public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-12 01:05:43 +02:00
Code Issues Packages Projects Releases Wiki Activity

vulkan: fix Flash Attention shmem write out of bounds (llama/29988)

Browse Source
This commit is contained in:
Ruben Ortlam
2026-10-06 10:38:22 +03:00
committed by Georgi Gerganov
parent 563a9f7ae7
commit cdc873029c
1 changed files with 6 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-vulkan/vulkan-shaders/flash_attn.comp
+6 -11
View File
@@ -130,20 +130,15 @@ void main() {
const float qd_inv = qd != 0.0f ? 1.0f / qd : 0.0f;
vals = round(vals * qd_inv);
Qf[buf_ib].qs[buf_iqs] = pack32(i8vec4(vals));
const float thread_sum = vals.x + vals.y + vals.z + vals.w;
// Q8_0 K does not use the row sum.
const float sum = FaTypeK == GGML_TYPE_Q8_0 ? 0.0f : subgroupClusteredAdd(thread_sum, 8);
// Q8_0 K only needs (qd, _); the asymmetric Q4_*/Q5_* family also stores
// the row-sum scaled by qd, used in k_dot_correction.
if (FaTypeK == GGML_TYPE_Q8_0) {
if (buf_iqs == 0) {
Qf[buf_ib].ds = FLOAT_TYPEV2(qd, 0.0f);
}
} else {
const float thread_sum = vals.x + vals.y + vals.z + vals.w;
const float sum = subgroupClusteredAdd(thread_sum, 8);
if (r < Br) {
Qf[buf_ib].qs[buf_iqs] = pack32(i8vec4(vals));
if (buf_iqs == 0) {
Qf[buf_ib].ds = FLOAT_TYPEV2(qd, sum * qd);
Qf[buf_ib].ds = FaTypeK == GGML_TYPE_Q8_0 ? FLOAT_TYPEV2(qd, 0.0f) : FLOAT_TYPEV2(qd, sum * qd);
}
}
#endif