vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

sync : ggml (conv 1d + 2d updates, UB fixes) (#3468)

Browse source 
* sync : ggml (conv 1d + 2d updates)

ggml-ci

* ggml : fix UB in q5_0 and q5_1 quantize code

ggml.c:1033:39: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior

ggml.c:1081:39: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior

ggml-ci

* tests : fix UB in test-quantize-perf
This commit is contained in:
Georgi Gerganov 2023-10-04 15:29:58 +03:00 • committed by GitHub
parent f72f8f22c9
commit f93af02488
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
6 changed files with 725 additions and 379 deletions
Download patch file Download diff file Expand all files Collapse all files

2
k_quants.c
View file

@ -69,7 +69,6 @@ inline static int32_t vaddvq_s32(int32x4_t v) {
// 2-6 bit quantization in super-blocks
//
//
// ===================== Helper functions
//
@ -348,7 +347,6 @@ void quantize_row_q2_K_reference(const float * restrict x, block_q2_K * restrict
const float q4scale = 15.f;
for (int i = 0; i < nb; i++) {
float max_scale = 0; // as we are deducting the min, scales are always positive
float max_min = 0;
for (int j = 0; j < QK_K/16; ++j) {