summaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2015-11-18drm/amdgpu: move bo_reserve out of amdgpu_vm_clear_boChunming Zhou1-15/+18
Change-Id: Ifbb0c06680494bfa04d0be5e5941d31ae2e5ef28 Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-18drm/amdgpu: add lock for interval tree in vmChunming Zhou2-2/+15
Change-Id: I62b892a22af37b32e6b4aefca80a25cf45426ed2 Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-18drm/amdgpu: keep the owner for VMIDsChristian König2-17/+24
We don't need the last VM use any more, keep the owner directly. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <davdi1.zhou@amd.com>
2015-11-18drm/amdgpu: move VM manager clean into the VM code againChristian König4-12/+20
It's not a good idea to duplicate that code. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <davdi1.zhou@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-18drm/amdgpu: cleanup VM coding styleChristian König1-50/+48
Fix the indentation and move the VM functions to the structures. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <davdi1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-18drm/amdgpu: remove unused VM manager fieldChristian König1-2/+0
Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <davdi1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
2015-11-18drm/amdgpu: cleanup scheduler command submissionChristian König2-39/+25
Unify the two code path again, cause they do pretty much the same thing. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <davdi1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
2015-11-18drm/amdgpu: fix typo in firmware nameChristian König1-1/+1
Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-11-18drm/i915: Consider SPLL as another shared pll, v2.Maarten Lankhorst4-41/+83
When diagnosing a unrelated bug for someone on irc, it would seem the hardware can be brought up by the BIOS with the embedded displayport using the SPLL for spread spectrum. Right now this is not handled well in i915, and it calculates the crtc needs to be reprogrammed on the first modeset without SSC, but the SPLL itself was kept active. Fix this by exposing SPLL as a shared pll that will not be returned by intel_get_shared_dpll; you have to know it exists to use it. Changes since v1: - Create a separate dpll_hw_state.spll for spll, and use separate pll functions for spll. Tested-by: Emil Renner Berthing <kernel@esmil.dk> Tested-by: Gabriel Feceoru <gabriel.feceoru@intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447681332-6318-1-git-send-email-maarten.lankhorst@linux.intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2015-11-18drm/i915: Fix gpu frequency change tracingMika Kuoppala1-1/+1
With gen < 9 we have had always 50Mhz units as our hw ratio. With gen >= 9 the hw ratio changed to 16.667Mhz (50/3). The result was that our gpu frequency tracing started to output values 3 times larger than expected due to hardcoded scaling value. Fix this by using Use intel_gpu_freq() when generating Mhz value from ratio for 'intel_gpu_freq_change' trace event. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92591 Cc: stable@vger.kernel.org # v4.3+ Reported-by: Eero Tamminen <eero.t.tamminen@intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447776866-29384-1-git-send-email-mika.kuoppala@intel.com
2015-11-17drm/vc4: Make sure that planes aren't scaled.Eric Anholt1-0/+10
We would scan out the memory around them if an upscale was attempted, and would just scan out incorrectly for downscaling. Signed-off-by: Eric Anholt <eric@anholt.net>
2015-11-17drm/vc4: Fix some failure to track __iomem decorations on pointers.Eric Anholt2-5/+6
Caught by the kbuild test robot. v2: Fix the +i I dropped in the first version. Signed-off-by: Eric Anholt <eric@anholt.net>
2015-11-17drm/vc4: checking for NULL instead of IS_ERRDan Carpenter1-2/+2
vc4_plane_init() returns an ERR_PTR on error, it doesn't return NULL. This was obviously intended because the next lines call PTR_ERR(primary_plane) already. Fixes: c8b75bca92cb ('Eric Anholt <eric@anholt.net>') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2015-11-17drm/vc4: fix itnull.cocci warningsJulia Lawall1-1/+1
