Skip to content

sycl: implement GGML_OP_TRI#19089

Merged
NeoZhangJianyu merged 4 commits intoggml-org:masterfrom
RachelMantel:sycl/tri
Jan 30, 2026
Merged

sycl: implement GGML_OP_TRI#19089
NeoZhangJianyu merged 4 commits intoggml-org:masterfrom
RachelMantel:sycl/tri

Conversation

@RachelMantel
Copy link
Contributor

@RachelMantel RachelMantel commented Jan 25, 2026

Implements GGML_OP_TRI for the SYCL backend (F32).

The implementation matches CPU semantics for all ggml_tri_type values
(lower/upper, with and without diagonal).

Tests:

  • ./build-sycl/bin/test-backend-ops test -b SYCL0 -o TRI (4/4 passed)

Part of #14909

@github-actions github-actions bot added ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language labels Jan 25, 2026
@RachelMantel
Copy link
Contributor Author

Just checking in to see if you had a chance to look at this.
Happy to make any changes if needed.

Copy link
Contributor

@NeoZhangJianyu NeoZhangJianyu left a comment

Choose a reason for hiding this comment

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

It's great!

Could you update the ops.md (https://github.com/ggml-org/llama.cpp/blob/master/docs/ops.md) for the supported OP?
Please refer to the guide in ops.md.

Thank you!

@github-actions github-actions bot added the documentation Improvements or additions to documentation label Jan 29, 2026
Copy link
Contributor

@NeoZhangJianyu NeoZhangJianyu left a comment

Choose a reason for hiding this comment

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

It's great work!

I will merge the PR after the CI is passed.

@RachelMantel
Copy link
Contributor Author

Fixed ops.md / CSV mismatch for GGML_OP_TRI. Latest commits are in; CI may need a re-run.

@RachelMantel
Copy link
Contributor Author

The remaining CI failures appear unrelated to this PR:

  • ggml-ci-x64-nvidia-vulkan-cm (known flaky Vulkan CI)
  • ubuntu-cpu-cmake (ppc64le) failing with a runner token error

All other checks are passing.

@NeoZhangJianyu NeoZhangJianyu merged commit c7358dd into ggml-org:master Jan 30, 2026
148 of 219 checks passed
CoraBlack pushed a commit to CoraBlack/llama.cpp that referenced this pull request Jan 30, 2026
* sycl: implement GGML_OP_TRI

* docs: update ops.md for SYCL TRI

* docs: regenerate ops.md

* docs: update SYCL support for GGML_OP_TRI
4b1tQu4ntN3k0 pushed a commit to 4b1tQu4ntN3k0/llama.cpp that referenced this pull request Feb 2, 2026
* sycl: implement GGML_OP_TRI

* docs: update ops.md for SYCL TRI

* docs: regenerate ops.md

* docs: update SYCL support for GGML_OP_TRI
shaofeiqi pushed a commit to qualcomm/llama.cpp that referenced this pull request Feb 6, 2026
* sycl: implement GGML_OP_TRI

* docs: update ops.md for SYCL TRI

* docs: regenerate ops.md

* docs: update SYCL support for GGML_OP_TRI
electimon pushed a commit to electimon/llama.cpp that referenced this pull request Mar 19, 2026
* sycl: implement GGML_OP_TRI

* docs: update ops.md for SYCL TRI

* docs: regenerate ops.md

* docs: update SYCL support for GGML_OP_TRI
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation Improvements or additions to documentation ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants