Commit Graph

50 Commits

Author SHA1 Message Date
Christian König
a27de35caa drm/amdgpu: remove the ring lock v2
It's not needed any more because all access goes through the scheduler now.

v2: Update commit message.

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>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-02-10 14:16:58 -05:00
Alex Deucher
18db89b402 drm/amdgpu: add a debugfs property to trigger a GPU reset
Ported from similar code in radeon.

Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-02-10 14:16:55 -05:00
Christian König
5907a0d8af drm/amdgpu: cleanup sync_seq handling
Not used any more without semaphores

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>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-02-10 14:16:52 -05:00
Chunming Zhou
cadf97b196 drm/amdgpu: clean up non-scheduler code path (v2)
Non-scheduler code is longer supported.

v2: agd: rebased on upstream

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Ken Wang  <Qingqing.Wang@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-02-10 14:16:50 -05:00
Chunming Zhou
b49c84a576 drm/amdgpu: add kmem cache for amdgpu fence
Change-Id: I5ad8dd156ccf27a6f18004aa0a215a0925b6e67b
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-16 11:05:50 -05:00
Christian König
c2776afe74 drm/amdgpu: use a timer for fence fallback
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>
2015-11-16 11:05:47 -05:00
Christian König
935c186aae drm/amdgpu: remove fence trace points
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>
2015-11-16 11:05:46 -05:00
Christian König
a95e264254 drm/amdgpu: group together common fence implementation
And also add some missing function documentation. No functional change.

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>
2015-11-04 12:29:23 -05:00
Christian König
24372447e5 drm/amdgpu: fix fence fallback check
Interrupts are notorious unreliable, enable the fallback at
a couple of more places.

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>
2015-11-04 12:29:22 -05:00
Christian König
6ef68c17d4 drm/amdgpu: remove amdgpu_fence_ref/unref
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>
2015-10-30 01:58:22 -04:00
Christian König
ee327caf1a drm/amdgpu: switch to common fence_wait_any_timeout v2
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)
2015-10-30 01:51:11 -04:00
Christian König
318cd340c5 drm/amdgpu: remove unneeded fence functions
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>
2015-10-30 01:49:30 -04:00
Junwei Zhang
ce16b0e5a3 drm/amdgpu: remove the unnecessary parameter adev for amdgpu_fence_wait_any()
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-10-21 11:35:13 -04:00
Christian König
0c418f1010 drm/amdgpu: remove the exclusive lock
Finally getting rid of it.

Signed-off-by: Christian König <christian.koenig@amd.com>
2015-10-21 11:35:12 -04:00
Christian König
b7e4dad3e1 drm/amdgpu: remove old lockup detection infrastructure
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>
2015-10-21 11:35:12 -04:00
Junwei Zhang
2440ff2c91 drm/amdgpu: add timer to fence to detect scheduler lockup
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>
2015-10-14 16:16:42 -04:00
Christian König
4f839a243d drm/amdgpu: more scheduler cleanups v2
Embed the scheduler into the ring structure instead of allocating it.
Use the ring name directly instead of the id.

v2: rebased, whitespace cleanup

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Chunming Zhou<david1.zhou@amd.com>
2015-09-23 17:23:39 -04:00
Christian König
5ec92a7692 drm/amdgpu: cleanup fence queue init v2
Move the fence related stuff into amdgpu_fence.c

v2: rework commit message, cause this is actually not a bug

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>
2015-09-23 17:23:38 -04:00
Christian König
9b398fa5c2 drm/amdgpu: rename fence->scheduler to sched v2
Just to be consistent with the other members.

v2: rename the ring member as well.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> (v1)
Reviewed-by: Chunming Zhou<david1.zhou@amd.com>
2015-09-23 17:23:37 -04:00
Christian König
8221d706f4 drm/amdgpu: partially revert "modify amdgpu_fence_wait_any() to amdgpu_fence_wait_multiple()" v2
That isn't used any more.

v2: rebase

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-09-02 12:19:52 -04:00
Junwei Zhang
113cd9dacb drm/amdgpu: adjust the judgement of removing fence callback
The fence in the array may be skipped if wait_all is false,
thus the related callback is not initialized with list head.
So removing this kind callback will cause NULL pointer reference.

Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-08-26 17:50:04 -04:00
Chunming Zhou
f38fdfddfa drm/amdgpu: add priv data to sched
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
2015-08-25 10:52:18 -04:00
Christian König
69f7dd652c drm/amdgpu: remove unused parameters to amd_sched_create
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-08-25 10:47:41 -04:00
Chunming Zhou
4ce9891ee1 drm/amdgpu: improve sa_bo->fence by kernel fence
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian K?nig <christian.koenig@amd.com>
2015-08-25 10:38:41 -04:00
Junwei Zhang
1aa4051b7f drm/amdgpu: modify amdgpu_fence_wait_any() to amdgpu_fence_wait_multiple()
Rename the function and update the related code with this modified function.
Add the new parameter of bool wait_all.

If wait_all is true, it will return when all fences are signaled or timeout.
If wait_all is false, it will return when any fence is signaled or timeout.

Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-08-25 10:38:28 -04:00
Christian König
00d2a2b21d 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>
2015-08-17 16:51:10 -04:00
Christian König
68ed3de434 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>
2015-08-17 16:51:10 -04:00
Christian König
02bc0650bc 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>
2015-08-17 16:51:09 -04:00
Christian König
3cdb8119d9 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>
2015-08-17 16:51:08 -04:00
Chunming Zhou
7484667c6a 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>
2015-08-17 16:50:59 -04:00
monk.liu
7f06c236b9 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>
2015-08-17 16:50:48 -04:00
monk.liu
e29551556e 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>
2015-08-17 16:50:48 -04:00
monk.liu
332dfe907b 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>
2015-08-17 16:50:47 -04:00
monk.liu
2e536084f2 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>
2015-08-17 16:50:47 -04:00
Jammy Zhou
4afcb30383 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>
2015-08-17 16:50:42 -04:00
Chunming Zhou
176e1ab1b5 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>
2015-08-17 16:50:38 -04:00
Chunming Zhou
e0d8f3c34e 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>
2015-08-17 16:50:38 -04:00
Chunming Zhou
c1b69ed0c6 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>
2015-08-17 16:50:33 -04:00
Alex Deucher
b80d8475c1 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>
2015-08-17 16:50:31 -04:00
Alex Deucher
5ceb54c68a 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>
2015-08-17 16:50:18 -04:00
Chunming Zhou
c6a4079bad 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>
2015-08-17 16:50:17 -04:00
Christian König
0753b45242 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>
2015-08-17 16:50:14 -04:00
Christian König
03507c4f2f drm/amdgpu: recreate fence from user seq
And use common fence infrastructure for the wait.

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>
2015-06-29 15:52:49 -04:00
monk.liu
332300b97e drm/amdgpu: fix wrong type
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-06-29 11:21:45 -04:00
Christian König
344c19f955 drm/amdgpu: simplify fence debugfs output a bit
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <aleander.deucher@amd.com>
2015-06-29 11:21:42 -04:00
Chunming Zhou
890ee23fc6 drm/amdgpu: disable user fence interrupt (v2)
amdgpu submits both kernel and user fences, but just need one interrupt,
disable user fence interrupt and don't effect user fence.

v2: fix merge error

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-06-05 11:02:06 -04:00
Jack Xiao
25f45e6338 drm/amdgpu: add zero timeout check in amdgpu_fence_wait_seq_timeout
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-06-05 10:59:25 -04:00
Christian König
9269a60686 drm/amdgpu: drop AMDGPU_FENCE_SIGNALED_SEQ
It's causing issues with VMID handling and comparing the
fence value two times actually doesn't make handling faster.

Port of radeon commit "d6d5c5b8364bcc4d52cddc68bcb0a330d2af20f3".

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
2015-06-03 21:03:50 -04:00
Jammy Zhou
86c2b79062 drm/amdgpu: rewording some left radeons
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-06-03 21:03:36 -04:00
Alex Deucher
d38ceaf99e drm/amdgpu: add core driver (v4)
This adds the non-asic specific core driver code.

v2: remove extra kconfig option
v3: implement minor fixes from Fengguang Wu
v4: fix cast in amdgpu_ucode.c

Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-06-03 21:03:15 -04:00