Connector cannot be null because it is a list entry, ie accessed at an offset from the positions of the list structure pointers themselves. Generated by: scripts/coccinelle/iterators/itnull.cocci Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Julia Lawall <julia.lawall@lip6.fr> Signed-off-by: Eric Anholt <eric@anholt.net>
2015-11-17drm/vc4: fix platform_no_drv_owner.cocci warningsJulia Lawall1-1/+0
drivers/gpu/drm/vc4/vc4_drv.c:248:3-8: No need to set .owner here. The core will do it. Remove .owner field if calls are used which set it automatically Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Julia Lawall <julia.lawall@lip6.fr> Signed-off-by: Eric Anholt <eric@anholt.net>
2015-11-17drm/vc4: vc4_plane_duplicate_state() can be statickbuild test robot1-4/+4
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2015-11-17drm/i915: Don't clobber the addfb2 ioctl paramsVille Syrjälä1-3/+4
We try to convert the old way of of specifying fb tiling (obj->tiling) into the new fb modifiers. We store the result in the passed in mode_cmd structure. But that structure comes directly from the addfb2 ioctl, and gets copied back out to userspace, which means we're clobbering the modifiers that the user provided (all 0 since the DRM_MODE_FB_MODIFIERS flag wasn't even set by the user). Hence if the user reuses the struct for another addfb2, the ioctl will be rejected since it's now asking for some modifiers w/o the flag set. Fix the problem by making a copy of the user provided structure. We can play any games we want with the copy. IGT-Version: 1.12-git (x86_64) (Linux: 4.4.0-rc1-stereo+ x86_64) ... Subtest basic-X-tiled: SUCCESS (0.001s) Test assertion failure function pitch_tests, file kms_addfb_basic.c:167: Failed assertion: drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0 Last errno: 22, Invalid argument Stack trace: #0 [__igt_fail_assert+0x101] #1 [pitch_tests+0x619] #2 [__real_main426+0x2f] #3 [main+0x23] #4 [__libc_start_main+0xf0] #5 [_start+0x29] #6 [<unknown>+0x29] Subtest framebuffer-vs-set-tiling failed. **** DEBUG **** Test assertion failure function pitch_tests, file kms_addfb_basic.c:167: Failed assertion: drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0 Last errno: 22, Invalid argument **** END **** Subtest framebuffer-vs-set-tiling: FAIL (0.003s) ... IGT-Version: 1.12-git (x86_64) (Linux: 4.4.0-rc1-stereo+ x86_64) Subtest framebuffer-vs-set-tiling: SUCCESS (0.000s) Cc: stable@vger.kernel.org # v4.1+ Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Fixes: 2a80eada326f ("drm/i915: Add fb format modifier support") Testcase: igt/kms_addfb_basic/clobbered-modifier Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447261890-3960-1-git-send-email-ville.syrjala@linux.intel.com
2015-11-17drm: Fix primary plane size for stereo doubled modes for legacy setcrtcVille Syrjälä1-6/+9
Properly double the hdisplay/vdisplay timings that we use as the primary plane size with stereo doubled modes. Otherwise the modeset gets rejected on machines where the primary plane must be fullscreen, and on the rest only the first eye would get a visible plane. Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: stable@vger.kernel.org #v3.19+ Fixes: 042652ed9599 ("drm/atomic-helper: implementatations for legacy interfaces") Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447686157-29607-1-git-send-email-ville.syrjala@linux.intel.com Testcase: igt/kms_3d Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2015-11-17drm/i915: Clear intel_crtc->atomic before updating it.Maarten Lankhorst1-0/+3
