thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

ggml : remove GGML_KQ_MASK_PAD constant (#17910)

Browse source 
* ggml : remove GGML_KQ_MASK_PAD constant

* cont : remove comment
This commit is contained in:
Georgi Gerganov 2025-12-10 20:53:16 +02:00 • committed by GitHub
parent 4df6e859e9
commit 4dff236a52
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 19 additions and 36 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml/src/ggml.c
View file

@ -5260,8 +5260,6 @@ struct ggml_tensor * ggml_flash_attn_ext(
if (mask) {
GGML_ASSERT(ggml_is_contiguous(mask));
GGML_ASSERT(mask->ne[1] >= GGML_PAD(q->ne[1], GGML_KQ_MASK_PAD) &&
"the Flash-Attention kernel requires the mask to be padded to GGML_KQ_MASK_PAD and at least n_queries big");
//GGML_ASSERT(ggml_can_repeat_rows(mask, qk));
GGML_ASSERT(q->ne[2] % mask->ne[2] == 0);