HIP: RDNA4 tensor core support for MMF #17077
JohannesGaessler
merged
17 commits into
ggml-org:master
from
zhang-hui-yulo:mmf_wmma_rdna4
mmf for rdna4
2f7cfcf6
align the padding for rdna4
d564a352
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
0ec241dc
forbit mul_mat_f for rdna4
bbee5feb
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
6b8ceebf
fix as comment
fd18344c
remove device kernels
7a09e22e
add constexpr for early return
c65dd59e
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
48a53b57
update based on review comment
b7c13eee
change based on the review comment
a0aa4917
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
8c2f9a30
JohannesGaessler
commented
on 2025-11-14
pass compile error
7a88d7cd
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
cfc149ae
JohannesGaessler
approved these changes
on 2025-11-14
keep code consistency
59a012f2
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
6802fbf7
Merge branch 'ggml-org:master' into mmf_wmma_rdna4
facded59
JohannesGaessler
approved these changes
on 2025-11-21
Reviewers
Assignees
No one assigned
Labels
Nvidia GPU
ggml
Milestone
No milestone
Login to write a write a comment.
Login via GitHub