diff options
author | Philip Yang <Philip.Yang@amd.com> | 2020-02-07 17:08:04 -0500 |
---|---|---|
committer | Alex Deucher <alexander.deucher@amd.com> | 2021-04-20 21:48:23 -0400 |
commit | 50ea50cf6f6d31d3235ad1853c5dbea766a3ed11 (patch) | |
tree | d8db08b809619b235b7d7c54079924694fd0d656 /drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | |
parent | b53fa124acdcec6d05bcdb36b55bf0f84471b1b7 (diff) | |
download | linux-50ea50cf6f6d31d3235ad1853c5dbea766a3ed11.tar.gz linux-50ea50cf6f6d31d3235ad1853c5dbea766a3ed11.tar.bz2 linux-50ea50cf6f6d31d3235ad1853c5dbea766a3ed11.zip |
drm/amdkfd: copy memory through gart table
Use sdma linear copy to migrate data between ram and vram. The sdma
linear copy command uses kernel buffer function queue to access system
memory through gart table.
Use reserved gart table window 0 to map system page address, and vram
page address is direct mapping. Use the same kernel buffer function to
fill in gart table mapping, so this is serialized with memory copy by
sdma job submit. We only need wait for the last memory copy sdma fence
for larger buffer migration.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_object.h')
0 files changed, 0 insertions, 0 deletions