Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 00/19] The series add support for UMQ submission
@ 2025-03-28  7:52 Sunil Khatri
  0 siblings, 0 replies; 2+ messages in thread
From: Sunil Khatri @ 2025-03-28  7:52 UTC (permalink / raw)
  To: igt-dev
  Cc: Alex Deucher, Christian König, Vitaly Prosyak, Strawbridge,
	Michael, Sunil Khatri

Here we are enabling UMQ submission framework along with some of the
test cases which validated gfx and compute with UMQ Submission

Sunil Khatri (19):
  drm-uapi/amdgpu: sync with drm-next
  drm-uapi/amdgpu: sync with drm-next
  lib/amdgpu: Add user mode queue support in ring context
  lib/amdgpu: Add support of amd user queues
  lib/amdgpu: add func amdgpu_bo_alloc_and_map_sync
  tests/amdgpu: Add user queue support for gfx and compute
  tests/amdgpu: Add UMQ submission tests for gfx and compute
  tests/amdgpu: Add amdgpu_sync_dependency_test with UMQ
  tests/amdgpu: use memory API's from amd_memory.h
  lib/amdgpu: add macro for adding cmds in user queue
  lib/amdgpu: use macro to add cmds in the user ring
  tests/amdgpu: Add amdgpu_cp_nops tests for UMQ
  drm-uapi/amdgpu: sync with drm-next
  lib/amdgpu: use right API to get the correct size
  lib/amdgpu: use a memory fence to serialize write
  tests/amdgpu: disable check for IP presense with no kernel queue
  lib/amdgpu: make the local functions as static
  lib/amdgpu: enable UMQ function under macro
  tests/amdgpu: Disable the UMQ tests under a macro

 include/drm-uapi/amdgpu_drm.h       | 261 ++++++++++++++++
 lib/amdgpu/amd_PM4.h                |   3 +
 lib/amdgpu/amd_command_submission.c | 313 +++++++++++++------
 lib/amdgpu/amd_command_submission.h |   8 +-
 lib/amdgpu/amd_compute.c            | 100 ++++--
 lib/amdgpu/amd_compute.h            |   2 +-
 lib/amdgpu/amd_ip_blocks.h          |  35 +++
 lib/amdgpu/amd_memory.c             |  19 ++
 lib/amdgpu/amd_memory.h             |   8 +
 lib/amdgpu/amd_user_queue.c         | 465 ++++++++++++++++++++++++++++
 lib/amdgpu/amd_user_queue.h         |  55 ++++
 lib/meson.build                     |   3 +-
 tests/amdgpu/amd_basic.c            | 174 ++++++++---
 tests/amdgpu/amd_cs_nop.c           | 149 ++++-----
 tests/amdgpu/amd_security.c         |   4 +-
 15 files changed, 1335 insertions(+), 264 deletions(-)
 create mode 100644 lib/amdgpu/amd_user_queue.c
 create mode 100644 lib/amdgpu/amd_user_queue.h

-- 
2.43.0


^ permalink raw reply	[flat|nested] 2+ messages in thread
* [PATCH v2 00/19] The series add support for UMQ submission
@ 2025-03-28  7:54 Sunil Khatri
  0 siblings, 0 replies; 2+ messages in thread
From: Sunil Khatri @ 2025-03-28  7:54 UTC (permalink / raw)
  To: igt-dev
  Cc: Alex Deucher, Christian König, Vitaly Prosyak, Strawbridge,
	Michael, Sunil Khatri

Here we are enabling UMQ submission framework along with some of the
test cases which validated gfx and compute with UMQ Submission

Sunil Khatri (19):
  drm-uapi/amdgpu: sync with drm-next
  drm-uapi/amdgpu: sync with drm-next
  lib/amdgpu: Add user mode queue support in ring context
  lib/amdgpu: Add support of amd user queues
  lib/amdgpu: add func amdgpu_bo_alloc_and_map_sync
  tests/amdgpu: Add user queue support for gfx and compute
  tests/amdgpu: Add UMQ submission tests for gfx and compute
  tests/amdgpu: Add amdgpu_sync_dependency_test with UMQ
  tests/amdgpu: use memory API's from amd_memory.h
  lib/amdgpu: add macro for adding cmds in user queue
  lib/amdgpu: use macro to add cmds in the user ring
  tests/amdgpu: Add amdgpu_cp_nops tests for UMQ
  drm-uapi/amdgpu: sync with drm-next
  lib/amdgpu: use right API to get the correct size
  lib/amdgpu: use a memory fence to serialize write
  tests/amdgpu: disable check for IP presense with no kernel queue
  lib/amdgpu: make the local functions as static
  lib/amdgpu: enable UMQ function under macro
  tests/amdgpu: Disable the UMQ tests under a macro

 include/drm-uapi/amdgpu_drm.h       | 261 ++++++++++++++++
 lib/amdgpu/amd_PM4.h                |   3 +
 lib/amdgpu/amd_command_submission.c | 313 +++++++++++++------
 lib/amdgpu/amd_command_submission.h |   8 +-
 lib/amdgpu/amd_compute.c            | 100 ++++--
 lib/amdgpu/amd_compute.h            |   2 +-
 lib/amdgpu/amd_ip_blocks.h          |  35 +++
 lib/amdgpu/amd_memory.c             |  19 ++
 lib/amdgpu/amd_memory.h             |   8 +
 lib/amdgpu/amd_user_queue.c         | 465 ++++++++++++++++++++++++++++
 lib/amdgpu/amd_user_queue.h         |  55 ++++
 lib/meson.build                     |   3 +-
 tests/amdgpu/amd_basic.c            | 174 ++++++++---
 tests/amdgpu/amd_cs_nop.c           | 149 ++++-----
 tests/amdgpu/amd_security.c         |   4 +-
 15 files changed, 1335 insertions(+), 264 deletions(-)
 create mode 100644 lib/amdgpu/amd_user_queue.c
 create mode 100644 lib/amdgpu/amd_user_queue.h

-- 
2.43.0


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2025-03-28  8:05 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-03-28  7:52 [PATCH v2 00/19] The series add support for UMQ submission Sunil Khatri
  -- strict thread matches above, loose matches on Subject: below --
2025-03-28  7:54 Sunil Khatri

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox