summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/i915
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/i915')
-rw-r--r--drivers/gpu/drm/i915/display/intel_ddi.c50
-rw-r--r--drivers/gpu/drm/i915/display/intel_display.c9
-rw-r--r--drivers/gpu/drm/i915/display/intel_dp.c20
-rw-r--r--drivers/gpu/drm/i915/display/intel_dp.h3
-rw-r--r--drivers/gpu/drm/i915/display/intel_dp_link_training.c25
-rw-r--r--drivers/gpu/drm/i915/display/intel_dp_link_training.h3
-rw-r--r--drivers/gpu/drm/i915/display/intel_dp_mst.c4
-rw-r--r--drivers/gpu/drm/i915/display/intel_overlay.c4
-rw-r--r--drivers/gpu/drm/i915/display/intel_sprite.c65
-rw-r--r--drivers/gpu/drm/i915/gem/i915_gem_domain.c45
-rw-r--r--drivers/gpu/drm/i915/gem/i915_gem_object.h1
-rw-r--r--drivers/gpu/drm/i915/gt/gen7_renderclear.c12
-rw-r--r--drivers/gpu/drm/i915/gt/intel_breadcrumbs.c47
-rw-r--r--drivers/gpu/drm/i915/gt/intel_ggtt.c47
-rw-r--r--drivers/gpu/drm/i915/i915_active.c28
-rw-r--r--drivers/gpu/drm/i915/i915_drv.h2
-rw-r--r--drivers/gpu/drm/i915/selftests/i915_gem_gtt.c2
17 files changed, 166 insertions, 201 deletions
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index d5ace48b1ace..dc13d1814d95 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2754,13 +2754,15 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
int n_entries, ln;
u32 val;
+ if (enc_to_dig_port(encoder)->tc_mode == TC_PORT_TBT_ALT)
+ return;
+
ddi_translations = icl_get_mg_buf_trans(encoder, crtc_state, &n_entries);
- /* The table does not have values for level 3 and level 9. */
- if (level >= n_entries || level == 3 || level == 9) {
+ if (level >= n_entries) {
drm_dbg_kms(&dev_priv->drm,
"DDI translation not found for level %d. Using %d instead.",
- level, n_entries - 2);
- level = n_entries - 2;
+ level, n_entries - 1);
+ level = n_entries - 1;
}
/* Set MG_TX_LINK_PARAMS cri_use_fs32 to 0. */
@@ -2891,6 +2893,9 @@ tgl_dkl_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
u32 val, dpcnt_mask, dpcnt_val;
int n_entries, ln;
+ if (enc_to_dig_port(encoder)->tc_mode == TC_PORT_TBT_ALT)
+ return;
+
ddi_translations = tgl_get_dkl_buf_trans(encoder, crtc_state, &n_entries);
if (level >= n_entries)
@@ -3532,6 +3537,23 @@ static void intel_ddi_disable_fec_state(struct intel_encoder *encoder,
intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
}
+static void intel_ddi_power_up_lanes(struct intel_encoder *encoder,
+ const struct intel_crtc_state *crtc_state)
+{
+ struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+ struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+ enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+ if (intel_phy_is_combo(i915, phy)) {
+ bool lane_reversal =
+ dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL;
+
+ intel_combo_phy_power_up_lanes(i915, phy, false,
+ crtc_state->lane_count,
+ lane_reversal);
+ }
+}
+
static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state,
struct intel_encoder *encoder,
const struct intel_crtc_state *crtc_state,
@@ -3621,14 +3643,7 @@ static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state,
* 7.f Combo PHY: Configure PORT_CL_DW10 Static Power Down to power up
* the used lanes of the DDI.
