COMMITS
January 2, 2026
J
Update TheRock CI SHA 20260102 (#3506)
Joseph Macaranas committed
December 31, 2025
V
[CK_BUILDER] Instance traits for conv bwd weight algorithms (#3498)
Ville Pietilä committed
D
Temporarily disable kernel instances that won't build on gfx1101 on Windows (#3499)
DarylHawkinsAMD committed
December 30, 2025
K
[CK_Builder] [testing] Integrate device random generators (#3427)
kabrahamAMD committed
B
Fix grouped conv wrw kernels names (#3494)
Bartłomiej Kocot committed
A
Grouped convolution backward data WMMA v3 implementation (#3460)
ApoorvaKalyani committed
December 29, 2025
Y
[CK_TILE] support split-k a16w4 gemm1 (#3389)
yadaish committed
J
[CK_BUILDER] Add GPU Reference Algorithm to CK Builder (#3381)
JH-Leon-KIM-AMD committed
K
Replace grouped conv bwd wei wmmaV3 bilin/scale bf16f32bf16 support with bf16bf16bf16 (#3470)
Kiefer van Teutem committed
Y
[CK_TILE] MX FLATMM Fix M Padding (#3489)
Yi DING committed
December 27, 2025
J
enable f8 tests (#3488)
joyeamd committed
December 25, 2025
Y
[CK_TILE] Align FMHA BWD Reference with Kernel Implementation (#3486)
Yi DING committed
E
[CK_TILE] Grouped gemm quant tensor layouts (#3414)
Erwin Terpstra committed
I
remove the LLVM_MAIN_REVISION usage (#3487)
Illia Silin committed
December 24, 2025
T
[CK TILE ENGINE] CI configuration with basic cases (#3475)
Thrupti Raj Lakshmana Gowda committed
K
Enable padding blockscale for abquant (#3453)
kensclin committed
P
[CK_TILE][FMHA] Add FP8 support for batch_prefill kernel (#3425)
Po Yen Chen committed
December 23, 2025
J
[CK_TILE] Minor splitk bugfix for gemms and conv (#3387)
jakpiase committed
J
[CK grouped gemm] Fix grouped gemm two stage HasMainK0BlockLoop (#3466)
Johannes Graner committed
K
[ck_builder] add utility functions to convolution (#3459)
kabrahamAMD committed
J
[CK_TILE] Add splitk support to ck tile conv bwd data (#3353)
jakpiase committed
L
Y
[CK_TILE] FMHA Ignore BWD Failed Cases in Smoke Test (#3480)
Yi DING committed
December 22, 2025
B
Fix grouped conv fwd wmma porting (#3479)
Bartłomiej Kocot committed
W
Post-merge cleanup for WMMA grouped conv fwd (#3468)
Wojciech Laskowski committed
B
Fix jenkinsfile for large tensor conv test (#3478)
Bartłomiej Kocot committed
December 20, 2025
J
[CK-TILE] Guard against compiler lexer diagnostic (#3444)
Jan Patrick Lehr committed
December 19, 2025
B
Improve XDL to WMMA porting for grouped conv fwd (#3456)
Bartłomiej Kocot committed
I
get LLVM_MAIN_REVISION macro from compiler header (#3469)
Illia Silin committed
G
Revert "details from org var (#3431)" (#3473)
Geo Min committed