horchen [Fri, 9 Jun 2017 11:56:48 +0000 (04:56 -0700)]
drm/amdgpu: add contiguous flag in ucode bo create
Under VF environment, the ucode would be settled to the visible VRAM,
As it would be pinned to the visible VRAM, it's better to add
contiguous flag,otherwise it need to move gpu address during the pin
process. This movement is not necessary.
Signed-off-by: horchen <horace.chen@amd.com> Reviewed-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Mon, 5 Jun 2017 14:11:59 +0000 (22:11 +0800)]
drm/amdgpu: fix missed gpu info firmware when cache firmware during S3
gpu_info firmware is released after data is used. But when system enters into
suspend, upper class driver will cache all firmware names. At that time,
gpu_info will be failing to load. It seems an upper class issue, that we should
not release gpu_info firmware until device finished.
Huang Rui [Wed, 10 May 2017 15:04:06 +0000 (23:04 +0800)]
drm/amdgpu: export test ib debugfs interface
As Christian and David's suggestion, submit the test ib ring debug interfaces.
It's useful for debugging with the command submission without VM case.
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Eric Huang [Thu, 8 Jun 2017 18:39:32 +0000 (14:39 -0400)]
drm/amd/powerplay: add GPU power display for vega10
Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Eric Huang [Thu, 8 Jun 2017 17:48:59 +0000 (13:48 -0400)]
drm/amd/powerplay: update vega10_ppsmc.h
Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 9 Jun 2017 12:22:31 +0000 (08:22 -0400)]
drm/amdgpu/gfx: fix MEC interrupt enablement for pipes != 0
The interrupt registers are not indexed.
Fixes: 763a47b8e (drm/amdgpu: teach amdgpu how to enable interrupts for any pipe v3) Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Xie [Thu, 8 Jun 2017 18:58:05 +0000 (14:58 -0400)]
drm/amdgpu: move comment to the right place
Signed-off-by: Alex Xie <AlexBin.Xie@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Xie [Thu, 8 Jun 2017 18:53:26 +0000 (14:53 -0400)]
drm/amdgpu: fix a typo in comment
Signed-off-by: Alex Xie <AlexBin.Xie@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Xie [Thu, 8 Jun 2017 18:21:28 +0000 (14:21 -0400)]
drm/amdgpu: remove duplicate function prototypes
There are two identical function prototypes in same header file
Signed-off-by: Alex Xie <AlexBin.Xie@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
v2: Fix logical mistake. If CPU update failed amdgpu_vm_bo_update_mapping()
would not return and instead fall through to SDMA update. Minor change due to
amdgpu_vm_bo_wait() prototype change
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If amdgpu.vm_update_context param is set to use CPU, then Page
Directories will be updated by CPU instead of SDMA
v2: Call amdgpu_vm_bo_wait before updating the page tables to ensure the
PD/PT BOs are free
v3: Minor changes - due to amdgpu_vm_bo_wait() prototype change, local
variable declaration order and function comments.
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Helper function useful for CPU update of VM page tables. Also useful if
kernel have to synchronously wait till VM page tables are updated.
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add VM update mode module param (amdgpu.vm_update_mode) that can used to
control how VM pde/pte are updated for Graphics and Compute.
BIT0 controls Graphics and BIT1 Compute.
BIT0 [= 0] Graphics updated by SDMA [= 1] by CPU
BIT1 [= 0] Compute updated by SDMA [= 1] by CPU
By default, only for large BAR system vm_update_mode = 2, indicating
that Graphics VMs will be updated via SDMA and Compute VMs will be
updated via CPU. And for all all other systems (by default)
vm_update_mode = 0
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Mon, 29 May 2017 09:05:20 +0000 (18:05 +0900)]
drm/amdgpu/radeon: Use radeon by default for CIK GPUs
Even if CONFIG_DRM_AMDGPU_CIK is enabled.
There is no feature parity yet for CIK, in particular amdgpu doesn't
support HDMI/DisplayPort audio without DC.
v2:
* Clarify the lack of feature parity being related to HDMI/DP audio.
* Fix "SI" typo in DRM_AMDGPU_CIK help entry.
Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Michel Dänzer [Mon, 29 May 2017 08:32:38 +0000 (17:32 +0900)]
drm/radeon: Make si_support and cik_support parameters always available
This will allow amdgpu-pro / other out-of-tree amdgpu builds to make use
of these options for using the out-of-tree amdgpu driver instead of the
in-tree radeon driver in a clean way.
Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Felix Kuehling [Mon, 5 Jun 2017 09:57:32 +0000 (18:57 +0900)]
drm/amdgpu: Update Kconfig help for SI and CIK support
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Felix Kuehling [Mon, 5 Jun 2017 09:53:55 +0000 (18:53 +0900)]
drm/amdgpu: Add module param to control SI support
If AMDGPU supports SI, add a module parameter to control SI
support. It's off by default in AMDGPU as long as SI suppost is
experimental, while it is on by default in radeon.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Acked-by: Michel Dänzer <michel.daenzer@amd.com>
[ Michel Dänzer: Squash in amdgpu_si_support initialization fix ] Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Felix Kuehling [Mon, 5 Jun 2017 09:52:51 +0000 (18:52 +0900)]
drm/radeon: Add module param to control SI support
If AMDGPU supports SI, add a module parameter to control SI
support in radeon. It's on by default in radeon, while it will be
off by default in AMDGPU as long as SI support is experimental.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Felix Kuehling [Mon, 5 Jun 2017 09:43:27 +0000 (18:43 +0900)]
drm/amdgpu: Add module param to control CIK support
If AMDGPU supports CIK, add a module parameter to control CIK
support. It's on by default in AMDGPU, while it is off by default
in radeon.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Alex Deucher [Wed, 7 Jun 2017 15:07:48 +0000 (11:07 -0400)]
drm/amdgpu/gfx9: Raven has two MECs
This was missed when Andres' queue patches were rebased.
Fixes: 42794b27 (drm/amdgpu: take ownership of per-pipe configuration v3) Reviewed-by: Alex Xie <AlexBin.Xie@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 7 Jun 2017 14:46:06 +0000 (10:46 -0400)]
drm/amdgpu: fix mec queue policy on single MEC asics
Fixes hangs on single MEC asics.
Fixes: 2ed286fb434 (drm/amdgpu: new queue policy, take first 2 queues of each pipe v2) Reviewed-by: Alex Xie <AlexBin.Xie@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 6 Jun 2017 21:41:20 +0000 (17:41 -0400)]
drm/amdgpu/gfx: create a common bitmask function (v2)
The same function was duplicated in all the gfx IPs. Use
a single implementation for all.
v2: use static inline (Alex Xie)
Reviewed-by: Alex Xie <AlexBin.Xie@amd.com> Suggested-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 31 May 2017 14:05:04 +0000 (10:05 -0400)]
drm/amdgpu/gfx8: drop per-APU CU limits
Always use the max for the family rather than the per sku limits.
This makes sure the mask is always the max size to avoid reporting
the wrong number of CUs.
Reviewed-by: Alex Xie <AlexBin.Xie@amd.com> Reviewed-by: Andres Rodriguez <andresx7@gmail.com> Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 5 Jun 2017 15:03:59 +0000 (11:03 -0400)]
drm/amdgpu/gfx9: new queue policy, take first 2 queues of each pipe
Instead of taking the first pipe and giving the rest to kfd, take the
first 2 queues of each pipe.
Effectively, amdgpu and amdkfd own the same number of queues. But
because the queues are spread over multiple pipes the hardware will be
able to better handle concurrent compute workloads.
amdgpu goes from 1 pipe to 4 pipes, i.e. from 1 compute threads to 4
amdkfd goes from 3 pipe to 4 pipes, i.e. from 3 compute threads to 4
gfx9 was missed when this patch set was rebased to include gfx9.