*/
- if (intel_phy_is_combo(dev_priv, phy)) {
- bool lane_reversal =
- dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL;
-
- intel_combo_phy_power_up_lanes(dev_priv, phy, false,
- crtc_state->lane_count,
- lane_reversal);
- }
+ intel_ddi_power_up_lanes(encoder, crtc_state);
/*
* 7.g Configure and enable DDI_BUF_CTL
@@ -3713,14 +3728,7 @@ static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state,
else
intel_prepare_dp_ddi_buffers(encoder, crtc_state);
- if (intel_phy_is_combo(dev_priv, phy)) {
- bool lane_reversal =
- dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL;
-
- intel_combo_phy_power_up_lanes(dev_priv, phy, false,
- crtc_state->lane_count,
- lane_reversal);
- }
+ intel_ddi_power_up_lanes(encoder, crtc_state);
intel_ddi_init_dp_buf_reg(encoder, crtc_state);
if (!is_mst)
@@ -4206,6 +4214,8 @@ static void intel_enable_ddi_hdmi(struct intel_atomic_state *state,
intel_de_write(dev_priv, reg, val);
}
+ intel_ddi_power_up_lanes(encoder, crtc_state);
+
/* In HDMI/DVI mode, the port width, and swing/emphasis values
* are ignored so nothing special needs to be done besides
* enabling the port.
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 53a00cf3fa32..61be6bed9162 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -2309,7 +2309,7 @@ intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
*/
ret = i915_vma_pin_fence(vma);
if (ret != 0 && INTEL_GEN(dev_priv) < 4) {
- i915_gem_object_unpin_from_display_plane(vma);
+ i915_vma_unpin(vma);
vma = ERR_PTR(ret);
goto err;
}
@@ -2327,12 +2327,9 @@ err:
void intel_unpin_fb_vma(struct i915_vma *vma, unsigned long flags)
{
- i915_gem_object_lock(vma->obj, NULL);
if (flags & PLANE_HAS_FENCE)
i915_vma_unpin_fence(vma);
- i915_gem_object_unpin_from_display_plane(vma);
- i915_gem_object_unlock(vma->obj);
-
+ i915_vma_unpin(vma);
i915_vma_put(vma);
}
@@ -4807,6 +4804,8 @@ u32 glk_plane_color_ctl(const struct intel_crtc_state *crtc_state,
plane_color_ctl |= PLANE_COLOR_YUV_RANGE_CORRECTION_DISABLE;
} else if (fb->format->is_yuv) {
plane_color_ctl |= PLANE_COLOR_INPUT_CSC_ENABLE;
+ if (plane_state->hw.color_range == DRM_COLOR_YCBCR_FULL_RANGE)
+ plane_color_ctl |= PLANE_COLOR_YUV_RANGE_CORRECTION_DISABLE;
}
return plane_color_ctl;
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 09123e8625c4..8a26307c4896 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -4637,24 +4637,6 @@ ivb_cpu_edp_set_signal_levels(struct intel_dp *intel_dp,
intel_de_posting_read(dev_priv, intel_dp->output_reg);
}
-void intel_dp_set_signal_levels(struct intel_dp *intel_dp,
- const struct intel_crtc_state *crtc_state)
-{
- struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
- u8 train_set = intel_dp->train_set[0];
-
- drm_dbg_kms(&dev_priv->drm, "Using vswing level %d%s\n",
- train_set & DP_TRAIN_VOLTAGE_SWING_MASK,
- train_set & DP_TRAIN_MAX_SWING_REACHED ? " (max)" : "");
- drm_dbg_kms(&dev_priv->drm, "Using pre-emphasis level %d%s\n",
- (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) >>
- DP_TRAIN_PRE_EMPHASIS_SHIFT,
- train_set & DP_TRAIN_MAX_PRE_EMPHASIS_REACHED ?
- " (max)" : "");
-
- intel_dp->set_signal_levels(intel_dp, crtc_state);
-}
-
void
intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state,
@@ -5703,7 +5685,7 @@ static void intel_dp_process_phy_request(struct intel_dp *intel_dp,
intel_dp_autotest_phy_ddi_disable(intel_dp, crtc_state);
- intel_dp_set_signal_levels(intel_dp, crtc_state);
+ intel_dp_set_signal_levels(intel_dp, crtc_state, DP_PHY_DPRX);
intel_dp_phy_pattern_update(intel_dp, crtc_state);
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index 05f7ddf7a795..6620f9efdcbb 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -96,9 +96,6 @@ void
intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state,
u8 dp_train_pat);
-void
-intel_dp_set_signal_levels(struct intel_dp *intel_dp,
- const struct intel_crtc_state *crtc_state);
void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
u8 *link_bw, u8 *rate_select);
bool intel_dp_source_supports_hbr2(struct intel_dp *intel_dp);
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 91d3979902d0..d8c6d7054d11 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -334,6 +334,27 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
return drm_dp_dpcd_write(&intel_dp->aux, reg, buf, len) == len;
}
+void intel_dp_set_signal_levels(struct intel_dp *intel_dp,
+ const struct intel_crtc_state *crtc_state,
+ enum drm_dp_phy dp_phy)
+{
+ struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+ u8 train_set = intel_dp->train_set[0];
+ char phy_name[10];
+
+ drm_dbg_kms(&dev_priv->drm, "Using vswing level %d%s, pre-emphasis level %d%s, at %s\n",
+ train_set & DP_TRAIN_VOLTAGE_SWING_MASK,
+ train_set & DP_TRAIN_MAX_SWING_REACHED ? " (max)" : "",
+ (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) >>
+ DP_TRAIN_PRE_EMPHASIS_SHIFT,
+ train_set & DP_TRAIN_MAX_PRE_EMPHASIS_REACHED ?
+ " (max)" : "",
+ intel_dp_phy_name(dp_phy, phy_name, sizeof(phy_name)));
+
+ if (intel_dp_phy_is_downstream_of_source(intel_dp, dp_phy))
+ intel_dp->set_signal_levels(intel_dp, crtc_state);
+}
+
static bool
intel_dp_reset_link_train(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state,
@@ -341,7 +362,7 @@ intel_dp_reset_link_train(struct intel_dp *intel_dp,
u8 dp_train_pat)
{
memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set));
- intel_dp_set_signal_levels(intel_dp, crtc_state);
+ intel_dp_set_signal_levels(intel_dp, crtc_state, dp_phy);
return intel_dp_set_link_train(intel_dp, crtc_state, dp_phy, dp_train_pat);
}
@@ -355,7 +376,7 @@ intel_dp_update_link_train(struct intel_dp *intel_dp,
DP_TRAINING_LANE0_SET_PHY_REPEATER(dp_phy);
int ret;
- intel_dp_set_signal_levels(intel_dp, crtc_state);
+ intel_dp_set_signal_levels(intel_dp, crtc_state, dp_phy);
ret = drm_dp_dpcd_write(&intel_dp->aux, reg,
intel_dp->train_set, crtc_state->lane_count);
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.h b/drivers/gpu/drm/i915/display/intel_dp_link_training.h
index 86905aa24db7..6a1f76bd8c75 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.h
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.h
@@ -17,6 +17,9 @@ void intel_dp_get_adjust_train(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state,
enum drm_dp_phy dp_phy,
const u8 link_status[DP_LINK_STATUS_SIZE]);
+void intel_dp_set_signal_levels(struct intel_dp *intel_dp,
+ const struct intel_crtc_state *crtc_state,
+ enum drm_dp_phy dp_phy);
void intel_dp_start_link_train(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state);
void intel_dp_stop_link_train(struct intel_dp *intel_dp,
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 27f04aed8764..3286b232be0b 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -69,7 +69,9 @@ static int intel_dp_mst_compute_link_config(struct intel_encoder *encoder,
slots = drm_dp_atomic_find_vcpi_slots(state, &intel_dp->mst_mgr,
connector->port,
- crtc_state->pbn, 0);
+ crtc_state->pbn,
+ drm_dp_get_vc_payload_bw(crtc_state->port_clock,
+ crtc_state->lane_count));
if (slots == -EDEADLK)
return slots;
if (slots >= 0)
diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
index 52b4f6193b4c..0095c8cac9b4 100644
--- a/drivers/gpu/drm/i915/display/intel_overlay.c
+++ b/drivers/gpu/drm/i915/display/intel_overlay.c
@@ -359,7 +359,7 @@ static void intel_overlay_release_old_vma(struct intel_overlay *overlay)
intel_frontbuffer_flip_complete(overlay->i915,
INTEL_FRONTBUFFER_OVERLAY(overlay->crtc->pipe));
- i915_gem_object_unpin_from_display_plane(vma);
+ i915_vma_unpin(vma);
i915_vma_put(vma);
}
@@ -860,7 +860,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
return 0;
out_unpin:
- i915_gem_object_unpin_from_display_plane(vma);
+ i915_vma_unpin(vma);
out_pin_section:
atomic_dec(&dev_priv->gpu_error.pending_fb_pin);
diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c
index 019a2d6d807a..3da2544fa1c0 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite.c
@@ -618,13 +618,19 @@ skl_program_scaler(struct intel_plane *plane,
/* Preoffset values for YUV to RGB Conversion */
#define PREOFF_YUV_TO_RGB_HI 0x1800
-#define PREOFF_YUV_TO_RGB_ME 0x1F00
+#define PREOFF_YUV_TO_RGB_ME 0x0000
#define PREOFF_YUV_TO_RGB_LO 0x1800
#define ROFF(x) (((x) & 0xffff) << 16)
#define GOFF(x) (((x) & 0xffff) << 0)
#define BOFF(x) (((x) & 0xffff) << 16)
+/*
+ * Programs the input color space conversion stage for ICL HDR planes.
