encoder_helper_add(hdmi->encoder, &vc4_hdmi_encoder_helper_funcs); hdmi->connector = vc4_hdmi_connector_init(drm, hdmi->encoder); diff --git a/drivers/gpu/drm/vc4/vc4_irq.c b/drivers/gpu/drm/vc4/vc4_irq.c new file mode 100644 index 000000000000..b68060e758db --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_irq.c @@ -0,0 +1,210 @@ +/* + * Copyright © 2014 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +/** DOC: Interrupt management for the V3D engine. + * + * We have an interrupt status register (V3D_INTCTL) which reports + * interrupts, and where writing 1 bits clears those interrupts. + * There are also a pair of interrupt registers + * (V3D_INTENA/V3D_INTDIS) where writing a 1 to their bits enables or + * disables that specific interrupt, and 0s written are ignored + * (reading either one returns the set of enabled interrupts). + * + * When we take a render frame interrupt, we need to wake the + * processes waiting for some frame to be done, and get the next frame + * submitted ASAP (so the hardware doesn't sit idle when there's work + * to do). + * + * When we take the binner out of memory interrupt, we need to + * allocate some new memory and pass it to the binner so that the + * current job can make progress. + */ + +#include "vc4_drv.h" +#include "vc4_regs.h" + +#define V3D_DRIVER_IRQS (V3D_INT_OUTOMEM | \ + V3D_INT_FRDONE) + +DECLARE_WAIT_QUEUE_HEAD(render_wait); + +static void +vc4_overflow_mem_work(struct work_struct *work) +{ + struct vc4_dev *vc4 = + container_of(work, struct vc4_dev, overflow_mem_work); + struct drm_device *dev = vc4->dev; + struct vc4_bo *bo; + + bo = vc4_bo_create(dev, 256 * 1024, true); + if (!bo) { + DRM_ERROR("Couldn't allocate binner overflow mem\n"); + return; + } + + /* If there's a job executing currently, then our previous + * overflow allocation is getting used in that job and we need + * to queue it to be released when the job is done. But if no + * job is executing at all, then we can free the old overflow + * object direcctly. + * + * No lock necessary for this pointer since we're the only + * ones that update the pointer, and our workqueue won't + * reenter. + */ + if (vc4->overflow_mem) { + struct vc4_exec_info *current_exec; + unsigned long irqflags; + + spin_lock_irqsave(&vc4->job_lock, irqflags); + current_exec = vc4_first_job(vc4); + if (current_exec) { + vc4->overflow_mem->seqno = vc4->finished_seqno + 1; + list_add_tail(&vc4->overflow_mem->unref_head, + ¤t_exec->unref_list); + vc4->overflow_mem = NULL; + } + spin_unlock_irqrestore(&vc4->job_lock, irqflags); + } + + if (vc4->overflow_mem) + drm_gem_object_unreference_unlocked(&vc4->overflow_mem->base.base); + vc4->overflow_mem = bo; + + V3D_WRITE(V3D_BPOA, bo->base.paddr); + V3D_WRITE(V3D_BPOS, bo->base.base.size); + V3D_WRITE(V3D_INTCTL, V3D_INT_OUTOMEM); + V3D_WRITE(V3D_INTENA, V3D_INT_OUTOMEM); +} + +static void +vc4_irq_finish_job(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct vc4_exec_info *exec = vc4_first_job(vc4); + + if (!exec) + return; + + vc4->finished_seqno++; + list_move_tail(&exec->head, &vc4->job_done_list); + vc4_submit_next_job(dev); + + wake_up_all(&vc4->job_wait_queue); + schedule_work(&vc4->job_done_work); +} + +irqreturn_t +vc4_irq(int irq, void *arg) +{ + struct drm_device *dev = arg; + struct vc4_dev *vc4 = to_vc4_dev(dev); + uint32_t intctl; + irqreturn_t status = IRQ_NONE; + + barrier(); + intctl = V3D_READ(V3D_INTCTL); + + /* Acknowledge the interrupts we're handling here. The render + * frame done interrupt will be cleared, while OUTOMEM will + * stay high until the underlying cause is cleared. + */ + V3D_WRITE(V3D_INTCTL, intctl); + + if (intctl & V3D_INT_OUTOMEM) { + /* Disable OUTOMEM until the work is done. */ + V3D_WRITE(V3D_INTDIS, V3D_INT_OUTOMEM); + schedule_work(&vc4->overflow_mem_work); + status = IRQ_HANDLED; + } + + if (intctl & V3D_INT_FRDONE) { + spin_lock(&vc4->job_lock); + vc4_irq_finish_job(dev); + spin_unlock(&vc4->job_lock); + status = IRQ_HANDLED; + } + + return status; +} + +void +vc4_irq_preinstall(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + init_waitqueue_head(&vc4->job_wait_queue); + INIT_WORK(&vc4->overflow_mem_work, vc4_overflow_mem_work); + + /* Clear any pending interrupts someone might have left around + * for us. + */ + V3D_WRITE(V3D_INTCTL, V3D_DRIVER_IRQS); +} + +int +vc4_irq_postinstall(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + /* Enable both the render done and out of memory interrupts. */ + V3D_WRITE(V3D_INTENA, V3D_DRIVER_IRQS); + + return 0; +} + +void +vc4_irq_uninstall(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + /* Disable sending interrupts for our driver's IRQs. */ + V3D_WRITE(V3D_INTDIS, V3D_DRIVER_IRQS); + + /* Clear any pending interrupts we might have left. */ + V3D_WRITE(V3D_INTCTL, V3D_DRIVER_IRQS); + + cancel_work_sync(&vc4->overflow_mem_work); +} + +/** Reinitializes interrupt registers when a GPU reset is performed. */ +void vc4_irq_reset(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + unsigned long irqflags; + + /* Acknowledge any stale IRQs. */ + V3D_WRITE(V3D_INTCTL, V3D_DRIVER_IRQS); + + /* + * Turn all our interrupts on. Binner out of memory is the + * only one we expect to trigger at this point, since we've + * just come from poweron and haven't supplied any overflow + * memory yet. + */ + V3D_WRITE(V3D_INTENA, V3D_DRIVER_IRQS); + + spin_lock_irqsave(&vc4->job_lock, irqflags); + vc4_irq_finish_job(dev); + spin_unlock_irqrestore(&vc4->job_lock, irqflags); +} diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c index 2e5597d10cc6..f95f2df5f8d1 100644 --- a/drivers/gpu/drm/vc4/vc4_kms.c +++ b/drivers/gpu/drm/vc4/vc4_kms.c @@ -15,6 +15,7 @@ */ #include "drm_crtc.h" +#include "drm_atomic.h" #include "drm_atomic_helper.h" #include "drm_crtc_helper.h" #include "drm_plane_helper.h" @@ -29,10 +30,152 @@ static void vc4_output_poll_changed(struct drm_device *dev) drm_fbdev_cma_hotplug_event(vc4->fbdev); } +struct vc4_commit { + struct drm_device *dev; + struct drm_atomic_state *state; + struct vc4_seqno_cb cb; +}; + +static void +vc4_atomic_complete_commit(struct vc4_commit *c) +{ + struct drm_atomic_state *state = c->state; + struct drm_device *dev = state->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + + drm_atomic_helper_commit_modeset_disables(dev, state); + + drm_atomic_helper_commit_planes(dev, state, false); + + drm_atomic_helper_commit_modeset_enables(dev, state); + + drm_atomic_helper_wait_for_vblanks(dev, state); + + drm_atomic_helper_cleanup_planes(dev, state); + + drm_atomic_state_free(state); + + up(&vc4->async_modeset); + + kfree(c); +} + +static void +vc4_atomic_complete_commit_seqno_cb(struct vc4_seqno_cb *cb) +{ + struct vc4_commit *c = container_of(cb, struct vc4_commit, cb); + + vc4_atomic_complete_commit(c); +} + +static struct vc4_commit *commit_init(struct drm_atomic_state *state) +{ + struct vc4_commit *c = kzalloc(sizeof(*c), GFP_KERNEL); + + if (!c) + return NULL; + c->dev = state->dev; + c->state = state; + + return c; +} + +/** + * vc4_atomic_commit - commit validated state object + * @dev: DRM device + * @state: the driver state object + * @async: asynchronous commit + * + * This function commits a with drm_atomic_helper_check() pre-validated state + * object. This can still fail when e.g. the framebuffer reservation fails. For + * now this doesn't implement asynchronous commits. + * + * RETURNS + * Zero for success or -errno. + */ +static int vc4_atomic_commit(struct drm_device *dev, + struct drm_atomic_state *state, + bool async) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + int ret; + int i; + uint64_t wait_seqno = 0; + struct vc4_commit *c; + + c = commit_init(state); + if (!c) + return -ENOMEM; + + /* Make sure that any outstanding modesets have finished. */ + ret = down_interruptible(&vc4->async_modeset); + if (ret) { + kfree(c); + return ret; + } + + ret = drm_atomic_helper_prepare_planes(dev, state); + if (ret) { + kfree(c); + up(&vc4->async_modeset); + return ret; + } + + for (i = 0; i < dev->mode_config.num_total_plane; i++) { + struct drm_plane *plane = state->planes[i]; + struct drm_plane_state *new_state = state->plane_states[i]; + + if (!plane) + continue; + + if ((plane->state->fb != new_state->fb) && new_state->fb) { + struct drm_gem_cma_object *cma_bo = + drm_fb_cma_get_gem_obj(new_state->fb, 0); + struct vc4_bo *bo = to_vc4_bo(&cma_bo->base); + + wait_seqno = max(bo->seqno, wait_seqno); + } + } + + /* + * This is the point of no return - everything below never fails except + * when the hw goes bonghits. Which means we can commit the new state on + * the software side now. + */ + + drm_atomic_helper_swap_state(dev, state); + + /* + * Everything below can be run asynchronously without the need to grab + * any modeset locks at all under one condition: It must be guaranteed + * that the asynchronous work has either been cancelled (if the driver + * supports it, which at least requires that the framebuffers get + * cleaned up with drm_atomic_helper_cleanup_planes()) or completed + * before the new state gets committed on the software side with + * drm_atomic_helper_swap_state(). + * + * This scheme allows new atomic state updates to be prepared and + * checked in parallel to the asynchronous completion of the previous + * update. Which is important since compositors need to figure out the + * composition of the next frame right after having submitted the + * current layout. + */ + + if (async) { + vc4_queue_seqno_cb(dev, &c->cb, wait_seqno, + vc4_atomic_complete_commit_seqno_cb); + } else { + vc4_wait_for_seqno(dev, wait_seqno, ~0ull, false); + vc4_atomic_complete_commit(c); + } + + return 0; +} + static const struct drm_mode_config_funcs vc4_mode_funcs = { .output_poll_changed = vc4_output_poll_changed, .atomic_check = drm_atomic_helper_check, - .atomic_commit = drm_atomic_helper_commit, + .atomic_commit = vc4_atomic_commit, .fb_create = drm_fb_cma_create, }; @@ -41,6 +184,8 @@ int vc4_kms_load(struct drm_device *dev) struct vc4_dev *vc4 = to_vc4_dev(dev); int ret; + sema_init(&vc4->async_modeset, 1); + ret = drm_vblank_init(dev, dev->mode_config.num_crtc); if (ret < 0) { dev_err(dev->dev, "failed to initialize vblank\n"); @@ -51,6 +196,8 @@ int vc4_kms_load(struct drm_device *dev) dev->mode_config.max_height = 2048; dev->mode_config.funcs = &vc4_mode_funcs; dev->mode_config.preferred_depth = 24; + dev->mode_config.async_page_flip = true; + dev->vblank_disable_allowed = true; drm_mode_config_reset(dev); diff --git a/drivers/gpu/drm/vc4/vc4_packet.h b/drivers/gpu/drm/vc4/vc4_packet.h new file mode 100644 index 000000000000..0f31cc06500f --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_packet.h @@ -0,0 +1,399 @@ +/* + * Copyright © 2014 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef VC4_PACKET_H +#define VC4_PACKET_H + +#include "vc4_regs.h" /* for VC4_MASK, VC4_GET_FIELD, VC4_SET_FIELD */ + +enum vc4_packet { + VC4_PACKET_HALT = 0, + VC4_PACKET_NOP = 1, + + VC4_PACKET_FLUSH = 4, + VC4_PACKET_FLUSH_ALL = 5, + VC4_PACKET_START_TILE_BINNING = 6, + VC4_PACKET_INCREMENT_SEMAPHORE = 7, + VC4_PACKET_WAIT_ON_SEMAPHORE = 8, + + VC4_PACKET_BRANCH = 16, + VC4_PACKET_BRANCH_TO_SUB_LIST = 17, + + VC4_PACKET_STORE_MS_TILE_BUFFER = 24, + VC4_PACKET_STORE_MS_TILE_BUFFER_AND_EOF = 25, + VC4_PACKET_STORE_FULL_RES_TILE_BUFFER = 26, + VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER = 27, + VC4_PACKET_STORE_TILE_BUFFER_GENERAL = 28, + VC4_PACKET_LOAD_TILE_BUFFER_GENERAL = 29, + + VC4_PACKET_GL_INDEXED_PRIMITIVE = 32, + VC4_PACKET_GL_ARRAY_PRIMITIVE = 33, + + VC4_PACKET_COMPRESSED_PRIMITIVE = 48, + VC4_PACKET_CLIPPED_COMPRESSED_PRIMITIVE = 49, + + VC4_PACKET_PRIMITIVE_LIST_FORMAT = 56, + + VC4_PACKET_GL_SHADER_STATE = 64, + VC4_PACKET_NV_SHADER_STATE = 65, + VC4_PACKET_VG_SHADER_STATE = 66, + + VC4_PACKET_CONFIGURATION_BITS = 96, + VC4_PACKET_FLAT_SHADE_FLAGS = 97, + VC4_PACKET_POINT_SIZE = 98, + VC4_PACKET_LINE_WIDTH = 99, + VC4_PACKET_RHT_X_BOUNDARY = 100, + VC4_PACKET_DEPTH_OFFSET = 101, + VC4_PACKET_CLIP_WINDOW = 102, + VC4_PACKET_VIEWPORT_OFFSET = 103, + VC4_PACKET_Z_CLIPPING = 104, + VC4_PACKET_CLIPPER_XY_SCALING = 105, + VC4_PACKET_CLIPPER_Z_SCALING = 106, + + VC4_PACKET_TILE_BINNING_MODE_CONFIG = 112, + VC4_PACKET_TILE_RENDERING_MODE_CONFIG = 113, + VC4_PACKET_CLEAR_COLORS = 114, + VC4_PACKET_TILE_COORDINATES = 115, + + /* Not an actual hardware packet -- this is what we use to put + * references to GEM bos in the command stream, since we need the u32 + * int the actual address packet in order to store the offset from the + * start of the BO. + */ + VC4_PACKET_GEM_HANDLES = 254, +} __attribute__ ((__packed__)); + +#define VC4_PACKET_HALT_SIZE 1 +#define VC4_PACKET_NOP_SIZE 1 +#define VC4_PACKET_FLUSH_SIZE 1 +#define VC4_PACKET_FLUSH_ALL_SIZE 1 +#define VC4_PACKET_START_TILE_BINNING_SIZE 1 +#define VC4_PACKET_INCREMENT_SEMAPHORE_SIZE 1 +#define VC4_PACKET_WAIT_ON_SEMAPHORE_SIZE 1 +#define VC4_PACKET_BRANCH_SIZE 5 +#define VC4_PACKET_BRANCH_TO_SUB_LIST_SIZE 5 +#define VC4_PACKET_STORE_MS_TILE_BUFFER_SIZE 1 +#define VC4_PACKET_STORE_MS_TILE_BUFFER_AND_EOF_SIZE 1 +#define VC4_PACKET_STORE_FULL_RES_TILE_BUFFER_SIZE 5 +#define VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER_SIZE 5 +#define VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE 7 +#define VC4_PACKET_LOAD_TILE_BUFFER_GENERAL_SIZE 7 +#define VC4_PACKET_GL_INDEXED_PRIMITIVE_SIZE 14 +#define VC4_PACKET_GL_ARRAY_PRIMITIVE_SIZE 10 +#define VC4_PACKET_COMPRESSED_PRIMITIVE_SIZE 1 +#define VC4_PACKET_CLIPPED_COMPRESSED_PRIMITIVE_SIZE 1 +#define VC4_PACKET_PRIMITIVE_LIST_FORMAT_SIZE 2 +#define VC4_PACKET_GL_SHADER_STATE_SIZE 5 +#define VC4_PACKET_NV_SHADER_STATE_SIZE 5 +#define VC4_PACKET_VG_SHADER_STATE_SIZE 5 +#define VC4_PACKET_CONFIGURATION_BITS_SIZE 4 +#define VC4_PACKET_FLAT_SHADE_FLAGS_SIZE 5 +#define VC4_PACKET_POINT_SIZE_SIZE 5 +#define VC4_PACKET_LINE_WIDTH_SIZE 5 +#define VC4_PACKET_RHT_X_BOUNDARY_SIZE 3 +#define VC4_PACKET_DEPTH_OFFSET_SIZE 5 +#define VC4_PACKET_CLIP_WINDOW_SIZE 9 +#define VC4_PACKET_VIEWPORT_OFFSET_SIZE 5 +#define VC4_PACKET_Z_CLIPPING_SIZE 9 +#define VC4_PACKET_CLIPPER_XY_SCALING_SIZE 9 +#define VC4_PACKET_CLIPPER_Z_SCALING_SIZE 9 +#define VC4_PACKET_TILE_BINNING_MODE_CONFIG_SIZE 16 +#define VC4_PACKET_TILE_RENDERING_MODE_CONFIG_SIZE 11 +#define VC4_PACKET_CLEAR_COLORS_SIZE 14 +#define VC4_PACKET_TILE_COORDINATES_SIZE 3 +#define VC4_PACKET_GEM_HANDLES_SIZE 9 + +/* Number of multisamples supported. */ +#define VC4_MAX_SAMPLES 4 +/* Size of a full resolution color or Z tile buffer load/store. */ +#define VC4_TILE_BUFFER_SIZE (64 * 64 * 4) + +/** @{ + * Bits used by packets like VC4_PACKET_STORE_TILE_BUFFER_GENERAL and + * VC4_PACKET_TILE_RENDERING_MODE_CONFIG. +*/ +#define VC4_TILING_FORMAT_LINEAR 0 +#define VC4_TILING_FORMAT_T 1 +#define VC4_TILING_FORMAT_LT 2 +/** @} */ + +/** @{ + * + * low bits of VC4_PACKET_STORE_FULL_RES_TILE_BUFFER and + * VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER. + */ +#define VC4_LOADSTORE_FULL_RES_EOF BIT(3) +#define VC4_LOADSTORE_FULL_RES_DISABLE_CLEAR_ALL BIT(2) +#define VC4_LOADSTORE_FULL_RES_DISABLE_ZS BIT(1) +#define VC4_LOADSTORE_FULL_RES_DISABLE_COLOR BIT(0) + +/** @{ + * + * low bits of VC4_PACKET_STORE_FULL_RES_TILE_BUFFER and + * VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER. + */ +#define VC4_LOADSTORE_FULL_RES_EOF BIT(3) +#define VC4_LOADSTORE_FULL_RES_DISABLE_CLEAR_ALL BIT(2) +#define VC4_LOADSTORE_FULL_RES_DISABLE_ZS BIT(1) +#define VC4_LOADSTORE_FULL_RES_DISABLE_COLOR BIT(0) + +/** @{ + * + * byte 2 of VC4_PACKET_STORE_TILE_BUFFER_GENERAL and + * VC4_PACKET_LOAD_TILE_BUFFER_GENERAL (low bits of the address) + */ + +#define VC4_LOADSTORE_TILE_BUFFER_EOF BIT(3) +#define VC4_LOADSTORE_TILE_BUFFER_DISABLE_FULL_VG_MASK BIT(2) +#define VC4_LOADSTORE_TILE_BUFFER_DISABLE_FULL_ZS BIT(1) +#define VC4_LOADSTORE_TILE_BUFFER_DISABLE_FULL_COLOR BIT(0) + +/** @} */ + +/** @{ + * + * byte 0-1 of VC4_PACKET_STORE_TILE_BUFFER_GENERAL and + * VC4_PACKET_LOAD_TILE_BUFFER_GENERAL + */ +#define VC4_STORE_TILE_BUFFER_DISABLE_VG_MASK_CLEAR BIT(15) +#define VC4_STORE_TILE_BUFFER_DISABLE_ZS_CLEAR BIT(14) +#define VC4_STORE_TILE_BUFFER_DISABLE_COLOR_CLEAR BIT(13) +#define VC4_STORE_TILE_BUFFER_DISABLE_SWAP BIT(12) + +#define VC4_LOADSTORE_TILE_BUFFER_FORMAT_MASK VC4_MASK(9, 8) +#define VC4_LOADSTORE_TILE_BUFFER_FORMAT_SHIFT 8 +#define VC4_LOADSTORE_TILE_BUFFER_RGBA8888 0 +#define VC4_LOADSTORE_TILE_BUFFER_BGR565_DITHER 1 +#define VC4_LOADSTORE_TILE_BUFFER_BGR565 2 +/** @} */ + +/** @{ + * + * byte 0 of VC4_PACKET_STORE_TILE_BUFFER_GENERAL and + * VC4_PACKET_LOAD_TILE_BUFFER_GENERAL + */ +#define VC4_STORE_TILE_BUFFER_MODE_MASK VC4_MASK(7, 6) +#define VC4_STORE_TILE_BUFFER_MODE_SHIFT 6 +#define VC4_STORE_TILE_BUFFER_MODE_SAMPLE0 (0 << 6) +#define VC4_STORE_TILE_BUFFER_MODE_DECIMATE_X4 (1 << 6) +#define VC4_STORE_TILE_BUFFER_MODE_DECIMATE_X16 (2 << 6) + +/** The values of the field are VC4_TILING_FORMAT_* */ +#define VC4_LOADSTORE_TILE_BUFFER_TILING_MASK VC4_MASK(5, 4) +#define VC4_LOADSTORE_TILE_BUFFER_TILING_SHIFT 4 + +#define VC4_LOADSTORE_TILE_BUFFER_BUFFER_MASK VC4_MASK(2, 0) +#define VC4_LOADSTORE_TILE_BUFFER_BUFFER_SHIFT 0 +#define VC4_LOADSTORE_TILE_BUFFER_NONE 0 +#define VC4_LOADSTORE_TILE_BUFFER_COLOR 1 +#define VC4_LOADSTORE_TILE_BUFFER_ZS 2 +#define VC4_LOADSTORE_TILE_BUFFER_Z 3 +#define VC4_LOADSTORE_TILE_BUFFER_VG_MASK 4 +#define VC4_LOADSTORE_TILE_BUFFER_FULL 5 +/** @} */ + +#define VC4_INDEX_BUFFER_U8 (0 << 4) +#define VC4_INDEX_BUFFER_U16 (1 << 4) + +/* This flag is only present in NV shader state. */ +#define VC4_SHADER_FLAG_SHADED_CLIP_COORDS BIT(3) +#define VC4_SHADER_FLAG_ENABLE_CLIPPING BIT(2) +#define VC4_SHADER_FLAG_VS_POINT_SIZE BIT(1) +#define VC4_SHADER_FLAG_FS_SINGLE_THREAD BIT(0) + +/** @{ byte 2 of config bits. */ +#define VC4_CONFIG_BITS_EARLY_Z_UPDATE BIT(1) +#define VC4_CONFIG_BITS_EARLY_Z BIT(0) +/** @} */ + +/** @{ byte 1 of config bits. */ +#define VC4_CONFIG_BITS_Z_UPDATE BIT(7) +/** same values in this 3-bit field as PIPE_FUNC_* */ +#define VC4_CONFIG_BITS_DEPTH_FUNC_SHIFT 4 +#define VC4_CONFIG_BITS_COVERAGE_READ_LEAVE BIT(3) + +#define VC4_CONFIG_BITS_COVERAGE_UPDATE_NONZERO (0 << 1) +#define VC4_CONFIG_BITS_COVERAGE_UPDATE_ODD (1 << 1) +#define VC4_CONFIG_BITS_COVERAGE_UPDATE_OR (2 << 1) +#define VC4_CONFIG_BITS_COVERAGE_UPDATE_ZERO (3 << 1) + +#define VC4_CONFIG_BITS_COVERAGE_PIPE_SELECT BIT(0) +/** @} */ + +/** @{ byte 0 of config bits. */ +#define VC4_CONFIG_BITS_RASTERIZER_OVERSAMPLE_NONE (0 << 6) +#define VC4_CONFIG_BITS_RASTERIZER_OVERSAMPLE_4X (1 << 6) +#define VC4_CONFIG_BITS_RASTERIZER_OVERSAMPLE_16X (2 << 6) + +#define VC4_CONFIG_BITS_AA_POINTS_AND_LINES BIT(4) +#define VC4_CONFIG_BITS_ENABLE_DEPTH_OFFSET BIT(3) +#define VC4_CONFIG_BITS_CW_PRIMITIVES BIT(2) +#define VC4_CONFIG_BITS_ENABLE_PRIM_BACK BIT(1) +#define VC4_CONFIG_BITS_ENABLE_PRIM_FRONT BIT(0) +/** @} */ + +/** @{ bits in the last u8 of VC4_PACKET_TILE_BINNING_MODE_CONFIG */ +#define VC4_BIN_CONFIG_DB_NON_MS BIT(7) + +#define VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_MASK VC4_MASK(6, 5) +#define VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_SHIFT 5 +#define VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_32 0 +#define VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_64 1 +#define VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_128 2 +#define VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_256 3 + +#define VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_MASK VC4_MASK(4, 3) +#define VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_SHIFT 3 +#define VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_32 0 +#define VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_64 1 +#define VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_128 2 +#define VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_256 3 + +#define VC4_BIN_CONFIG_AUTO_INIT_TSDA BIT(2) +#define VC4_BIN_CONFIG_TILE_BUFFER_64BIT BIT(1) +#define VC4_BIN_CONFIG_MS_MODE_4X BIT(0) +/** @} */ + +/** @{ bits in the last u16 of VC4_PACKET_TILE_RENDERING_MODE_CONFIG */ +#define VC4_RENDER_CONFIG_DB_NON_MS BIT(12) +#define VC4_RENDER_CONFIG_EARLY_Z_COVERAGE_DISABLE BIT(11) +#define VC4_RENDER_CONFIG_EARLY_Z_DIRECTION_G BIT(10) +#define VC4_RENDER_CONFIG_COVERAGE_MODE BIT(9) +#define VC4_RENDER_CONFIG_ENABLE_VG_MASK BIT(8) + +/** The values of the field are VC4_TILING_FORMAT_* */ +#define VC4_RENDER_CONFIG_MEMORY_FORMAT_MASK VC4_MASK(7, 6) +#define VC4_RENDER_CONFIG_MEMORY_FORMAT_SHIFT 6 + +#define VC4_RENDER_CONFIG_DECIMATE_MODE_1X (0 << 4) +#define VC4_RENDER_CONFIG_DECIMATE_MODE_4X (1 << 4) +#define VC4_RENDER_CONFIG_DECIMATE_MODE_16X (2 << 4) + +#define VC4_RENDER_CONFIG_FORMAT_MASK VC4_MASK(3, 2) +#define VC4_RENDER_CONFIG_FORMAT_SHIFT 2 +#define VC4_RENDER_CONFIG_FORMAT_BGR565_DITHERED 0 +#define VC4_RENDER_CONFIG_FORMAT_RGBA8888 1 +#define VC4_RENDER_CONFIG_FORMAT_BGR565 2 + +#define VC4_RENDER_CONFIG_TILE_BUFFER_64BIT BIT(1) +#define VC4_RENDER_CONFIG_MS_MODE_4X BIT(0) + +#define VC4_PRIMITIVE_LIST_FORMAT_16_INDEX (1 << 4) +#define VC4_PRIMITIVE_LIST_FORMAT_32_XY (3 << 4) +#define VC4_PRIMITIVE_LIST_FORMAT_TYPE_POINTS (0 << 0) +#define VC4_PRIMITIVE_LIST_FORMAT_TYPE_LINES (1 << 0) +#define VC4_PRIMITIVE_LIST_FORMAT_TYPE_TRIANGLES (2 << 0) +#define VC4_PRIMITIVE_LIST_FORMAT_TYPE_RHT (3 << 0) + +enum vc4_texture_data_type { + VC4_TEXTURE_TYPE_RGBA8888 = 0, + VC4_TEXTURE_TYPE_RGBX8888 = 1, + VC4_TEXTURE_TYPE_RGBA4444 = 2, + VC4_TEXTURE_TYPE_RGBA5551 = 3, + VC4_TEXTURE_TYPE_RGB565 = 4, + VC4_TEXTURE_TYPE_LUMINANCE = 5, + VC4_TEXTURE_TYPE_ALPHA = 6, + VC4_TEXTURE_TYPE_LUMALPHA = 7, + VC4_TEXTURE_TYPE_ETC1 = 8, + VC4_TEXTURE_TYPE_S16F = 9, + VC4_TEXTURE_TYPE_S8 = 10, + VC4_TEXTURE_TYPE_S16 = 11, + VC4_TEXTURE_TYPE_BW1 = 12, + VC4_TEXTURE_TYPE_A4 = 13, + VC4_TEXTURE_TYPE_A1 = 14, + VC4_TEXTURE_TYPE_RGBA64 = 15, + VC4_TEXTURE_TYPE_RGBA32R = 16, + VC4_TEXTURE_TYPE_YUV422R = 17, +}; + +#define VC4_TEX_P0_OFFSET_MASK VC4_MASK(31, 12) +#define VC4_TEX_P0_OFFSET_SHIFT 12 +#define VC4_TEX_P0_CSWIZ_MASK VC4_MASK(11, 10) +#define VC4_TEX_P0_CSWIZ_SHIFT 10 +#define VC4_TEX_P0_CMMODE_MASK VC4_MASK(9, 9) +#define VC4_TEX_P0_CMMODE_SHIFT 9 +#define VC4_TEX_P0_FLIPY_MASK VC4_MASK(8, 8) +#define VC4_TEX_P0_FLIPY_SHIFT 8 +#define VC4_TEX_P0_TYPE_MASK VC4_MASK(7, 4) +#define VC4_TEX_P0_TYPE_SHIFT 4 +#define VC4_TEX_P0_MIPLVLS_MASK VC4_MASK(3, 0) +#define VC4_TEX_P0_MIPLVLS_SHIFT 0 + +#define VC4_TEX_P1_TYPE4_MASK VC4_MASK(31, 31) +#define VC4_TEX_P1_TYPE4_SHIFT 31 +#define VC4_TEX_P1_HEIGHT_MASK VC4_MASK(30, 20) +#define VC4_TEX_P1_HEIGHT_SHIFT 20 +#define VC4_TEX_P1_ETCFLIP_MASK VC4_MASK(19, 19) +#define VC4_TEX_P1_ETCFLIP_SHIFT 19 +#define VC4_TEX_P1_WIDTH_MASK VC4_MASK(18, 8) +#define VC4_TEX_P1_WIDTH_SHIFT 8 + +#define VC4_TEX_P1_MAGFILT_MASK VC4_MASK(7, 7) +#define VC4_TEX_P1_MAGFILT_SHIFT 7 +# define VC4_TEX_P1_MAGFILT_LINEAR 0 +# define VC4_TEX_P1_MAGFILT_NEAREST 1 + +#define VC4_TEX_P1_MINFILT_MASK VC4_MASK(6, 4) +#define VC4_TEX_P1_MINFILT_SHIFT 4 +# define VC4_TEX_P1_MINFILT_LINEAR 0 +# define VC4_TEX_P1_MINFILT_NEAREST 1 +# define VC4_TEX_P1_MINFILT_NEAR_MIP_NEAR 2 +# define VC4_TEX_P1_MINFILT_NEAR_MIP_LIN 3 +# define VC4_TEX_P1_MINFILT_LIN_MIP_NEAR 4 +# define VC4_TEX_P1_MINFILT_LIN_MIP_LIN 5 + +#define VC4_TEX_P1_WRAP_T_MASK VC4_MASK(3, 2) +#define VC4_TEX_P1_WRAP_T_SHIFT 2 +#define VC4_TEX_P1_WRAP_S_MASK VC4_MASK(1, 0) +#define VC4_TEX_P1_WRAP_S_SHIFT 0 +# define VC4_TEX_P1_WRAP_REPEAT 0 +# define VC4_TEX_P1_WRAP_CLAMP 1 +# define VC4_TEX_P1_WRAP_MIRROR 2 +# define VC4_TEX_P1_WRAP_BORDER 3 + +#define VC4_TEX_P2_PTYPE_MASK VC4_MASK(31, 30) +#define VC4_TEX_P2_PTYPE_SHIFT 30 +# define VC4_TEX_P2_PTYPE_IGNORED 0 +# define VC4_TEX_P2_PTYPE_CUBE_MAP_STRIDE 1 +# define VC4_TEX_P2_PTYPE_CHILD_IMAGE_DIMENSIONS 2 +# define VC4_TEX_P2_PTYPE_CHILD_IMAGE_OFFSETS 3 + +/* VC4_TEX_P2_PTYPE_CUBE_MAP_STRIDE bits */ +#define VC4_TEX_P2_CMST_MASK VC4_MASK(29, 12) +#define VC4_TEX_P2_CMST_SHIFT 12 +#define VC4_TEX_P2_BSLOD_MASK VC4_MASK(0, 0) +#define VC4_TEX_P2_BSLOD_SHIFT 0 + +/* VC4_TEX_P2_PTYPE_CHILD_IMAGE_DIMENSIONS */ +#define VC4_TEX_P2_CHEIGHT_MASK VC4_MASK(22, 12) +#define VC4_TEX_P2_CHEIGHT_SHIFT 12 +#define VC4_TEX_P2_CWIDTH_MASK VC4_MASK(10, 0) +#define VC4_TEX_P2_CWIDTH_SHIFT 0 + +/* VC4_TEX_P2_PTYPE_CHILD_IMAGE_OFFSETS */ +#define VC4_TEX_P2_CYOFF_MASK VC4_MASK(22, 12) +#define VC4_TEX_P2_CYOFF_SHIFT 12 +#define VC4_TEX_P2_CXOFF_MASK VC4_MASK(10, 0) +#define VC4_TEX_P2_CXOFF_SHIFT 0 + +#endif /* VC4_PACKET_H */ diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c index 887f3caad0be..0addbad15832 100644 --- a/drivers/gpu/drm/vc4/vc4_plane.c +++ b/drivers/gpu/drm/vc4/vc4_plane.c @@ -29,6 +29,14 @@ struct vc4_plane_state { u32 *dlist; u32 dlist_size; /* Number of dwords in allocated for the display list */ u32 dlist_count; /* Number of used dwords in the display list. */ + + /* Offset in the dlist to pointer word 0. */ + u32 pw0_offset; + + /* Offset where the plane's dlist was last stored in the + hardware at vc4_crtc_atomic_flush() time. + */ + u32 *hw_dlist; }; static inline struct vc4_plane_state * @@ -207,6 +215,8 @@ static int vc4_plane_mode_set(struct drm_plane *plane, /* Position Word 3: Context. Written by the HVS. */ vc4_dlist_write(vc4_state, 0xc0c0c0c0); + vc4_state->pw0_offset = vc4_state->dlist_count; + /* Pointer Word 0: RGB / Y Pointer */ vc4_dlist_write(vc4_state, bo->paddr + offset); @@ -258,6 +268,8 @@ u32 vc4_plane_write_dlist(struct drm_plane *plane, u32 __iomem *dlist) struct vc4_plane_state *vc4_state = to_vc4_plane_state(plane->state); int i; + vc4_state->hw_dlist = dlist; + /* Can't memcpy_toio() because it needs to be 32-bit writes. */ for (i = 0; i < vc4_state->dlist_count; i++) writel(vc4_state->dlist[i], &dlist[i]); @@ -272,6 +284,34 @@ u32 vc4_plane_dlist_size(struct drm_plane_state *state) return vc4_state->dlist_count; } +/* Updates the plane to immediately (well, once the FIFO needs + * refilling) scan out from at a new framebuffer. + */ +void vc4_plane_async_set_fb(struct drm_plane *plane, struct drm_framebuffer *fb) +{ + struct vc4_plane_state *vc4_state = to_vc4_plane_state(plane->state); + struct drm_gem_cma_object *bo = drm_fb_cma_get_gem_obj(fb, 0); + uint32_t addr; + + /* We're skipping the address adjustment for negative origin, + * because this is only called on the primary plane. + */ + WARN_ON_ONCE(plane->state->crtc_x < 0 || plane->state->crtc_y < 0); + addr = bo->paddr + fb->offsets[0]; + + /* Write the new address into the hardware immediately. The + * scanout will start from this address as soon as the FIFO + * needs to refill with pixels. + */ + writel(addr, &vc4_state->hw_dlist[vc4_state->pw0_offset]); + + /* Also update the CPU-side dlist copy, so that any later + * atomic updates that don't do a new modeset on our plane + * also use our updated address. + */ + vc4_state->dlist[vc4_state->pw0_offset] = addr; +} + static const struct drm_plane_helper_funcs vc4_plane_helper_funcs = { .prepare_fb = NULL, .cleanup_fb = NULL, @@ -317,7 +357,7 @@ struct drm_plane *vc4_plane_init(struct drm_device *dev, ret = drm_universal_plane_init(dev, plane, 0xff, &vc4_plane_funcs, formats, ARRAY_SIZE(formats), - type); + type, NULL); drm_plane_helper_add(plane, &vc4_plane_helper_funcs); diff --git a/drivers/gpu/drm/vc4/vc4_qpu_defines.h b/drivers/gpu/drm/vc4/vc4_qpu_defines.h new file mode 100644 index 000000000000..d5c2f3c85ebb --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_qpu_defines.h @@ -0,0 +1,264 @@ +/* + * Copyright © 2014 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef VC4_QPU_DEFINES_H +#define VC4_QPU_DEFINES_H + +enum qpu_op_add { + QPU_A_NOP, + QPU_A_FADD, + QPU_A_FSUB, + QPU_A_FMIN, + QPU_A_FMAX, + QPU_A_FMINABS, + QPU_A_FMAXABS, + QPU_A_FTOI, + QPU_A_ITOF, + QPU_A_ADD = 12, + QPU_A_SUB, + QPU_A_SHR, + QPU_A_ASR, + QPU_A_ROR, + QPU_A_SHL, + QPU_A_MIN, + QPU_A_MAX, + QPU_A_AND, + QPU_A_OR, + QPU_A_XOR, + QPU_A_NOT, + QPU_A_CLZ, + QPU_A_V8ADDS = 30, + QPU_A_V8SUBS = 31, +}; + +enum qpu_op_mul { + QPU_M_NOP, + QPU_M_FMUL, + QPU_M_MUL24, + QPU_M_V8MULD, + QPU_M_V8MIN, + QPU_M_V8MAX, + QPU_M_V8ADDS, + QPU_M_V8SUBS, +}; + +enum qpu_raddr { + QPU_R_FRAG_PAYLOAD_ZW = 15, /* W for A file, Z for B file */ + /* 0-31 are the plain regfile a or b fields */ + QPU_R_UNIF = 32, + QPU_R_VARY = 35, + QPU_R_ELEM_QPU = 38, + QPU_R_NOP, + QPU_R_XY_PIXEL_COORD = 41, + QPU_R_MS_REV_FLAGS = 41, + QPU_R_VPM = 48, + QPU_R_VPM_LD_BUSY, + QPU_R_VPM_LD_WAIT, + QPU_R_MUTEX_ACQUIRE, +}; + +enum qpu_waddr { + /* 0-31 are the plain regfile a or b fields */ + QPU_W_ACC0 = 32, /* aka r0 */ + QPU_W_ACC1, + QPU_W_ACC2, + QPU_W_ACC3, + QPU_W_TMU_NOSWAP, + QPU_W_ACC5, + QPU_W_HOST_INT, + QPU_W_NOP, + QPU_W_UNIFORMS_ADDRESS, + QPU_W_QUAD_XY, /* X for regfile a, Y for regfile b */ + QPU_W_MS_FLAGS = 42, + QPU_W_REV_FLAG = 42, + QPU_W_TLB_STENCIL_SETUP = 43, + QPU_W_TLB_Z, + QPU_W_TLB_COLOR_MS, + QPU_W_TLB_COLOR_ALL, + QPU_W_TLB_ALPHA_MASK, + QPU_W_VPM, + QPU_W_VPMVCD_SETUP, /* LD for regfile a, ST for regfile b */ + QPU_W_VPM_ADDR, /* LD for regfile a, ST for regfile b */ + QPU_W_MUTEX_RELEASE, + QPU_W_SFU_RECIP, + QPU_W_SFU_RECIPSQRT, + QPU_W_SFU_EXP, + QPU_W_SFU_LOG, + QPU_W_TMU0_S, + QPU_W_TMU0_T, + QPU_W_TMU0_R, + QPU_W_TMU0_B, + QPU_W_TMU1_S, + QPU_W_TMU1_T, + QPU_W_TMU1_R, + QPU_W_TMU1_B, +}; + +enum qpu_sig_bits { + QPU_SIG_SW_BREAKPOINT, + QPU_SIG_NONE, + QPU_SIG_THREAD_SWITCH, + QPU_SIG_PROG_END, + QPU_SIG_WAIT_FOR_SCOREBOARD, + QPU_SIG_SCOREBOARD_UNLOCK, + QPU_SIG_LAST_THREAD_SWITCH, + QPU_SIG_COVERAGE_LOAD, + QPU_SIG_COLOR_LOAD, + QPU_SIG_COLOR_LOAD_END, + QPU_SIG_LOAD_TMU0, + QPU_SIG_LOAD_TMU1, + QPU_SIG_ALPHA_MASK_LOAD, + QPU_SIG_SMALL_IMM, + QPU_SIG_LOAD_IMM, + QPU_SIG_BRANCH +}; + +enum qpu_mux { + /* hardware mux values */ + QPU_MUX_R0, + QPU_MUX_R1, + QPU_MUX_R2, + QPU_MUX_R3, + QPU_MUX_R4, + QPU_MUX_R5, + QPU_MUX_A, + QPU_MUX_B, + + /* non-hardware mux values */ + QPU_MUX_IMM, +}; + +enum qpu_cond { + QPU_COND_NEVER, + QPU_COND_ALWAYS, + QPU_COND_ZS, + QPU_COND_ZC, + QPU_COND_NS, + QPU_COND_NC, + QPU_COND_CS, + QPU_COND_CC, +}; + +enum qpu_pack_mul { + QPU_PACK_MUL_NOP, + /* replicated to each 8 bits of the 32-bit dst. */ + QPU_PACK_MUL_8888 = 3, + QPU_PACK_MUL_8A, + QPU_PACK_MUL_8B, + QPU_PACK_MUL_8C, + QPU_PACK_MUL_8D, +}; + +enum qpu_pack_a { + QPU_PACK_A_NOP, + /* convert to 16 bit float if float input, or to int16. */ + QPU_PACK_A_16A, + QPU_PACK_A_16B, + /* replicated to each 8 bits of the 32-bit dst. */ + QPU_PACK_A_8888, + /* Convert to 8-bit unsigned int. */ + QPU_PACK_A_8A, + QPU_PACK_A_8B, + QPU_PACK_A_8C, + QPU_PACK_A_8D, + + /* Saturating variants of the previous instructions. */ + QPU_PACK_A_32_SAT, /* int-only */ + QPU_PACK_A_16A_SAT, /* int or float */ + QPU_PACK_A_16B_SAT, + QPU_PACK_A_8888_SAT, + QPU_PACK_A_8A_SAT, + QPU_PACK_A_8B_SAT, + QPU_PACK_A_8C_SAT, + QPU_PACK_A_8D_SAT, +}; + +enum qpu_unpack_r4 { + QPU_UNPACK_R4_NOP, + QPU_UNPACK_R4_F16A_TO_F32, + QPU_UNPACK_R4_F16B_TO_F32, + QPU_UNPACK_R4_8D_REP, + QPU_UNPACK_R4_8A, + QPU_UNPACK_R4_8B, + QPU_UNPACK_R4_8C, + QPU_UNPACK_R4_8D, +}; + +#define QPU_MASK(high, low) \ + ((((uint64_t)1 << ((high) - (low) + 1)) - 1) << (low)) + +#define QPU_GET_FIELD(word, field) \ + ((uint32_t)(((word) & field ## _MASK) >> field ## _SHIFT)) + +#define QPU_SIG_SHIFT 60 +#define QPU_SIG_MASK QPU_MASK(63, 60) + +#define QPU_UNPACK_SHIFT 57 +#define QPU_UNPACK_MASK QPU_MASK(59, 57) + +/** + * If set, the pack field means PACK_MUL or R4 packing, instead of normal + * regfile a packing. + */ +#define QPU_PM ((uint64_t)1 << 56) + +#define QPU_PACK_SHIFT 52 +#define QPU_PACK_MASK QPU_MASK(55, 52) + +#define QPU_COND_ADD_SHIFT 49 +#define QPU_COND_ADD_MASK QPU_MASK(51, 49) +#define QPU_COND_MUL_SHIFT 46 +#define QPU_COND_MUL_MASK QPU_MASK(48, 46) + +#define QPU_SF ((uint64_t)1 << 45) + +#define QPU_WADDR_ADD_SHIFT 38 +#define QPU_WADDR_ADD_MASK QPU_MASK(43, 38) +#define QPU_WADDR_MUL_SHIFT 32 +#define QPU_WADDR_MUL_MASK QPU_MASK(37, 32) + +#define QPU_OP_MUL_SHIFT 29 +#define QPU_OP_MUL_MASK QPU_MASK(31, 29) + +#define QPU_RADDR_A_SHIFT 18 +#define QPU_RADDR_A_MASK QPU_MASK(23, 18) +#define QPU_RADDR_B_SHIFT 12 +#define QPU_RADDR_B_MASK QPU_MASK(17, 12) +#define QPU_SMALL_IMM_SHIFT 12 +#define QPU_SMALL_IMM_MASK QPU_MASK(17, 12) + +#define QPU_ADD_A_SHIFT 9 +#define QPU_ADD_A_MASK QPU_MASK(11, 9) +#define QPU_ADD_B_SHIFT 6 +#define QPU_ADD_B_MASK QPU_MASK(8, 6) +#define QPU_MUL_A_SHIFT 3 +#define QPU_MUL_A_MASK QPU_MASK(5, 3) +#define QPU_MUL_B_SHIFT 0 +#define QPU_MUL_B_MASK QPU_MASK(2, 0) + +#define QPU_WS ((uint64_t)1 << 44) + +#define QPU_OP_ADD_SHIFT 24 +#define QPU_OP_ADD_MASK QPU_MASK(28, 24) + +#endif /* VC4_QPU_DEFINES_H */ diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h index 9e4e904c668e..4e52a0a88551 100644 --- a/drivers/gpu/drm/vc4/vc4_regs.h +++ b/drivers/gpu/drm/vc4/vc4_regs.h @@ -154,7 +154,7 @@ #define V3D_PCTRS14 0x006f4 #define V3D_PCTR15 0x006f8 #define V3D_PCTRS15 0x006fc -#define V3D_BGE 0x00f00 +#define V3D_DBGE 0x00f00 #define V3D_FDBGO 0x00f04 #define V3D_FDBGB 0x00f08 #define V3D_FDBGR 0x00f0c diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c new file mode 100644 index 000000000000..8a2a312e2c1b --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_render_cl.c @@ -0,0 +1,634 @@ +/* + * Copyright © 2014-2015 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +/** + * DOC: Render command list generation + * + * In the VC4 driver, render command list generation is performed by the + * kernel instead of userspace. We do this because validating a + * user-submitted command list is hard to get right and has high CPU overhead, + * while the number of valid configurations for render command lists is + * actually fairly low. + */ + +#include "uapi/drm/vc4_drm.h" +#include "vc4_drv.h" +#include "vc4_packet.h" + +struct vc4_rcl_setup { + struct drm_gem_cma_object *color_read; + struct drm_gem_cma_object *color_write; + struct drm_gem_cma_object *zs_read; + struct drm_gem_cma_object *zs_write; + struct drm_gem_cma_object *msaa_color_write; + struct drm_gem_cma_object *msaa_zs_write; + + struct drm_gem_cma_object *rcl; + u32 next_offset; +}; + +static inline void rcl_u8(struct vc4_rcl_setup *setup, u8 val) +{ + *(u8 *)(setup->rcl->vaddr + setup->next_offset) = val; + setup->next_offset += 1; +} + +static inline void rcl_u16(struct vc4_rcl_setup *setup, u16 val) +{ + *(u16 *)(setup->rcl->vaddr + setup->next_offset) = val; + setup->next_offset += 2; +} + +static inline void rcl_u32(struct vc4_rcl_setup *setup, u32 val) +{ + *(u32 *)(setup->rcl->vaddr + setup->next_offset) = val; + setup->next_offset += 4; +} + +/* + * Emits a no-op STORE_TILE_BUFFER_GENERAL. + * + * If we emit a PACKET_TILE_COORDINATES, it must be followed by a store of + * some sort before another load is triggered. + */ +static void vc4_store_before_load(struct vc4_rcl_setup *setup) +{ + rcl_u8(setup, VC4_PACKET_STORE_TILE_BUFFER_GENERAL); + rcl_u16(setup, + VC4_SET_FIELD(VC4_LOADSTORE_TILE_BUFFER_NONE, + VC4_LOADSTORE_TILE_BUFFER_BUFFER) | + VC4_STORE_TILE_BUFFER_DISABLE_COLOR_CLEAR | + VC4_STORE_TILE_BUFFER_DISABLE_ZS_CLEAR | + VC4_STORE_TILE_BUFFER_DISABLE_VG_MASK_CLEAR); + rcl_u32(setup, 0); /* no address, since we're in None mode */ +} + +/* + * Calculates the physical address of the start of a tile in a RCL surface. + * + * Unlike the other load/store packets, + * VC4_PACKET_LOAD/STORE_FULL_RES_TILE_BUFFER don't look at the tile + * coordinates packet, and instead just store to the address given. + */ +static uint32_t vc4_full_res_offset(struct vc4_exec_info *exec, + struct drm_gem_cma_object *bo, + struct drm_vc4_submit_rcl_surface *surf, + uint8_t x, uint8_t y) +{ + return bo->paddr + surf->offset + VC4_TILE_BUFFER_SIZE * + (DIV_ROUND_UP(exec->args->width, 32) * y + x); +} + +/* + * Emits a PACKET_TILE_COORDINATES if one isn't already pending. + * + * The tile coordinates packet triggers a pending load if there is one, are + * used for clipping during rendering, and determine where loads/stores happen + * relative to their base address. + */ +static void vc4_tile_coordinates(struct vc4_rcl_setup *setup, + uint32_t x, uint32_t y) +{ + rcl_u8(setup, VC4_PACKET_TILE_COORDINATES); + rcl_u8(setup, x); + rcl_u8(setup, y); +} + +static void emit_tile(struct vc4_exec_info *exec, + struct vc4_rcl_setup *setup, + uint8_t x, uint8_t y, bool first, bool last) +{ + struct drm_vc4_submit_cl *args = exec->args; + bool has_bin = args->bin_cl_size != 0; + + /* Note that the load doesn't actually occur until the + * tile coords packet is processed, and only one load + * may be outstanding at a time. + */ + if (setup->color_read) { + if (args->color_read.flags & + VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { + rcl_u8(setup, VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER); + rcl_u32(setup, + vc4_full_res_offset(exec, setup->color_read, + &args->color_read, x, y) | + VC4_LOADSTORE_FULL_RES_DISABLE_ZS); + } else { + rcl_u8(setup, VC4_PACKET_LOAD_TILE_BUFFER_GENERAL); + rcl_u16(setup, args->color_read.bits); + rcl_u32(setup, setup->color_read->paddr + + args->color_read.offset); + } + } + + if (setup->zs_read) { + if (args->zs_read.flags & + VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { + rcl_u8(setup, VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER); + rcl_u32(setup, + vc4_full_res_offset(exec, setup->zs_read, + &args->zs_read, x, y) | + VC4_LOADSTORE_FULL_RES_DISABLE_COLOR); + } else { + if (setup->color_read) { + /* Exec previous load. */ + vc4_tile_coordinates(setup, x, y); + vc4_store_before_load(setup); + } + + rcl_u8(setup, VC4_PACKET_LOAD_TILE_BUFFER_GENERAL); + rcl_u16(setup, args->zs_read.bits); + rcl_u32(setup, setup->zs_read->paddr + + args->zs_read.offset); + } + } + + /* Clipping depends on tile coordinates having been + * emitted, so we always need one here. + */ + vc4_tile_coordinates(setup, x, y); + + /* Wait for the binner before jumping to the first + * tile's lists. + */ + if (first && has_bin) + rcl_u8(setup, VC4_PACKET_WAIT_ON_SEMAPHORE); + + if (has_bin) { + rcl_u8(setup, VC4_PACKET_BRANCH_TO_SUB_LIST); + rcl_u32(setup, (exec->tile_bo->paddr + + exec->tile_alloc_offset + + (y * exec->bin_tiles_x + x) * 32)); + } + + if (setup->msaa_color_write) { + bool last_tile_write = (!setup->msaa_zs_write && + !setup->zs_write && + !setup->color_write); + uint32_t bits = VC4_LOADSTORE_FULL_RES_DISABLE_ZS; + + if (!last_tile_write) + bits |= VC4_LOADSTORE_FULL_RES_DISABLE_CLEAR_ALL; + else if (last) + bits |= VC4_LOADSTORE_FULL_RES_EOF; + rcl_u8(setup, VC4_PACKET_STORE_FULL_RES_TILE_BUFFER); + rcl_u32(setup, + vc4_full_res_offset(exec, setup->msaa_color_write, + &args->msaa_color_write, x, y) | + bits); + } + + if (setup->msaa_zs_write) { + bool last_tile_write = (!setup->zs_write && + !setup->color_write); + uint32_t bits = VC4_LOADSTORE_FULL_RES_DISABLE_COLOR; + + if (setup->msaa_color_write) + vc4_tile_coordinates(setup, x, y); + if (!last_tile_write) + bits |= VC4_LOADSTORE_FULL_RES_DISABLE_CLEAR_ALL; + else if (last) + bits |= VC4_LOADSTORE_FULL_RES_EOF; + rcl_u8(setup, VC4_PACKET_STORE_FULL_RES_TILE_BUFFER); + rcl_u32(setup, + vc4_full_res_offset(exec, setup->msaa_zs_write, + &args->msaa_zs_write, x, y) | + bits); + } + + if (setup->zs_write) { + bool last_tile_write = !setup->color_write; + + if (setup->msaa_color_write || setup->msaa_zs_write) + vc4_tile_coordinates(setup, x, y); + + rcl_u8(setup, VC4_PACKET_STORE_TILE_BUFFER_GENERAL); + rcl_u16(setup, args->zs_write.bits | + (last_tile_write ? + 0 : VC4_STORE_TILE_BUFFER_DISABLE_COLOR_CLEAR)); + rcl_u32(setup, + (setup->zs_write->paddr + args->zs_write.offset) | + ((last && last_tile_write) ? + VC4_LOADSTORE_TILE_BUFFER_EOF : 0)); + } + + if (setup->color_write) { + if (setup->msaa_color_write || setup->msaa_zs_write || + setup->zs_write) { + vc4_tile_coordinates(setup, x, y); + } + + if (last) + rcl_u8(setup, VC4_PACKET_STORE_MS_TILE_BUFFER_AND_EOF); + else + rcl_u8(setup, VC4_PACKET_STORE_MS_TILE_BUFFER); + } +} + +static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec, + struct vc4_rcl_setup *setup) +{ + struct drm_vc4_submit_cl *args = exec->args; + bool has_bin = args->bin_cl_size != 0; + uint8_t min_x_tile = args->min_x_tile; + uint8_t min_y_tile = args->min_y_tile; + uint8_t max_x_tile = args->max_x_tile; + uint8_t max_y_tile = args->max_y_tile; + uint8_t xtiles = max_x_tile - min_x_tile + 1; + uint8_t ytiles = max_y_tile - min_y_tile + 1; + uint8_t x, y; + uint32_t size, loop_body_size; + + size = VC4_PACKET_TILE_RENDERING_MODE_CONFIG_SIZE; + loop_body_size = VC4_PACKET_TILE_COORDINATES_SIZE; + + if (args->flags & VC4_SUBMIT_CL_USE_CLEAR_COLOR) { + size += VC4_PACKET_CLEAR_COLORS_SIZE + + VC4_PACKET_TILE_COORDINATES_SIZE + + VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE; + } + + if (setup->color_read) { + if (args->color_read.flags & + VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { + loop_body_size += VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER_SIZE; + } else { + loop_body_size += VC4_PACKET_LOAD_TILE_BUFFER_GENERAL_SIZE; + } + } + if (setup->zs_read) { + if (args->zs_read.flags & + VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { + loop_body_size += VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER_SIZE; + } else { + if (setup->color_read && + !