From: Aaron Teo Date: Fri, 25 Jul 2025 17:09:03 +0000 (+0800) Subject: ggml-cpu : disable GGML_NNPA by default due to instability (llama/14880) X-Git-Tag: upstream/0.0.2404~38 X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=85a7bc59332dfc44e30a105650fbb9a45e29beed;p=pkg%2Fggml%2Fsources%2Fggml ggml-cpu : disable GGML_NNPA by default due to instability (llama/14880) * docs: update s390x document for sentencepiece Signed-off-by: Aaron Teo (cherry picked from commit e086c5e3a7ab3463d8e0906efcfa39352db0a48d) * docs: update huggingface links + reword Signed-off-by: Aaron Teo (cherry picked from commit 8410b085ea8c46e22be38266147a1e94757ef108) * ggml-cpu: disable ggml-nnpa compile flag by default fixes #14877 Signed-off-by: Aaron Teo (cherry picked from commit 412f4c7c88894b8f55846b4719c76892a23cfe09) * docs: update s390x build docs to reflect nnpa disable Signed-off-by: Aaron Teo (cherry picked from commit c1eeae1d0c2edc74ab9fbeff2707b0d357cf0b4d) --------- Signed-off-by: Aaron Teo --- diff --git a/CMakeLists.txt b/CMakeLists.txt index 8ca1053c..20467c54 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -131,7 +131,7 @@ option(GGML_RVV "ggml: enable rvv" ON) option(GGML_RV_ZFH "ggml: enable riscv zfh" OFF) option(GGML_XTHEADVECTOR "ggml: enable xtheadvector" OFF) option(GGML_VXE "ggml: enable vxe" ON) -option(GGML_NNPA "ggml: enable nnpa" ON) +option(GGML_NNPA "ggml: enable nnpa" OFF) # temp disabled by default, see: https://github.com/ggml-org/llama.cpp/issues/14877 option(GGML_CPU_ALL_VARIANTS "ggml: build all variants of the CPU backend (requires GGML_BACKEND_DL)" OFF) set(GGML_CPU_ARM_ARCH "" CACHE STRING "ggml: CPU architecture for ARM") diff --git a/src/ggml-cpu/CMakeLists.txt b/src/ggml-cpu/CMakeLists.txt index 2cc42d4b..f188d163 100644 --- a/src/ggml-cpu/CMakeLists.txt +++ b/src/ggml-cpu/CMakeLists.txt @@ -458,6 +458,7 @@ function(ggml_add_cpu_backend_variant_impl tag_name) list(APPEND ARCH_FLAGS -march=z16) elseif (${S390X_M} MATCHES "9175|9176") # NOTE: Only available from GCC 15.1.0 onwards. Any z17 machine with compile issues must first verify their GCC version. + # binutils must also be updated to the latest for the -march=z17 flag to work. Otherwise, use -march=arch15. message(STATUS "z17 target") list(APPEND ARCH_FLAGS -march=z17) else()