This reverts commit 515986100d.
The dma_buf field in struct drm_gem_object is not stable over the
object instance's lifetime. The field becomes NULL when user space
releases the final GEM handle on the buffer object. This resulted
in a NULL-pointer deref.
Workarounds in commit 5307dce878 ("drm/gem: Acquire references on
GEM handles for framebuffers") and commit f6bfc9afc7 ("drm/framebuffer:
Acquire internal references on GEM handles") only solved the problem
partially. They especially don't work for buffer objects without a DRM
framebuffer associated.
Hence, this revert to going back to using .import_attach->dmabuf.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Simona Vetter <simona.vetter@ffwll.ch>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250715082635.34974-1-tzimmermann@suse.de
During process kill, drm_sched_entity_flush() will kill the vm
entities. The following job submissions of this process will fail, and
the resources of these jobs have not been released, nor have the fences
been signalled, causing tasks to hang and timeout.
Fix by check entity status in amdgpu_vm_ready() and avoid submit jobs to
stopped entity.
v2: add amdgpu_vm_ready() check before amdgpu_vm_clear_freed() in
function amdgpu_cs_vm_handling().
Fixes: 1f02f2044b ("drm/amdgpu: Avoid extra evict-restore process.")
Signed-off-by: Liu01 Tong <Tong.Liu01@amd.com>
Signed-off-by: Lin.Cao <lincao12@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f101c13a8720c73e67f8f9d511fbbeda95bcedb1)
If vm belongs to another process, this is fclose after fork,
wait may enable signaling KFD eviction fence and cause parent process queue evicted.
[677852.634569] amdkfd_fence_enable_signaling+0x56/0x70 [amdgpu]
[677852.634814] __dma_fence_enable_signaling+0x3e/0xe0
[677852.634820] dma_fence_wait_timeout+0x3a/0x140
[677852.634825] amddma_resv_wait_timeout+0x7f/0xf0 [amdkcl]
[677852.634831] amdgpu_vm_wait_idle+0x2d/0x60 [amdgpu]
[677852.635026] amdgpu_flush+0x34/0x50 [amdgpu]
[677852.635208] filp_flush+0x38/0x90
[677852.635213] filp_close+0x14/0x30
[677852.635216] do_close_on_exec+0xdd/0x130
[677852.635221] begin_new_exec+0x1da/0x490
[677852.635225] load_elf_binary+0x307/0xea0
[677852.635231] ? srso_alias_return_thunk+0x5/0xfbef5
[677852.635235] ? ima_bprm_check+0xa2/0xd0
[677852.635240] search_binary_handler+0xda/0x260
[677852.635245] exec_binprm+0x58/0x1a0
[677852.635249] bprm_execve.part.0+0x16f/0x210
[677852.635254] bprm_execve+0x45/0x80
[677852.635257] do_execveat_common.isra.0+0x190/0x200
Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Gang Ba <Gang.Ba@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
We need to know the wptr and sequence number associated
with a fence so that we can re-emit the unprocessed state
after a ring reset. Pre-allocate storage space for
the ring buffer contents and add helpers to save off
and re-emit the unprocessed state so that it can be
re-emitted after the queue is reset.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Avoid dereferencing struct drm_gem_object.import_attach for the
imported dma-buf. The dma_buf field in the GEM object instance refers
to the same buffer. Prepares to make import_attach an implementation
detail of PRIME.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test.
v2:
- keep amdgpu_bo_print_info() as-is (Christian)
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Convert from generic DRM_* to dev_* calls to have device context info.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
To avoid a cast when calling drm_dev_wedged_event(), replace pid and
task name inside of struct amdgpu_task_info with struct
drm_wedge_task_info.
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250617124949.2151549-6-andrealmeid@igalia.com
Signed-off-by: André Almeida <andrealmeid@igalia.com>
To avoid repetitive code in amdgpu, create a function that prints the
content of struct amdgpu_task_info.
