ollama/ml
Shalini Salomi Bodapati 176e9097cf ggml: include PowerPC CPU backend in sync and fix PPC build breakage
This change addresses the PowerPC build breakage noted in the review
discussion of #13427.

Specifically:
- Include ggml-cpu PowerPC backend sources in the rsync filter so they
  are propagated from llama/vendor into Ollama’s vendored ggml snapshot.
- Apply upstream ggml fixes required for PowerPC builds, including the
  vector macro collision fix and conditional POWER11 backend enablement.

Without syncing the PowerPC backend sources, ggml-cpu is incomplete in
Ollama and PPC builds remain broken even with the upstream fixes applied.

Signed-off-by: Shalini Salomi Bodapati <Shalini.Salomi.Bodapati@ibm.com>
2025-12-22 02:28:42 -06:00
..
backend ggml: include PowerPC CPU backend in sync and fix PPC build breakage 2025-12-22 02:28:42 -06:00
nn model: add rnj-1 inference support (#13354) 2025-12-08 16:49:17 -08:00
backend.go ggml: Enable flash attention for vision encoders 2025-12-04 15:19:06 -08:00
device.go CUDA: filter devices on secondary discovery (#13317) 2025-12-03 12:58:16 -08:00
path.go cpu: always ensure LibOllamaPath included (#12890) 2025-10-31 14:37:29 -07:00