DeepSpeed/csrc
baodi 1fdad1fa52
make xpu ops compatible with oneapi 2025.0 (#6760)
Compatibility update for xpu ops

This PR introduces changes that will make xpu ops compatible with the
OneAPI 2025.0 toolkit. This is an important update that will allow us to
develop and ship our most demanding models on this innovative hardware.

---------

Signed-off-by: baodii <di.bao@intel.com>
Co-authored-by: Logan Adams <114770087+loadams@users.noreply.github.com>
Co-authored-by: Logan Adams <loadams@microsoft.com>
2024-11-19 17:38:27 +00:00
..
adagrad Fix Type Name Inconsistency & Typo in cpu_adam (#6732) 2024-11-11 23:31:45 +00:00
adam Fix Type Name Inconsistency & Typo in cpu_adam (#6732) 2024-11-11 23:31:45 +00:00
aio AIO File Offsets (#6641) 2024-11-12 16:34:17 +00:00
cpu [CPU] Allow deepspeed.comm.inference_all_reduce in torch.compile graph (#5604) 2024-07-15 22:24:11 +00:00
deepspeed4science/evoformer_attn Update clang-format version from 16 to 18. (#5839) 2024-08-06 09:14:21 -07:00
fp_quantizer wrap include cuda_bf16.h with ifdef BF16_AVAILABLE (#6520) 2024-09-10 16:08:50 +00:00
gds AIO File Offsets (#6641) 2024-11-12 16:34:17 +00:00
includes Fix Type Name Inconsistency & Typo in cpu_adam (#6732) 2024-11-11 23:31:45 +00:00
lamb Switch from HIP_PLATFORM_HCC to HIP_PLATFORM_AMD (#4539) 2023-10-19 21:01:48 +00:00
lion Fix Type Name Inconsistency & Typo in cpu_adam (#6732) 2024-11-11 23:31:45 +00:00
quantization Fixed the Windows build. (#5596) 2024-05-31 22:11:10 +00:00
random_ltd Rocm warp size fix (#5402) 2024-05-17 20:35:58 +00:00
sparse_attention Update DeepSpeed copyright license to Apache 2.0 (#3111) 2023-03-30 17:14:38 -07:00
spatial Switch from HIP_PLATFORM_HCC to HIP_PLATFORM_AMD (#4539) 2023-10-19 21:01:48 +00:00
transformer [Bug Fix] Support threads_per_head < 64 for wavefront size of 64 (#6622) 2024-11-04 21:51:27 +00:00
utils Update DeepSpeed copyright license to Apache 2.0 (#3111) 2023-03-30 17:14:38 -07:00
xpu make xpu ops compatible with oneapi 2025.0 (#6760) 2024-11-19 17:38:27 +00:00