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250617124949.2151549-3-andrealmeid@igalia.com
Signed-off-by: André Almeida <andrealmeid@igalia.com>
Allow NULL pointers at amdgpu_vm_put_task_info() as it common practice
for "put" or "free" functions. This avoid an extra check for NULL for
callers.
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250617124949.2151549-2-andrealmeid@igalia.com
Signed-off-by: André Almeida <andrealmeid@igalia.com>
Switch from a bool to an enum and allow more options
for enforce isolation. There are now 3 modes of operation:
- Disabled (0)
- Enabled (serialization and cleaner shader) (1)
- Enabled in legacy mode (no serialization or cleaner shader) (2)
This provides better flexibility for more use cases.
Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This was leftover from MES bring up when we had MES
user queues in the kernel. It's no longer used so
remove it.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Note when the cleaner shader is executed.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Instead of emitting the cleaner shader for every job which has the
enforce_isolation flag set only emit it for the first submission from
every client.
v2: add missing NULL check
v3: fix another NULL pointer deref
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Curret kfd does not allocate pasid values, instead uses pasid value for each
vm from graphic driver. So should not prevent graphic driver from releasing
pasid values since the values are allocated by graphic driver, not kfd driver
anymore. This patch does not stop graphic driver release pasid values.
Fixes: 8544374c0f ("drm/amdkfd: Have kfd driver use same PASID values from graphic driver")
Signed-off-by: Xiaogang Chen <xiaogang.chen@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
SVM migration unmap pages from GPU and then update mapping to GPU to
recover page fault. Currently unmap clears the PDE entry for range
length >= huge page and free PTB bo, update mapping to alloc new PT bo.
There is race bug that the freed entry bo maybe still on the pt_free
list, reused when updating mapping and then freed, leave invalid PDE
entry and cause GPU page fault.
By setting the update to clear only one PDE entry or clear PTB, to
avoid unmap to free PTE bo. This fixes the race bug and improve the
unmap and map to GPU performance. Update mapping to huge page will
still free the PTB bo.
With this change, the vm->pt_freed list and work is not needed. Add
WARN_ON(unlocked) in amdgpu_vm_pt_free_dfs to catch if unmap to free the
PTB.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Before, every time fdinfo is queried we try to lock all the BOs in the
VM and calculate memory usage from scratch. This works okay if the
fdinfo is rarely read and the VMs don't have a ton of BOs. If either of
these conditions is not true, we get a massive performance hit.
In this new revision, we track the BOs as they change states. This way
when the fdinfo is queried we only need to take the status lock and copy
out the usage stats with minimal impact to the runtime performance. With
this new approach however, we would no longer be able to track active
buffers.
Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241219151411.1150-6-Yunxiang.Li@amd.com
Signed-off-by: Christian König <christian.koenig@amd.com>
amdgpu_vm_bo_invalidate doesn't use the adev parameter and not all
callers have a reference to adev handy, so remove it for cleanliness.
Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241219151411.1150-5-Yunxiang.Li@amd.com
Signed-off-by: Christian König <christian.koenig@amd.com>
Third time's the charm, I hope?
Fixes: d3116756a7 ("drm/ttm: rename bo->mem and make it a pointer")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3837
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emitting the cleaner shader must come after the check if a VM switch is
necessary or not.
Otherwise we will emit the cleaner shader every time and not just when it is
necessary because we switched between applications.
This can otherwise crash on gang submit and probably decreases performance
quite a bit.
v2: squash in fix from Srini (Alex)
Signed-off-by: Christian König <christian.koenig@amd.com>
Fixes: ee7a846ea2 ("drm/amdgpu: Emit cleaner shader at end of IB submission")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Requested by both Bas and Friedrich. Mapping PTEs as PRT doesn't need to
sync for anything.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Convert fdinfo memory stats to use the common drm_print_memory_stats
helper.
This achieves alignment with the common keys as documented in
drm-usage-stats.rst, adding specifically drm-total- key the driver was
missing until now.
Additionally I made the code stop skipping total size for objects which
currently do not have a backing store, and I added resident, active and
purgeable reporting.
