llama.cpp
ggml: use __builtin_amdgcn_sudot4 in __dp4a for gfx11
#4787
Merged

ggml: use __builtin_amdgcn_sudot4 in __dp4a for gfx11 #4787

ggerganov merged 1 commit into ggml-org:master from fix-perf-for-gfx1101
kzhuravl
kzhuravl ggml: use __builtin_amdgcn_sudot4 in __dp4a for gfx11
28c3c833
kzhuravl
kzhuravl
kzhuravl commented on 2024-01-05
ggerganov ggerganov requested a review from SlyEcho SlyEcho 2 years ago
SlyEcho
SlyEcho approved these changes on 2024-01-06
ggerganov ggerganov merged 63ee677e into master 2 years ago
kzhuravl
kzhuravl kzhuravl deleted the fix-perf-for-gfx1101 branch 2 years ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone