The basic problem here is that it's not allowed to page fault while
holding the reservation lock.

So it can happen that multiple processes try to validate an userptr
at the same time.

Work around that by putting the HMM range object into the mutex
protected bo list for now.

v2: make sure range is set to NULL in case of an error
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

A user reported a bug on CAPE VERDE system where uvd_v3_1
IP component failed to initialize as there is an issue with
BO move code from one memory to other.

In function amdgpu_mem_visible() called by amdgpu_bo_move(),
when there are no blocks to compare or if we have a single
block then break the loop.

Fixes: 312b4dc1 ("drm/amdgpu: Fix VRAM BO swap issue")
Signed-off-by: NArunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

DRM buddy manager allocates the contiguous memory requests in
a single block or multiple blocks. So for the ttm move operation
(incase of low vram memory) we should consider all the blocks to
compute the total memory size which compared with the struct
ttm_resource num_pages in order to verify that the blocks are
contiguous for the eviction process.

v2: Added a Fixes tag
v3: Rewrite the code to save a bit of calculations and
    variables (Christian)

Fixes: c9cad937 ("drm/amdgpu: add drm buddy support to amdgpu")
Signed-off-by: NArunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Apply new intersect and compatible callback instead
of having a generic placement range verfications.

v2: Added a separate callback for compatiblilty
    checks (Christian)
v3: Cleanups and removal of workarounds
Signed-off-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NArunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220820073304.178444-6-Arunpravin.PaneerSelvam@amd.com

For typesafety, use container_of() instead of implicit cast from struct
ttm_tt to struct amdgpu_ttm_tt.
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NRajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Make sure we can at least move and release BOs without backing store.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichael J. Ruhl <michael.j.ruhl@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220707102453.3633-3-christian.koenig@amd.com

Certain GL unit tests for large textures can cause problems
with the OOM killer since there is no way to link this memory
to a process.  This was originally mitigated (but not necessarily
eliminated) by limiting the GTT size.  The problem is this limit
is often too low for many modern games so just make the limit 1/2
of system memory. The OOM accounting needs to be addressed, but
we shouldn't prevent common 3D applications from being usable
just to potentially mitigate that corner case.

Set default GTT size to max(3G, 1/2 of system ram) by default.

v2: drop previous logic and default to 3/4 of ram
v3: default to half of ram to align with ttm
v4: fix spelling in comment (Kent)

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1942Reviewed-by: NMarek Olšák <marek.olsak@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Certain GL unit tests for large textures can cause problems
with the OOM killer since there is no way to link this memory
to a process.  This was originally mitigated (but not necessarily
eliminated) by limiting the GTT size.  The problem is this limit
is often too low for many modern games so just make the limit 1/2
of system memory. The OOM accounting needs to be addressed, but
we shouldn't prevent common 3D applications from being usable
just to potentially mitigate that corner case.

Set default GTT size to max(3G, 1/2 of system ram) by default.

v2: drop previous logic and default to 3/4 of ram
v3: default to half of ram to align with ttm
v4: fix spelling in comment (Kent)

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1942Reviewed-by: NMarek Olšák <marek.olsak@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add a AMDGPU_GEM_CREATE_DISCARDABLE flag to note that the content of a BO
doesn't needs to be preserved during eviction.

KFD was already using a similar functionality for SVM BOs so replace the
internal flag with the new UAPI.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NMarek Olšák <marek.olsak@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Switch to using the TTM resource manager debugfs helpers. It's
exactly the same functionality but the debugfs code is shared with
other drivers.

The TTM resource managers need to stay valid for as long as the
drm debugfs_root is valid.
Signed-off-by: NZack Rusin <zackr@vmware.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Nirmoy Das <nirmoy.das@amd.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: amd-gfx@lists.freedesktop.org
Link: https://patchwork.freedesktop.org/patch/msgid/20220412033526.369115-4-zack@kde.orgReviewed-by: NChristian König <christian.koenig@amd.com>

Add an usage for submissions independent of implicit sync but still
interesting for memory management.

v2: cleanup the kerneldoc a bit
v3: separate amdgpu changes from this
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20220407085946.744568-10-christian.koenig@amd.com

This change adds the dma_resv_usage enum and allows us to specify why a
dma_resv object is queried for its containing fences.

Additional to that a dma_resv_usage_rw() helper function is added to aid
retrieving the fences for a read or write userspace submission.

