public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-19 15:46:13 +02:00
Code Issues Packages Projects Releases Wiki Activity

ggml: allow casting between f32 and i32 (llama/15783)

Browse Source 
* ggml: allow casting between f32 and i32

* fix cuda

* add vulkan

* fix CPU non-cont

* add non-cont test case

* add note

* extend test number range

* correct note

* add cont version for vulkan
This commit is contained in:
Xuan-Son Nguyen
2025-09-20 13:42:51 +03:00
committed by Georgi Gerganov
parent 0175a1df8d
commit 40bcd1a469
11 changed files with 236 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/include/ggml-cpu.h
+1
View File
@@ -134,6 +134,7 @@ extern "C" {
GGML_BACKEND_API ggml_backend_reg_t ggml_backend_cpu_reg(void);
GGML_BACKEND_API void ggml_cpu_fp32_to_fp32(const float *, float *, int64_t);
GGML_BACKEND_API void ggml_cpu_fp32_to_i32 (const float *, int32_t *, int64_t);
GGML_BACKEND_API void ggml_cpu_fp32_to_fp16(const float *, ggml_fp16_t *, int64_t);
GGML_BACKEND_API void ggml_cpu_fp16_to_fp32(const ggml_fp16_t *, float *, int64_t);
GGML_BACKEND_API void ggml_cpu_fp32_to_bf16(const float *, ggml_bf16_t *, int64_t);
ggml/include/ggml.h
+1
View File
@@ -1404,6 +1404,7 @@ extern "C" {
struct ggml_tensor * a,
struct ggml_tensor * b);
// note: casting from f32 to i32 will discard the fractional part
GGML_API struct ggml_tensor * ggml_cast(
struct ggml_context * ctx,
struct ggml_tensor * a,