Commit Graph

694033 Commits

Author SHA1 Message Date
Christian König
e1eb899b45 drm/amdgpu: add IOCTL interface for per VM BOs v3
Add the IOCTL interface so that applications can allocate per VM BOs.

Still WIP since not all corner cases are tested yet, but this reduces average
CS overhead for 10K BOs from 21ms down to 48us.

v2: add some extra checks, remove the WIP tag
v3: rename new flag to AMDGPU_GEM_CREATE_VM_ALWAYS_VALID

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:46:35 -04:00
Christian König
73fb16e7eb drm/amdgpu: add support for per VM BOs v2
Per VM BOs are handled like VM PDs and PTs. They are always valid and don't
need to be specified in the BO lists.

v2: validate PDs/PTs first

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:46:26 -04:00
Christian König
4f5839c56e drm/amdgpu: restrict userptr even more
Don't allow them to be GEM imported into another process.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:46:17 -04:00
Christian König
0f2fc435d8 drm/amdgpu: fix new PD update code for Vega10 v2
We need to refer to the parent instead of the root BO for multi
level page tables on Vega10. Also don't set the PDE_PTE bit.

v2: Don't set the PDE_PTE bit either.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-and-Tested-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:46:08 -04:00
Christian König
705e519e0e drm/amdgpu: move hw generation check into amdgpu_doorbell_init v2
This way we can safely call it on SI as well.

v2: fix type in commit message

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:45:59 -04:00
Christian König
570144c652 drm/amdgpu: cleanup the VM code a bit more
The src isn't used any more after GART hack removal.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:45:47 -04:00
Christian König
ea09729c93 drm/amdgpu: rework page directory filling v2
Keep track off relocated PDs/PTs instead of walking and checking all PDs.

v2: fix root PD handling

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>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-31 13:45:24 -04:00
Himanshu Jha
ebe02de2c6 drm/amd/powerplay/hwmgr: Remove null check before kfree
kfree on NULL pointer is a no-op and therefore checking is redundant.

Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by:  Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:57:33 -04:00
Himanshu Jha
c5927537dd drm/amd: Remove null check before kfree
Kfree on NULL pointer is a no-op and therefore checking is redundant.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:57:32 -04:00
Christian König
3f3333f8a0 drm/amdgpu: track evicted page tables v2
Instead of validating all page tables when one was evicted,
track which one needs a validation.

v2: simplify amdgpu_vm_ready as well

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>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:28:04 -04:00
Christian König
00b5cc83c4 drm/amdgpu: fix comment on amdgpu_bo_va
Except for the reference count all other members are protected
by the VM PD being reserved.

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>
2017-08-29 15:28:04 -04:00
Christian König
cb7b6ec2f8 drm/amdgpu: add bo_va cleared flag again v2
We changed this to use an extra list a while back, but for the next
series I need a separate flag again.

v2: reorder to avoid unlocked list access

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>
2017-08-29 15:28:03 -04:00
Christian König
3d7d4d3a1b drm/amdgpu: rework moved handling in the VM v2
Instead of using the vm_state use a separate flag to note
that the BO was moved.

v2: reorder patches to avoid temporary lockless access

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>
2017-08-29 15:28:02 -04:00
Tom St Denis
08cab989f7 drm/amd/amdgpu: Add write() method to VRAM debugfs entry (v2)
Allows writing data to vram via debugfs.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

(v2):  Call get_user before holding spinlock.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:28:02 -04:00
Rex Zhu
841e3be124 drm/amd/powerplay: notify smu once display changed on Rv.
when User turn off display or screen idle timeout,
smu need this message to start S0i2 entry.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:28:01 -04:00
Rex Zhu
3b4ca9e649 drm/amd/powerplay: add dummy pp table for raven. (v2)
As there is no PPTable in RV, it is difficult to
cleanly decouple PPTABLE functionality in existing
codes.