Acked-by: Tom St Denis <tom.stdenis@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 5 Jun 2017 14:58:56 +0000 (10:58 -0400)]
drm/amdgpu/gfx9: allocate queues horizontally across pipes
Pipes provide better concurrency than queues, therefore we want to make
sure that apps use queues from different pipes whenever possible.
Optimize for the trivial case where an app will consume rings in order,
therefore we don't want adjacent rings to belong to the same pipe.
gfx9 was missed when these patches were rebased.
Reviewed-by: Tom St Denis <tom.stdenis@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Eric Huang [Fri, 2 Jun 2017 14:57:24 +0000 (10:57 -0400)]
drm/amd/powerplay: add sclk and mclk overdrive for vega10
For overclocking sclk and mclk.
Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Thu, 1 Jun 2017 07:33:26 +0000 (15:33 +0800)]
drm/amdgpu: update to use RREG32_SOC15/WREG32_SOC15 for gmc9
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Thu, 1 Jun 2017 07:30:04 +0000 (15:30 +0800)]
drm/amdgpu: update to use RREG32_SOC15/WREG32_SOC15 for mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Thu, 1 Jun 2017 07:15:28 +0000 (15:15 +0800)]
drm/amdgpu: update to use RREG32_SOC15/WREG32_SOC15 for gfxhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 02:35:42 +0000 (10:35 +0800)]
drm/amdgpu: fix the gart table cleared issue for S3
Something writes over the first 8 MB so reserve this
on vega10 until we root cause it.
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 3 May 2017 01:52:06 +0000 (09:52 +0800)]
drm/amdgpu: add ip block number prints
User is able to follow the ip block number to write the ip_block_mask for
selecting the one which user would like to enable.
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 3 May 2017 01:40:17 +0000 (09:40 +0800)]
drm/amdgpu: add ip name print for selecting ips with ip_block_mask
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 15:49:46 +0000 (23:49 +0800)]
drm/amdgpu: remove mmhub ip
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 15:46:26 +0000 (23:46 +0800)]
drm/amdgpu: remove gfxhub ip
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 15:35:44 +0000 (23:35 +0800)]
drm/amdgpu: export mmhub get clockgating into gmc
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 15:13:34 +0000 (23:13 +0800)]
drm/amdgpu: export mmhub set clockgating into gmc
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 14:59:18 +0000 (22:59 +0800)]
drm/amdgpu: export mmhub sw_init into gmc
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 14:57:18 +0000 (22:57 +0800)]
drm/amdgpu: export gfxhub sw_init into gmc
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 14:32:35 +0000 (22:32 +0800)]
drm/amdgpu: fix to miss program invalidation at resume
This patch moves invalidation into gart enable function from hw_init.
Because we would like align the sequence calling between init and resume.
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 14:17:11 +0000 (22:17 +0800)]
drm/amdgpu: abstract setup vmid config for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 13:52:00 +0000 (21:52 +0800)]
drm/amdgpu: abstract disable identity aperture for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 13:39:10 +0000 (21:39 +0800)]
drm/amdgpu: abstract system domain enablement for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 10:07:48 +0000 (18:07 +0800)]
drm/amdgpu: abstract cache initialization for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 09:19:01 +0000 (17:19 +0800)]
drm/amdgpu: abstract TLB initialization for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 09:04:28 +0000 (17:04 +0800)]
drm/amdgpu: abstract system aperture initialization for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 08:40:14 +0000 (16:40 +0800)]
drm/amdgpu: abstract gart aperture initialization for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Wed, 31 May 2017 08:20:48 +0000 (16:20 +0800)]
drm/amdgpu: abstract gart table initialization for gfxhub/mmhub
Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Xie [Thu, 1 Jun 2017 13:42:59 +0000 (09:42 -0400)]
drm/amdgpu: Move compute vm bug logic to amdgpu_vm.c
In review, Christian would like to keep the logic
inside amdgpu_vm.c with a cost of slightly slower.
