]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
sycl : Improve SYCL doc (#23025)
authorTodd Malsbary <redacted>
Thu, 4 Jun 2026 05:02:54 +0000 (22:02 -0700)
committerGitHub <redacted>
Thu, 4 Jun 2026 05:02:54 +0000 (08:02 +0300)
commitf478f1b6d766683702d3dcba67b6eb02a39d5ec6
tree806cdd6ea77ccf0c456260c1d7a837331af865db
parent94a220cd6745e6e3f8de62870b66fd5b9bc92700
sycl : Improve SYCL doc (#23025)

* Tidy up SYCL doc a bit

- Add explicit links to referenced items
- Fix spelling errors

Signed-off-by: Todd Malsbary <redacted>
* Correct documented default for GGML_SYCL_GRAPH

The default is ON, not OFF:

  $ cmake -LAH -B build | grep GGML_SYCL_GRAPH
  ...
  GGML_SYCL_GRAPH:BOOL=ON

Signed-off-by: Todd Malsbary <redacted>
* Move docker instructions from SYCL.md to docker.md

This makes them directly accesible from the Quick Start section
of the top-level README.md.

Signed-off-by: Todd Malsbary <redacted>
* Refer to intel.Dockerfile for ARGs and their defaults

The defaults are always changing; this avoids accuracy errors
from duplicating the information.

Signed-off-by: Todd Malsbary <redacted>
* Remove mention of Nvidia in SYCL row of backend table

This support was removed in 2026.02 - refer to the SYCL.md News.

Signed-off-by: Todd Malsbary <redacted>
---------

Signed-off-by: Todd Malsbary <redacted>
docs/backend/SYCL.md
docs/docker.md