v2: agd: squash in clean build fix

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:28:01 -04:00
Rex Zhu
e154162ef7 drm/amd/powerplay: refine pp code for raven
delete useless code.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:28:00 -04:00
Alex Deucher
ffe6d881e9 drm/amdgpu/gfx9: adjust mqd allocation size
To allocate additional space for the dynamic cu masks.
Confirmed with the hw team that we only need 1 dword
for the mask.  The mask is the same for each SE so
you only need 1 dword.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:28:00 -04:00
Alex Deucher
29696bd680 drm/amdgpu/gfx9: update mqd to include dynamic CU mask
Necessary for proper operation with KIQ.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:59 -04:00
Alex Deucher
31bf29ab39 drm/amdgpu/gfx8: drop cz mqd
It was unused and according to hw team, it's the same for
all asics in a gfx family so remove it.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:58 -04:00
Alex Deucher
925d5d798f drm/amdgpu/gfx8: apply dynamic cu mask to APUs as well
Confirmed with the hw team.  It's the same for all asics.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:58 -04:00
Alex Deucher
ecf9d34485 drm/amdgpu/powerplay/vega10: fix typo in register base index
Probably a copy pasta.  No functional difference, both have
the same value.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reported-by: Michael von Khurja <mvonkhurja@techpowerup.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:57 -04:00
Christian König
6ac7defb5c drm/amdgpu: cleanup GWS, GDS and OA allocation
Those are certainly not kernel allocations, instead set the NO_CPU_ACCESS flag.

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>
2017-08-29 15:27:57 -04:00
Christian König
34d7be5dc2 drm/amdgpu: fix and cleanup VM ready check
Stop checking the mapped BO itself, cause that one is
certainly not a page table.

Additional to that move the code into amdgpu_vm.c

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>
2017-08-29 15:27:56 -04:00
Christian König
87f64a76b3 drm/amdgpu: fix amdgpu_vm_bo_map trace point
That somehow got lost.

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>
2017-08-29 15:27:55 -04:00
Kent Russell
5b41d94cc4 drm/amdgpu: Move VBIOS version to sysfs
sysfs is more stable, and doesn't require root to access

Signed-off-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:55 -04:00
Kent Russell
db95e21855 drm/amdgpu: Add debugfs file for VBIOS and version
Add 2 debugfs files, one that contains the VBIOS version, and one that
contains the VBIOS itself. These won't change after initialization,
so we can add the VBIOS version when we parse the atombios information.

This ensures that we can find out the VBIOS version, even when the dmesg
buffer fills up, and makes it easier to associate which VBIOS version is
for which GPU on mGPU configurations. Set the size to 20 characters in
case of some weird VBIOS version that exceeds the expected 17 character
format (3-8-3\0). The VBIOS dump also allows for easy debugging

    v2: Move to debugfs, clarify commit message, add VBIOS dump file

Signed-off-by: Kent Russell <kent.russell@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>
2017-08-29 15:27:54 -04:00
Tom St Denis
96bec19835 drm/ttm: Remove needless 'extern' on functions in header.
Minor tidy up.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:54 -04:00
Tom St Denis
f7871fd193 drm/radeon: use new TTM populate/dma map helper functions
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:53 -04:00
Tom St Denis
7405e0dad4 drm/amd/amdgpu: Use new TTM populate/map helper function
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:52 -04:00
Tom St Denis
a4dec819c8 drm/ttm: Add helper functions to populate/map in one call (v2)
These functions replace a section of common code found
in radeon/amdgpu drivers (and possibly others) as part
of the ttm_tt_*populate() callbacks.

v2: squash in fix for sw iommu from Tom

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:52 -04:00
Harry Wentland
e719d5169f drm/amd/include: Add hdmi_redriver_set to atomfirmware
We'll need this for a some upcoming display changes

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:51 -04:00
Tom St Denis
ca3670aa37 drm/amd/amdgpu: Remove AMDGPU tracepoint and use new TTM tracepoint (v2)
Switches the AMDGPU driver over to the TTM tracepoint and removes
our old one.  Now you can enable traces before loading the module
and trace all mappings.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

(v2): Use struct device instead of pci in trace.
2017-08-29 15:27:51 -04:00
Tom St Denis
a92e145059 drm/ttm: Add DMA map/unmap tracepoint (v3)
Also exports two functions that vendor drivers can call
to trace DMA mappings.  This is meant to help translate
IOMMU mappings of bus addresses back to physical pages.

