summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2015-11-03drm/amdgpu: don't oops on failure to load (v2)Dave Airlie1-0/+3
In two places amdgpu tries to tear down something it hasn't initalised when failing. This is what happens when you enable experimental support on topaz which then fails in ring init. This patch allows it to fail cleanly. v2 (agd): split out scheduler change into a separate patch Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2015-11-03drm/amdgpu: don't VT switch on suspendAlex Deucher1-0/+1
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-02drm/amdgpu: Make amdgpu_mn functions inlineHarry Wentland1-2/+2
Unused amdgpu_mn functions threw warnings for every file that includes amdgpu.h. It makes sense to inline this amdgpu_mn stubs to avoid the warning. Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Cc: stable@vger.kernel.org
2015-10-30drm/amdgpu: remove amdgpu_fence_ref/unrefChristian König4-35/+4
Just move the remaining users to fence_put/get. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30drm/amdgpu: use common fence for syncChristian König2-11/+12
Stop using the driver internal functions. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30drm/amdgpu: use the new fence_is_laterChristian König1-17/+13
Instead of coding the check ourself. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30drm/amdgpu: use common fences for VMID management v2Christian König4-18/+21
v2: add missing NULL check. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30drm/amdgpu: move ring_from_fence to common codeChristian König3-21/+27
Going to need that elsewhere as well. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30drm/amdgpu: switch to common fence_wait_any_timeout v2Christian König3-109/+13
No need to duplicate the functionality any more. v2: fix handling if no fence is available. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1)
2015-10-30drm/amdgpu: remove unneeded fence functionsChristian König1-12/+1
amdgpu_fence_default_wait isn't needed any more the default wait does the same thing and amdgpu_test_signaled is dead as well. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30dma-buf/fence: add fence_is_later()Christian König1-3/+19
Return true when fence 1 is later than fence 2 without checking if any of them are signaled. Useful for driver specific resource handling based on fences. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-30dma-buf/fence: add fence_wait_any_timeout function v2Christian König2-1/+100
Waiting for the first fence in an array of fences to signal. This is useful for device driver specific resource managers and also Vulkan needs something similar. v2: more parameter checks, handling for timeout==0, remove NULL entry support, better callback removal. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
2015-10-29drm/amdgpu: ignore scheduler fences from the same entityChristian König1-0/+6
We are going to submit them before the job anyway. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-10-28drm/amdgpu: add GFX 8.1 register headersAlex Deucher3-0/+30967
Minor differences compared to GFX 8.0 Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-28drm/amdgpu: add some additional CZ revisionsAlex Deucher1-0/+7
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2015-10-27drm/radeon/dce4+: optimize pageflip (v2)Alex Deucher1-28/+4
Taking the grph update lock is only necessary when updating the the secondary address (for single pipe stereo). v2: fix comment as per Michel's suggestion Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-23drm/amdgpu/dce11: optimize pageflipAlex Deucher1-29/+5
Taking the grph update lock is only necessary when updating the the secondary address (for single pipe stereo). Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-23drm/amdgpu/dce10: optimize pageflipAlex Deucher1-30/+6
Taking the grph update lock is only necessary when updating the the secondary address (for single pipe stereo). Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-23drm/amdgpu/dce8: optimize pageflipAlex Deucher1-31/+7
Taking the grph update lock is only necessary when updating the the secondary address (for single pipe stereo). Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: change VM size default to 64GBChristian König1-2/+2
That's still small enough to not waste to much memory on PD/PTs. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add Stoney pci idsSamuel Li1-0/+2
Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: update the core VI support for StoneySamuel Li3-7/+34
Add core VI enablement for Stoney. Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add VCE support for Stoney (v2)Samuel Li2-2/+8
Stoney is VCE 3.x single. v2: Stoney is single pipe like Fiji Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add UVD support for StoneySamuel Li1-0/+5
Stoney is UVD 6.x. Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add GFX support for Stoney (v2)Samuel Li1-18/+382
Stoney is GFX 8.1. v2: update to latest golden settings Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add SDMA support for Stoney (v2)Samuel Li1-0/+28
Stoney is SDMA 3.x. v2: update to latest golden register settings Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add DCE support for StoneySamuel Li1-1/+18
Stoney is DCE 11.x. Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: Update SMC/DPM for StoneySamuel Li2-16/+65
Stoney is SMC 8.x. Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add GMC support for StoneySamuel Li1-0/+12
