linux/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c, branch v5.8Mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/
https://git.shady.money/linux/atom?h=v5.82020-04-28T20:20:30Zdrm/amdgpu: rename direct to immediate for VM updates2020-04-28T20:20:30ZChristian Königchristian.koenig@amd.com2020-04-01T10:50:56Zurn:sha1:eaad0c3aa978e7ed654e574691b56cc24d2d409a
To avoid confusion with direct ring submissions rename bottom
of pipe VM table changes to immediate updates.
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>
drm/amdgpu: rework synchronization of VM updates v42020-02-05T04:30:39ZChristian Königchristian.koenig@amd.com2020-01-23T13:49:45Zurn:sha1:9f3cc18d19b7b791b8f59bcae51f88e759f73146
If provided we only sync to the BOs reservation
object and no longer to the root PD.
v2: update comment, cleanup amdgpu_bo_sync_wait_resv
v3: use correct reservation object while clearing
v4: fix typo in amdgpu_bo_sync_wait_resv
Signed-off-by: Christian König <christian.koenig@amd.com>
Tested-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: explicitly sync VM update to PDs/PTs2020-02-05T04:30:39ZChristian Königchristian.koenig@amd.com2019-12-16T13:22:35Zurn:sha1:81417bea8755c914134aaf3f46b56be52f4e273a
Explicitly sync VM updates to the moving fence in PDs and PTs.
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>
drm/amdgpu: trace if a PD/PT update is done directly2019-09-16T20:30:32ZChristian Königchristian.koenig@amd.com2019-04-03T11:30:56Zurn:sha1:1313dacfadc502655942def39079c6a36b3dddb7
This is usfull for debugging.
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>
drm/amdgpu: allow direct submission in the VM backends v22019-09-16T15:42:55ZChristian Königchristian.koenig@amd.com2019-09-16T15:33:28Zurn:sha1:47ca7efa4c13f8698cddc0b9c96938e7febe9258
This allows us to update page tables directly while in a page fault.
v2: use direct/delayed entities and still wait for moves
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>
drm/amdgpu: move VM table mapping into the backend as well2019-03-28T03:40:50ZChristian Königchristian.koenig@amd.com2019-03-21T15:34:18Zurn:sha1:ecf96b52bf98a22f14bd33c7deee0aad8eb6b569
Clean that up further and also fix another case where the BO
wasn't kmapped for CPU based updates.
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>
drm/amdgpu: new VM update backends2019-03-21T18:59:15ZChristian Königchristian.koenig@amd.com2019-03-18T12:16:03Zurn:sha1:6dd09027a22ac78f6ab1881aa7e2e21ad61c02b8
Separate out all functions for SDMA and CPU based page table
updates into separate backends.
This way we can keep most of the complexity of those from the
core VM code.
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>