llama.cpp/ggml
Jeff Bolz a0f3897d53
vulkan: fix top_k bug when there are ties in the input (#17659)
* vulkan: Reduce temporary memory usage for TOP_K

- Compute row size for the temp buffer based on the output of the first pass.
- Update shader addressing math to use the output row size
- Pass the output row size as "ncols_output", what used to be "ncols_output" is now "k"

For the common case of K=40 and src0=(200000,1,1,1), this reduces the temporary buffer
from about 3.2MB to 500KB.

* vulkan: fix top_k bug when there are ties in the input

I noticed by inspection a bug in the vulkan top_k shader where if the least
value in the top_k appears multiple times we could end up writing those extra
copies out rather than some larger values (if the larger values are on higher
numbered threads).

I rewrote the test verification to handle this case, where the final index set
is not necessarily the same.

* Update tests/test-backend-ops.cpp

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2025-12-05 22:03:19 +01:00
..
cmake ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094) 2025-08-07 13:45:41 +02:00
include rpc : fix alloc size logic (#17116) 2025-12-05 19:39:04 +02:00
src vulkan: fix top_k bug when there are ties in the input (#17659) 2025-12-05 22:03:19 +01:00
.gitignore vulkan : cmake integration (#8119) 2024-07-13 18:12:39 +02:00
CMakeLists.txt build : move _WIN32_WINNT definition to headers (#17736) 2025-12-04 07:04:02 +01:00