+ * Note that it is assumed that this stage always happens after YUV
+ * range correction. Thus, the input to this stage is assumed to be
+ * in full-range YCbCr.
+ */
static void
icl_program_input_csc(struct intel_plane *plane,
const struct intel_crtc_state *crtc_state,
@@ -672,52 +678,7 @@ icl_program_input_csc(struct intel_plane *plane,
0x0, 0x7800, 0x7F10,
},
};
-
- /* Matrix for Limited Range to Full Range Conversion */
- static const u16 input_csc_matrix_lr[][9] = {
- /*
- * BT.601 Limted range YCbCr -> full range RGB
- * The matrix required is :
- * [1.164384, 0.000, 1.596027,
- * 1.164384, -0.39175, -0.812813,
- * 1.164384, 2.017232, 0.0000]
- */
- [DRM_COLOR_YCBCR_BT601] = {
- 0x7CC8, 0x7950, 0x0,
- 0x8D00, 0x7950, 0x9C88,
- 0x0, 0x7950, 0x6810,
- },
- /*
- * BT.709 Limited range YCbCr -> full range RGB
- * The matrix required is :
- * [1.164384, 0.000, 1.792741,
- * 1.164384, -0.213249, -0.532909,
- * 1.164384, 2.112402, 0.0000]
- */
- [DRM_COLOR_YCBCR_BT709] = {
- 0x7E58, 0x7950, 0x0,
- 0x8888, 0x7950, 0xADA8,
- 0x0, 0x7950, 0x6870,
- },
- /*
- * BT.2020 Limited range YCbCr -> full range RGB
- * The matrix required is :
- * [1.164, 0.000, 1.678,
- * 1.164, -0.1873, -0.6504,
- * 1.164, 2.1417, 0.0000]
- */
- [DRM_COLOR_YCBCR_BT2020] = {
- 0x7D70, 0x7950, 0x0,
- 0x8A68, 0x7950, 0xAC00,
- 0x0, 0x7950, 0x6890,
- },
- };
- const u16 *csc;
-
- if (plane_state->hw.color_range == DRM_COLOR_YCBCR_FULL_RANGE)
- csc = input_csc_matrix[plane_state->hw.color_encoding];
- else
- csc = input_csc_matrix_lr[plane_state->hw.color_encoding];
+ const u16 *csc = input_csc_matrix[plane_state->hw.color_encoding];
intel_de_write_fw(dev_priv, PLANE_INPUT_CSC_COEFF(pipe, plane_id, 0),
ROFF(csc[0]) | GOFF(csc[1]));
@@ -734,14 +695,8 @@ icl_program_input_csc(struct intel_plane *plane,
intel_de_write_fw(dev_priv, PLANE_INPUT_CSC_PREOFF(pipe, plane_id, 0),
PREOFF_YUV_TO_RGB_HI);
- if (plane_state->hw.color_range == DRM_COLOR_YCBCR_FULL_RANGE)
- intel_de_write_fw(dev_priv,
- PLANE_INPUT_CSC_PREOFF(pipe, plane_id, 1),
- 0);
- else
- intel_de_write_fw(dev_priv,
- PLANE_INPUT_CSC_PREOFF(pipe, plane_id, 1),
- PREOFF_YUV_TO_RGB_ME);
+ intel_de_write_fw(dev_priv, PLANE_INPUT_CSC_PREOFF(pipe, plane_id, 1),
+ PREOFF_YUV_TO_RGB_ME);
intel_de_write_fw(dev_priv, PLANE_INPUT_CSC_PREOFF(pipe, plane_id, 2),
PREOFF_YUV_TO_RGB_LO);
intel_de_write_fw(dev_priv,
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index fcce6909f201..3d435bfff764 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -387,48 +387,6 @@ err:
return vma;
}
-static void i915_gem_object_bump_inactive_ggtt(struct drm_i915_gem_object *obj)
-{
- struct drm_i915_private *i915 = to_i915(obj->base.dev);
- struct i915_vma *vma;
-
- if (list_empty(&obj->vma.list))
- return;
-
- mutex_lock(&i915->ggtt.vm.mutex);
- spin_lock(&obj->vma.lock);
- for_each_ggtt_vma(vma, obj) {
- if (!drm_mm_node_allocated(&vma->node))
- continue;
-
- GEM_BUG_ON(vma->vm != &i915->ggtt.vm);
- list_move_tail(&vma->vm_link, &vma->vm->bound_list);
- }
- spin_unlock(&obj->vma.lock);
- mutex_unlock(&i915->ggtt.vm.mutex);
-
- if (i915_gem_object_is_shrinkable(obj)) {
- unsigned long flags;
-
- spin_lock_irqsave(&i915->mm.obj_lock, flags);
-
- if (obj->mm.madv == I915_MADV_WILLNEED &&