Legacy keys have been preserved, with the outlook of only potentially
removing only the drm-memory- when the time gets right.
The example output now looks like this:
pos: 0
flags: 02100002
mnt_id: 24
ino: 1239
drm-driver: amdgpu
drm-client-id: 4
drm-pdev: 0000:04:00.0
pasid: 32771
drm-total-cpu: 0
drm-shared-cpu: 0
drm-active-cpu: 0
drm-resident-cpu: 0
drm-purgeable-cpu: 0
drm-total-gtt: 2392 KiB
drm-shared-gtt: 0
drm-active-gtt: 0
drm-resident-gtt: 2392 KiB
drm-purgeable-gtt: 0
drm-total-vram: 44564 KiB
drm-shared-vram: 31952 KiB
drm-active-vram: 0
drm-resident-vram: 44564 KiB
drm-purgeable-vram: 0
drm-memory-vram: 44564 KiB
drm-memory-gtt: 2392 KiB
drm-memory-cpu: 0 KiB
amd-memory-visible-vram: 44564 KiB
amd-evicted-vram: 0 KiB
amd-evicted-visible-vram: 0 KiB
amd-requested-vram: 44564 KiB
amd-requested-visible-vram: 11952 KiB
amd-requested-gtt: 2392 KiB
drm-engine-compute: 46464671 ns
v2:
* Track purgeable via AMDGPU_GEM_CREATE_DISCARDABLE.
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Rob Clark <robdclark@chromium.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This patch tries to solve the basic problem we also need to sync to
the KFD fences of the BO because otherwise it can be that we clear
PTEs while the KFD queues are still running.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This was only used as workaround for recovering the page tables after
VRAM was lost and is no longer necessary after the function
amdgpu_vm_bo_reset_state_machine() started to do the same.
Compute never used shadows either, so the only proplematic case left is
SVM and that is most likely not recoverable in any way when VRAM is
lost.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
That is clearly not something we should do upstream. The SDMA is
mandatory for the driver to work correctly.
We could do this for emulation and bringup, but in those cases the
engineer should probably enabled CPU based updates manually.
This reverts commit 62eefd10ac.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit adds a description for the 'ts' parameter in the
amdgpu_vm_handle_fault function's comment block.
Fixes the below with gcc W=1:
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:2781: warning: Function parameter or struct member 'ts' not described in 'amdgpu_vm_handle_fault'
Cc: Xiaogang.Chen <Xiaogang.Chen@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202408251419.vgZHg3GV-lkp@intel.com/
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When it needs to get a value within a certain interval, using clamp()
makes the code easier to understand than min(max()).
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Li Zetao <lizetao1@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
CPU based update doesn't produce a fence, handle such cases properly.
Fixes: d8a3f0a034 ("drm/amdgpu: implement TLB flush fence")
Signed-off-by: Lang Yu <lang.yu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rework how VM operations synchronize to submissions. Provide an
amdgpu_sync container to the backends instead of an reservation
object and fill in the amdgpu_sync object in the higher layers
of the code.
No intended functional change, just prepares for upcomming changes.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When app unmap vm ranges(munmap) kfd/svm starts drain pending page fault and
not handle any incoming pages fault of this process until a deferred work item
got executed by default system wq. The time period of "not handle page fault"
can be long and is unpredicable. That is advese to kfd performance on page
faults recovery.
This patch uses time stamp of incoming page fault to decide to drop or recover
page fault. When app unmap vm ranges kfd records each gpu device's ih ring
current time stamp. These time stamps are used at kfd page fault recovery
routine.
Any page fault happened on unmapped ranges after unmap events is application
bug that accesses vm range after unmap. It is not driver work to cover that.
By using time stamp of page fault do not need drain page faults at deferred
work. So, the time period that kfd does not handle page faults is reduced
and can be controlled.
Signed-off-by: Xiaogang.Chen <Xiaogang.Chen@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This patch adds a new parameter 'enforce_isolation' to the amdgpu_job
structure. This parameter is used to determine whether shader isolation
should be enforced for a job. The enforce_isolation parameter is then
stored in the amdgpu_job structure and used when flushing the VM.
The enforce_isolation field of the amdgpu_job structure is set directly
after the job is allocated
This change allows more fine-grained control over shader isolation,
making it possible to enforce isolation on a per-job basis rather than
globally. This can be useful in scenarios where only certain jobs
require isolation.
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
This commit introduces the emission of a cleaner shader at the end of
the IB submission process. This is achieved by adding a new function
pointer, `emit_cleaner_shader`, to the `amdgpu_ring_funcs` structure. If
the `emit_cleaner_shader` function is set in the ring functions, it is
called during the VM flush process.
The cleaner shader is only emitted if the `enable_cleaner_shader` flag
is set in the `amdgpu_device` structure. This allows the cleaner shader
emission to be controlled on a per-device basis.
By emitting a cleaner shader at the end of the IB submission, we can
ensure that the VM state is properly cleaned up after each submission.
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
avoid using SDMA if it is unavailable.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Tim Huang <tim.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Page table of compute VM in the VRAM will lost after gpu reset.
VRAM won't be restored since compute VM has no shadows.
[How]
Use higher 32-bit of vm->generation to record a vram_lost_counter.
Reset the VM state machine when vm->genertaion is not equal to
the new generation token.
v2: Check vm->generation instead of calling drm_sched_entity_error
in amdgpu_vm_validate.
v3: Use new generation token instead of vram_lost_counter for check.
Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
(cherry picked from commit 47c0388b05)
[Why]
Page table of compute VM in the VRAM will lost after gpu reset.
VRAM won't be restored since compute VM has no shadows.
[How]
Use higher 32-bit of vm->generation to record a vram_lost_counter.
Reset the VM state machine when vm->genertaion is not equal to
the new generation token.
v2: Check vm->generation instead of calling drm_sched_entity_error
in amdgpu_vm_validate.
v3: Use new generation token instead of vram_lost_counter for check.
Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
To address the problem with hitches moving when bulk move
sublists are lru-bumped, register the list cursors with the
ttm_lru_bulk_move structure when traversing its list, and
when lru-bumping the list, move the cursor hitch to the tail.
This also means it's mandatory for drivers to call
ttm_lru_bulk_move_init() and ttm_lru_bulk_move_fini() when
initializing and finalizing the bulk move structure, so add
those calls to the amdgpu- and xe driver.
Compared to v1 this is slightly more code but less fragile
and hopefully easier to understand.
Changes in previous series:
- Completely rework the functionality
- Avoid a NULL pointer dereference assigning manager->mem_type
- Remove some leftover code causing build problems
v2:
- For hitch bulk tail moves, store the mem_type in the cursor
instead of with the manager.
v3:
- Remove leftover mem_type member from change in v2.
v6:
- Add some lockdep asserts (Matthew Brost)
- Avoid NULL pointer dereference (Matthew Brost)
- No need to check bo->resource before dereferencing
bo->bulk_move (Matthew Brost)
Cc: Christian König <christian.koenig@amd.com>
Cc: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: <dri-devel@lists.freedesktop.org>
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240705153206.68526-5-thomas.hellstrom@linux.intel.com
Signed-off-by: Christian König <christian.koenig@amd.com>
Align kerneldoc with the function argument name.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Fixes: 26e20235ce ("drm/amdgpu: Add amdgpu_bo_is_vm_bo helper")
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Help code readability by replacing a bunch of:
bo->tbo.base.resv == vm->root.bo->tbo.base.resv
With:
amdgpu_vm_is_bo_always_valid(vm, bo)
No functional changes.
v2:
* Rename helper and move to amdgpu_vm. (Christian)
v3:
* Use Christian's kerneldoc.
v4:
* Fixed logic inversion in amdgpu_vm_bo_get_memory.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Verify the parameters of
amdgpu_vm_bo_(map/replace_map/clearing_mappings) in one common place.