Used by the umr amdgpu debugger for instance.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

(v2): Use dev_name() to get PCI path instead.
(v3): Use correct types for dma/phys addresses
2017-08-29 15:27:50 -04:00
Evan Quan
727030b0c6 drm/amdgpu: support polaris10/11/12 new cp firmwares
Newer versions of the CP firmware require changes in how the driver
initializes the hw block.
Change the firmware name for new firmware to maintain compatibility with
older kernels.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:50 -04:00
Colin Ian King
fd4b5f54e1 drm/amdgpu: remove duplicate return statement
Remove a redundant identical return statement, it has no use.

Detected by CoverityScan, CID#1454586 ("Structurally dead code")

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:49 -04:00
Christophe JAILLET
06f10a537e drm/amdgpu: check memory allocation failure
Check memory allocation failure and return -ENOMEM in such a case.

'num_post_dep_syncobjs' still has to be set to 0 before the test in order
to have it initialized if 'amdgpu_cs_parser_fini()' is called to free
resources.

The calling graph would be, in such a case!
   failure in amdgpu_cs_process_syncobj_out_dep()
      ---> error code returned by amdgpu_cs_dependencies()
         --> amdgpu_cs_parser_fini() is called

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:48 -04:00
Roger He
a3ce364558 drm/amd/amdgpu: fix BANK_SELECT on Vega10 (v2)
BANK_SELECT should always be FRAGMENT_SIZE + 3 due to 8-entry (2^3)
per cache line in L2 TLB for Vega10.

v2: agd: fix warning

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:48 -04:00
Christian König
1cacc86a63 drm/amdgpu: inline amdgpu_ttm_do_bind again
The function is called only once and doesn't do anything special.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:47 -04:00
Christian König
9b0655e3ad drm/amdgpu: fix amdgpu_ttm_bind
Use ttm_bo_mem_space instead of manually allocating GART space.

This allows us to evict BOs when there isn't enought GART space any more.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:46 -04:00
Christian König
febb84a603 drm/amdgpu: remove the GART copy hack
This isn't used since we don't map evicted BOs to GART any more.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:46 -04:00
Monk Liu
172423bcc7 drm/ttm:fix wrong decoding of bo_count
we observe abnormal number from:
/sys/devices/virtual/drm/amdttm/buffer_objects/bo_count

bo_count is atomic_inc which is "int" type,
shouldn't explicitly turn it to unsigned long.

Signed-off-by: Monk Liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:45 -04:00
Monk Liu
7e96a13523 drm/ttm: fix missing inc bo_count
Signed-off-by: Monk Liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:45 -04:00
Alex Deucher
b249e18df1 drm/amdgpu: set sched_hw_submission higher for KIQ (v3)
KIQ doesn't really use the GPU scheduler.  The base
drivers generally use the KIQ ring directly rather than
submitting IBs.  However, amdgpu_sched_hw_submission
(which defaults to 2) limits the number of outstanding
fences to 2.  KFD uses the KIQ for TLB flushes and the
2 fence limit hurts performance when there are several KFD
processes running.

v2: move some expressions to one line
    change KIQ sched_hw_submission to at least 16
v3: bump to 256

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:44 -04:00
Alex Deucher
c3db7b5a55 drm/amdgpu: move default gart size setting into gmc modules
Move the asic specific code into the IP modules.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:43 -04:00
Alex Deucher
a4da14cc62 drm/amdgpu: refine default gart size
Be more explicit and add comments explaining each case.
Also s/gart/GART/ in the parameter string as per Felix'
suggestion.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:43 -04:00
Evan Quan
84d43463a2 drm/amd/powerplay: ACG frequency added in PPTable
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:42 -04:00
Christian König
f0694d3b8a drm/amdgpu: discard commands of killed processes
When a process is killed we shouldn't submit all waiting jobs, but instead
clean up as fast as possible.

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>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:42 -04:00
Christian König
cf273a59ca drm/amdgpu: fix and cleanup shadow handling
Set the shadow flag on the shadow and not the parent, always bind shadow BOs
during allocation instead of manually, use the reservation_object wrappers
to grab the lock.

This fixes a couple of issues with binding the shadow BOs as well as correctly
evicting them when memory becomes tight.

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>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:40 -04:00
Alex Deucher
83e74db6a8 drm/amdgpu: add automatic per asic settings for gart_size
We need a larger gart for asics that do not support GPUVM on all
engines (e.g., MM) to make sure we have enough space for all
gtt buffers in physical mode.  Change the default size based on
the asic type.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-08-29 15:27:40 -04:00