This is then deployed to the different query functions of the dma_resv
object and all of their users. When the write paratermer was previously
true we now use DMA_RESV_USAGE_WRITE and DMA_RESV_USAGE_READ otherwise.

v2: add KERNEL/OTHER in separate patch
v3: some kerneldoc suggestions by Daniel
v4: some more kerneldoc suggestions by Daniel, fix missing cases lost in
    the rebase pointed out by Bas.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20220407085946.744568-2-christian.koenig@amd.com

Instead of providing the bulk move structure for each LRU update set
this as property of the BO. This should avoid costly bulk move rebuilds
with some games under RADV.

v2: some name polishing, add a few more kerneldoc words.
v3: add some lockdep
v4: fix bugs, handle pin/unpin as well
v5: improve kerneldoc
Signed-off-by: NChristian König <christian.koenig@amd.com>
Tested-by: NBas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20220321132601.2161-5-christian.koenig@amd.com

This is provided by TTM now.

Also switch man->size to bytes instead of pages and fix the double
printing of size and usage in debugfs.

v2: fix size checking as well
Signed-off-by: NChristian König <christian.koenig@amd.com>
Tested-by: NBas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: NMatthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220214093439.2989-8-christian.koenig@amd.com

This adds support to discover the  buffer objects that belong to a
process being checkpointed. The data corresponding to these buffer
objects is returned to user space plugin running under criu master
context which then stores this info to recreate these buffer objects
during a restore operation.
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NDavid Yat Sin <david.yatsin@amd.com>
Signed-off-by: NRajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

We probably never trigger this, but the logic inside the check is
inverted.
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>

We ran into the problem that clearing really larger buffer (60GiB) caused an
SDMA timeout.

Restructure the function to use the dst window instead of mapping the whole
buffer into the GART and then fill only 2MiB/256MiB chunks at a time.

v2: rebase on restructured window map.
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>

Instead of limiting the size before we call the mapping
function let the function itself limit the size.
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>

That should never happen, but make sure that we only warn instead of
crash.
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>

We probably never trigger this, but the logic inside the check is
inverted.
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>

Instead of calling the debug operation directly.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211124124430.20859-10-christian.koenig@amd.com

Debug VRAM access through SDMA has several broken parts resulting in
silent MMIO fallback.

BO kernel creation takes the location of the cpu addr pointer, not
the pointer itself for address kmap.

drm_dev_enter return true on success so change access check.

The source BO is reserved but not pinned so find the address using the
cursor offset relative to its memory domain start.
Signed-off-by: NJonathan Kim <jonathan.kim@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

That's just a leftover from old radeon days and was preventing CS and GART
bindings before the hardware was initialized. But nowdays that is
perfectly valid.

The only thing we need to warn about are GART binding before the table
is even allocated.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NGuchun Chen <guchun.chen@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Some suggested cleanups to declutter ttm when doing debug VRAM access over
SDMA.
Signed-off-by: NJonathan Kim <jonathan.kim@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

For better performance during VRAM access for debugged processes, do
read/write copies over SDMA.

In order to fulfill post mortem debugging on a broken device, fallback to
stable MMIO access when gpu recovery is disabled or when job submission
time outs are set to max.  Failed SDMA access should automatically fall
back to MMIO access.

Use a pre-allocated GTT bounce buffer pre-mapped into GART to avoid
page-table updates and TLB flushes on access.
Signed-off-by: NJonathan Kim <jonathan.kim@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Patch: 3efb17ae7e92 ("drm/amdgpu: Call amdgpu_device_unmap_mmio() if device
is unplugged to prevent crash in GPU initialization failure") makes call to
amdgpu_device_unmap_mmio() conditioned on device unplugged. This patch unmaps
MMIO mappings even when device is not unplugged.

v2: Add condition of drm_dev_enter() to deleted unmaps in patch
"drm/amdgpu: Unmap all MMIO mappings"
Signed-off-by: NLeslie Shi <Yuliang.Shi@amd.com>
Reviewed-by: NAndrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Partially revert commit 5f319c5c.

First of all this is illegal use of RCU to call dma_fence_enable_sw_signaling()
since we don't hold a reference to the fence in question and can crash badly.

Then the code doesn't seem to have the intended effect since only the
exclusive fence is handled, but the KFD fences are always added as shared fence.