(args->color_read.flags & + VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES)) { + loop_body_size += VC4_PACKET_TILE_COORDINATES_SIZE; + loop_body_size += VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE; + } + loop_body_size += VC4_PACKET_LOAD_TILE_BUFFER_GENERAL_SIZE; + } + } + + if (has_bin) { + size += VC4_PACKET_WAIT_ON_SEMAPHORE_SIZE; + loop_body_size += VC4_PACKET_BRANCH_TO_SUB_LIST_SIZE; + } + + if (setup->msaa_color_write) + loop_body_size += VC4_PACKET_STORE_FULL_RES_TILE_BUFFER_SIZE; + if (setup->msaa_zs_write) + loop_body_size += VC4_PACKET_STORE_FULL_RES_TILE_BUFFER_SIZE; + + if (setup->zs_write) + loop_body_size += VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE; + if (setup->color_write) + loop_body_size += VC4_PACKET_STORE_MS_TILE_BUFFER_SIZE; + + /* We need a VC4_PACKET_TILE_COORDINATES in between each store. */ + loop_body_size += VC4_PACKET_TILE_COORDINATES_SIZE * + ((setup->msaa_color_write != NULL) + + (setup->msaa_zs_write != NULL) + + (setup->color_write != NULL) + + (setup->zs_write != NULL) - 1); + + size += xtiles * ytiles * loop_body_size; + + setup->rcl = &vc4_bo_create(dev, size, true)->base; + if (!setup->rcl) + return -ENOMEM; + list_add_tail(&to_vc4_bo(&setup->rcl->base)->unref_head, + &exec->unref_list); + + rcl_u8(setup, VC4_PACKET_TILE_RENDERING_MODE_CONFIG); + rcl_u32(setup, + (setup->color_write ? (setup->color_write->paddr + + args->color_write.offset) : + 0)); + rcl_u16(setup, args->width); + rcl_u16(setup, args->height); + rcl_u16(setup, args->color_write.bits); + + /* The tile buffer gets cleared when the previous tile is stored. If + * the clear values changed between frames, then the tile buffer has + * stale clear values in it, so we have to do a store in None mode (no + * writes) so that we trigger the tile buffer clear. + */ + if (args->flags & VC4_SUBMIT_CL_USE_CLEAR_COLOR) { + rcl_u8(setup, VC4_PACKET_CLEAR_COLORS); + rcl_u32(setup, args->clear_color[0]); + rcl_u32(setup, args->clear_color[1]); + rcl_u32(setup, args->clear_z); + rcl_u8(setup, args->clear_s); + + vc4_tile_coordinates(setup, 0, 0); + + rcl_u8(setup, VC4_PACKET_STORE_TILE_BUFFER_GENERAL); + rcl_u16(setup, VC4_LOADSTORE_TILE_BUFFER_NONE); + rcl_u32(setup, 0); /* no address, since we're in None mode */ + } + + for (y = min_y_tile; y <= max_y_tile; y++) { + for (x = min_x_tile; x <= max_x_tile; x++) { + bool first = (x == min_x_tile && y == min_y_tile); + bool last = (x == max_x_tile && y == max_y_tile); + + emit_tile(exec, setup, x, y, first, last); + } + } + + BUG_ON(setup->next_offset != size); + exec->ct1ca = setup->rcl->paddr; + exec->ct1ea = setup->rcl->paddr + setup->next_offset; + + return 0; +} + +static int vc4_full_res_bounds_check(struct vc4_exec_info *exec, + struct drm_gem_cma_object *obj, + struct drm_vc4_submit_rcl_surface *surf) +{ + struct drm_vc4_submit_cl *args = exec->args; + u32 render_tiles_stride = DIV_ROUND_UP(exec->args->width, 32); + + if (surf->offset > obj->base.size) { + DRM_ERROR("surface offset %d > BO size %zd\n", + surf->offset, obj->base.size); + return -EINVAL; + } + + if ((obj->base.size - surf->offset) / VC4_TILE_BUFFER_SIZE < + render_tiles_stride * args->max_y_tile + args->max_x_tile) { + DRM_ERROR("MSAA tile %d, %d out of bounds " + "(bo size %zd, offset %d).\n", + args->max_x_tile, args->max_y_tile, + obj->base.size, + surf->offset); + return -EINVAL; + } + + return 0; +} + +static int vc4_rcl_msaa_surface_setup(struct vc4_exec_info *exec, + struct drm_gem_cma_object **obj, + struct drm_vc4_submit_rcl_surface *surf) +{ + if (surf->flags != 0 || surf->bits != 0) { + DRM_ERROR("MSAA surface had nonzero flags/bits\n"); + return -EINVAL; + } + + if (surf->hindex == ~0) + return 0; + + *obj = vc4_use_bo(exec, surf->hindex); + if (!*obj) + return -EINVAL; + + if (surf->offset & 0xf) { + DRM_ERROR("MSAA write must be 16b aligned.\n"); + return -EINVAL; + } + + return vc4_full_res_bounds_check(exec, *obj, surf); +} + +static int vc4_rcl_surface_setup(struct vc4_exec_info *exec, + struct drm_gem_cma_object **obj, + struct drm_vc4_submit_rcl_surface *surf) +{ + uint8_t tiling = VC4_GET_FIELD(surf->bits, + VC4_LOADSTORE_TILE_BUFFER_TILING); + uint8_t buffer = VC4_GET_FIELD(surf->bits, + VC4_LOADSTORE_TILE_BUFFER_BUFFER); + uint8_t format = VC4_GET_FIELD(surf->bits, + VC4_LOADSTORE_TILE_BUFFER_FORMAT); + int cpp; + int ret; + + if (surf->flags & ~VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { + DRM_ERROR("Extra flags set\n"); + return -EINVAL; + } + + if (surf->hindex == ~0) + return 0; + + *obj = vc4_use_bo(exec, surf->hindex); + if (!*obj) + return -EINVAL; + + if (surf->flags & VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { + if (surf == &exec->args->zs_write) { + DRM_ERROR("general zs write may not be a full-res.\n"); + return -EINVAL; + } + + if (surf->bits != 0) { + DRM_ERROR("load/store general bits set with " + "full res load/store.\n"); + return -EINVAL; + } + + ret = vc4_full_res_bounds_check(exec, *obj, surf); + if (!ret) + return ret; + + return 0; + } + + if (surf->bits & ~(VC4_LOADSTORE_TILE_BUFFER_TILING_MASK | + VC4_LOADSTORE_TILE_BUFFER_BUFFER_MASK | + VC4_LOADSTORE_TILE_BUFFER_FORMAT_MASK)) { + DRM_ERROR("Unknown bits in load/store: 0x%04x\n", + surf->bits); + return -EINVAL; + } + + if (tiling > VC4_TILING_FORMAT_LT) { + DRM_ERROR("Bad tiling format\n"); + return -EINVAL; + } + + if (buffer == VC4_LOADSTORE_TILE_BUFFER_ZS) { + if (format != 0) { + DRM_ERROR("No color format should be set for ZS\n"); + return -EINVAL; + } + cpp = 4; + } else if (buffer == VC4_LOADSTORE_TILE_BUFFER_COLOR) { + switch (format) { + case VC4_LOADSTORE_TILE_BUFFER_BGR565: + case VC4_LOADSTORE_TILE_BUFFER_BGR565_DITHER: + cpp = 2; + break; + case VC4_LOADSTORE_TILE_BUFFER_RGBA8888: + cpp = 4; + break; + default: + DRM_ERROR("Bad tile buffer format\n"); + return -EINVAL; + } + } else { + DRM_ERROR("Bad load/store buffer %d.\n", buffer); + return -EINVAL; + } + + if (surf->offset & 0xf) { + DRM_ERROR("load/store buffer must be 16b aligned.\n"); + return -EINVAL; + } + + if (!vc4_check_tex_size(exec, *obj, surf->offset, tiling, + exec->args->width, exec->args->height, cpp)) { + return -EINVAL; + } + + return 0; +} + +static int +vc4_rcl_render_config_surface_setup(struct vc4_exec_info *exec, + struct vc4_rcl_setup *setup, + struct drm_gem_cma_object **obj, + struct drm_vc4_submit_rcl_surface *surf) +{ + uint8_t tiling = VC4_GET_FIELD(surf->bits, + VC4_RENDER_CONFIG_MEMORY_FORMAT); + uint8_t format = VC4_GET_FIELD(surf->bits, + VC4_RENDER_CONFIG_FORMAT); + int cpp; + + if (surf->flags != 0) { + DRM_ERROR("No flags supported on render config.\n"); + return -EINVAL; + } + + if (surf->bits & ~(VC4_RENDER_CONFIG_MEMORY_FORMAT_MASK | + VC4_RENDER_CONFIG_FORMAT_MASK | + VC4_RENDER_CONFIG_MS_MODE_4X | + VC4_RENDER_CONFIG_DECIMATE_MODE_4X)) { + DRM_ERROR("Unknown bits in render config: 0x%04x\n", + surf->bits); + return -EINVAL; + } + + if (surf->hindex == ~0) + return 0; + + *obj = vc4_use_bo(exec, surf->hindex); + if (!*obj) + return -EINVAL; + + if (tiling > VC4_TILING_FORMAT_LT) { + DRM_ERROR("Bad tiling format\n"); + return -EINVAL; + } + + switch (format) { + case VC4_RENDER_CONFIG_FORMAT_BGR565_DITHERED: + case VC4_RENDER_CONFIG_FORMAT_BGR565: + cpp = 2; + break; + case VC4_RENDER_CONFIG_FORMAT_RGBA8888: + cpp = 4; + break; + default: + DRM_ERROR("Bad tile buffer format\n"); + return -EINVAL; + } + + if (!vc4_check_tex_size(exec, *obj, surf->offset, tiling, + exec->args->width, exec->args->height, cpp)) { + return -EINVAL; + } + + return 0; +} + +int vc4_get_rcl(struct drm_device *dev, struct vc4_exec_info *exec) +{ + struct vc4_rcl_setup setup = {0}; + struct drm_vc4_submit_cl *args = exec->args; + bool has_bin = args->bin_cl_size != 0; + int ret; + + if (args->min_x_tile > args->max_x_tile || + args->min_y_tile > args->max_y_tile) { + DRM_ERROR("Bad render tile set (%d,%d)-(%d,%d)\n", + args->min_x_tile, args->min_y_tile, + args->max_x_tile, args->max_y_tile); + return -EINVAL; + } + + if (has_bin && + (args->max_x_tile > exec->bin_tiles_x || + args->max_y_tile > exec->bin_tiles_y)) { + DRM_ERROR("Render tiles (%d,%d) outside of bin config " + "(%d,%d)\n", + args->max_x_tile, args->max_y_tile, + exec->bin_tiles_x, exec->bin_tiles_y); + return -EINVAL; + } + + ret = vc4_rcl_render_config_surface_setup(exec, &setup, + &setup.color_write, + &args->color_write); + if (ret) + return ret; + + ret = vc4_rcl_surface_setup(exec, &setup.color_read, &args->color_read); + if (ret) + return ret; + + ret = vc4_rcl_surface_setup(exec, &setup.zs_read, &args->zs_read); + if (ret) + return ret; + + ret = vc4_rcl_surface_setup(exec, &setup.zs_write, &args->zs_write); + if (ret) + return ret; + + ret = vc4_rcl_msaa_surface_setup(exec, &setup.msaa_color_write, + &args->msaa_color_write); + if (ret) + return ret; + + ret = vc4_rcl_msaa_surface_setup(exec, &setup.msaa_zs_write, + &args->msaa_zs_write); + if (ret) + return ret; + + /* We shouldn't even have the job submitted to us if there's no + * surface to write out. + */ + if (!setup.color_write && !setup.zs_write && + !setup.msaa_color_write && !setup.msaa_zs_write) { + DRM_ERROR("RCL requires color or Z/S write\n"); + return -EINVAL; + } + + return vc4_create_rcl_bo(dev, exec, &setup); +} diff --git a/drivers/gpu/drm/vc4/vc4_trace.h b/drivers/gpu/drm/vc4/vc4_trace.h new file mode 100644 index 000000000000..ad7b1ea720c2 --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_trace.h @@ -0,0 +1,63 @@ +/* + * Copyright (C) 2015 Broadcom + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#if !defined(_VC4_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) +#define _VC4_TRACE_H_ + +#include <linux/stringify.h> +#include <linux/types.h> +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM vc4 +#define TRACE_INCLUDE_FILE vc4_trace + +TRACE_EVENT(vc4_wait_for_seqno_begin, + TP_PROTO(struct drm_device *dev, uint64_t seqno, uint64_t timeout), + TP_ARGS(dev, seqno, timeout), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u64, seqno) + __field(u64, timeout) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->seqno = seqno; + __entry->timeout = timeout; + ), + + TP_printk("dev=%u, seqno=%llu, timeout=%llu", + __entry->dev, __entry->seqno, __entry->timeout) +); + +TRACE_EVENT(vc4_wait_for_seqno_end, + TP_PROTO(struct drm_device *dev, uint64_t seqno), + TP_ARGS(dev, seqno), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u64, seqno) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->seqno = seqno; + ), + + TP_printk("dev=%u, seqno=%llu", + __entry->dev, __entry->seqno) +); + +#endif /* _VC4_TRACE_H_ */ + +/* This part must be outside protection */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#include <trace/define_trace.h> diff --git a/drivers/gpu/drm/vc4/vc4_trace_points.c b/drivers/gpu/drm/vc4/vc4_trace_points.c new file mode 100644 index 000000000000..e6278f25716b --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_trace_points.c @@ -0,0 +1,14 @@ +/* + * Copyright (C) 2015 Broadcom + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include "vc4_drv.h" + +#ifndef __CHECKER__ +#define CREATE_TRACE_POINTS +#include "vc4_trace.h" +#endif diff --git a/drivers/gpu/drm/vc4/vc4_v3d.c b/drivers/gpu/drm/vc4/vc4_v3d.c new file mode 100644 index 000000000000..424d515ffcda --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_v3d.c @@ -0,0 +1,262 @@ +/* + * Copyright (c) 2014 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark <robdclark@gmail.