- !atomic_read(&obj->mm.shrink_pin))
- list_move_tail(&obj->mm.link, &i915->mm.shrink_list);
-
- spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
- }
-}
-
-void
-i915_gem_object_unpin_from_display_plane(struct i915_vma *vma)
-{
- /* Bump the LRU to try and avoid premature eviction whilst flipping */
- i915_gem_object_bump_inactive_ggtt(vma->obj);
-
- i915_vma_unpin(vma);
-}
-
/**
* Moves a single object to the CPU read, and possibly write domain.
* @obj: object to act on
@@ -569,9 +527,6 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
else
err = i915_gem_object_set_to_cpu_domain(obj, write_domain);
- /* And bump the LRU for this access */
- i915_gem_object_bump_inactive_ggtt(obj);
-
i915_gem_object_unlock(obj);
if (write_domain)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index be14486f63a7..4556afe18f16 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -486,7 +486,6 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
u32 alignment,
const struct i915_ggtt_view *view,
unsigned int flags);
-void i915_gem_object_unpin_from_display_plane(struct i915_vma *vma);
void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
diff --git a/drivers/gpu/drm/i915/gt/gen7_renderclear.c b/drivers/gpu/drm/i915/gt/gen7_renderclear.c
index 94465374ca2f..e961ad6a3129 100644
--- a/drivers/gpu/drm/i915/gt/gen7_renderclear.c
+++ b/drivers/gpu/drm/i915/gt/gen7_renderclear.c
@@ -390,6 +390,16 @@ static void emit_batch(struct i915_vma * const vma,
&cb_kernel_ivb,
desc_count);
+ /* Reset inherited context registers */
+ gen7_emit_pipeline_invalidate(&cmds);
+ batch_add(&cmds, MI_LOAD_REGISTER_IMM(2));
+ batch_add(&cmds, i915_mmio_reg_offset(CACHE_MODE_0_GEN7));
+ batch_add(&cmds, 0xffff0000);
+ batch_add(&cmds, i915_mmio_reg_offset(CACHE_MODE_1));
+ batch_add(&cmds, 0xffff0000 | PIXEL_SUBSPAN_COLLECT_OPT_DISABLE);
+ gen7_emit_pipeline_flush(&cmds);
+
+ /* Switch to the media pipeline and our base address */
gen7_emit_pipeline_invalidate(&cmds);
batch_add(&cmds, PIPELINE_SELECT | PIPELINE_SELECT_MEDIA);
batch_add(&cmds, MI_NOOP);
@@ -399,9 +409,11 @@ static void emit_batch(struct i915_vma * const vma,
gen7_emit_state_base_address(&cmds, descriptors);
gen7_emit_pipeline_invalidate(&cmds);
+ /* Set the clear-residual kernel state */
gen7_emit_vfe_state(&cmds, bv, urb_size - 1, 0, 0);
gen7_emit_interface_descriptor_load(&cmds, descriptors, desc_count);
+ /* Execute the kernel on all HW threads */
for (i = 0; i < num_primitives(bv); i++)
gen7_emit_media_object(&cmds, i);
diff --git a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
index 0625cbb3b431..1d1757584f49 100644
--- a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
+++ b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
@@ -187,18 +187,6 @@ static void add_retire(struct intel_breadcrumbs *b, struct intel_timeline *tl)
intel_engine_add_retire(b->irq_engine, tl);
}
-static bool __signal_request(struct i915_request *rq)
-{
- GEM_BUG_ON(test_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags));
-
- if (!__dma_fence_signal(&rq->fence)) {
- i915_request_put(rq);
- return false;
- }
-
- return true;
-}
-
static struct llist_node *
slist_add(struct llist_node *node, struct llist_node *head)
{