Fixes: dc54d3d174 ("drm/amdgpu: implement AMDGPU_VA_OP_CLEAR v2")
Cc: stable@vger.kernel.org
Reported-by: Vlad Stolyarov <hexed@google.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The idea behind this patch is to delay the freeing of PT entry objects
until the TLB flush is done.
This patch:
- Adds a tlb_flush_waitlist in amdgpu_vm_update_params which will keep the
objects that need to be freed after tlb_flush.
- Adds PT entries in this list in amdgpu_vm_ptes_update after finding
the PT entry.
- Changes functionality of amdgpu_vm_pt_free_dfs from (df_search + free)
to simply freeing of the BOs, also renames it to
amdgpu_vm_pt_free_list to reflect this same.
- Exports function amdgpu_vm_pt_free_list to be called directly.
- Calls amdgpu_vm_pt_free_list directly from amdgpu_vm_update_range.
V2: rebase
V4: Addressed review comments from Christian
- add only locked PTEs entries in TLB flush waitlist.
- do not create a separate function for list flush.
- do not create a new lock for TLB flush.
- there is no need to wait on tlb_flush_fence exclusively.
V5: Addressed review comments from Christian
- change the amdgpu_vm_pt_free_dfs's functionality to simple freeing
of the objects and rename it.
- add all the PTE objects in params->tlb_flush_waitlist
- let amdgpu_vm_pt_free_root handle the freeing of BOs independently
- call amdgpu_vm_pt_free_list directly
V6: Rebase
V7: Rebase
V8: Added a NULL check to fix this backtrace issue:
[ 415.351447] BUG: kernel NULL pointer dereference, address: 0000000000000008
[ 415.359245] #PF: supervisor write access in kernel mode
[ 415.365081] #PF: error_code(0x0002) - not-present page
[ 415.370817] PGD 101259067 P4D 101259067 PUD 10125a067 PMD 0
[ 415.377140] Oops: 0002 [#1] PREEMPT SMP NOPTI
[ 415.382004] CPU: 0 PID: 25481 Comm: test_with_MPI.e Tainted: G OE 5.18.2-mi300-build-140423-ubuntu-22.04+ #24
[ 415.394437] Hardware name: AMD Corporation Sh51p/Sh51p, BIOS RMO1001AS 02/21/2024
[ 415.402797] RIP: 0010:amdgpu_vm_ptes_update+0x6fd/0xa10 [amdgpu]
[ 415.409648] Code: 4c 89 ff 4d 8d 66 30 e8 f1 ed ff ff 48 85 db 74 42 48 39 5d a0 74 40 48 8b 53 20 48 8b 4b 18 48 8d 43 18 48 8d 75 b0 4c 89 ff <48
> 89 51 08 48 89 0a 49 8b 56 30 48 89 42 08 48 89 53 18 4c 89 63
[ 415.430621] RSP: 0018:ffffc9000401f990 EFLAGS: 00010287
[ 415.436456] RAX: ffff888147bb82f0 RBX: ffff888147bb82d8 RCX: 0000000000000000
[ 415.444426] RDX: 0000000000000000 RSI: ffffc9000401fa30 RDI: ffff888161f80000
[ 415.452397] RBP: ffffc9000401fa80 R08: 0000000000000000 R09: ffffc9000401fa00
[ 415.460368] R10: 00000007f0cc0000 R11: 00000007f0c85000 R12: ffffc9000401fb20
[ 415.468340] R13: 00000007f0d00000 R14: ffffc9000401faf0 R15: ffff888161f80000
[ 415.476312] FS: 00007f132ff89840(0000) GS:ffff889f87c00000(0000) knlGS:0000000000000000
[ 415.485350] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 415.491767] CR2: 0000000000000008 CR3: 0000000161d46003 CR4: 0000000000770ef0
[ 415.499738] PKRU: 55555554
[ 415.502750] Call Trace:
[ 415.505482] <TASK>
[ 415.507825] amdgpu_vm_update_range+0x32a/0x880 [amdgpu]
[ 415.513869] amdgpu_vm_clear_freed+0x117/0x250 [amdgpu]
[ 415.519814] amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu+0x18c/0x250 [amdgpu]
[ 415.527729] kfd_ioctl_unmap_memory_from_gpu+0xed/0x340 [amdgpu]
[ 415.534551] kfd_ioctl+0x3b6/0x510 [amdgpu]
V9: Addressed review comments from Christian
- No NULL check reqd for root PT freeing
- Free PT list regardless of needs_flush
- Move adding BOs in list in a separate function
V10: Added Christian's RB
V11: squash in list fix
Cc: Christian König <Christian.Koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Felix Kuehling <felix.kuehling@amd.com>
Cc: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Acked-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Reviewed-by: Christian König <Christian.Koenig@amd.com>
Tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Shashank Sharma <shashank.sharma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The problem is that when (for example) 4k pages are replaced
with a single 2M page we need to wait for change to be flushed
out by invalidating the TLB before the PT can be freed.