The loop is still optimized out with this patch.
v2: remove the if statement. Now it is not slower.
Signed-off-by: Alex Xie <AlexBin.Xie@amd.com> Reviewed-by: Christian König <christian.koeng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: implement lru amdgpu_queue_mgr policy for compute v4
Use an LRU policy to map usermode rings to HW compute queues.
Most compute clients use one queue, and usually the first queue
available. This results in poor pipe/queue work distribution when
multiple compute apps are running. In most cases pipe 0 queue 0 is
the only queue that gets used.
In order to better distribute work across multiple HW queues, we adopt
a policy to map the usermode ring ids to the LRU HW queue.
This fixes a large majority of multi-app compute workloads sharing the
same HW queue, even though 7 other queues are available.
v2: use ring->funcs->type instead of ring->hw_ip
v3: remove amdgpu_queue_mapper_funcs
v4: change ring_lru_list_lock to spinlock, grab only once in lru_get()
Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Andres Rodriguez [Thu, 16 Feb 2017 05:47:32 +0000 (00:47 -0500)]
drm/amdgpu: untie user ring ids from kernel ring ids v6
Add amdgpu_queue_mgr, a mechanism that allows disjointing usermode's
ring ids from the kernel's ring ids.
The queue manager maintains a per-file descriptor map of user ring ids
to amdgpu_ring pointers. Once a map is created it is permanent (this is
required to maintain FIFO execution guarantees for a context's ring).
Different queue map policies can be configured for each HW IP.
Currently all HW IPs use the identity mapper, i.e. kernel ring id is
equal to the user ring id.
The purpose of this mechanism is to distribute the load across multiple
queues more effectively for HW IPs that support multiple rings.
Userspace clients are unable to check whether a specific resource is in
use by a different client. Therefore, it is up to the kernel driver to
make the optimal choice.
v2: remove amdgpu_queue_mapper_funcs
v3: made amdgpu_queue_mgr per context instead of per-fd
v4: add context_put on error paths
v5: rebase and include new IPs UVD_ENC & VCN_*
v6: drop unused amdgpu_ring_is_valid_index (Alex)
Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Andres Rodriguez [Sat, 25 Feb 2017 01:50:20 +0000 (20:50 -0500)]
drm/amdgpu: workaround tonga HW bug in HQD programming sequence
Tonga based asics may experience hangs when an HQD's EOP parameters
are modified.
Workaround this HW issue by avoiding writes to these registers for
tonga asics.
Based on the following ROCm commit:
2a0fb8 - drm/amdgpu: Synchronize KFD HQD load protocol with CP scheduler
From the ROCm git repository:
https://github.com/RadeonOpenCompute/ROCK-Kernel-Driver.git
CC: Jay Cornwall <Jay.Cornwall@amd.com> Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Andres Rodriguez [Fri, 24 Feb 2017 20:28:43 +0000 (15:28 -0500)]
drm/amdgpu: condense mqd programming sequence
The MQD structure matches the reg layout. Take advantage of this to
simplify HQD programming.
Note that the ACTIVE field still needs to be programmed last.
Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: new queue policy, take first 2 queues of each pipe v2
Instead of taking the first pipe and giving the rest to kfd, take the
first 2 queues of each pipe.
Effectively, amdgpu and amdkfd own the same number of queues. But
because the queues are spread over multiple pipes the hardware will be
able to better handle concurrent compute workloads.
amdgpu goes from 1 pipe to 4 pipes, i.e. from 1 compute threads to 4
amdkfd goes from 3 pipe to 4 pipes, i.e. from 3 compute threads to 4
v2: fix policy comment
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: avoid KIQ clashing with compute or KFD queues v2
Instead of picking an arbitrary queue for KIQ, search for one according
to policy. The queue must be unused.
Also report the KIQ as an unavailable resource to KFD.