Only keep the handling to throw away the content of SVM BOs.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211122123926.385017-1-christian.koenig@amd.comReviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>

amdgpu_ttm_backend_bind() only needed for TTM_PL_TT
and AMDGPU_PL_PREEMPT.
Signed-off-by: NNirmoy Das <nirmoy.das@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The userptr can be unmapped by application and still registered to
driver, restore userptr work return user pages will get -EFAULT bad
address error. Pretend this error as succeed. GPU access this userptr
will have VM fault later, it is better than application soft hangs with
stalled user mode queues.

v2: squash in warning fix (Alex)
Signed-off-by: NPhilip Yang <Philip.Yang@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

In order to better track where in the kernel the dma-buf code is used,
put the symbols in the namespace DMA_BUF and modify all users of the
symbols to properly import the namespace to not break the build at the
same time.

Now the output of modinfo shows the use of these symbols, making it
easier to watch for users over time:

$ modinfo drivers/misc/fastrpc.ko | grep import
import_ns:      DMA_BUF

Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Mauro Carvalho Chehab <mchehab@kernel.org>
Cc: dri-devel@lists.freedesktop.org
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: NChristian König <christian.koenig@amd.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NSumit Semwal <sumit.semwal@linaro.org>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20211010124628.17691-1-gregkh@linuxfoundation.orgSigned-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Simplifying the code a bit.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20211005113742.1101-13-christian.koenig@amd.com

When creating unregistered new svm range to recover retry fault, avoid
new svm range to overlap with ranges or userptr ranges managed by TTM,
otherwise svm migration will trigger TTM or userptr eviction, to evict
user queues unexpectedly.

Change helper amdgpu_ttm_tt_affect_userptr to return userptr which is
inside the range. Add helper svm_range_check_vm_userptr to scan all
userptr of the vm, and return overlap userptr bo start, last.
Signed-off-by: NPhilip Yang <Philip.Yang@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Unify BO evicting functionality for possible memory
types in amdgpu_ttm.c.
Signed-off-by: NNirmoy Das <nirmoy.das@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It covers more than just ttm_bo_type_sg usage, like with say dma-buf,
since one other user is userptr in amdgpu, and in the future we might
have some more. Hence EXTERNAL is likely a more suitable name.

v2(Christian):
  - Rename these to TTM_TT_FLAGS_*
  - Fix up all the holes in the flag values
Suggested-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NMatthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210929132629.353541-1-matthew.auld@intel.comSigned-off-by: NChristian König <christian.koenig@amd.com>

Now that setting page->index shouldn't be needed anymore, we are just
left with setting page->mapping, and here it looks like amdgpu is the
only user, where pointing the page->mapping at the dev_mapping is used
to verify that the pages do indeed belong to the device, if userspace
later tries to touch them.

v2(Christian):
  - Drop the functions altogether and just inline modifying
    the page->mapping
Signed-off-by: NMatthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210927114114.152310-3-matthew.auld@intel.comSigned-off-by: NChristian König <christian.koenig@amd.com>

The memory backing old_mem is already freed at that point, move the
check a bit more up.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Fixes: bfa3357e ("drm/ttm: allocate resource object instead of embedding it v2")
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1699Acked-by: NNirmoy Das <nirmoy.das@amd.com>
Reviewed-by: NMichel Dänzer <mdaenzer@redhat.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

The memory backing old_mem is already freed at that point, move the
check a bit more up.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Fixes: bfa3357e ("drm/ttm: allocate resource object instead of embedding it v2")
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1699Acked-by: NNirmoy Das <nirmoy.das@amd.com>
Reviewed-by: NMichel Dänzer <mdaenzer@redhat.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Move the functionality into ttm_tt_fini and ttm_bo_tt_destroy instead.

We don't need this any more since we removed the unbind from the destroy
code paths in the drivers.

Also add a warning to ttm_tt_fini() if we try to fini a still populated TT
object.

v2: instead of reverting the patch move the functionality to different
places.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NThomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210728130552.2074-5-christian.koenig@amd.com

Doing this in amdgpu_ttm_backend_destroy() is to late.

It turned out that this is not a good idea at all because it leaves pointers
to freed up system memory pages in the GART tables of the drivers.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NThomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210728130552.2074-2-christian.koenig@amd.com

drm/amdgpu: drop redundant null-pointer checks in amdgpu_ttm_tt_populate() and amdgpu_ttm_tt_unpopulate()

The varialbe gtt in the function amdgpu_ttm_tt_populate() and
amdgpu_ttm_tt_unpopulate() is guaranteed to be not NULL in the context.
Thus the null-pointer checks are redundant and can be dropped.
Reported-by: NTOTE Robot <oslab@tsinghua.edu.cn>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NTuo Li <islituo@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>