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "linux/component.h" +#include "vc4_drv.h" +#include "vc4_regs.h" + +#ifdef CONFIG_DEBUG_FS +#define REGDEF(reg) { reg, #reg } +static const struct { + uint32_t reg; + const char *name; +} vc4_reg_defs[] = { + REGDEF(V3D_IDENT0), + REGDEF(V3D_IDENT1), + REGDEF(V3D_IDENT2), + REGDEF(V3D_SCRATCH), + REGDEF(V3D_L2CACTL), + REGDEF(V3D_SLCACTL), + REGDEF(V3D_INTCTL), + REGDEF(V3D_INTENA), + REGDEF(V3D_INTDIS), + REGDEF(V3D_CT0CS), + REGDEF(V3D_CT1CS), + REGDEF(V3D_CT0EA), + REGDEF(V3D_CT1EA), + REGDEF(V3D_CT0CA), + REGDEF(V3D_CT1CA), + REGDEF(V3D_CT00RA0), + REGDEF(V3D_CT01RA0), + REGDEF(V3D_CT0LC), + REGDEF(V3D_CT1LC), + REGDEF(V3D_CT0PC), + REGDEF(V3D_CT1PC), + REGDEF(V3D_PCS), + REGDEF(V3D_BFC), + REGDEF(V3D_RFC), + REGDEF(V3D_BPCA), + REGDEF(V3D_BPCS), + REGDEF(V3D_BPOA), + REGDEF(V3D_BPOS), + REGDEF(V3D_BXCF), + REGDEF(V3D_SQRSV0), + REGDEF(V3D_SQRSV1), + REGDEF(V3D_SQCNTL), + REGDEF(V3D_SRQPC), + REGDEF(V3D_SRQUA), + REGDEF(V3D_SRQUL), + REGDEF(V3D_SRQCS), + REGDEF(V3D_VPACNTL), + REGDEF(V3D_VPMBASE), + REGDEF(V3D_PCTRC), + REGDEF(V3D_PCTRE), + REGDEF(V3D_PCTR0), + REGDEF(V3D_PCTRS0), + REGDEF(V3D_PCTR1), + REGDEF(V3D_PCTRS1), + REGDEF(V3D_PCTR2), + REGDEF(V3D_PCTRS2), + REGDEF(V3D_PCTR3), + REGDEF(V3D_PCTRS3), + REGDEF(V3D_PCTR4), + REGDEF(V3D_PCTRS4), + REGDEF(V3D_PCTR5), + REGDEF(V3D_PCTRS5), + REGDEF(V3D_PCTR6), + REGDEF(V3D_PCTRS6), + REGDEF(V3D_PCTR7), + REGDEF(V3D_PCTRS7), + REGDEF(V3D_PCTR8), + REGDEF(V3D_PCTRS8), + REGDEF(V3D_PCTR9), + REGDEF(V3D_PCTRS9), + REGDEF(V3D_PCTR10), + REGDEF(V3D_PCTRS10), + REGDEF(V3D_PCTR11), + REGDEF(V3D_PCTRS11), + REGDEF(V3D_PCTR12), + REGDEF(V3D_PCTRS12), + REGDEF(V3D_PCTR13), + REGDEF(V3D_PCTRS13), + REGDEF(V3D_PCTR14), + REGDEF(V3D_PCTRS14), + REGDEF(V3D_PCTR15), + REGDEF(V3D_PCTRS15), + REGDEF(V3D_DBGE), + REGDEF(V3D_FDBGO), + REGDEF(V3D_FDBGB), + REGDEF(V3D_FDBGR), + REGDEF(V3D_FDBGS), + REGDEF(V3D_ERRSTAT), +}; + +int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused) +{ + struct drm_info_node *node = (struct drm_info_node *)m->private; + struct drm_device *dev = node->minor->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + int i; + + for (i = 0; i < ARRAY_SIZE(vc4_reg_defs); i++) { + seq_printf(m, "%s (0x%04x): 0x%08x\n", + vc4_reg_defs[i].name, vc4_reg_defs[i].reg, + V3D_READ(vc4_reg_defs[i].reg)); + } + + return 0; +} + +int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused) +{ + struct drm_info_node *node = (struct drm_info_node *)m->private; + struct drm_device *dev = node->minor->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + uint32_t ident1 = V3D_READ(V3D_IDENT1); + uint32_t nslc = VC4_GET_FIELD(ident1, V3D_IDENT1_NSLC); + uint32_t tups = VC4_GET_FIELD(ident1, V3D_IDENT1_TUPS); + uint32_t qups = VC4_GET_FIELD(ident1, V3D_IDENT1_QUPS); + + seq_printf(m, "Revision: %d\n", + VC4_GET_FIELD(ident1, V3D_IDENT1_REV)); + seq_printf(m, "Slices: %d\n", nslc); + seq_printf(m, "TMUs: %d\n", nslc * tups); + seq_printf(m, "QPUs: %d\n", nslc * qups); + seq_printf(m, "Semaphores: %d\n", + VC4_GET_FIELD(ident1, V3D_IDENT1_NSEM)); + + return 0; +} +#endif /* CONFIG_DEBUG_FS */ + +/* + * Asks the firmware to turn on power to the V3D engine. + * + * This may be doable with just the clocks interface, though this + * packet does some other register setup from the firmware, too. + */ +int +vc4_v3d_set_power(struct vc4_dev *vc4, bool on) +{ + if (on) + return pm_generic_poweroff(&vc4->v3d->pdev->dev); + else + return pm_generic_resume(&vc4->v3d->pdev->dev); +} + +static void vc4_v3d_init_hw(struct drm_device *dev) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + + /* Take all the memory that would have been reserved for user + * QPU programs, since we don't have an interface for running + * them, anyway. + */ + V3D_WRITE(V3D_VPMBASE, 0); +} + +static int vc4_v3d_bind(struct device *dev, struct device *master, void *data) +{ + struct platform_device *pdev = to_platform_device(dev); + struct drm_device *drm = dev_get_drvdata(master); + struct vc4_dev *vc4 = to_vc4_dev(drm); + struct vc4_v3d *v3d = NULL; + int ret; + + v3d = devm_kzalloc(&pdev->dev, sizeof(*v3d), GFP_KERNEL); + if (!v3d) + return -ENOMEM; + + v3d->pdev = pdev; + + v3d->regs = vc4_ioremap_regs(pdev, 0); + if (IS_ERR(v3d->regs)) + return PTR_ERR(v3d->regs); + + vc4->v3d = v3d; + + if (V3D_READ(V3D_IDENT0) != V3D_EXPECTED_IDENT0) { + DRM_ERROR("V3D_IDENT0 read 0x%08x instead of 0x%08x\n", + V3D_READ(V3D_IDENT0), V3D_EXPECTED_IDENT0); + return -EINVAL; + } + + /* Reset the binner overflow address/size at setup, to be sure + * we don't reuse an old one. + */ + V3D_WRITE(V3D_BPOA, 0); + V3D_WRITE(V3D_BPOS, 0); + + vc4_v3d_init_hw(drm); + + ret = drm_irq_install(drm, platform_get_irq(pdev, 0)); + if (ret) { + DRM_ERROR("Failed to install IRQ handler\n"); + return ret; + } + + return 0; +} + +static void vc4_v3d_unbind(struct device *dev, struct device *master, + void *data) +{ + struct drm_device *drm = dev_get_drvdata(master); + struct vc4_dev *vc4 = to_vc4_dev(drm); + + drm_irq_uninstall(drm); + + /* Disable the binner's overflow memory address, so the next + * driver probe (if any) doesn't try to reuse our old + * allocation. + */ + V3D_WRITE(V3D_BPOA, 0); + V3D_WRITE(V3D_BPOS, 0); + + vc4->v3d = NULL; +} + +static const struct component_ops vc4_v3d_ops = { + .bind = vc4_v3d_bind, + .unbind = vc4_v3d_unbind, +}; + +static int vc4_v3d_dev_probe(struct platform_device *pdev) +{ + return component_add(&pdev->dev, &vc4_v3d_ops); +} + +static int vc4_v3d_dev_remove(struct platform_device *pdev) +{ + component_del(&pdev->dev, &vc4_v3d_ops); + return 0; +} + +static const struct of_device_id vc4_v3d_dt_match[] = { + { .compatible = "brcm,vc4-v3d" }, + {} +}; + +struct platform_driver vc4_v3d_driver = { + .probe = vc4_v3d_dev_probe, + .remove = vc4_v3d_dev_remove, + .driver = { + .name = "vc4_v3d", + .of_match_table = vc4_v3d_dt_match, + }, +}; diff --git a/drivers/gpu/drm/vc4/vc4_validate.c b/drivers/gpu/drm/vc4/vc4_validate.c new file mode 100644 index 000000000000..0fb5b994b9dd --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_validate.c @@ -0,0 +1,900 @@ +/* + * Copyright © 2014 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +/** + * Command list validator for VC4. + * + * The VC4 has no IOMMU between it and system memory. So, a user with + * access to execute command lists could escalate privilege by + * overwriting system memory (drawing to it as a framebuffer) or + * reading system memory it shouldn't (reading it as a texture, or + * uniform data, or vertex data). + * + * This validates command lists to ensure that all accesses are within + * the bounds of the GEM objects referenced. It explicitly whitelists + * packets, and looks at the offsets in any address fields to make + * sure they're constrained within the BOs they reference. + * + * Note that because of the validation that's happening anyway, this + * is where GEM relocation processing happens. + */ + +#include "uapi/drm/vc4_drm.h" +#include "vc4_drv.h" +#include "vc4_packet.h" + +#define VALIDATE_ARGS \ + struct vc4_exec_info *exec, \ + void *validated, \ + void *untrusted + +/** Return the width in pixels of a 64-byte microtile. */ +static uint32_t +utile_width(int cpp) +{ + switch (cpp) { + case 1: + case 2: + return 8; + case 4: + return 4; + case 8: + return 2; + default: + DRM_ERROR("unknown cpp: %d\n", cpp); + return 1; + } +} + +/** Return the height in pixels of a 64-byte microtile. */ +static uint32_t +utile_height(int cpp) +{ + switch (cpp) { + case 1: + return 8; + case 2: + case 4: + case 8: + return 4; + default: + DRM_ERROR("unknown cpp: %d\n", cpp); + return 1; + } +} + +/** + * The texture unit decides what tiling format a particular miplevel is using + * this function, so we lay out our miptrees accordingly. + */ +static bool +size_is_lt(uint32_t width, uint32_t height, int cpp) +{ + return (width <= 4 * utile_width(cpp) || + height <= 4 * utile_height(cpp)); +} + +struct drm_gem_cma_object * +vc4_use_bo(struct vc4_exec_info *exec, uint32_t hindex) +{ + struct drm_gem_cma_object *obj; + struct vc4_bo *bo; + + if (hindex >= exec->bo_count) { + DRM_ERROR("BO index %d greater than BO count %d\n", + hindex, exec->bo_count); + return NULL; + } + obj = exec->bo[hindex]; + bo = to_vc4_bo(&obj->base); + + if (bo->validated_shader) { + DRM_ERROR("Trying to use shader BO as something other than " + "a shader\n"); + return NULL; + } + + return obj; +} + +static struct drm_gem_cma_object * +vc4_use_handle(struct vc4_exec_info *exec, uint32_t gem_handles_packet_index) +{ + return vc4_use_bo(exec, exec->bo_index[gem_handles_packet_index]); +} + +static bool +validate_bin_pos(struct vc4_exec_info *exec, void *untrusted, uint32_t pos) +{ + /* Note that the untrusted pointer passed to these functions is + * incremented past the packet byte. + */ + return (untrusted - 1 == exec->bin_u + pos); +} + +static uint32_t +gl_shader_rec_size(uint32_t pointer_bits) +{ + uint32_t attribute_count = pointer_bits & 7; + bool extended = pointer_bits & 8; + + if (attribute_count == 0) + attribute_count = 8; + + if (extended) + return 100 + attribute_count * 4; + else + return 36 + attribute_count * 8; +} + +bool +vc4_check_tex_size(struct vc4_exec_info *exec, struct drm_gem_cma_object *fbo, + uint32_t offset, uint8_t tiling_format, + uint32_t width, uint32_t height, uint8_t cpp) +{ + uint32_t aligned_width, aligned_height, stride, size; + uint32_t utile_w = utile_width(cpp); + uint32_t utile_h = utile_height(cpp); + + /* The shaded vertex format stores signed 12.4 fixed point + * (-2048,2047) offsets from the viewport center, so we should + * never have a render target larger than 4096. The texture + * unit can only sample from 2048x2048, so it's even more + * restricted. This lets us avoid worrying about overflow in + * our math. + */ + if (width > 4096 || height > 4096) { + DRM_ERROR("Surface dimesions (%d,%d) too large", width, height); + return false; + } + + switch (tiling_format) { + case VC4_TILING_FORMAT_LINEAR: + aligned_width = round_up(width, utile_w); + aligned_height = height; + break; + case VC4_TILING_FORMAT_T: + aligned_width = round_up(width, utile_w * 8); + aligned_height = round_up(height, utile_h * 8); + break; + case VC4_TILING_FORMAT_LT: + aligned_width = round_up(width, utile_w); + aligned_height = round_up(height, utile_h); + break; + default: + DRM_ERROR("buffer tiling %d unsupported\n", tiling_format); + return false; + } + + stride = aligned_width * cpp; + size = stride * aligned_height; + + if (size + offset < size || + size + offset > fbo->base.size) { + DRM_ERROR("Overflow in %dx%d (%dx%d) fbo size (%d + %d > %zd)\n", + width, height, + aligned_width, aligned_height, + size, offset, fbo->base.size); + return false; + } + + return true; +} + +static int +validate_flush(VALIDATE_ARGS) +{ + if (!validate_bin_pos(exec, untrusted, exec->args->bin_cl_size - 1)) { + DRM_ERROR("Bin CL must end with VC4_PACKET_FLUSH\n"); + return -EINVAL; + } + exec->found_flush = true; + + return 0; +} + +static int +validate_start_tile_binning(VALIDATE_ARGS) +{ + if (exec->found_start_tile_binning_packet) { + DRM_ERROR("Duplicate VC4_PACKET_START_TILE_BINNING\n"); + return -EINVAL; + } + exec->found_start_tile_binning_packet = true; + + if (!exec->found_tile_binning_mode_config_packet) { + DRM_ERROR("missing VC4_PACKET_TILE_BINNING_MODE_CONFIG\n"); + return -EINVAL; + } + + return 0; +} + +static int +validate_increment_semaphore(VALIDATE_ARGS) +{ + if (!validate_bin_pos(exec, untrusted, exec->args->bin_cl_size - 2)) { + DRM_ERROR("Bin CL must end with " + "VC4_PACKET_INCREMENT_SEMAPHORE\n"); + return -EINVAL; + } + exec->found_increment_semaphore_packet = true; + + return 0; +} + +static int +validate_indexed_prim_list(VALIDATE_ARGS) +{ + struct drm_gem_cma_object *ib; + uint32_t length = *(uint32_t *)(untrusted + 1); + uint32_t offset = *(uint32_t *)(untrusted + 5); + uint32_t max_index = *(uint32_t *)(untrusted + 9); + uint32_t index_size = (*(uint8_t *)(untrusted + 0) >> 4) ? 