]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 4f02636)
ggml : optimize for ppc64le using VSX intrinsics (ggml/784)
authorHong Bo PENG <redacted>
Sun, 12 May 2024 09:17:18 +0000 (17:17 +0800)
committerGeorgi Gerganov <redacted>
Tue, 14 May 2024 16:08:09 +0000 (19:08 +0300)
commit0d26d8ccd8caebab75af697c0275f599075fdacf
treeb5dc95b6829b3130ede9e2b5407b8703d49556cdtree
parent4f0263633b40e94e8b69fd6e7e4395cfedfd5c12commit | diff
ggml : optimize for ppc64le using VSX intrinsics (ggml/784)

* optimize for ppc64le using VSX intrinsics

* 1. code clean up by removing comments about overflow concern.

2. fix typo in suffix of scaling.

* Continue to fix typo in suffix of scaling for QK_K <> 256

---------

Co-authored-by: Georgi Gerganov <redacted>
ggml-quants.c diff | blob | history
Packaging of ggml-org/llama.cpp