Skip to content

xe: softmax: correct src/dst scale in vectorized kernel #3451

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

pv-pterab-s
Copy link
Contributor

Corrects missing scale implementation in the vectorized version of the reusable softmax kernel. Adds explicit tests to check that the individual vectorized and "small" kernel implementations function.

@pv-pterab-s pv-pterab-s requested review from a team as code owners June 19, 2025 16:57
@github-actions github-actions bot added platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel component:tests Codeowner: @oneapi-src/onednn-arch labels Jun 19, 2025
@pv-pterab-s pv-pterab-s force-pushed the pryorgal/softmax_reusable_scale_fix branch 2 times, most recently from fd83dd8 to 82ff55d Compare June 20, 2025 17:47
@pv-pterab-s pv-pterab-s force-pushed the pryorgal/softmax_reusable_scale_fix branch from 82ff55d to 3bf6078 Compare June 20, 2025 21:59
@pv-pterab-s
Copy link
Contributor Author

make test
enable os_win
enable os_lnx
enable os_mac
disable test_device_cpu
enable test_device_gpu
disable build_cpu_runtime_omp
disable build_cpu_runtime_sycl
disable build_cpu_runtime_tbb
enable build_graph
enable compiler_icx-previous
enable compiler_gnu9
enable compiler_clang14
enable compiler_vs2022
disable build_gpu_runtime_sycl
disable benchdnn_all
enable benchdnn_softmax
enable benchdnn_graph
enable arch_gpu_xe-hpc
disable arch_gpu_xe-hpg-atsm
disable arch_gpu_xe-hpg-dg2
disable arch_gpu_xe-lp
disable arch_gpu_xe-lpg
disable arch_gpu_xe-lpg+
disable arch_gpu_xe2-hpg-bmg
disable arch_gpu_xe2-lpg

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
component:tests Codeowner: @oneapi-src/onednn-arch platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants