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
whisper.cpp/ggml/src/ggml-sycl
History
Daniel Bevenius 5c524bb879
ggml : fix padding in timestep embedding kernels (llama/15932) 
* ggml : remove adding extra dim timestep embedding

This commit updates the ggml_timestep_embedding function to no longer
add an extra dimension when the specified dimension is odd.

The motivation for this change is that this introduces an unnecessary
dimension when the dimension is odd, which caused an issue in the
kernels which were not expecting this extra dimension and it resulted in
uninitialized memory for the second to last dimension.

* ggml-cuda : fix padding in timestep embedding kernel

This commit removes the zeroing out of the last dimension now that we
are not adding the extra padding dimension.

* ggml-metal : fix padding in timestep embedding kernel

This commit fixes the zero padding for odd dimensions in
the timestep embedding kernel

* ggml-opencl : fix padding in timestep embedding kernel

This commit fixes the zero padding for odd dimensions in
the timestep embedding kernel.

* ggml-sycl : fix padding in timestep embedding kernel

This commit fixes the zero padding for odd dimensions in
the timestep embedding kernel.

* ggml-vulkan : fix padding in timestep embedding kernel

This commit fixes the zero padding for odd dimensions in
the timestep embedding kernel.

* ggml-cpu : fix padding in timestep embedding function

This commit removes the zeroing out of the last dimension now that we
are not adding the extra padding dimension.
 		2025-09-20 13:45:30 +03:00
..
dpct Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
CMakeLists.txt SYCL: Bump oneMath commit (llama/14152) 2025-06-18 12:40:34 +03:00
backend.hpp sycl: refactor quantization to q8_1 (llama/14815) 2025-08-18 20:30:45 +03:00
binbcast.cpp SYCL: Add COUNT_EQUAL operator support (llama/15991) 2025-09-20 13:45:30 +03:00
binbcast.hpp SYCL: Add COUNT_EQUAL operator support (llama/15991) 2025-09-20 13:45:30 +03:00
common.cpp SYCL: Remove misleading ggml_sycl_op_flatten function (llama/12387) 2025-03-31 14:56:53 +03:00
common.hpp sycl: GGML_SYCL_DISABLE_OPT on by default for all Intel Devices (llama/13973) 2025-07-01 17:54:53 +03:00
concat.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
concat.hpp SYCL: Refactor ggml_sycl_compute_forward (llama/11121) 2025-01-14 10:38:01 +02:00
conv.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
conv.hpp SYCL: Refactor ggml_sycl_compute_forward (llama/11121) 2025-01-14 10:38:01 +02:00
convert.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
convert.hpp sycl: addressing non-contiguous src1 mul_mats (nc and batched) (llama/13343) 2025-05-13 13:59:21 +03:00
cpy.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
cpy.hpp SYCL: Add set_rows support for quantized types (llama/14883) 2025-08-18 20:30:45 +03:00
dequantize.hpp sycl: Add reorder to Q6_K mmvq implementation (llama/13885) 2025-06-10 12:40:33 +03:00
dmmv.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
dmmv.hpp whisper : reorganize source code + improve CMake (#2256) 2024-06-26 19:34:09 +03:00
element_wise.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
element_wise.hpp ggml : implement GEGLU_ERF and GEGLU_QUICK ops (llama/14445) 2025-07-12 19:23:56 +03:00
gemm.hpp sycl: Batched mulmat rework for oneDNN dispatch (llama/14617) 2025-07-20 00:23:50 +03:00
getrows.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
getrows.hpp SYCL: Remove misleading ggml_sycl_op_flatten function (llama/12387) 2025-03-31 14:56:53 +03:00
ggml-sycl.cpp SYCL: Add COUNT_EQUAL operator support (llama/15991) 2025-09-20 13:45:30 +03:00
gla.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
gla.hpp SYCL: Add gated linear attention kernel (llama/11175) 2025-02-03 22:00:57 +02:00
im2col.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
im2col.hpp SYCL: Remove misleading ggml_sycl_op_flatten function (llama/12387) 2025-03-31 14:56:53 +03:00
mmq.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
mmq.hpp whisper : reorganize source code + improve CMake (#2256) 2024-06-26 19:34:09 +03:00
mmvq.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
mmvq.hpp whisper : reorganize source code + improve CMake (#2256) 2024-06-26 19:34:09 +03:00
norm.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
norm.hpp SYCL: Remove misleading ggml_sycl_op_flatten function (llama/12387) 2025-03-31 14:56:53 +03:00
outprod.cpp sycl: Add more debug prints (llama/13640) 2025-05-27 18:03:00 +03:00
outprod.hpp SYCL: Refactor ggml_sycl_compute_forward (llama/11121) 2025-01-14 10:38:01 +02:00
presets.hpp Optimize RWKV6 Operator Naming and Implement Multi-core CPU/ SYCL Acceleration (llama/10133) 2024-11-15 15:21:04 +02:00
quantize.hpp sycl: refactor quantization to q8_1 (llama/14815) 2025-08-18 20:30:45 +03:00
quants.hpp sycl: fixed semantics of block offset calculation (llama/14814) 2025-07-28 13:02:32 +03:00
rope.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
rope.hpp SYCL: Add non-contiguous support in ROPE (llama/12993) 2025-04-24 20:39:16 +03:00
set_rows.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
set_rows.hpp SYCL: Initial set_rows kernel implementation (llama/14562) 2025-07-12 19:23:56 +03:00
softmax.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
softmax.hpp SYCL : SOFTMAX F16 mask support and other fixes (llama/11261) 2025-02-03 22:00:57 +02:00
sycl_hw.cpp sycl: GGML_SYCL_DISABLE_OPT on by default for all Intel Devices (llama/13973) 2025-07-01 17:54:53 +03:00
sycl_hw.hpp sycl: GGML_SYCL_DISABLE_OPT on by default for all Intel Devices (llama/13973) 2025-07-01 17:54:53 +03:00
tsembd.cpp ggml : fix padding in timestep embedding kernels (llama/15932) 2025-09-20 13:45:30 +03:00
tsembd.hpp SYCL: Refactor ggml_sycl_compute_forward (llama/11121) 2025-01-14 10:38:01 +02:00
vecdotq.hpp sycl: fixed semantics of block offset calculation (llama/14814) 2025-07-28 13:02:32 +03:00
wkv.cpp Revert "sycl: add usage of enqueue_functions extension (llama/14244)" (llama/15910) 2025-09-20 13:45:28 +03:00
wkv.hpp llama: Add support for RWKV v7 architecture (llama/12412) 2025-03-27 11:06:03 +02:00