If an atomic update fails intel_crtc->atomic may have have some values left from the last atomic check. One example is atomic->wait_for_vblank, which results in spurious errors in kms_flip. [ 1551.892708] ------------[ cut here ]------------ [ 1551.892721] WARNING: CPU: 3 PID: 4179 at ../drivers/gpu/drm/drm_irq.c:1199 drm_wait_one_vblank+0x197/0x1a0 [drm]() [ 1551.892722] vblank not available on crtc 2, ret=-22 [ 1551.892751] Modules linked in: snd_hda_intel i915 drm_kms_helper drm intel_gtt i2c_algo_bit cfbfillrect syscopyarea cfbimgblt sysfillrect sysimgblt fb_sys_fops cfbcopyarea agpgart cfg80211 binfmt_misc snd_hda_codec_hdmi intel_rapl iosf_mbi x86_pkg_temp_thermal coretemp kvm_intel snd_hda_codec_realtek kvm snd_hda_codec_generic iTCO_wdt aesni_intel aes_x86_64 glue_helper lrw snd_hda_codec gf128mul ablk_helper cryptd snd_hwdep psmouse snd_hda_core pcspkr snd_pcm snd_timer snd lpc_ich i2c_i801 mfd_core soundcore wmi evdev [last unloaded: drm] [ 1551.892753] CPU: 3 PID: 4179 Comm: kms_pipe_crc_ba Tainted: G U W 4.3.0-reg+ #6 [ 1551.892754] Hardware name: /DZ77BH-55K, BIOS BHZ7710H.86A.0100.2013.0517.0942 05/17/2013 [ 1551.892758] ffffffffa03128d8 ffff8800cec73890 ffffffff812c0f3c ffff8800cec738d8 [ 1551.892760] ffff8800cec738c8 ffffffff8104ff36 ffff880116ae2290 0000000000000002 [ 1551.892762] ffff8800d39fcda0 ffff8800d038b4d0 ffff8800d42b5550 ffff8800cec73928 [ 1551.892763] Call Trace: [ 1551.892768] [<ffffffff812c0f3c>] dump_stack+0x4e/0x82 [ 1551.892771] [<ffffffff8104ff36>] warn_slowpath_common+0x86/0xc0 [ 1551.892773] [<ffffffff8104ffbc>] warn_slowpath_fmt+0x4c/0x50 [ 1551.892781] [<ffffffffa02e6708>] ? drm_vblank_get+0x78/0xd0 [drm] [ 1551.892787] [<ffffffffa02e6d47>] drm_wait_one_vblank+0x197/0x1a0 [drm] [ 1551.892813] [<ffffffffa03d052f>] intel_post_plane_update+0xef/0x120 [i915] [ 1551.892832] [<ffffffffa03d11d2>] intel_atomic_commit+0x4c2/0x1600 [i915] [ 1551.892862] [<ffffffffa02ff0c7>] ? drm_atomic_check_only+0x147/0x5e0 [drm] [ 1551.892872] [<ffffffffa02feeb7>] ? drm_atomic_add_affected_connectors+0x27/0xf0 [drm] [ 1551.892881] [<ffffffffa02ff597>] drm_atomic_commit+0x37/0x60 [drm] [ 1551.892887] [<ffffffffa034301a>] restore_fbdev_mode+0x28a/0x2c0 [drm_kms_helper] [ 1551.892895] [<ffffffffa0345253>] drm_fb_helper_restore_fbdev_mode_unlocked+0x33/0x80 [drm_kms_helper] [ 1551.892900] [<ffffffffa03452cd>] drm_fb_helper_set_par+0x2d/0x50 [drm_kms_helper] [ 1551.892920] [<ffffffffa03e7a9a>] intel_fbdev_set_par+0x1a/0x60 [i915] [ 1551.892923] [<ffffffff8131a5a7>] fb_set_var+0x1a7/0x3f0 [ 1551.892927] [<ffffffff8109732f>] ? trace_hardirqs_on_caller+0x12f/0x1c0 [ 1551.892931] [<ffffffff81314f32>] fbcon_blank+0x212/0x2f0 [ 1551.892935] [<ffffffff81373f4a>] do_unblank_screen+0xba/0x1d0 [ 1551.892937] [<ffffffff8136b725>] vt_ioctl+0x13d5/0x1450 [ 1551.892940] [<ffffffff8107cdd1>] ? preempt_count_sub+0x41/0x50 [ 1551.892943] [<ffffffff8135d8a3>] tty_ioctl+0x423/0xe30 [ 1551.892947] [<ffffffff8119f721>] do_vfs_ioctl+0x301/0x560 [ 1551.892949] [<ffffffff8119b1e3>] ? putname+0x53/0x60 [ 1551.892952] [<ffffffff811ab376>] ? __fget_light+0x66/0x90 [ 1551.892955] [<ffffffff8119f9f9>] SyS_ioctl+0x79/0x90 [ 1551.892958] [<ffffffff81552e97>] entry_SYSCALL_64_fastpath+0x12/0x6f [ 1551.892961] ---[ end trace 3e764d4b6628c91c ]--- Testcase: kms_flip Reported-and-tested-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: stable@vger.kernel.org #v4.3 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Daniel Stone <daniels@collabora.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/5649C2BA.6080300@mblankhorst.nl
2015-11-17drm/core: Fix old_fb handling in pan_display_atomic.Maarten Lankhorst1-22/+8
Don't touch plane->old_fb/fb without having the right locks held. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447237751-9663-6-git-send-email-maarten.lankhorst@ubuntu.com
2015-11-17drm/core: Fix old_fb handling in restore_fbdev_mode_atomic.Maarten Lankhorst1-15/+6
Don't touch plane->old_fb/fb without having the right locks held. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447237751-9663-5-git-send-email-maarten.lankhorst@ubuntu.com
2015-11-17drm/atomic: add a drm_atomic_clean_old_fb helper.Maarten Lankhorst1-18/+40
This is useful for all the boilerplate code about cleaning old_fb. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447237751-9663-4-git-send-email-maarten.lankhorst@ubuntu.com