@@ -269,9 +257,11 @@ static void signal_irq_work(struct irq_work *work)
release = remove_signaling_context(b, ce);
spin_unlock(&ce->signal_lock);
- if (__signal_request(rq))
+ if (__dma_fence_signal(&rq->fence))
/* We own signal_node now, xfer to local list */
signal = slist_add(&rq->signal_node, signal);
+ else
+ i915_request_put(rq);
if (release) {
add_retire(b, ce->timeline);
@@ -358,6 +348,17 @@ void intel_breadcrumbs_free(struct intel_breadcrumbs *b)
kfree(b);
}
+static void irq_signal_request(struct i915_request *rq,
+ struct intel_breadcrumbs *b)
+{
+ if (!__dma_fence_signal(&rq->fence))
+ return;
+
+ i915_request_get(rq);
+ if (llist_add(&rq->signal_node, &b->signaled_requests))
+ irq_work_queue(&b->irq_work);
+}
+
static void insert_breadcrumb(struct i915_request *rq)
{
struct intel_breadcrumbs *b = READ_ONCE(rq->engine)->breadcrumbs;
@@ -367,17 +368,13 @@ static void insert_breadcrumb(struct i915_request *rq)
if (test_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags))
return;
- i915_request_get(rq);
-
/*
* If the request is already completed, we can transfer it
* straight onto a signaled list, and queue the irq worker for
* its signal completion.
*/
if (__i915_request_is_complete(rq)) {
- if (__signal_request(rq) &&
- llist_add(&rq->signal_node, &b->signaled_requests))
- irq_work_queue(&b->irq_work);
+ irq_signal_request(rq, b);
return;
}
@@ -408,6 +405,8 @@ static void insert_breadcrumb(struct i915_request *rq)
break;
}
}
+
+ i915_request_get(rq);
list_add_rcu(&rq->signal_link, pos);
GEM_BUG_ON(!check_signal_order(ce, rq));
GEM_BUG_ON(test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &rq->fence.flags));
@@ -448,19 +447,25 @@ bool i915_request_enable_breadcrumb(struct i915_request *rq)
void i915_request_cancel_breadcrumb(struct i915_request *rq)
{
+ struct intel_breadcrumbs *b = READ_ONCE(rq->engine)->breadcrumbs;
struct intel_context *ce = rq->context;
bool release;
- if (!test_and_clear_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags))
+ spin_lock(&ce->signal_lock);
+ if (!test_and_clear_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags)) {
+ spin_unlock(&ce->signal_lock);
return;
+ }
- spin_lock(&ce->signal_lock);
list_del_rcu(&rq->signal_link);
- release = remove_signaling_context(rq->engine->breadcrumbs, ce);
+ release = remove_signaling_context(b, ce);
spin_unlock(&ce->signal_lock);
if (release)
intel_context_put(ce);
+ if (__i915_request_is_complete(rq))
+ irq_signal_request(rq, b);
+
i915_request_put(rq);
}
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index cf94525be2c1..db8c66dde655 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -526,16 +526,39 @@ static int init_ggtt(struct i915_ggtt *ggtt)
mutex_init(&ggtt->error_mutex);
if (ggtt->mappable_end) {
- /* Reserve a mappable slot for our lockless error capture */
- ret = drm_mm_insert_node_in_range(&ggtt->vm.mm,
- &ggtt->error_capture,
- PAGE_SIZE, 0,
- I915_COLOR_UNEVICTABLE,
- 0, ggtt->mappable_end,
- DRM_MM_INSERT_LOW);
- if (ret)
- return ret;
+ /*
+ * Reserve a mappable slot for our lockless error capture.
+ *
+ * We strongly prefer taking address 0x0 in order to protect
+ * other critical buffers against accidental overwrites,
+ * as writing to address 0 is a very common mistake.
