Commit 55c509d
authored
ggml : refactor llamafile_sgemm PPC code (ggml-org#14673)
Remove un-necessary templates from class definition and packing functions
Reduce deeply nested conditionals, if-else switching in mnapck function
Replace repetitive code with inline functions in Packing functions
2 ~ 7% improvement in Q8 Model
15 ~ 50% improvement in Q4 Model
Signed-off-by: Shalini Salomi Bodapati <[email protected]>1 parent 9c9e4fc commit 55c509d
1 file changed
+343
-1094
lines changed
0 commit comments