2015-11-17drm/core: Fix old_fb handling in drm_mode_atomic_ioctl.Maarten Lankhorst1-1/+2
plane_mask should be cleared inside the retry loop, because it gets reset on every retry. Without this fix the plane->fb refcounting might get out of sync on retries, resulting in either leaked memory or use-after-free. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: stable@vger.kernel.org #v4.3 Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447237751-9663-3-git-send-email-maarten.lankhorst@ubuntu.com
2015-11-17drm/core: Set legacy_cursor_update in drm_atomic_helper_disable_plane.Maarten Lankhorst1-3/+3
legacy_cursor_update was being set in restore_fbdev_mode_atomic which was probably unintended. Fix this by only setting it in the function that needs it. This oversight was introduced in commit bbb1e52402b2a288b09ae37e8182599931c7e9df Author: Rob Clark <robdclark@gmail.com> Date: Tue Aug 25 15:35:58 2015 -0400 drm/fb-helper: atomic restore_fbdev_mode()... Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> [Jani: checkpatch fix] Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447237751-9663-2-git-send-email-maarten.lankhorst@ubuntu.com
2015-11-17drm/amdgpu: remove the unnecessary parameter adev for amdgpu_sa_bo_new()Junwei Zhang4-8/+6
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-17drm/amdgpu: wait interruptible when semaphores are disabled v2Christian König1-2/+8
Otherwise debugging locked up processes isn't possible. v2: rebased Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1)
2015-11-16drm/amdgpu: update pd while updating vm as wellChunming Zhou1-0/+3
Change-Id: I93a861cd6707f7d91672b9e19757cc50008cd7a2 Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-16drm/amdgpu: fix handling order in scheduler CSChristian König1-5/+5
We need to clear parser.ibs and num_ibs before amd_sched_fence_create, otherwise the IB could be freed twice if fence creates fails. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-11-16drm/amdgpu: fix incorrect mutex usage v3Christian König5-48/+37
Before this patch the scheduler fence was created when we push the job into the queue, so we could only get the fence after pushing it. The mutex now was necessary to prevent the thread pushing the jobs to the hardware from running faster than the thread pushing the jobs into the queue. Otherwise the thread pushing jobs into the queue would have accessed possible freed up memory when it tries to get a reference to the fence. So what you get in the end is thread A: mutex_lock(&job->lock); ... Kick of thread B. ... mutex_unlock(&job->lock); And thread B: mutex_lock(&job->lock); .... mutex_unlock(&job->lock); kfree(job); I'm actually not sure if I'm still up to date on this, but this usage pattern used to be not allowed with mutexes. See here as well https://lwn.net/Articles/575460/. v2: remove unrelated changes, fix missing owner v3: rebased, add more commit message Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: cleanup scheduler fence get/put danceChristian König2-2/+0
The code was correct, but getting two references when the ownership is linearly moved on is a bit awkward and just overhead. Signed: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: add command submission workflow tracepointChunming Zhou5-4/+76
OGL needs these tracepoints to investigate performance issue. Change-Id: I5e58187d061253f7d665dfce8e4e163ba91d3e2b Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
2015-11-16drm/amdgpu: update Fiji's tiling mode tableFlora Cui1-1/+291
Change-Id: I925c15015390113f7e27746ec5751eaa6a92c2a7 Signed-off-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: fix bug that can't enter thermal interrupt for bonaire.Rex Zhu1-4/+4
Set reversed bit to enable/disable thermal interrupt. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-11-16drm/amdgpu: fix seq_printf format stringArnd Bergmann1-3/+3