+ *
+ * Since 0 may already be in use by the system (e.g. the BIOS
+ * framebuffer), we let the reservation fail quietly and hope
+ * 0 remains reserved always.
+ *
+ * If we fail to reserve 0, and then fail to find any space
+ * for an error-capture, remain silent. We can afford not
+ * to reserve an error_capture node as we have fallback
+ * paths, and we trust that 0 will remain reserved. However,
+ * the only likely reason for failure to insert is a driver
+ * bug, which we expect to cause other failures...
+ */
+ ggtt->error_capture.size = I915_GTT_PAGE_SIZE;
+ ggtt->error_capture.color = I915_COLOR_UNEVICTABLE;
+ if (drm_mm_reserve_node(&ggtt->vm.mm, &ggtt->error_capture))
+ drm_mm_insert_node_in_range(&ggtt->vm.mm,
+ &ggtt->error_capture,
+ ggtt->error_capture.size, 0,
+ ggtt->error_capture.color,
+ 0, ggtt->mappable_end,
+ DRM_MM_INSERT_LOW);
}
+ if (drm_mm_node_allocated(&ggtt->error_capture))
+ drm_dbg(&ggtt->vm.i915->drm,
+ "Reserved GGTT:[%llx, %llx] for use by error capture\n",
+ ggtt->error_capture.start,
+ ggtt->error_capture.start + ggtt->error_capture.size);
/*
* The upper portion of the GuC address space has a sizeable hole
@@ -548,9 +571,9 @@ static int init_ggtt(struct i915_ggtt *ggtt)
/* Clear any non-preallocated blocks */
drm_mm_for_each_hole(entry, &ggtt->vm.mm, hole_start, hole_end) {
- drm_dbg_kms(&ggtt->vm.i915->drm,
- "clearing unused GTT space: [%lx, %lx]\n",
- hole_start, hole_end);
+ drm_dbg(&ggtt->vm.i915->drm,
+ "clearing unused GTT space: [%lx, %lx]\n",
+ hole_start, hole_end);
ggtt->vm.clear_range(&ggtt->vm, hole_start,
hole_end - hole_start);
}
diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index 10a865f3dc09..9ed19b8bca60 100644
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -631,24 +631,26 @@ static int flush_lazy_signals(struct i915_active *ref)
int __i915_active_wait(struct i915_active *ref, int state)
{
- int err;
-
might_sleep();
- if (!i915_active_acquire_if_busy(ref))
- return 0;
-
/* Any fence added after the wait begins will not be auto-signaled */
- err = flush_lazy_signals(ref);
- i915_active_release(ref);
- if (err)
- return err;
+ if (i915_active_acquire_if_busy(ref)) {
+ int err;
- if (!i915_active_is_idle(ref) &&
- ___wait_var_event(ref, i915_active_is_idle(ref),
- state, 0, 0, schedule()))
- return -EINTR;
+ err = flush_lazy_signals(ref);
+ i915_active_release(ref);
+ if (err)
+ return err;
+ if (___wait_var_event(ref, i915_active_is_idle(ref),
+ state, 0, 0, schedule()))
+ return -EINTR;
+ }
+
+ /*
+ * After the wait is complete, the caller may free the active.
+ * We have to flush any concurrent retirement before returning.
+ */
flush_work(&ref->work);
return 0;
}
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 632c713227dc..c6964f82a1bb 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1346,7 +1346,7 @@ intel_subplatform(const struct intel_runtime_info *info, enum intel_platform p)
{
const unsigned int pi = __platform_mask_index(info, p);
- return info->platform_mask[pi] & INTEL_SUBPLATFORM_BITS;
+ return info->platform_mask[pi] & ((1 << INTEL_SUBPLATFORM_BITS) - 1);
}
static __always_inline bool
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index c53a222e3dec..713770fb2b92 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -1880,7 +1880,7 @@ static int igt_cs_tlb(void *arg)
vma = i915_vma_instance(out, vm, NULL);
if (IS_ERR(vma)) {
err = PTR_ERR(vma);
- goto out_put_batch;
+ goto out_put_out;
}
err = i915_vma_pin(vma, 0, 0,
generated by cgit v1.2.3 (git 2.39.0) at 2024-11-24 14:18:41 +0300