2 : 1; + struct vc4_shader_state *shader_state; + + /* Check overflow condition */ + if (exec->shader_state_count == 0) { + DRM_ERROR("shader state must precede primitives\n"); + return -EINVAL; + } + shader_state = &exec->shader_state[exec->shader_state_count - 1]; + + if (max_index > shader_state->max_index) + shader_state->max_index = max_index; + + ib = vc4_use_handle(exec, 0); + if (!ib) + return -EINVAL; + + if (offset > ib->base.size || + (ib->base.size - offset) / index_size < length) { + DRM_ERROR("IB access overflow (%d + %d*%d > %zd)\n", + offset, length, index_size, ib->base.size); + return -EINVAL; + } + + *(uint32_t *)(validated + 5) = ib->paddr + offset; + + return 0; +} + +static int +validate_gl_array_primitive(VALIDATE_ARGS) +{ + uint32_t length = *(uint32_t *)(untrusted + 1); + uint32_t base_index = *(uint32_t *)(untrusted + 5); + uint32_t max_index; + struct vc4_shader_state *shader_state; + + /* Check overflow condition */ + if (exec->shader_state_count == 0) { + DRM_ERROR("shader state must precede primitives\n"); + return -EINVAL; + } + shader_state = &exec->shader_state[exec->shader_state_count - 1]; + + if (length + base_index < length) { + DRM_ERROR("primitive vertex count overflow\n"); + return -EINVAL; + } + max_index = length + base_index - 1; + + if (max_index > shader_state->max_index) + shader_state->max_index = max_index; + + return 0; +} + +static int +validate_gl_shader_state(VALIDATE_ARGS) +{ + uint32_t i = exec->shader_state_count++; + + if (i >= exec->shader_state_size) { + DRM_ERROR("More requests for shader states than declared\n"); + return -EINVAL; + } + + exec->shader_state[i].addr = *(uint32_t *)untrusted; + exec->shader_state[i].max_index = 0; + + if (exec->shader_state[i].addr & ~0xf) { + DRM_ERROR("high bits set in GL shader rec reference\n"); + return -EINVAL; + } + + *(uint32_t *)validated = (exec->shader_rec_p + + exec->shader_state[i].addr); + + exec->shader_rec_p += + roundup(gl_shader_rec_size(exec->shader_state[i].addr), 16); + + return 0; +} + +static int +validate_tile_binning_config(VALIDATE_ARGS) +{ + struct drm_device *dev = exec->exec_bo->base.dev; + struct vc4_bo *tile_bo; + uint8_t flags; + uint32_t tile_state_size, tile_alloc_size; + uint32_t tile_count; + + if (exec->found_tile_binning_mode_config_packet) { + DRM_ERROR("Duplicate VC4_PACKET_TILE_BINNING_MODE_CONFIG\n"); + return -EINVAL; + } + exec->found_tile_binning_mode_config_packet = true; + + exec->bin_tiles_x = *(uint8_t *)(untrusted + 12); + exec->bin_tiles_y = *(uint8_t *)(untrusted + 13); + tile_count = exec->bin_tiles_x * exec->bin_tiles_y; + flags = *(uint8_t *)(untrusted + 14); + + if (exec->bin_tiles_x == 0 || + exec->bin_tiles_y == 0) { + DRM_ERROR("Tile binning config of %dx%d too small\n", + exec->bin_tiles_x, exec->bin_tiles_y); + return -EINVAL; + } + + if (flags & (VC4_BIN_CONFIG_DB_NON_MS | + VC4_BIN_CONFIG_TILE_BUFFER_64BIT)) { + DRM_ERROR("unsupported binning config flags 0x%02x\n", flags); + return -EINVAL; + } + + /* The tile state data array is 48 bytes per tile, and we put it at + * the start of a BO containing both it and the tile alloc. + */ + tile_state_size = 48 * tile_count; + + /* Since the tile alloc array will follow us, align. */ + exec->tile_alloc_offset = roundup(tile_state_size, 4096); + + *(uint8_t *)(validated + 14) = + ((flags & ~(VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_MASK | + VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_MASK)) | + VC4_BIN_CONFIG_AUTO_INIT_TSDA | + VC4_SET_FIELD(VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE_32, + VC4_BIN_CONFIG_ALLOC_INIT_BLOCK_SIZE) | + VC4_SET_FIELD(VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE_128, + VC4_BIN_CONFIG_ALLOC_BLOCK_SIZE)); + + /* Initial block size. */ + tile_alloc_size = 32 * tile_count; + + /* + * The initial allocation gets rounded to the next 256 bytes before + * the hardware starts fulfilling further allocations. + */ + tile_alloc_size = roundup(tile_alloc_size, 256); + + /* Add space for the extra allocations. This is what gets used first, + * before overflow memory. It must have at least 4096 bytes, but we + * want to avoid overflow memory usage if possible. + */ + tile_alloc_size += 1024 * 1024; + + tile_bo = vc4_bo_create(dev, exec->tile_alloc_offset + tile_alloc_size, + true); + exec->tile_bo = &tile_bo->base; + if (!exec->tile_bo) + return -ENOMEM; + list_add_tail(&tile_bo->unref_head, &exec->unref_list); + + /* tile alloc address. */ + *(uint32_t *)(validated + 0) = (exec->tile_bo->paddr + + exec->tile_alloc_offset); + /* tile alloc size. */ + *(uint32_t *)(validated + 4) = tile_alloc_size; + /* tile state address. */ + *(uint32_t *)(validated + 8) = exec->tile_bo->paddr; + + return 0; +} + +static int +validate_gem_handles(VALIDATE_ARGS) +{ + memcpy(exec->bo_index, untrusted, sizeof(exec->bo_index)); + return 0; +} + +#define VC4_DEFINE_PACKET(packet, func) \ + [packet] = { packet ## _SIZE, #packet, func } + +static const struct cmd_info { + uint16_t len; + const char *name; + int (*func)(struct vc4_exec_info *exec, void *validated, + void *untrusted); +} cmd_info[] = { + VC4_DEFINE_PACKET(VC4_PACKET_HALT, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_NOP, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_FLUSH, validate_flush), + VC4_DEFINE_PACKET(VC4_PACKET_FLUSH_ALL, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_START_TILE_BINNING, + validate_start_tile_binning), + VC4_DEFINE_PACKET(VC4_PACKET_INCREMENT_SEMAPHORE, + validate_increment_semaphore), + + VC4_DEFINE_PACKET(VC4_PACKET_GL_INDEXED_PRIMITIVE, + validate_indexed_prim_list), + VC4_DEFINE_PACKET(VC4_PACKET_GL_ARRAY_PRIMITIVE, + validate_gl_array_primitive), + + VC4_DEFINE_PACKET(VC4_PACKET_PRIMITIVE_LIST_FORMAT, NULL), + + VC4_DEFINE_PACKET(VC4_PACKET_GL_SHADER_STATE, validate_gl_shader_state), + + VC4_DEFINE_PACKET(VC4_PACKET_CONFIGURATION_BITS, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_FLAT_SHADE_FLAGS, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_POINT_SIZE, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_LINE_WIDTH, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_RHT_X_BOUNDARY, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_DEPTH_OFFSET, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_CLIP_WINDOW, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_VIEWPORT_OFFSET, NULL), + VC4_DEFINE_PACKET(VC4_PACKET_CLIPPER_XY_SCALING, NULL), + /* Note: The docs say this was also 105, but it was 106 in the + * initial userland code drop. + */ + VC4_DEFINE_PACKET(VC4_PACKET_CLIPPER_Z_SCALING, NULL), + + VC4_DEFINE_PACKET(VC4_PACKET_TILE_BINNING_MODE_CONFIG, + validate_tile_binning_config), + + VC4_DEFINE_PACKET(VC4_PACKET_GEM_HANDLES, validate_gem_handles), +}; + +int +vc4_validate_bin_cl(struct drm_device *dev, + void *validated, + void *unvalidated, + struct vc4_exec_info *exec) +{ + uint32_t len = exec->args->bin_cl_size; + uint32_t dst_offset = 0; + uint32_t src_offset = 0; + + while (src_offset < len) { + void *dst_pkt = validated + dst_offset; + void *src_pkt = unvalidated + src_offset; + u8 cmd = *(uint8_t *)src_pkt; + const struct cmd_info *info; + + if (cmd >= ARRAY_SIZE(cmd_info)) { + DRM_ERROR("0x%08x: packet %d out of bounds\n", + src_offset, cmd); + return -EINVAL; + } + + info = &cmd_info[cmd]; + if (!info->name) { + DRM_ERROR("0x%08x: packet %d invalid\n", + src_offset, cmd); + return -EINVAL; + } + + if (src_offset + info->len > len) { + DRM_ERROR("0x%08x: packet %d (%s) length 0x%08x " + "exceeds bounds (0x%08x)\n", + src_offset, cmd, info->name, info->len, + src_offset + len); + return -EINVAL; + } + + if (cmd != VC4_PACKET_GEM_HANDLES) + memcpy(dst_pkt, src_pkt, info->len); + + if (info->func && info->func(exec, + dst_pkt + 1, + src_pkt + 1)) { + DRM_ERROR("0x%08x: packet %d (%s) failed to validate\n", + src_offset, cmd, info->name); + return -EINVAL; + } + + src_offset += info->len; + /* GEM handle loading doesn't produce HW packets. */ + if (cmd != VC4_PACKET_GEM_HANDLES) + dst_offset += info->len; + + /* When the CL hits halt, it'll stop reading anything else. */ + if (cmd == VC4_PACKET_HALT) + break; + } + + exec->ct0ea = exec->ct0ca + dst_offset; + + if (!exec->found_start_tile_binning_packet) { + DRM_ERROR("Bin CL missing VC4_PACKET_START_TILE_BINNING\n"); + return -EINVAL; + } + + /* The bin CL must be ended with INCREMENT_SEMAPHORE and FLUSH. The + * semaphore is used to trigger the render CL to start up, and the + * FLUSH is what caps the bin lists with + * VC4_PACKET_RETURN_FROM_SUB_LIST (so they jump back to the main + * render CL when they get called to) and actually triggers the queued + * semaphore increment. + */ + if (!exec->found_increment_semaphore_packet || !exec->found_flush) { + DRM_ERROR("Bin CL missing VC4_PACKET_INCREMENT_SEMAPHORE + " + "VC4_PACKET_FLUSH\n"); + return -EINVAL; + } + + return 0; +} + +static bool +reloc_tex(struct vc4_exec_info *exec, + void *uniform_data_u, + struct vc4_texture_sample_info *sample, + uint32_t texture_handle_index) + +{ + struct drm_gem_cma_object *tex; + uint32_t p0 = *(uint32_t *)(uniform_data_u + sample->p_offset[0]); + uint32_t p1 = *(uint32_t *)(uniform_data_u + sample->p_offset[1]); + uint32_t p2 = (sample->p_offset[2] != ~0 ? + *(uint32_t *)(uniform_data_u + sample->p_offset[2]) : 0); + uint32_t p3 = (sample->p_offset[3] != ~0 ? + *(uint32_t *)(uniform_data_u + sample->p_offset[3]) : 0); + uint32_t *validated_p0 = exec->uniforms_v + sample->p_offset[0]; + uint32_t offset = p0 & VC4_TEX_P0_OFFSET_MASK; + uint32_t miplevels = VC4_GET_FIELD(p0, VC4_TEX_P0_MIPLVLS); + uint32_t width = VC4_GET_FIELD(p1, VC4_TEX_P1_WIDTH); + uint32_t height = VC4_GET_FIELD(p1, VC4_TEX_P1_HEIGHT); + uint32_t cpp, tiling_format, utile_w, utile_h; + uint32_t i; + uint32_t cube_map_stride = 0; + enum vc4_texture_data_type type; + + tex = vc4_use_bo(exec, texture_handle_index); + if (!tex) + return false; + + if (sample->is_direct) { + uint32_t remaining_size = tex->base.size - p0; + + if (p0 > tex->base.size - 4) { + DRM_ERROR("UBO offset greater than UBO size\n"); + goto fail; + } + if (p1 > remaining_size - 4) { + DRM_ERROR("UBO clamp would allow reads " + "outside of UBO\n"); + goto fail; + } + *validated_p0 = tex->paddr + p0; + return true; + } + + if (width == 0) + width = 2048; + if (height == 0) + height = 2048; + + if (p0 & VC4_TEX_P0_CMMODE_MASK) { + if (VC4_GET_FIELD(p2, VC4_TEX_P2_PTYPE) == + VC4_TEX_P2_PTYPE_CUBE_MAP_STRIDE) + cube_map_stride = p2 & VC4_TEX_P2_CMST_MASK; + if (VC4_GET_FIELD(p3, VC4_TEX_P2_PTYPE) == + VC4_TEX_P2_PTYPE_CUBE_MAP_STRIDE) { + if (cube_map_stride) { + DRM_ERROR("Cube map stride set twice\n"); + goto fail; + } + + cube_map_stride = p3 & VC4_TEX_P2_CMST_MASK; + } + if (!cube_map_stride) { + DRM_ERROR("Cube map stride not set\n"); + goto fail; + } + } + + type = (VC4_GET_FIELD(p0, VC4_TEX_P0_TYPE) | + (VC4_GET_FIELD(p1, VC4_TEX_P1_TYPE4) << 4)); + + switch (type) { + case VC4_TEXTURE_TYPE_RGBA8888: + case VC4_TEXTURE_TYPE_RGBX8888: + case VC4_TEXTURE_TYPE_RGBA32R: + cpp = 4; + break; + case VC4_TEXTURE_TYPE_RGBA4444: + case VC4_TEXTURE_TYPE_RGBA5551: + case VC4_TEXTURE_TYPE_RGB565: + case VC4_TEXTURE_TYPE_LUMALPHA: + case VC4_TEXTURE_TYPE_S16F: + case VC4_TEXTURE_TYPE_S16: + cpp = 2; + break; + case VC4_TEXTURE_TYPE_LUMINANCE: + case VC4_TEXTURE_TYPE_ALPHA: + case VC4_TEXTURE_TYPE_S8: + cpp = 1; + break; + case VC4_TEXTURE_TYPE_ETC1: + case VC4_TEXTURE_TYPE_BW1: + case VC4_TEXTURE_TYPE_A4: + case VC4_TEXTURE_TYPE_A1: + case VC4_TEXTURE_TYPE_RGBA64: + case VC4_TEXTURE_TYPE_YUV422R: + default: + DRM_ERROR("Texture format %d unsupported\n", type); + goto fail; + } + utile_w = utile_width(cpp); + utile_h = utile_height(cpp); + + if (type == VC4_TEXTURE_TYPE_RGBA32R) { + tiling_format = VC4_TILING_FORMAT_LINEAR; + } else { + if (size_is_lt(width, height, cpp)) + tiling_format = VC4_TILING_FORMAT_LT; + else + tiling_format = VC4_TILING_FORMAT_T; + } + + if (!vc4_check_tex_size(exec, tex, offset + cube_map_stride * 5, + tiling_format, width, height, cpp)) { + goto fail; + } + + /* The mipmap levels are stored before the base of the texture. Make + * sure there is actually space in the BO. + */ + for (i = 1; i <= miplevels; i++) { + uint32_t level_width = max(width >> i, 1u); + uint32_t level_height = max(height >> i, 1u); + uint32_t aligned_width, aligned_height; + uint32_t level_size; + + /* Once the levels get small enough, they drop from T to LT. */ + if (tiling_format == VC4_TILING_FORMAT_T && + size_is_lt(level_width, level_height, cpp)) { + tiling_format = VC4_TILING_FORMAT_LT; + } + + switch (tiling_format) { + case VC4_TILING_FORMAT_T: + aligned_width = round_up(level_width, utile_w * 8); + aligned_height = round_up(level_height, utile_h * 8); + break; + case VC4_TILING_FORMAT_LT: + aligned_width = round_up(level_width, utile_w); + aligned_height = round_up(level_height, utile_h); + break; + default: + aligned_width = round_up(level_width, utile_w); + aligned_height = level_height; + break; + } + + level_size = aligned_width * cpp * aligned_height; + + if (offset < level_size) { + DRM_ERROR("Level %d (%dx%d -> %dx%d) size %db " + "overflowed buffer bounds (offset %d)\n", + i, level_width, level_height, + aligned_width, aligned_height, + level_size, offset); + goto fail; + } + + offset -= level_size; + } + + *validated_p0 = tex->paddr + p0; + + return true; + fail: + DRM_INFO("Texture p0 at %d: 0x%08x\n", sample->p_offset[0], p0); + DRM_INFO("Texture p1 at %d: 0x%08x\n", sample->p_offset[1], p1); + DRM_INFO("Texture p2 at %d: 0x%08x\n", sample->p_offset[2], p2); + DRM_INFO("Texture p3 at %d: 0x%08x\n", sample->p_offset[3], p3); + return false; +} + +static int +validate_gl_shader_rec(struct drm_device *dev, + struct vc4_exec_info *exec, + struct vc4_shader_state *state) +{ + uint32_t *src_handles; + void *pkt_u, *pkt_v; + static const uint32_t shader_reloc_offsets[] = { + 4, /* fs */ + 16, /* vs */ + 28, /* cs */ + }; + uint32_t shader_reloc_count = ARRAY_SIZE(shader_reloc_offsets); + struct drm_gem_cma_object *bo[shader_reloc_count + 8]; + uint32_t nr_attributes, nr_relocs, packet_size; + int i; + + nr_attributes = state->addr & 0x7; + if (nr_attributes == 0) + nr_attributes = 8; + packet_size = gl_shader_rec_size(state->addr); + + nr_relocs = ARRAY_SIZE(shader_reloc_offsets) + nr_attributes; + if (nr_relocs * 4 > exec->shader_rec_size) { + DRM_ERROR("overflowed shader recs reading %d handles " + "from %d bytes left\n", + nr_relocs, exec->shader_rec_size); + return -EINVAL; + } + src_handles = exec->shader_rec_u; + exec->shader_rec_u += nr_relocs * 4; + exec->shader_rec_size -= nr_relocs * 4; + + if (packet_size > exec->shader_rec_size) { + DRM_ERROR("overflowed shader recs copying %db packet " + "from %d bytes left\n", + packet_size, exec->shader_rec_size); + return -EINVAL; + } + pkt_u = exec->shader_rec_u; + pkt_v = exec->shader_rec_v; + memcpy(pkt_v, pkt_u, packet_size); + exec->shader_rec_u += packet_size; + /* Shader recs have to be aligned to 16 bytes (due to the attribute + * flags being in the low bytes), so round the next validated shader + * rec address up. This should be safe, since we've got so many + * relocations in a shader rec packet. + */ + BUG_ON(roundup(packet_size, 16) - packet_size > nr_relocs * 4); + exec->shader_rec_v += roundup(packet_size, 16); + exec->shader_rec_size -= packet_size; + + if (!