Stoney is GMC 8.x. Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: add Stoney chip familySamuel Li2-0/+2
Stoney is based on Carrizo with some IP upgrades. Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-21drm/amdgpu: fix the broken vm->mutex V2Chunming Zhou3-46/+19
fix the vm->mutex and ww_mutex confilcts. vm->mutex is always token first, then ww_mutex. V2: remove unneccessary checking for pt bo. Change-Id: Iea56e183752c02831126d06d2f5b7a474a6e4743 Signed-off-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-10-21drm/amdgpu: remove the unnecessary parameter adev for amdgpu_fence_wait_any()Junwei Zhang3-10/+4
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-10-21drm/amdgpu: remove the exclusive lockChristian König6-42/+8
Finally getting rid of it. Signed-off-by: Christian König <christian.koenig@amd.com>
2015-10-21drm/amdgpu: remove old lockup detection infrastructureChristian König15-169/+1
It didn't worked to well anyway. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
2015-10-21drm: fix trivial typosGeliang Tang4-4/+4
s/regsiter/register/ Signed-off-by: Geliang Tang <geliangtang@163.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-19drm/amdgpu/dce: simplify suspend/resumeAlex Deucher3-33/+14
We were basically opencoding the same thing in both hw_init and resume and hw_fini and suspend. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-19drm/amdgpu/gfx8: set TC_WB_ACTION_EN in RELEASE_MEM packetAlex Deucher1-0/+1
This is the recommended setting from the hw team for newer versions of the firmware. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2015-10-15drm/radeon: Use rdev->gem.mutex to protect hyperz/cmask ownersDaniel Vetter1-2/+8
This removes the last depency of radeon for dev->struct_mutex! Also the locking scheme for hyperz/cmask owners seems a bit unsound, there's no protection in the preclose handler (and that never did hold dev->struct_mutex while being called). So grab the same lock there, too. There's also all the checks in the cs checker, but since the overall design seems to never stall for the previous owner I figured it's ok if I leave this racy. It was racy even before I touched it after all too. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: fix lockup when clean pending fencesJunwei Zhang1-3/+3
The first lockup fence will lock the fence list of scheduler. Then cancel the delayed workqueues for all clean pending fences without waiting the workqueues to finish. Change-Id: I9bec826de1aa49d587b0662f3fb4a95333979429 Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-10-14drm/amdgpu: add timer to fence to detect scheduler lockupJunwei Zhang3-3/+61
Change-Id: I67e987db0efdca28faa80b332b75571192130d33 Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: David Zhou <david1.zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-10-14drm/amdgpu: add VM CS mapping trace pointChristian König2-1/+19
Output all VM mappings a command submission uses. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: add option to clear VM page tables after every submitChristian König3-1/+21
This makes it much easier to find when userspace misses to send some buffers. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: add option to stop on VM faultChristian König4-6/+80
Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: only print meaningful VM faultsChristian König2-4/+12
Port of radeon commit 9b7d786b900baf7c0d1a7e211570aef1cb27590f. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: also trace already allocated VMIDsChristian König1-1/+3
Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: Drop unnecessary #include <linux/vga_switcheroo.h>Lukas Wunner2-2/+0
This was added to two radeon files even though they don't use any vga_switcheroo symbols, the amdgpu fork inherited them: Added to amdgpu_acpi.c by commit d7a2952f1ade ("drm/radeon: Add support for the ATIF ACPI method to the radeon driver"). Added to amdgpu_bios.c by commit 6a9ee8af344e ("vga_switcheroo: initial implementation (v15)"). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/radeon: Drop unnecessary #include <linux/vga_switcheroo.h>Lukas Wunner3-3/+0
This was added to three files even though they don't use any vga_switcheroo symbols: Added to radeon_acpi.c by commit d7a2952f1ade ("drm/radeon: Add support for the ATIF ACPI method to the radeon driver"). Added to radeon_asic.c by commit 0a10c85129c2 ("drm/radeon: create radeon_asic.c"). Added to radeon_bios.c by commit 6a9ee8af344e ("vga_switcheroo: initial implementation (v15)"). Signed-off-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: clean up pageflip interrupt handlingAlex Deucher3-159/+51
Check to make sure we aren't touching a non-existent display controller and simplify the code. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: rework sdma structuresAlex Deucher9-258/+245
Rework the sdma structures in the driver to consolidate all of the sdma info into a single structure and allow for asics that may have different numbers of sdma instances. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-10-14drm/amdgpu: unpin cursor BOs on suspend and pin them again on resumeAlex Deucher1-1/+30
Everything is evicted from VRAM before suspend, so we need to make sure all BOs are unpinned and re-pinned after resume. Fixes broken mouse cursor after resume introduced by commit b9729b17. Port of radeon commit: f3cbb17bcf676a2fc6aedebe9fbebd59e550c51a Reviewed-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 2024-11-25 10:06:34 +0300