llama.cpp/ggml
Chenguang Li c1c354e44c
CANN: Refactor ND to NZ workspace to be per-device (#15763)
* CANN:Refactor ND to NZ workspace to be per-device in Ascend backend

- Replaced the previous single global ND→NZ workspace with a per-device
  cache using unordered_map keyed by device ID.
- Functions `release_nz_workspace`, `relloc_nz_workspace`, and
  `get_nz_workspace` now manage workspace independently for each device,
  preventing memory conflicts in multi-device / pipeline parallel scenarios.
- This change fixes potential precision issues caused by workspace
  overwrites when multiple devices perform ND→NZ conversions concurrently.

Co-authored-by: hipudding <huafengchun@gmail.com>

* refactor

Signed-off-by: noemotiovon <757486878@qq.com>

* rename

Signed-off-by: noemotiovon <757486878@qq.com>

* fix review comments

Signed-off-by: noemotiovon <757486878@qq.com>

---------

Signed-off-by: noemotiovon <757486878@qq.com>
Co-authored-by: hipudding <huafengchun@gmail.com>
2025-09-04 20:20:14 +08:00
..
cmake ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094) 2025-08-07 13:45:41 +02:00
include ggml: add ops for WAN video model (cuda && cpu) (#15669) 2025-09-04 10:38:49 +02:00
src CANN: Refactor ND to NZ workspace to be per-device (#15763) 2025-09-04 20:20:14 +08:00
.gitignore vulkan : cmake integration (#8119) 2024-07-13 18:12:39 +02:00
CMakeLists.txt ggml-cpu : optimize RVV kernels (#15720) 2025-09-03 16:16:21 +08:00