In testing I ran into KCQ initialization issues when using pipes 2/3 of
MEC2 for the KIQ. Therefore the policy disallows grabbing one of these.
v2: fix (ring.me + 1) to (ring.me -1) in amdgpu_amdkfd_device_init
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: remove hardcoded queue_mask in PACKET3_SET_RESOURCES
The assumption that we are only using the first pipe no longer holds.
Instead, calculate the queue_mask from the queue_bitmap.
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: allocate queues horizontally across pipes
Pipes provide better concurrency than queues, therefore we want to make
sure that apps use queues from different pipes whenever possible.
Optimize for the trivial case where an app will consume rings in order,
therefore we don't want adjacent rings to belong to the same pipe.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: remove duplicate magic constants from amdgpu_amdkfd_gfx*.c
This information is already available in adev.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: allow split HQD on per-queue granularity v5
Update the KGD to KFD interface to allow sharing pipes with queue
granularity instead of pipe granularity.
This allows for more interesting pipe/queue splits.
v2: fix overflow check for res.queue_mask
v3: fix shift overflow when setting res.queue_mask
v4: fix comment in is_pipeline_enabled()
v5: clamp res.queue_mask to the first MEC only
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: teach amdgpu how to enable interrupts for any pipe v3
The current implementation is hardcoded to enable ME1/PIPE0 interrupts
only.
This patch allows amdgpu to enable interrupts for any pipe of ME1.
v2: added gfx9 support
v3: use soc15_grbm_select for gfx9
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: allow split of queues with kfd at queue granularity v4
Previously the queue/pipe split with kfd operated with pipe
granularity. This patch allows amdgpu to take ownership of an arbitrary
set of queues.
It also consolidates the last few magic numbers in the compute
initialization process into mec_init.
v2: support for gfx9
v3: renamed AMDGPU_MAX_QUEUES to AMDGPU_MAX_COMPUTE_QUEUES
v4: fix off-by-one in num_mec checks in *_compute_queue_acquire
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: take ownership of per-pipe configuration v3
Make amdgpu the owner of all per-pipe state of the HQDs.
This change will allow us to split the queues between kfd and amdgpu
with a queue granularity instead of pipe granularity.
This patch fixes kfd allocating an HDP_EOP region for its 3 pipes which
goes unused.
v2: support for gfx9
v3: fix gfx7 HPD intitialization
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Take ownership of pipe initialization away from KFD.
Note that hpd_eop_gpu_addr was already large enough to accomodate all
pipes.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rename straggler instances of r(adeon)dev to a(mdgpu)dev
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Andres Rodriguez [Thu, 13 Apr 2017 17:55:41 +0000 (13:55 -0400)]
drm/amdgpu: fix kgd_hqd_load failing to update shadow_wptr
The return value from copy_form_user is 0 for the success case.
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: unify MQD programming sequence for kfd and amdgpu v2
Use the same gfx_*_mqd_commit function for kfd and amdgpu codepaths.
This removes the last duplicates of this programming sequence.
v2: fix cp_hqd_pq_wptr value
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: remove duplicate definition of cik_mqd
The gfxv7 contains a slightly different version of cik_mqd called
bonaire_mqd. This can introduce subtle bugs if fixes are not applied in
both places.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: detect timeout error when deactivating hqd
Handle HQD deactivation timeouts instead of ignoring them.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Andres Rodriguez [Wed, 12 Apr 2017 21:19:54 +0000 (17:19 -0400)]
drm/amdgpu: refactor MQD/HQD initialization v3
The MQD programming sequence currently exists in 3 different places.
Refactor it to absorb all the duplicates.
The success path remains mostly identical except for a slightly
different order in the non-kiq case. This shouldn't matter if the HQD
is disabled.
The error handling paths have been updated to deal with the new code
structure.
v2: the non-kiq path for gfxv8 was dropped in the rebase
v3: split MEC_HPD_SIZE rename, dropped doorbell changes
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net> Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Andres Rodriguez <andresx7@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
projects / mirror_ubuntu-jammy-kernel.git / log