llama.cpp
[SYCL] Fix WARP_SIZE=16 bug of Intel GPU
#8266
Merged

[SYCL] Fix WARP_SIZE=16 bug of Intel GPU #8266

airMeng merged 8 commits into ggml-org:master from luoyu-intel:sycl-acc
luoyu-intel
github-actions github-actions added testing
github-actions github-actions added ggml
github-actions github-actions added SYCL
airMeng airMeng requested a review from AidanBeltonS AidanBeltonS 1 year ago
luoyu-intel luoyu-intel force pushed from c9045c1c to 3bf8c2c9 1 year ago
airMeng
airMeng approved these changes on 2024-07-03
airMeng
NeoZhangJianyu
NeoZhangJianyu commented on 2024-07-03
NeoZhangJianyu
NeoZhangJianyu approved these changes on 2024-07-03
qnixsynapse
luoyu-intel
qnixsynapse
Alcpz
mofosyne mofosyne added Review Complexity : Medium
luoyu-intel fix group_norm ut
c675aaf0
luoyu-intel split softmax
e50517b6
luoyu-intel fix softmax
d70305b3
luoyu-intel revert qx_k
0012f2c1
luoyu-intel add concat support condition
870b607c
luoyu-intel revert debug code
d7cf5f5a
luoyu-intel move QK_WARP_SIZE to presets.hpp
ac8a4bd9
luoyu-intel luoyu-intel force pushed from 4887fdce to ac8a4bd9 1 year ago
luoyu-intel rebase work_space api
87098db6
airMeng airMeng merged a9554e20 into master 1 year ago
luoyu-intel luoyu-intel deleted the sycl-acc branch 1 year ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone