public_git
/
whisper.cpp
mirror of https://github.com/ggml-org/whisper.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

vulkan: Add missing bounds checking to scalar/coopmat1 mul_mat_id (llama/15334)

This commit is contained in:
Jeff Bolz 2025-08-16 03:58:38 -05:00 committed by Georgi Gerganov
parent e27f4f205d
commit 5b62995350
1 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-vulkan/vulkan-shaders/mul_mm.comp
View File

@ -801,7 +801,7 @@ void main() {
} }
#else #else
const uint row_i = ic * BN + loadc_b + l; const uint row_i = ic * BN + loadc_b + l;
if (row_i < _ne1) { if (row_i < _ne1 && block + loadr_b < end_k) {
const u16vec2 row_idx = row_ids[row_i]; const u16vec2 row_idx = row_ids[row_i];
buf_b[(loadc_b + l) * SHMEM_STRIDE + loadr_b] = TO_FLOAT_TYPE(data_b[pos_b + row_idx.y * p.batch_stride_b + (row_idx.x % p.ne11) * p.stride_b + loadr_b]); buf_b[(loadc_b + l) * SHMEM_STRIDE + loadr_b] = TO_FLOAT_TYPE(data_b[pos_b + row_idx.y * p.batch_stride_b + (row_idx.x % p.ne11) * p.stride_b + loadr_b]);
} else { } else {
@ -875,10 +875,12 @@ void main() {
const u16vec2 row_idx = row_ids[row_i]; const u16vec2 row_idx = row_ids[row_i];
if (dr + cm_row * TM + store_r < p.M) {
data_d[row_idx.y * p.batch_stride_d + row_idx.x * p.stride_d + dr + cm_row * TM + store_r] = D_TYPE(coopmat_stage[warp_i * TM * TN + (col + store_c) * TM + store_r]); data_d[row_idx.y * p.batch_stride_d + row_idx.x * p.stride_d + dr + cm_row * TM + store_r] = D_TYPE(coopmat_stage[warp_i * TM * TN + (col + store_c) * TM + store_r]);
} }
} }
} }
}
#else #else
const bool is_aligned = p.stride_d % 4 == 0; // Assumption: D_TYPE == float const bool is_aligned = p.stride_d % 4 == 0; // Assumption: D_TYPE == float
@ -925,7 +927,9 @@ void main() {
#endif // MUL_MAT_ID #endif // MUL_MAT_ID
[[unroll]] for (uint cr = 0; cr < TM; cr++) { [[unroll]] for (uint cr = 0; cr < TM; cr++) {
#ifdef MUL_MAT_ID #ifdef MUL_MAT_ID
if (dr_warp + cr < p.M) {
data_d[row_idx.y * p.batch_stride_d + row_idx.x * p.stride_d + dr_warp + cr] = D_TYPE(sums[(wsic * TN + cc) * (WMITER * TM) + wsir * TM + cr]); data_d[row_idx.y * p.batch_stride_d + row_idx.x * p.stride_d + dr_warp + cr] = D_TYPE(sums[(wsic * TN + cc) * (WMITER * TM) + wsir * TM + cr]);
}
#else #else
if (dr_warp + cr < p.M && dc_warp + cc < p.N) { if (dr_warp + cr < p.M && dc_warp + cc < p.N) {
data_d[offsets + (dc_warp + cc) * p.stride_d + dr_warp + cr] = D_TYPE(sums[(wsic * TN + cc) * (WMITER * TM) + wsir * TM + cr]); data_d[offsets + (dc_warp + cc) * p.stride_d + dr_warp + cr] = D_TYPE(sums[(wsic * TN + cc) * (WMITER * TM) + wsir * TM + cr]);