Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

CUDA: experimental native mxfp4 support for blackwell (#17906)

Browse Source 
* CUDA: experimental native mxfp4 support for blackwell

* optimize load_tiles

* optimize quantize_mxfp4

* cleanup

* first pass review: formatting

* use interleaved layout for mma

* mmq: add assert for size

* use __nv_fp4x4_e2m1

* use iter_k as 512, cleanup

* Use 1200 as blackwell instead of 1000

* address review comments

* mmq: fix stride

* quantize.cu: use reference impl of e8m0 scale

* address review comments

* add 120f-virtual + minor fixes

---------

Co-authored-by: Aman Gupta <aman>
This commit is contained in:
Aman Gupta
2025-12-24 22:28:26 +08:00
committed by GitHub
co-authored by Aman Gupta <aman>
parent 54132f1b1f
commit c8a2417d7b
8 changed files with 426 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/mma.cuh
+21
View File
@@ -900,6 +900,27 @@ namespace ggml_cuda_mma {
#endif // AMPERE_MMA_AVAILABLE
}
static __device__ __forceinline__ void mma_block_scaled(tile<16, 8, float> & D,
const tile<16, 8, int> & A,
const tile<8, 8, int> & B,
uint32_t a_scale,
uint32_t b_scale) {
#ifdef BLACKWELL_MMA_AVAILABLE
const int * Axi = (const int *) A.x;
const int * Bxi = (const int *) B.x;
float * Dxi = (float *) D.x;
asm volatile(
"mma.sync.aligned.kind::mxf4.block_scale.scale_vec::2X.m16n8k64.row.col.f32.e2m1.e2m1.f32.ue8m0 "
"{%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3}, "
"%10, {0, 0}, %11, {0, 0};"
: "+f"(Dxi[0]), "+f"(Dxi[1]), "+f"(Dxi[2]), "+f"(Dxi[3])
: "r"(Axi[0]), "r"(Axi[1]), "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[0]), "r"(Bxi[1]), "r"(a_scale), "r"(b_scale));
#else
GGML_UNUSED_VARS(D, A, B, a_scale, b_scale);
#endif // BLACKWELL_MMA_AVAILABLE
}
static __device__ __forceinline__ void mma(
tile<16, 8, float> & D, const tile<16, 8, half2> & A, const tile<8, 8, half2> & B) {
#ifdef TURING_MMA_AVAILABLE