(*(uint16_t *)pkt_u & VC4_SHADER_FLAG_FS_SINGLE_THREAD)) { + DRM_ERROR("Multi-threaded fragment shaders not supported.\n"); + return -EINVAL; + } + + for (i = 0; i < shader_reloc_count; i++) { + if (src_handles[i] > exec->bo_count) { + DRM_ERROR("Shader handle %d too big\n", src_handles[i]); + return -EINVAL; + } + + bo[i] = exec->bo[src_handles[i]]; + if (!bo[i]) + return -EINVAL; + } + for (i = shader_reloc_count; i < nr_relocs; i++) { + bo[i] = vc4_use_bo(exec, src_handles[i]); + if (!bo[i]) + return -EINVAL; + } + + for (i = 0; i < shader_reloc_count; i++) { + struct vc4_validated_shader_info *validated_shader; + uint32_t o = shader_reloc_offsets[i]; + uint32_t src_offset = *(uint32_t *)(pkt_u + o); + uint32_t *texture_handles_u; + void *uniform_data_u; + uint32_t tex; + + *(uint32_t *)(pkt_v + o) = bo[i]->paddr + src_offset; + + if (src_offset != 0) { + DRM_ERROR("Shaders must be at offset 0 of " + "the BO.\n"); + return -EINVAL; + } + + validated_shader = to_vc4_bo(&bo[i]->base)->validated_shader; + if (!validated_shader) + return -EINVAL; + + if (validated_shader->uniforms_src_size > + exec->uniforms_size) { + DRM_ERROR("Uniforms src buffer overflow\n"); + return -EINVAL; + } + + texture_handles_u = exec->uniforms_u; + uniform_data_u = (texture_handles_u + + validated_shader->num_texture_samples); + + memcpy(exec->uniforms_v, uniform_data_u, + validated_shader->uniforms_size); + + for (tex = 0; + tex < validated_shader->num_texture_samples; + tex++) { + if (!reloc_tex(exec, + uniform_data_u, + &validated_shader->texture_samples[tex], + texture_handles_u[tex])) { + return -EINVAL; + } + } + + *(uint32_t *)(pkt_v + o + 4) = exec->uniforms_p; + + exec->uniforms_u += validated_shader->uniforms_src_size; + exec->uniforms_v += validated_shader->uniforms_size; + exec->uniforms_p += validated_shader->uniforms_size; + } + + for (i = 0; i < nr_attributes; i++) { + struct drm_gem_cma_object *vbo = + bo[ARRAY_SIZE(shader_reloc_offsets) + i]; + uint32_t o = 36 + i * 8; + uint32_t offset = *(uint32_t *)(pkt_u + o + 0); + uint32_t attr_size = *(uint8_t *)(pkt_u + o + 4) + 1; + uint32_t stride = *(uint8_t *)(pkt_u + o + 5); + uint32_t max_index; + + if (state->addr & 0x8) + stride |= (*(uint32_t *)(pkt_u + 100 + i * 4)) & ~0xff; + + if (vbo->base.size < offset || + vbo->base.size - offset < attr_size) { + DRM_ERROR("BO offset overflow (%d + %d > %d)\n", + offset, attr_size, vbo->base.size); + return -EINVAL; + } + + if (stride != 0) { + max_index = ((vbo->base.size - offset - attr_size) / + stride); + if (state->max_index > max_index) { + DRM_ERROR("primitives use index %d out of " + "supplied %d\n", + state->max_index, max_index); + return -EINVAL; + } + } + + *(uint32_t *)(pkt_v + o) = vbo->paddr + offset; + } + + return 0; +} + +int +vc4_validate_shader_recs(struct drm_device *dev, + struct vc4_exec_info *exec) +{ + uint32_t i; + int ret = 0; + + for (i = 0; i < exec->shader_state_count; i++) { + ret = validate_gl_shader_rec(dev, exec, &exec->shader_state[i]); + if (ret) + return ret; + } + + return ret; +} diff --git a/drivers/gpu/drm/vc4/vc4_validate_shaders.c b/drivers/gpu/drm/vc4/vc4_validate_shaders.c new file mode 100644 index 000000000000..f67124b4c534 --- /dev/null +++ b/drivers/gpu/drm/vc4/vc4_validate_shaders.c @@ -0,0 +1,513 @@ +/* + * Copyright © 2014 Broadcom + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +/** + * DOC: Shader validator for VC4. + * + * The VC4 has no IOMMU between it and system memory, so a user with + * access to execute shaders could escalate privilege by overwriting + * system memory (using the VPM write address register in the + * general-purpose DMA mode) or reading system memory it shouldn't + * (reading it as a texture, or uniform data, or vertex data). + * + * This walks over a shader BO, ensuring that its accesses are + * appropriately bounded, and recording how many texture accesses are + * made and where so that we can do relocations for them in the + * uniform stream. + */ + +#include "vc4_drv.h" +#include "vc4_qpu_defines.h" + +struct vc4_shader_validation_state { + struct vc4_texture_sample_info tmu_setup[2]; + int tmu_write_count[2]; + + /* For registers that were last written to by a MIN instruction with + * one argument being a uniform, the address of the uniform. + * Otherwise, ~0. + * + * This is used for the validation of direct address memory reads. + */ + uint32_t live_min_clamp_offsets[32 + 32 + 4]; + bool live_max_clamp_regs[32 + 32 + 4]; +}; + +static uint32_t +waddr_to_live_reg_index(uint32_t waddr, bool is_b) +{ + if (waddr < 32) { + if (is_b) + return 32 + waddr; + else + return waddr; + } else if (waddr <= QPU_W_ACC3) { + return 64 + waddr - QPU_W_ACC0; + } else { + return ~0; + } +} + +static uint32_t +raddr_add_a_to_live_reg_index(uint64_t inst) +{ + uint32_t sig = QPU_GET_FIELD(inst, QPU_SIG); + uint32_t add_a = QPU_GET_FIELD(inst, QPU_ADD_A); + uint32_t raddr_a = QPU_GET_FIELD(inst, QPU_RADDR_A); + uint32_t raddr_b = QPU_GET_FIELD(inst, QPU_RADDR_B); + + if (add_a == QPU_MUX_A) + return raddr_a; + else if (add_a == QPU_MUX_B && sig != QPU_SIG_SMALL_IMM) + return 32 + raddr_b; + else if (add_a <= QPU_MUX_R3) + return 64 + add_a; + else + return ~0; +} + +static bool +is_tmu_submit(uint32_t waddr) +{ + return (waddr == QPU_W_TMU0_S || + waddr == QPU_W_TMU1_S); +} + +static bool +is_tmu_write(uint32_t waddr) +{ + return (waddr >= QPU_W_TMU0_S && + waddr <= QPU_W_TMU1_B); +} + +static bool +record_texture_sample(struct vc4_validated_shader_info *validated_shader, + struct vc4_shader_validation_state *validation_state, + int tmu) +{ + uint32_t s = validated_shader->num_texture_samples; + int i; + struct vc4_texture_sample_info *temp_samples; + + temp_samples = krealloc(validated_shader->texture_samples, + (s + 1) * sizeof(*temp_samples), + GFP_KERNEL); + if (!temp_samples) + return false; + + memcpy(&temp_samples[s], + &validation_state->tmu_setup[tmu], + sizeof(*temp_samples)); + + validated_shader->num_texture_samples = s + 1; + validated_shader->texture_samples = temp_samples; + + for (i = 0; i < 4; i++) + validation_state->tmu_setup[tmu].p_offset[i] = ~0; + + return true; +} + +static bool +check_tmu_write(uint64_t inst, + struct vc4_validated_shader_info *validated_shader, + struct vc4_shader_validation_state *validation_state, + bool is_mul) +{ + uint32_t waddr = (is_mul ? + QPU_GET_FIELD(inst, QPU_WADDR_MUL) : + QPU_GET_FIELD(inst, QPU_WADDR_ADD)); + uint32_t raddr_a = QPU_GET_FIELD(inst, QPU_RADDR_A); + uint32_t raddr_b = QPU_GET_FIELD(inst, QPU_RADDR_B); + int tmu = waddr > QPU_W_TMU0_B; + bool submit = is_tmu_submit(waddr); + bool is_direct = submit && validation_state->tmu_write_count[tmu] == 0; + uint32_t sig = QPU_GET_FIELD(inst, QPU_SIG); + + if (is_direct) { + uint32_t add_b = QPU_GET_FIELD(inst, QPU_ADD_B); + uint32_t clamp_reg, clamp_offset; + + if (sig == QPU_SIG_SMALL_IMM) { + DRM_ERROR("direct TMU read used small immediate\n"); + return false; + } + + /* Make sure that this texture load is an add of the base + * address of the UBO to a clamped offset within the UBO. + */ + if (is_mul || + QPU_GET_FIELD(inst, QPU_OP_ADD) != QPU_A_ADD) { + DRM_ERROR("direct TMU load wasn't an add\n"); + return false; + } + + /* We assert that the the clamped address is the first + * argument, and the UBO base address is the second argument. + * This is arbitrary, but simpler than supporting flipping the + * two either way. + */ + clamp_reg = raddr_add_a_to_live_reg_index(inst); + if (clamp_reg == ~0) { + DRM_ERROR("direct TMU load wasn't clamped\n"); + return false; + } + + clamp_offset = validation_state->live_min_clamp_offsets[clamp_reg]; + if (clamp_offset == ~0) { + DRM_ERROR("direct TMU load wasn't clamped\n"); + return false; + } + + /* Store the clamp value's offset in p1 (see reloc_tex() in + * vc4_validate.c). + */ + validation_state->tmu_setup[tmu].p_offset[1] = + clamp_offset; + + if (!(add_b == QPU_MUX_A && raddr_a == QPU_R_UNIF) && + !(add_b == QPU_MUX_B && raddr_b == QPU_R_UNIF)) { + DRM_ERROR("direct TMU load didn't add to a uniform\n"); + return false; + } + + validation_state->tmu_setup[tmu].is_direct = true; + } else { + if (raddr_a == QPU_R_UNIF || (sig != QPU_SIG_SMALL_IMM && + raddr_b == QPU_R_UNIF)) { + DRM_ERROR("uniform read in the same instruction as " + "texture setup.\n"); + return false; + } + } + + if (validation_state->tmu_write_count[tmu] >= 4) { + DRM_ERROR("TMU%d got too many parameters before dispatch\n", + tmu); + return false; + } + validation_state->tmu_setup[tmu].p_offset[validation_state->tmu_write_count[tmu]] = + validated_shader->uniforms_size; + validation_state->tmu_write_count[tmu]++; + /* Since direct uses a RADDR uniform reference, it will get counted in + * check_instruction_reads() + */ + if (!is_direct) + validated_shader->uniforms_size += 4; + + if (submit) { + if (!record_texture_sample(validated_shader, + validation_state, tmu)) { + return false; + } + + validation_state->tmu_write_count[tmu] = 0; + } + + return true; +} + +static bool +check_reg_write(uint64_t inst, + struct vc4_validated_shader_info *validated_shader, + struct vc4_shader_validation_state *validation_state, + bool is_mul) +{ + uint32_t waddr = (is_mul ? + QPU_GET_FIELD(inst, QPU_WADDR_MUL) : + QPU_GET_FIELD(inst, QPU_WADDR_ADD)); + + switch (waddr) { + case QPU_W_UNIFORMS_ADDRESS: + /* XXX: We'll probably need to support this for reladdr, but + * it's definitely a security-related one. + */ + DRM_ERROR("uniforms address load unsupported\n"); + return false; + + case QPU_W_TLB_COLOR_MS: + case QPU_W_TLB_COLOR_ALL: + case QPU_W_TLB_Z: + /* These only interact with the tile buffer, not main memory, + * so they're safe. + */ + return true; + + case QPU_W_TMU0_S: + case QPU_W_TMU0_T: + case QPU_W_TMU0_R: + case QPU_W_TMU0_B: + case QPU_W_TMU1_S: + case QPU_W_TMU1_T: + case QPU_W_TMU1_R: + case QPU_W_TMU1_B: + return check_tmu_write(inst, validated_shader, validation_state, + is_mul); + + case QPU_W_HOST_INT: + case QPU_W_TMU_NOSWAP: + case QPU_W_TLB_ALPHA_MASK: + case QPU_W_MUTEX_RELEASE: + /* XXX: I haven't thought about these, so don't support them + * for now. + */ + DRM_ERROR("Unsupported waddr %d\n", waddr); + return false; + + case QPU_W_VPM_ADDR: + DRM_ERROR("General VPM DMA unsupported\n"); + return false; + + case QPU_W_VPM: + case QPU_W_VPMVCD_SETUP: + /* We allow VPM setup in general, even including VPM DMA + * configuration setup, because the (unsafe) DMA can only be + * triggered by QPU_W_VPM_ADDR writes. + */ + return true; + + case QPU_W_TLB_STENCIL_SETUP: + return true; + } + + return true; +} + +static void +track_live_clamps(uint64_t inst, + struct vc4_validated_shader_info *validated_shader, + struct vc4_shader_validation_state *validation_state) +{ + uint32_t op_add = QPU_GET_FIELD(inst, QPU_OP_ADD); + uint32_t waddr_add = QPU_GET_FIELD(inst, QPU_WADDR_ADD); + uint32_t waddr_mul = QPU_GET_FIELD(inst, QPU_WADDR_MUL); + uint32_t cond_add = QPU_GET_FIELD(inst, QPU_COND_ADD); + uint32_t add_a = QPU_GET_FIELD(inst, QPU_ADD_A); + uint32_t add_b = QPU_GET_FIELD(inst, QPU_ADD_B); + uint32_t raddr_a = QPU_GET_FIELD(inst, QPU_RADDR_A); + uint32_t raddr_b = QPU_GET_FIELD(inst, QPU_RADDR_B); + uint32_t sig = QPU_GET_FIELD(inst, QPU_SIG); + bool ws = inst & QPU_WS; + uint32_t lri_add_a, lri_add, lri_mul; + bool add_a_is_min_0; + + /* Check whether OP_ADD's A argumennt comes from a live MAX(x, 0), + * before we clear previous live state. + */ + lri_add_a = raddr_add_a_to_live_reg_index(inst); + add_a_is_min_0 = (lri_add_a != ~0 && + validation_state->live_max_clamp_regs[lri_add_a]); + + /* Clear live state for registers written by our instruction. */ + lri_add = waddr_to_live_reg_index(waddr_add, ws); + lri_mul = waddr_to_live_reg_index(waddr_mul, !ws); + if (lri_mul != ~0) { + validation_state->live_max_clamp_regs[lri_mul] = false; + validation_state->live_min_clamp_offsets[lri_mul] = ~0; + } + if (lri_add != ~0) { + validation_state->live_max_clamp_regs[lri_add] = false; + validation_state->live_min_clamp_offsets[lri_add] = ~0; + } else { + /* Nothing further to do for live tracking, since only ADDs + * generate new live clamp registers. + */ + return; + } + + /* Now, handle remaining live clamp tracking for the ADD operation. */ + + if (cond_add != QPU_COND_ALWAYS) + return; + + if (op_add == QPU_A_MAX) { + /* Track live clamps of a value to a minimum of 0 (in either + * arg). + */ + if (sig != QPU_SIG_SMALL_IMM || raddr_b != 0 || + (add_a != QPU_MUX_B && add_b != QPU_MUX_B)) { + return; + } + + validation_state->live_max_clamp_regs[lri_add] = true; + } else if (op_add == QPU_A_MIN) { + /* Track live clamps of a value clamped to a minimum of 0 and + * a maximum of some uniform's offset. + */ + if (!add_a_is_min_0) + return; + + if (!