OSDN Git Service
Daniel Vetter [Thu, 9 Jul 2015 21:32:49 +0000 (23:32 +0200)]
drm/amdgpu: Don't take dev->struct_mutex in bo_force_delete
It really doesn't protect anything which doesn't have other locks
already. Also this is run from driver unload code so not much need for
locks anyway.
Same changes as for radeon really.
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Daniel Vetter [Thu, 9 Jul 2015 21:32:48 +0000 (23:32 +0200)]
drm/radeon: Don't take dev->struct_mutex in pm functions
We already grab 2 device-global locks (write-sema rdev->pm.mclk_lock
and rdev->ring_lock), adding another global mutex won't serialize this
code more. And since there's really nothing interesting that gets
protected in radeon by dev->struct mutex (we only have the global z
buffer owners and it's still serializing gem bo destruction in the drm
core - which is irrelevant since radeon uses ttm anyway internally)
this doesn't add protection. Remove it.
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Daniel Vetter [Thu, 9 Jul 2015 21:32:47 +0000 (23:32 +0200)]
drm/radeon: Don't take dev->struct_mutex in bo_force_delete
It really doesn't protect anything which doesn't have other locks
already. Also this is run from driver unload code so not much need for
locks anyway.
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Fri, 7 Aug 2015 18:22:40 +0000 (20:22 +0200)]
drm/amdgpu: remove VI hw bug workaround v3
The workaround simply doesn't work because VM mappings
are controlled by userspace not the kernel.
Additional to that this is just a performance problem
which happens if you have holes in your VM mapping.
v2: adjust virtual addr alignment as well.
v3: fix trivial warning
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com> (v1)
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> (v2)
Christian König [Fri, 7 Aug 2015 14:15:36 +0000 (16:15 +0200)]
drm/amdgpu: cleanup amdgpu_fence_ring_wait_seq
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Fri, 7 Aug 2015 13:57:21 +0000 (15:57 +0200)]
drm/amdgpu: remove duplicate amdgpu_fence_process implementation
Looks like that somehow got missed while during porting the radeon changes.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Fri, 7 Aug 2015 11:53:36 +0000 (13:53 +0200)]
drm/amdgpu: remove amdgpu_fence_wait
It was just a wrapper for fence_wait anyway.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Thu, 6 Aug 2015 18:44:47 +0000 (20:44 +0200)]
drm/amdgpu: use the reservation obj wait for the UVD msg
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Thu, 6 Aug 2015 15:49:25 +0000 (17:49 +0200)]
drm/amdgpu: remove amdgpu_fence_signaled
The common kernel function does the same thing.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Sun, 2 Aug 2015 03:18:04 +0000 (11:18 +0800)]
drm/amd: add scheduler fence implementation (v2)
scheduler fence is based on kernel fence framework.
v2: squash in Christian's build fix
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 04:57:31 +0000 (12:57 +0800)]
drm/amdgpu: use kernel submit helper in vm
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Thu, 6 Aug 2015 07:19:12 +0000 (15:19 +0800)]
drm/amdgpu: use amd_sched_job in its backend ops
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Christian König [Wed, 5 Aug 2015 19:22:10 +0000 (21:22 +0200)]
drm/amdgpu: cleanup and fix scheduler fence handling v2
v2: rebased
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>
Christian König [Wed, 5 Aug 2015 16:33:21 +0000 (18:33 +0200)]
drm/amdgpu: merge amd_sched_entity and amd_context_entity v2
Avoiding a couple of casts.
v2: rename c_entity to entity as well
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Wed, 5 Aug 2015 16:18:52 +0000 (18:18 +0200)]
drm/amdgpu: fix coding style in a couple of places
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Wed, 5 Aug 2015 16:11:14 +0000 (18:11 +0200)]
drm/amdgpu: remove unused parent entity
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Chunming Zhou [Wed, 5 Aug 2015 11:52:14 +0000 (19:52 +0800)]
drm/amdgpu: process sched job exactly triggered by fence signal
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Wed, 5 Aug 2015 11:07:08 +0000 (19:07 +0800)]
Revert "drm/amdgpu: return new seq_no for amd_sched_push_job"
This reverts commit
d1d33da8eb86b8ca41dd9ed95738030df5267b95.
