Implement batched gemm bias permute for RDNA4#3534

Merged

ErwinTerpstra merged 18 commits intodevelopfrom

eterpstr/96-implement-device_batched_gemm_bias_permute-for-rdna4

Jan 17, 2026

Commits on Dec 19, 2025

feat: test setup for batched contraction (aka batched gemm multiple d e permute)
ErwinTerpstra
committed
wip: device struct for WMMA batched contraction multiple d based on new gridwise op
ErwinTerpstra
committed

Commits on Jan 7, 2026

feat: working batched contraction on RDNA, non-naive tensor descriptors for gridwise_gemm_wmma_cshuffle_v3, test setup for odd cases
ErwinTerpstra
committed

Commits on Jan 9, 2026

fix: don't compile batched contraction test on unsupported architectures
ErwinTerpstra
committed

Commits on Jan 15, 2026

Commits on Jan 16, 2026