Solve this by moving the TLB flush into a DMA-fence object which
can be used to delay the freeing of the PT BOs until it is signaled.
V2: (Shashank)
- rebase
- set dma_fence_error only in case of error
- add tlb_flush fence only when PT/PD BO is locked (Felix)
- use vm->pasid when f is NULL (Mukul)
V4: - add a wait for (f->dependency) in tlb_fence_work (Christian)
- move the misplaced fence_create call to the end (Philip)
V5: - free the f->dependency properly
V6: (Shashank)
- light code movement, moved all the clean-up in previous patch
- introduce params.needs_flush and its usage in this patch
- rebase without TLB HW sequence patch
V7:
- Keep the vm->last_update_fence and tlb_cb code until
we can fix the HW sequencing (Christian)
- Move all the tlb_fence related code in a separate function so that
its easier to read and review
V9: Addressed review comments from Christian
- start PT update only when we have callback memory allocated
V10:
- handle device unlock in OOM case (Christian, Mukul)
- added Christian's R-B
Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Reviewed-by: Shashank Sharma <shashank.sharma@amd.com>
Reviewed-by: Christian Koenig <christian.koenig@amd.com>
Signed-off-by: Christian Koenig <christian.koenig@amd.com>
Signed-off-by: Shashank Sharma <shashank.sharma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Currently page fault information is stored per
vm and which could be freed or stale during
reset. Add it pagefault information in the
vm_manager which is a global space for vm's
and remains valid across.
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
First of all calculating the number of dw to patch into a
conditional execution is not something HW generation specific.
This is just standard ring buffer calculations. While at it also
reduce the BUG_ON() into WARN_ON().
Then instead of a random bit pattern use 0 as default value for
the number of dw skipped, this way it's not mandatory any more
to patch the conditional execution.
And last make the address to check a parameter of the
conditional execution instead of getting this from the ring.
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>
This patch changes the handling and lifecycle of vm->task_info object.
The major changes are:
- vm->task_info is a dynamically allocated ptr now, and its uasge is
reference counted.
- introducing two new helper funcs for task_info lifecycle management
- amdgpu_vm_get_task_info: reference counts up task_info before
returning this info
- amdgpu_vm_put_task_info: reference counts down task_info
- last put to task_info() frees task_info from the vm.
This patch also does logistical changes required for existing usage
of vm->task_info.
V2: Do not block all the prints when task_info not found (Felix)
V3: Fixed review comments from Felix
- Fix wrong indentation
- No debug message for -ENOMEM
- Add NULL check for task_info
- Do not duplicate the debug messages (ti vs no ti)
- Get first reference of task_info in vm_init(), put last
in vm_fini()
V4: Fixed review comments from Felix
- fix double reference increment in create_task_info
- change amdgpu_vm_get_task_info_pasid
- additional changes in amdgpu_gem.c while porting
Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Shashank Sharma <shashank.sharma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