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Conflicts:
drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
Christian König [Tue, 4 Aug 2015 15:51:05 +0000 (17:51 +0200)]
drm/amdgpu: cleanup amdgpu_ctx inti/fini v2
Cleanup the kernel context handling.
v2: rebased
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com> (v1)
Christian König [Tue, 4 Aug 2015 14:58:36 +0000 (16:58 +0200)]
drm/amdgpu: stop leaking the ctx id into the scheduler v2
Id's are for the IOCTL ABI only.
v2: remove tgid as well
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Tue, 4 Aug 2015 14:20:31 +0000 (16:20 +0200)]
drm/amdgpu: cleanup ctx_mgr init/fini
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Tue, 4 Aug 2015 09:54:48 +0000 (11:54 +0200)]
drm/amdgpu: fix bo list handling in CS
We didn't initialized the mutex in the cloned bo list resulting in nice
warnings from lockdep. Also fixes error handling in this function.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Mon, 3 Aug 2015 18:39:12 +0000 (20:39 +0200)]
drm/amdgpu: reorder the code to avoid forward declerations
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Chunming Zhou [Tue, 4 Aug 2015 03:30:09 +0000 (11:30 +0800)]
drm/amdgpu: move sched job process from isr to fence callback
This way can avoid interrupt lost, and can process sched job exactly.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Jammy Zhou [Mon, 3 Aug 2015 02:27:57 +0000 (10:27 +0800)]
drm/amdgpu: add amd_sched_next_queued_seq function
This function is used to get the next queued sequence number
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 31 Jul 2015 09:54:29 +0000 (17:54 +0800)]
drm/amdgpu: make last_handled_seq atomic
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 31 Jul 2015 09:18:15 +0000 (17:18 +0800)]
drm/amdgpu: add amd_sched_commit
This function is to update last_emitted_v_seq and wake up the waiters.
It should be called by driver in the run_job backend function
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 31 Jul 2015 08:47:28 +0000 (16:47 +0800)]
drm/amdgpu: return new seq_no for amd_sched_push_job
It is clean to update last_queued_v_seq in the scheduler module
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Thu, 30 Jul 2015 09:19:52 +0000 (17:19 +0800)]
drm/amdgpu: some code refinement v2
Fix the code alignment, etc.
v2: rebase the code
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 12:02:43 +0000 (20:02 +0800)]
drm/amdgpu: fix null pointer by previous cleanup
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 03:43:19 +0000 (11:43 +0800)]
drm/amdgpu: add kernel fence in ib_submit_kernel_helper
every sbumission should be able to get a fence.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 05:22:35 +0000 (13:22 +0800)]
drm/amdgpu: use kernel fence for sdma ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 05:19:43 +0000 (13:19 +0800)]
drm/amdgpu: use kernel fence for gfx ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 08:19:29 +0000 (16:19 +0800)]
drm/amdgpu: use kernel fence in amdgpu_test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 05:28:16 +0000 (13:28 +0800)]
drm/amdgpu: use kernel fence for vce ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 05:11:04 +0000 (13:11 +0800)]
drm/amdgpu: change uvd ib test to use kernel fence directly
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 10:19:38 +0000 (18:19 +0800)]
drm/amdgpu: use kernel fence for last_pt_update
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Mon, 3 Aug 2015 03:38:09 +0000 (11:38 +0800)]
drm/amdgpu: use kernel fence diretly in amdgpu_bo_fence
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Christian König [Fri, 31 Jul 2015 12:31:49 +0000 (14:31 +0200)]
drm/amdgpu: clean up amd sched wait_ts and wait_signal
Remove code not used at the moment.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Thu, 30 Jul 2015 09:53:42 +0000 (11:53 +0200)]
drm/amdgpu: stop using addr to check for BO move v3
It is theoretically possible that a swapped out BO gets the
same GTT address, but different backing pages while being swapped in.
Instead just use another VA state to note updated areas.
Ported from not upstream yet radeon commit with the same name.
v2: fix some bugs in the original implementation found in the radeon code.
v3: squash in VCE/UVD fix
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
monk.liu [Thu, 30 Jul 2015 10:28:12 +0000 (18:28 +0800)]
drm/amdgpu: move wait_queue_head from adev to ring (v2)
thus unnecessary wake up could be avoid between rings
v2:
move wait_queue_head to fence_drv from ring
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
monk.liu [Thu, 30 Jul 2015 10:26:18 +0000 (18:26 +0800)]
drm/amdgpu: re-implement fence_default_wait
use fence_wait_any to implement fence_default_wait
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
monk.liu [Thu, 30 Jul 2015 07:19:05 +0000 (15:19 +0800)]
drm/amdgpu: new implement for fence_wait_any (v2)
origninal method will sleep/schedule at the granurarity of HZ/2 and
based on seq signal method, the new implement is based on kernel fance
interface, no unnecessary schedule at all
v2: replace logic of original amdgpu_fence_wait_any
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
monk.liu [Thu, 30 Jul 2015 06:56:18 +0000 (14:56 +0800)]
drm/amdgpu: use kernel fence interface when possible
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Chunming Zhou [Fri, 3 Jul 2015 06:18:26 +0000 (14:18 +0800)]
drm/amdgpu: use scheduler for VCE ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Fri, 3 Jul 2015 06:08:18 +0000 (14:08 +0800)]
drm/amdgpu: use scheduler for UVD ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Mon, 1 Jun 2015 07:33:20 +0000 (15:33 +0800)]
drm/amdgpu: use gpu scheduler for sdma ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Mon, 1 Jun 2015 07:16:03 +0000 (15:16 +0800)]
drm/amdgpu: Use gpu scheduler for gfx ring ib test
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Wed, 29 Jul 2015 02:33:14 +0000 (10:33 +0800)]
drm/amdgpu: add helper function for kernel submission
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Thu, 30 Jul 2015 09:59:43 +0000 (17:59 +0800)]
drm/amdgpu: fix seq in ctx_add_fence
if enabling scheduler, then the queued seq is assigned
when pushing job before emitting job.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Chunming Zhou [Tue, 28 Jul 2015 09:31:04 +0000 (17:31 +0800)]
drm/amdgpu: wait forever for wait emit
the job must be emitted by scheduler, otherwise scheduler is abnormal.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Jammy Zhou [Thu, 30 Jul 2015 08:44:05 +0000 (16:44 +0800)]
drm/amdgpu: add amdgpu.sched_hw_submission option
This option can be used to specify the max number of submissions in the
active HW queue. The default value is 2 now.
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Jammy Zhou [Thu, 30 Jul 2015 08:36:58 +0000 (16:36 +0800)]
drm/amdgpu: add amdgpu.sched_jobs option
This option can be used to specify the max job number in the job queue,
and it is 16 by default.
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Jammy Zhou [Thu, 30 Jul 2015 08:24:53 +0000 (16:24 +0800)]
drm/amdgpu: silent the message for GPU scheduler creation
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Mon, 27 Jul 2015 13:40:35 +0000 (15:40 +0200)]
drm/amdgpu: fix syncing to VM updates
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Tue, 28 Jul 2015 08:11:52 +0000 (16:11 +0800)]
drm/amdgpu: add check for callback
it is possible that the callback isn't defined sometimes.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
Jammy Zhou [Wed, 27 May 2015 10:23:34 +0000 (18:23 +0800)]
drm/amdgpu: add enable_scheduler module option
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Fri, 24 Jul 2015 02:49:47 +0000 (10:49 +0800)]
drm/amdgpu: protect fence_process from multiple context
fence_process may be called from kthread, user thread and interrupt context.
it is possible to called concurrently, then will wake up fence queue multiple times.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 09:43:41 +0000 (17:43 +0800)]
drm/amdgpu: add sched isr to fence process
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 08:52:10 +0000 (16:52 +0800)]
drm/amdgpu: dispatch job for vm
use kernel context to submit command for vm
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Mon, 6 Jul 2015 05:42:58 +0000 (13:42 +0800)]
drm/amdgpu: add kernel ctx support (v2)
v2: rebase against kfd changes
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 08:04:39 +0000 (16:04 +0800)]
drm/amdgpu: prepare job before push to sw queue for pte ring
user mode will still use pte ring as a normal ring.
if the prepare job generates another command(update pte) on its ring in scheduler,
then will kill scheduler which is going to waiting later job but pending running job.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 07:53:04 +0000 (15:53 +0800)]
drm/amdgpu: make sure the fence is emitted before ring to get it.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 07:13:53 +0000 (15:13 +0800)]
drm/amdgpu: use scheduler user seq instead of previous user seq
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 06:36:51 +0000 (14:36 +0800)]
drm/amdgpu: dispatch jobs in cs
BO validation is moved to scheduler except usrptr which must be validated
in user process
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 05:47:05 +0000 (13:47 +0800)]
drm/amdgpu: add bo list copy
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 05:45:14 +0000 (13:45 +0800)]
drm/amdgpu: add backend implementation of gpu scheduler (v2)
v2: fix rebase breakage
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 05:19:45 +0000 (13:19 +0800)]
drm/amdgpu: disable hw semaphore with scheduler
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Chunming Zhou [Tue, 21 Jul 2015 05:17:19 +0000 (13:17 +0800)]
drm/amdgpu: add context entity init
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Alex Deucher [Mon, 17 Aug 2015 02:55:02 +0000 (22:55 -0400)]
drm/amdgpu: add scheduler initialization
1. Add kernel parameter option, default 0
2. Add scheduler initialization for amdgpu
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Jammy Zhou [Fri, 22 May 2015 10:55:07 +0000 (18:55 +0800)]
drm/amd: add basic scheduling framework
run queue:
A set of entities scheduling commands for the same ring. It
implements the scheduling policy that selects the next entity to
emit commands from.
entity:
A scheduler entity is a wrapper around a job queue or a group of
other entities. This can be used to build hierarchies of entities.
For example all job queue entities belonging to the same process
may be placed in a higher level entity and scheduled against other
process entities.
Entities take turns emitting jobs from their job queue to the
corresponding hardware ring, in accordance with the scheduler policy.
Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Acked-by: Christian K?nig <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
David Zhang [Sat, 11 Jul 2015 15:13:40 +0000 (23:13 +0800)]
drm/amdgpu: Enable the Fiji DID 0x7300 support
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Alex Deucher [Thu, 6 Aug 2015 15:51:23 +0000 (11:51 -0400)]
drm/amdgpu: remove VM workaround for Fiji
The bug is fixed in fiji.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 27 Jul 2015 18:24:14 +0000 (14:24 -0400)]
drm/amdgpu: add support for VCE 3.x on Fiji
VCE on fiji is single pipe only.
Reviewed-by: David Zhang <david1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
David Zhang [Wed, 8 Jul 2015 09:32:15 +0000 (17:32 +0800)]
drm/amdgpu: Add Fiji support to the UVD 6.0 IP module
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Wed, 8 Jul 2015 09:29:27 +0000 (17:29 +0800)]
drm/amdgpu: Add Fiji support to the SDMA 3.0 IP module
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Thu, 30 Jul 2015 23:42:11 +0000 (19:42 -0400)]
drm/amdgpu: Add Fiji support to the GFX 8.0 IP module (v2)
v2: agd5f: fix the rb setup.
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Tue, 7 Jul 2015 17:28:20 +0000 (01:28 +0800)]
drm/amdgpu: Add Fiji support to the DCE 10.0 IP module (v2)
v2: agd5f: fix up XDMA golden settings
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Tue, 7 Jul 2015 17:23:25 +0000 (01:23 +0800)]
drm/amdgpu: Add Fiji support to SMC and DPM (v2)
v2: agd5f: prepare for release
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Wed, 8 Jul 2015 11:13:25 +0000 (19:13 +0800)]
drm/amdgpu: Add Fiji smu 7.1.3 headers (v2)
v2: agd5f: prepare for release
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Wed, 8 Jul 2015 13:40:31 +0000 (21:40 +0800)]
drm/amdgpu: Add Fiji support to IH module
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Tue, 7 Jul 2015 17:11:52 +0000 (01:11 +0800)]
drm/amdgpu: Add Fiji support to the GMC 8.5 IP module
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
David Zhang [Tue, 7 Jul 2015 17:05:16 +0000 (01:05 +0800)]
drm/amdgpu: Add Fiji DID 0x7300 common support
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Alex Deucher [Fri, 31 Jul 2015 21:20:14 +0000 (17:20 -0400)]
drm/amdgpu: handle conditional support for CIK properly
gfx7 support is not necessary or available if CIK support
is not enabled.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
monk.liu [Fri, 17 Jul 2015 10:39:25 +0000 (18:39 +0800)]
drm/amdgpu: optimize amdgpu_parser_init
use kmalloc_array instead of kcalloc where appropriate and other
cleanups.