(add_b == QPU_MUX_A && raddr_a == QPU_R_UNIF) && + !(add_b == QPU_MUX_B && raddr_b == QPU_R_UNIF && + sig != QPU_SIG_SMALL_IMM)) { + return; + } + + validation_state->live_min_clamp_offsets[lri_add] = + validated_shader->uniforms_size; + } +} + +static bool +check_instruction_writes(uint64_t inst, + struct vc4_validated_shader_info *validated_shader, + struct vc4_shader_validation_state *validation_state) +{ + uint32_t waddr_add = QPU_GET_FIELD(inst, QPU_WADDR_ADD); + uint32_t waddr_mul = QPU_GET_FIELD(inst, QPU_WADDR_MUL); + bool ok; + + if (is_tmu_write(waddr_add) && is_tmu_write(waddr_mul)) { + DRM_ERROR("ADD and MUL both set up textures\n"); + return false; + } + + ok = (check_reg_write(inst, validated_shader, validation_state, + false) && + check_reg_write(inst, validated_shader, validation_state, + true)); + + track_live_clamps(inst, validated_shader, validation_state); + + return ok; +} + +static bool +check_instruction_reads(uint64_t inst, + struct vc4_validated_shader_info *validated_shader) +{ + uint32_t raddr_a = QPU_GET_FIELD(inst, QPU_RADDR_A); + uint32_t raddr_b = QPU_GET_FIELD(inst, QPU_RADDR_B); + uint32_t sig = QPU_GET_FIELD(inst, QPU_SIG); + + if (raddr_a == QPU_R_UNIF || + (raddr_b == QPU_R_UNIF && sig != QPU_SIG_SMALL_IMM)) { + /* This can't overflow the uint32_t, because we're reading 8 + * bytes of instruction to increment by 4 here, so we'd + * already be OOM. + */ + validated_shader->uniforms_size += 4; + } + + return true; +} + +struct vc4_validated_shader_info * +vc4_validate_shader(struct drm_gem_cma_object *shader_obj) +{ + bool found_shader_end = false; + int shader_end_ip = 0; + uint32_t ip, max_ip; + uint64_t *shader; + struct vc4_validated_shader_info *validated_shader; + struct vc4_shader_validation_state validation_state; + int i; + + memset(&validation_state, 0, sizeof(validation_state)); + + for (i = 0; i < 8; i++) + validation_state.tmu_setup[i / 4].p_offset[i % 4] = ~0; + for (i = 0; i < ARRAY_SIZE(validation_state.live_min_clamp_offsets); i++) + validation_state.live_min_clamp_offsets[i] = ~0; + + shader = shader_obj->vaddr; + max_ip = shader_obj->base.size / sizeof(uint64_t); + + validated_shader = kcalloc(1, sizeof(*validated_shader), GFP_KERNEL); + if (!validated_shader) + return NULL; + + for (ip = 0; ip < max_ip; ip++) { + uint64_t inst = shader[ip]; + uint32_t sig = QPU_GET_FIELD(inst, QPU_SIG); + + switch (sig) { + case QPU_SIG_NONE: + case QPU_SIG_WAIT_FOR_SCOREBOARD: + case QPU_SIG_SCOREBOARD_UNLOCK: + case QPU_SIG_COLOR_LOAD: + case QPU_SIG_LOAD_TMU0: + case QPU_SIG_LOAD_TMU1: + case QPU_SIG_PROG_END: + case QPU_SIG_SMALL_IMM: + if (!check_instruction_writes(inst, validated_shader, + &validation_state)) { + DRM_ERROR("Bad write at ip %d\n", ip); + goto fail; + } + + if (!check_instruction_reads(inst, validated_shader)) + goto fail; + + if (sig == QPU_SIG_PROG_END) { + found_shader_end = true; + shader_end_ip = ip; + } + + break; + + case QPU_SIG_LOAD_IMM: + if (!check_instruction_writes(inst, validated_shader, + &validation_state)) { + DRM_ERROR("Bad LOAD_IMM write at ip %d\n", ip); + goto fail; + } + break; + + default: + DRM_ERROR("Unsupported QPU signal %d at " + "instruction %d\n", sig, ip); + goto fail; + } + + /* There are two delay slots after program end is signaled + * that are still executed, then we're finished. + */ + if (found_shader_end && ip == shader_end_ip + 2) + break; + } + + if (ip == max_ip) { + DRM_ERROR("shader failed to terminate before " + "shader BO end at %zd\n", + shader_obj->base.size); + goto fail; + } + + /* Again, no chance of integer overflow here because the worst case + * scenario is 8 bytes of uniforms plus handles per 8-byte + * instruction. + */ + validated_shader->uniforms_src_size = + (validated_shader->uniforms_size + + 4 * validated_shader->num_texture_samples); + + return validated_shader; + +fail: + if (validated_shader) { + kfree(validated_shader->texture_samples); + kfree(validated_shader); + } + return NULL; +} diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c index f545913a56c7..a165f03eaa79 100644 --- a/drivers/gpu/drm/virtio/virtgpu_display.c +++ b/drivers/gpu/drm/virtio/virtgpu_display.c @@ -215,7 +215,7 @@ static const struct drm_framebuffer_funcs virtio_gpu_fb_funcs = { int virtio_gpu_framebuffer_init(struct drm_device *dev, struct virtio_gpu_framebuffer *vgfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj) { int ret; @@ -374,16 +374,6 @@ static const struct drm_connector_helper_funcs virtio_gpu_conn_helper_funcs = { .best_encoder = virtio_gpu_best_encoder, }; -static void virtio_gpu_conn_save(struct drm_connector *connector) -{ - DRM_DEBUG("\n"); -} - -static void virtio_gpu_conn_restore(struct drm_connector *connector) -{ - DRM_DEBUG("\n"); -} - static enum drm_connector_status virtio_gpu_conn_detect( struct drm_connector *connector, bool force) @@ -409,8 +399,6 @@ static void virtio_gpu_conn_destroy(struct drm_connector *connector) static const struct drm_connector_funcs virtio_gpu_connector_funcs = { .dpms = drm_atomic_helper_connector_dpms, - .save = virtio_gpu_conn_save, - .restore = virtio_gpu_conn_restore, .detect = virtio_gpu_conn_detect, .fill_modes = drm_helper_probe_single_connector_modes, .destroy = virtio_gpu_conn_destroy, @@ -443,7 +431,7 @@ static int vgdev_output_init(struct virtio_gpu_device *vgdev, int index) if (IS_ERR(plane)) return PTR_ERR(plane); drm_crtc_init_with_planes(dev, crtc, plane, NULL, - &virtio_gpu_crtc_funcs); + &virtio_gpu_crtc_funcs, NULL); drm_mode_crtc_set_gamma_size(crtc, 256); drm_crtc_helper_add(crtc, &virtio_gpu_crtc_helper_funcs); plane->crtc = crtc; @@ -453,7 +441,7 @@ static int vgdev_output_init(struct virtio_gpu_device *vgdev, int index) drm_connector_helper_add(connector, &virtio_gpu_conn_helper_funcs); drm_encoder_init(dev, encoder, &virtio_gpu_enc_funcs, - DRM_MODE_ENCODER_VIRTUAL); + DRM_MODE_ENCODER_VIRTUAL, NULL); drm_encoder_helper_add(encoder, &virtio_gpu_enc_helper_funcs); encoder->possible_crtcs = 1 << index; @@ -465,7 +453,7 @@ static int vgdev_output_init(struct virtio_gpu_device *vgdev, int index) static struct drm_framebuffer * virtio_gpu_user_framebuffer_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd) + const struct drm_mode_fb_cmd2 *mode_cmd) { struct drm_gem_object *obj = NULL; struct virtio_gpu_framebuffer *virtio_gpu_fb; diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h index 79f0abe69b64..8f486f4c7023 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.h +++ b/drivers/gpu/drm/virtio/virtgpu_drv.h @@ -328,7 +328,7 @@ void virtio_gpu_dequeue_fence_func(struct work_struct *work); /* virtio_gpu_display.c */ int virtio_gpu_framebuffer_init(struct drm_device *dev, struct virtio_gpu_framebuffer *vgfb, - struct drm_mode_fb_cmd2 *mode_cmd, + const struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object *obj); int virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev); void virtio_gpu_modeset_fini(struct virtio_gpu_device *vgdev); diff --git a/drivers/gpu/drm/virtio/virtgpu_fb.c b/drivers/gpu/drm/virtio/virtgpu_fb.c index 6a81e084593b..2242a80866a9 100644 --- a/drivers/gpu/drm/virtio/virtgpu_fb.c +++ b/drivers/gpu/drm/virtio/virtgpu_fb.c @@ -32,7 +32,6 @@ struct virtio_gpu_fbdev { struct drm_fb_helper helper; struct virtio_gpu_framebuffer vgfb; - struct list_head fbdev_list; struct virtio_gpu_device *vgdev; struct delayed_work work; }; diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c index 4a74129c5708..572fb351feab 100644 --- a/drivers/gpu/drm/virtio/virtgpu_plane.c +++ b/drivers/gpu/drm/virtio/virtgpu_plane.c @@ -107,7 +107,7 @@ struct drm_plane *virtio_gpu_plane_init(struct virtio_gpu_device *vgdev, &virtio_gpu_plane_funcs, virtio_gpu_formats, ARRAY_SIZE(virtio_gpu_formats), - DRM_PLANE_TYPE_PRIMARY); + DRM_PLANE_TYPE_PRIMARY, NULL); if (ret) goto err_plane_init; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 9fcd7f82995c..9394c3535e85 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -930,7 +930,7 @@ vmw_kms_new_framebuffer(struct vmw_private *dev_priv, static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev, struct drm_file *file_priv, - struct drm_mode_fb_cmd2 *mode_cmd2) + const struct drm_mode_fb_cmd2 *mode_cmd2) { struct vmw_private *dev_priv = vmw_priv(dev); struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; @@ -1331,14 +1331,6 @@ static int vmw_du_update_layout(struct vmw_private *dev_priv, unsigned num, return 0; } -void vmw_du_crtc_save(struct drm_crtc *crtc) -{ -} - -void vmw_du_crtc_restore(struct drm_crtc *crtc) -{ -} - void vmw_du_crtc_gamma_set(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b, uint32_t start, uint32_t size) @@ -1360,14 +1352,6 @@ int vmw_du_connector_dpms(struct drm_connector *connector, int mode) return 0; } -void vmw_du_connector_save(struct drm_connector *connector) -{ -} - -void vmw_du_connector_restore(struct drm_connector *connector) -{ -} - enum drm_connector_status vmw_du_connector_detect(struct drm_connector *connector, bool force) { @@ -1554,7 +1538,7 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector, drm_mode_probed_add(connector, mode); } - drm_mode_connector_list_update(connector, true); + drm_mode_connector_list_update(connector); /* Move the prefered mode first, help apps pick the right mode. */ drm_mode_sort(&connector->modes); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c index bb63e4d795fa..2aff5e51d926 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c @@ -295,8 +295,6 @@ static int vmw_ldu_crtc_set_config(struct drm_mode_set *set) } static struct drm_crtc_funcs vmw_legacy_crtc_funcs = { - .save = vmw_du_crtc_save, - .restore = vmw_du_crtc_restore, .cursor_set = vmw_du_crtc_cursor_set, .cursor_move = vmw_du_crtc_cursor_move, .gamma_set = vmw_du_crtc_gamma_set, @@ -329,8 +327,6 @@ static void vmw_ldu_connector_destroy(struct drm_connector *connector) static struct drm_connector_funcs vmw_legacy_connector_funcs = { .dpms = vmw_du_connector_dpms, - .save = vmw_du_connector_save, - .restore = vmw_du_connector_restore, .detect = vmw_du_connector_detect, .fill_modes = vmw_du_connector_fill_modes, .set_property = vmw_du_connector_set_property, @@ -367,7 +363,7 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit) connector->status = vmw_du_connector_detect(connector, true); drm_encoder_init(dev, encoder, &vmw_legacy_encoder_funcs, - DRM_MODE_ENCODER_VIRTUAL); + DRM_MODE_ENCODER_VIRTUAL, NULL); drm_mode_connector_attach_encoder(connector, encoder); encoder->possible_crtcs = (1 << unit); encoder->possible_clones = 0; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c index b96d1ab610c5..6bb7af37934a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c @@ -531,8 +531,6 @@ out_no_fence: } static struct drm_crtc_funcs vmw_screen_object_crtc_funcs = { - .save = vmw_du_crtc_save, - .restore = vmw_du_crtc_restore, .cursor_set = vmw_du_crtc_cursor_set, .cursor_move = vmw_du_crtc_cursor_move, .gamma_set = vmw_du_crtc_gamma_set, @@ -565,10 +563,6 @@ static void vmw_sou_connector_destroy(struct drm_connector *connector) static struct drm_connector_funcs vmw_sou_connector_funcs = { .dpms = vmw_du_connector_dpms, - .save = vmw_du_connector_save, - .restore = vmw_du_connector_restore, - .detect = vmw_du_connector_detect, - .fill_modes = vmw_du_connector_fill_modes, .set_property = vmw_du_connector_set_property, .destroy = vmw_sou_connector_destroy, }; @@ -603,7 +597,7 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit) connector->status = vmw_du_connector_detect(connector, true); drm_encoder_init(dev, encoder, &vmw_screen_object_encoder_funcs, - DRM_MODE_ENCODER_VIRTUAL); + DRM_MODE_ENCODER_VIRTUAL, NULL); drm_mode_connector_attach_encoder(connector, encoder); encoder->possible_crtcs = (1 << unit); encoder->possible_clones = 0; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c index b1fc1c02792d..45e72c2f15cd 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c @@ -1041,8 +1041,6 @@ out_finish: * Screen Target CRTC dispatch table */ static struct drm_crtc_funcs vmw_stdu_crtc_funcs = { - .save = vmw_du_crtc_save, - .restore = vmw_du_crtc_restore, .cursor_set = vmw_du_crtc_cursor_set, .cursor_move = vmw_du_crtc_cursor_move, .gamma_set = vmw_du_crtc_gamma_set, @@ -1101,8 +1099,6 @@ static void vmw_stdu_connector_destroy(struct drm_connector *connector) static struct drm_connector_funcs vmw_stdu_connector_funcs = { .dpms = vmw_du_connector_dpms, - .save = vmw_du_connector_save, - .restore = vmw_du_connector_restore, .detect = vmw_du_connector_detect, .fill_modes = vmw_du_connector_fill_modes, .set_property = vmw_du_connector_set_property, @@ -1149,7 +1145,7 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit) connector->status = vmw_du_connector_detect(connector, false); drm_encoder_init(dev, encoder, &vmw_stdu_encoder_funcs, - DRM_MODE_ENCODER_VIRTUAL); + DRM_MODE_ENCODER_VIRTUAL, NULL); drm_mode_connector_attach_encoder(connector, encoder); encoder->possible_crtcs = (1 << unit); encoder->possible_clones = 0; |