The amdgpu driver has a debugfs interface that shows the amount of VRAM in use, but the newly added code causes a build error on all 32-bit architectures: drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c:1076:17: warning: format '%lu' expects argument of type 'long unsigned int', but argument 4 has type 'long long int' [-Wformat=] This fixes the format string to use "%llu" for printing 64-bit numbers, which works everywhere, as long as we also cast to 'u64'. Unlike atomic64_t, u64 is defined as 'unsigned long long' on all architectures. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Fixes: a2ef8a974931 ("drm/amdgpu: add vram usage into debugfs") Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: fix quirk for MSI R7 370 Armor 2XMaxim Sheviakov1-1/+1
There was a typo in the original. bug: https://bugs.freedesktop.org/show_bug.cgi?id=92865 Signed-off-by: Maxim Sheviakov <mrader3940@yandex.ru> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: Fix default page access routingJay Cornwall2-0/+2
The VM default page (used when a VM translation fails) is allocated in system memory. The VM is misconfigured to interpret the physical address as referencing a VRAM physical page. Route default page accesses to system memory. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Jay Cornwall <jay@jcornwall.me> Cc: <stable@vger.kernel.org> # v4.2+ Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: unconditionally set sysfs_initializedAlex Deucher1-2/+1
Avoids spew on resume for systems where sysfs may fail even on init. bug: https://bugzilla.kernel.org/show_bug.cgi?id=106851 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2015-11-16drm/amdgpu: cleanup amdgpu_cs_parser handlingChristian König2-60/+33
No need any more to allocate that structure dynamically, just put it on the stack. This is a start to cleanup some of the scheduler fallouts. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: fix leaking the IBs on errorChristian König1-13/+12
Fixing a memory leak when the scheduler is enabled. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amd: add kmem cache for sched fenceChunming Zhou3-2/+23
Change-Id: I45bb8ff10ef05dc3b15e31a77fbcf31117705f11 Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-16drm/amdgpu: add kmem cache for amdgpu fenceChunming Zhou1-2/+20
Change-Id: I5ad8dd156ccf27a6f18004aa0a215a0925b6e67b Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-16drm/amdgpu: update fiji_mgcg_cgcg_init tableFlora Cui1-5/+0
Change-Id: If44b8057741c78208f1976f60f31b535c944d0bd Signed-off-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-11-16drm/amdgpu: use common fence for amdgpu_vm_fenceChristian König3-8/+8
Just cleanup the function parameters. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: use fence_is_later() for vm_flush as well v2Christian König1-12/+9
v2: remove superfluous check Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1) Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-11-16drm/amdgpu: use a timer for fence fallbackChristian König2-46/+35
Less overhead than a work item and also adds proper cleanup handling. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: remove fence trace pointsChristian König2-44/+0
Mostly unused and replaced by the common trace points. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: update Fiji's mmPA_SC_RASTER_CONFIG valueFlora Cui1-1/+4
Change-Id: I6d138306a878450e5bf8a77a2f1aacc380a39fe5 Signed-off-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: Only prompt for enabling PAT when we'd allow write-combiningMichel Dänzer1-2/+3
No use bothering users about this for whom we disable write-combining for other reasons anyway. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: Always disable RADEON_GEM_GTT_UC along with RADEON_GEM_GTT_WCMichel Dänzer1-2/+2
Write-combining is a CPU feature. From the GPU POV, these both simply mean no GPU<->CPU cache coherency. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: Disable uncacheable CPU mappings of GTT with RV6xxMichel Dänzer1-0/+6
They reportedly cause random GPU hangs. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91268 Reviewed-by: Christian König <christian.koenig@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-27 23:58:54 +0300