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Wed, 22 Jul 2015 03:29:01 +0000 (11:29 +0800)]
drm/amdgpu: merge amdgpu_family.h into amd_shared.h (v2)
Make the definitions common for all driver components
v2: fix kfd
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Tue, 21 Jul 2015 13:17:16 +0000 (21:17 +0800)]
drm/amdgpu: add some pptable definitions
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Tue, 21 Jul 2015 09:41:48 +0000 (17:41 +0800)]
drm/amdgpu: add some common definitions to amd_shared.h
Add GPU family definitions and timeout value for IP components.
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Sat, 11 Jul 2015 10:08:46 +0000 (12:08 +0200)]
drm/amdgpu: allow userspace to read more debug registers
Feel free to suggest more.
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Alex Deucher [Mon, 17 Aug 2015 02:51:50 +0000 (22:51 -0400)]
drm/amdgpu: move some atombios definitions to common folder (v2)
the definitions can be shared by different IP components.
v2: fix include path
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Tue, 21 Jul 2015 16:02:21 +0000 (18:02 +0200)]
drm/amdgpu: fix UVD/VCE fence handling
We need to return the sequence number to userspace
even when we don't use user fences.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 5 Aug 2015 16:41:48 +0000 (12:41 -0400)]
drm/amdgpu: add fence suspend/resume functions
Added to:
- handle draining the ring on suspend
- properly enable/disable interrupts on suspend and resume
Fix breakages from:
commit
467ee3be53d240d08beed2e82a941e820c1ac323
Author: Chunming Zhou <david1.zhou@amd.com>
Date: Mon Jun 1 14:14:32 2015 +0800
drm/amdgpu: always enable EOP interrupt v2
Tested-by: Audrey Grodzovsky <audrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Mon, 1 Jun 2015 06:14:32 +0000 (14:14 +0800)]
drm/amdgpu: always enable EOP interrupt v2
v2 (chk): always enable EOP interrupt, independent of scheduler,
remove now unused delayed_irq handling.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Mon, 20 Jul 2015 14:09:40 +0000 (16:09 +0200)]
drm/amdgpu: rework vm_grab_id interface
This makes assigning VM IDs independent from the use of VM IDs.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Mon, 20 Jul 2015 13:47:30 +0000 (15:47 +0200)]
drm/amdgpu: no updates shouldn't cause vm flush v2
v2 (chk): split fix from original patch
Signed-off-by: monk.liu <monk.liu@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Sat, 18 Jul 2015 17:20:05 +0000 (19:20 +0200)]
drm/amdgpu: fix signed overrun in amdgpu_ctx_get_fence
Otherwise the first 16 fences of a context will always signal immediately.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Thu, 16 Jul 2015 10:01:06 +0000 (12:01 +0200)]
drm/amdgpu: fix context memory leak
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Tue, 7 Jul 2015 16:44:01 +0000 (18:44 +0200)]
drm/amdgpu: remove amdgpu_fence_recreate
It's not used any more.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Tue, 7 Jul 2015 15:24:49 +0000 (17:24 +0200)]
drm/amdgpu: add user fence context map v2
This is a prerequisite for the GPU scheduler to make the order
of submission independent from the order of execution.
v2: properly implement the locking
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Christian König [Mon, 6 Jul 2015 20:06:40 +0000 (22:06 +0200)]
drm/amdgpu: deal with foreign fences in amdgpu_sync
This also requires some error handling from the callers of that function.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>