Skip to content
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

Fix missing RVV gemm kernel descriptions and regenerate tests/benchmarks #7888

Merged
merged 1 commit into from
Mar 10, 2025

Conversation

copybara-service[bot]
Copy link
Contributor

Fix missing RVV gemm kernel descriptions and regenerate tests/benchmarks

xnn_qs8_qc8w_gemm_minmax_fp32_ukernel_1x4v__rvv,
xnn_init_qs8_qc8w_conv_minmax_fp32_scalar_params,
xnn_pack_qs8_gemm_goi_w,
/*mr=*/1, /*nr=*/4 * xnn_init_hardware_config()->vlenb / sizeof(int8_t), /*kr=*/1, /*sr=*/1,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@dsharlet I looked into this while fixing PR #7932. Sizeof() here needs to be int32_t for qc8w.

Changing https://github.com/google/XNNPACK/blob/master/tools/generate-gemm-test.py#L741 to be "qs8": "int32_t", resolves the problem, and the tests pass successfully. I'm not sure if this line should be further qualified by the 'activation' although there is currently no qs8-gemm rvv implementation, and so perhaps that single change is sufficient.

@copybara-service copybara-service bot force-pushed the test_731115584 branch 2 times, most recently from 2179a5c to e90b94f Compare March 10, 2025 04:21
@copybara-service copybara-service bot merged commit f481f28 into master Mar 10, 2025
@copybara-service copybara-service bot deleted the test_731115584 branch March 10, 2025 04:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants