Avoid unnecessary XGMI hight pstate trigger when mapping none-vram memory for peer device
Signed-off-by: Nshaoyunl <shaoyun.liu@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit 915d3eec.

This depends on an HMM fix which is not upstream yet.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The buffers should be cleared when possible but we also don't want
buffer creation to fail in the rare case where the ring isn't ready
during the call. This could happen during some suspend/resume sequences.

Cc: Christian König <ckoenig.leichtzumerken@gmail.com>
Signed-off-by: NNicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The dumb_create API isn't intended for high performance rendering
and it's more useful for userspace (ie. IGT) to have them precleared.

The bonus here is that we also won't needlessly leak whatever was
previously in VRAM, but it also probably wasn't sensitive if it was
going through this API.
Signed-off-by: NNicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Let's start to allocate VM PDs/PTs on demand instead of pre-allocating
them during mapping.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Acked-by: NHuang Rui <ray.huang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Use HMM helper function hmm_vma_fault() to get physical pages backing
userptr and start CPU page table update track of those pages. Then use
hmm_vma_range_done() to check if those pages are updated before
amdgpu_cs_submit for gfx or before user queues are resumed for kfd.

If userptr pages are updated, for gfx, amdgpu_cs_ioctl will restart
from scratch, for kfd, restore worker is rescheduled to retry.

HMM simplify the CPU page table concurrent update check, so remove
guptasklock, mmu_invalidations, last_set_pages fields from
amdgpu_ttm_tt struct.

HMM does not pin the page (increase page ref count), so remove related
operations like release_pages(), put_page(), mark_page_dirty().
Signed-off-by: NPhilip Yang <Philip.Yang@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

- move all adjustments into one place
- specify GDS/GWS/OA alignment in basic units of the heaps
- it looks like GDS alignment was 1 instead of 4
Signed-off-by: NMarek Olšák <marek.olsak@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Let's support simultaneous submissions to multiple engines.

v2: rename the field to num_shared and fix up all users
v3: rebased
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

That only worked by pure coincident. Completely remove the shifting and
always apply correct PAGE_SHIFT.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Correct sign extend the GMC addresses to 48bit.

v2: sign extending turned out easier than thought.
v3: clean up the defines and move them into amdgpu_gmc.h as well
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Just rename functions, no functional change.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

We still use legacy type of gem_object_create, it should update to ttm_bo_type
now.
Signed-off-by: NHuang Rui <ray.huang@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Don't need validation list any more
Signed-off-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: NDavid Zhou <david1.zhou@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Per VM BOs share the reservation object with the PD and so need to
reserve a shared fence slot for the update.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

In per-vm-bo case, there could be no bo list.
But gds bo created from user space  must be passed to bo list.
So adding a check to prevent to creat gds bo as per-vm-bo.
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Move logic of getting supported domain to a helper
function
Signed-off-by: NDeepak Sharma <Deepak.Sharma@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

When vram size <= THRESHOLD(256M) lets use GTT for dumb buffer
allocation. As SG will be enabled with vram size <= 256M
scan out will not be an issue.

v2: Use amdgpu_display_supported_domains to get supported domain.
Signed-off-by: NDeepak Sharma <Deepak.Sharma@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Quite useful to know.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It is meaningless anyway.

v2: remove unused variable (Alex)
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Just note if a BO was imported/exported.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

if domain is illegal, we should return error.
v2:
  remove duplicated domain checking.
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

When GEM needs to fallback to GTT for VRAM BOs we still want the
preferred domain to be untouched so that the BO has a cance to move back
to VRAM in the future.
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

After that, we can easily add new parameter when need.

v2:
a) rebase.
b) Initialize struct amdgpu_bo_param, future new
member could only be used in some one case, but all member
should have its own initial value.
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com> (v1)
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Cc: christian.koenig@amd.com
Cc: Felix.Kuehling@amd.com
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit 7d1ca1325260a9e9329b10a21e3692e6f188936f.

Makes fallback handling to complicated. This is just a feature for the
GEM interface and shouldn't leak into the core BO create function.

The intended change to preserve the preferred domains is implemented in
a follow up patch.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Acked-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

v2: add sanity checking
v3: make code open
v4: also handle visible to invisible fallback
v5: Since two fallback cases, re-use goto retry
v6: avoid bo is unref when retry, and only user BO can fallback
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com> (v5)
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Cc: felix.kuehling@amd.com
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit bbaf1871ea1d7b7021e350a3e8bb1ec8408c2030.

Felix reports this commit causes regression.
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

v2: add sanity checking
v3: make code open
v4: also handle visible to invisible fallback
v5: Since two fallback cases, re-use goto retry
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

We unmapped imported DMA-bufs when the GEM handle was dropped, not when the
hardware was done with the buffere.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

We unmapped imported DMA-bufs when the GEM handle was dropped, not when the
hardware was done with the buffere.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Drop the "kernel" and sg parameter and give the BO type to create
explicit to amdgpu_bo_create instead of figuring it out from the
parameters.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NRoger He <Hongbo.He@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Currently, if amdgpu_vm_bo_update() fails, the returned error
is being ignored.

Fix this by properly checking _r_ after calling amdgpu_vm_bo_update.
Also, remove redundant code just before label _error_.

Addresses-Coverity-ID: 1464280 ("Unused value")
Fixes: 0abc6878 ("drm/amdgpu: update VM PDs after the PTs")
Signed-off-by: NGustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit 2046d46d.

Not needed any more.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

And rename it to struct gmc_funcs.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NSamuel Li <Samuel.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Necessary for the next patch.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

amdgpu_device.c was getting pretty cluttered.
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Userspace buggy userspace can spam the logs.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Similar to the CPU address space the VA on Vega10 has a hole in it.

v2: use dev_dbg instead of dev_err
v3: add some more comments to explain how the hw works
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Give moving a BO into place an operation context to work with.

v2: rebased
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Tested-by: NDieter Nützel <Dieter@nuetzel-hh.de>
Tested-by: NMichel Dänzer <michel.daenzer@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This can be triggered by userspace, e.g. trying to allocate too large a
BO, so it shouldn't log anything by default.

Callers need to handle failure anyway.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

After starting VNC server or running CTS test, kernel will hang and
can see below call trace:

[961816] INFO: task khugepaged:42 blocked for more than 120 seconds.
[968581]       Tainted: G           OE   4.13.0 #1
[973495] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables
          this message.
[980962] khugepaged      D    0    42      2 0x00000000
[980967] Call Trace:
[980977]  __schedule+0x28d/0x890
[980982]  schedule+0x36/0x80
[980986]  rwsem_down_read_failed+0x139/0x1c0
[980991]  ? update_curr+0x100/0x1c0
[981004]  call_rwsem_down_read_failed+0x18/0x30
[981007]  down_read+0x20/0x40
[981012]  khugepaged_scan_mm_slot+0x78/0x1ac0
[981018]  ? __switch_to+0x23e/0x4a0
[981022]  ? finish_task_switch+0x79/0x240
[981026]  khugepaged+0x146/0x480
[981031]  ? remove_wait_queue+0x60/0x60
[981035]  kthread+0x109/0x140
[981037]  ? khugepaged_scan_mm_slot+0x1ac0/0x1ac0
[981039]  ? kthread_park+0x60/0x60
[981044]  ret_from_fork+0x25/0x30

After checking code and found 'commit b72cf4fc ("drm/amdgpu: move
taking mmap_sem into get_user_pages v2")' forget to drop one case of
up_read.
Signed-off-by: NXiangliang.Yu <Xiangliang.Yu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>