when using "-march=native" option with gcc 4.8.x & 4.9.x, VIA nano CPU gets detected as "core2" instead of "x86-64" CPUs, then fails to compile. Using "-march=x86-64" succeed but does not uses the full instruction set of the CPU. Thus one has to compile using: CFLAGS="-march=x86-64 -mcx16 -msahf -mfxsr --param l1-cache-size=64 --param l1-cache-line-size=64 --param l2-cache-size=1024 -mmmx -msse -msse2 -msse3 -mssse3" in order to get optimized compilated code, which is quite painful. Reproducible: Always Steps to Reproduce: 1. > cat /proc/cpuinfo processor : 0 vendor_id : CentaurHauls cpu family : 6 model : 15 model name : VIA Nano processor U2250 (1.6GHz Capable) stepping : 3 cpu MHz : 800.000 cache size : 1024 KB fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat clflush acpi mmx fxsr sse sse2 ss tm pbe syscall nx lm constant_tsc rep_good nopl pni monitor vmx est tm2 ssse3 cx16 xtpr rng rng_en ace ace_en ace2 phe phe_en lahf_lm bugs : bogomips : 3191.71 clflush size : 64 cache_alignment : 128 address sizes : 36 bits physical, 48 bits virtual power management: 2. > echo 'int main(){return 0;}' > test.c && gcc -march=native -O2 -pipe test.c -o test && rm test.c test Actual Results: Compilation fails with the following error message and informations: [...] test.c:1:0: error: CPU you selected does not support x86-64 instruction set int main(){return 0;} ^ Expected Results: Compilation to succeed. running gcc with "-v -Q" options gives more informations about the problem: gcc selects "-march=core2" and "-mtune=i386": GNU C (GCC) version 4.8.4 (x86_64-unknown-linux-gnu) compiled by GNU C version 4.8.4, GMP version 5.1.3, MPFR version 3.1.2-p10, MPC version 1.0.2 GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072 options passed: -v -iprefix /usr/local/src/gcc-4.8.4/host-x86_64-unknown-linux-gnu/gcc/../lib/gcc/x86_64-unknown-linux-gnu/4.8.4/ test.c -march=core2 -mcx16 -msahf -mno-movbe -mno-aes -mno-pclmul -mno-popcnt -mno-abm -mno-lwp -mno-fma -mno-fma4 -mno-xop -mno-bmi -mno-bmi2 -mno-tbm -mno-avx -mno-avx2 -mno-sse4.2 -mno-sse4.1 -mno-lzcnt -mno-rtm -mno-hle -mno-rdrnd -mno-f16c -mno-fsgsbase -mno-rdseed -mno-prfchw -mno-adx -mfxsr -mno-xsave -mno-xsaveopt --param l1-cache-size=64 --param l1-cache-line-size=64 --param l2-cache-size=1024 -mtune=i386 -O2 -fno-use-linker-plugin options enabled: -faggressive-loop-optimizations -falign-labels -fasynchronous-unwind-tables -fauto-inc-dec -fbranch-count-reg -fcaller-saves -fcombine-stack-adjustments -fcommon -fcompare-elim -fcprop-registers -fcrossjumping -fcse-follow-jumps -fdefer-pop -fdelete-null-pointer-checks -fdevirtualize -fdwarf2-cfi-asm -fearly-inlining -feliminate-unused-debug-types -fexpensive-optimizations -fforward-propagate -ffunction-cse -fgcse -fgcse-lm -fgnu-runtime -fgnu-unique -fguess-branch-probability -fhoist-adjacent-loads -fident -fif-conversion -fif-conversion2 -findirect-inlining -finline -finline-atomics -finline-functions-called-once -finline-small-functions -fipa-cp -fipa-profile -fipa-pure-const -fipa-reference -fipa-sra -fira-hoist-pressure -fira-share-save-slots -fira-share-spill-slots -fivopts -fkeep-static-consts -fleading-underscore -fmath-errno -fmerge-constants -fmerge-debug-strings -fmove-loop-invariants -fomit-frame-pointer -foptimize-register-move -foptimize-sibling-calls -foptimize-strlen -fpartial-inlining -fpeephole -fpeephole2 -fprefetch-loop-arrays -free -freg-struct-return -fregmove -freorder-blocks -freorder-functions -frerun-cse-after-loop -fsched-critical-path-heuristic -fsched-dep-count-heuristic -fsched-group-heuristic -fsched-interblock -fsched-last-insn-heuristic -fsched-rank-heuristic -fsched-spec -fsched-spec-insn-heuristic -fsched-stalled-insns-dep -fshow-column -fshrink-wrap -fsigned-zeros -fsplit-ivs-in-unroller -fsplit-wide-types -fstrict-aliasing -fstrict-overflow -fstrict-volatile-bitfields -fsync-libcalls -fthread-jumps -ftoplevel-reorder -ftrapping-math -ftree-bit-ccp -ftree-builtin-call-dce -ftree-ccp -ftree-ch -ftree-coalesce-vars -ftree-copy-prop -ftree-copyrename -ftree-cselim -ftree-dce -ftree-dominator-opts -ftree-dse -ftree-forwprop -ftree-fre -ftree-loop-if-convert -ftree-loop-im -ftree-loop-ivcanon -ftree-loop-optimize -ftree-parallelize-loops= -ftree-phiprop -ftree-pre -ftree-pta -ftree-reassoc -ftree-scev-cprop -ftree-sink -ftree-slp-vectorize -ftree-slsr -ftree-sra -ftree-switch-conversion -ftree-tail-merge -ftree-ter -ftree-vect-loop-version -ftree-vrp -funit-at-a-time -funwind-tables -fvar-tracking -fvar-tracking-assignments -fzero-initialized-in-bss -m128bit-long-double -m64 -m80387 -maccumulate-outgoing-args -malign-stringops -mcx16 -mfancy-math-387 -mfp-ret-in-387 -mfxsr -mglibc -mieee-fp -mlong-double-80 -mmmx -mno-sse4 -mpush-args -mred-zone -msahf -msse -msse2 -msse3 -mssse3 -mtls-direct-seg-refs where we expect to get "-march=x86-64": (here's the output of a patched version) GNU C (GCC) version 4.8.4 (x86_64-unknown-linux-gnu) compiled by GNU C version 4.8.4, GMP version 5.1.3, MPFR version 3.1.2-p10, MPC version 1.0.2 GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072 options passed: -v -iprefix /usr/local/src/gcc-4.8.4/host-x86_64-unknown-linux-gnu/gcc/../lib/gcc/x86_64-unknown-linux-gnu/4.8.4/ test.c -march=x86-64 -mcx16 -msahf -mno-movbe -mno-aes -mno-pclmul -mno-popcnt -mno-abm -mno-lwp -mno-fma -mno-fma4 -mno-xop -mno-bmi -mno-bmi2 -mno-tbm -mno-avx -mno-avx2 -msse3 -mssse3 -mno-sse4.2 -mno-sse4.1 -mno-lzcnt -mno-rtm -mno-hle -mno-rdrnd -mno-f16c -mno-fsgsbase -mno-rdseed -mno-prfchw -mno-adx -mfxsr -mno-xsave -mno-xsaveopt --param l1-cache-size=64 --param l1-cache-line-size=64 --param l2-cache-size=1024 -mtune=generic -O2 -fno-use-linker-plugin options enabled: -faggressive-loop-optimizations -falign-labels -fasynchronous-unwind-tables -fauto-inc-dec -fbranch-count-reg -fcaller-saves -fcombine-stack-adjustments -fcommon -fcompare-elim -fcprop-registers -fcrossjumping -fcse-follow-jumps -fdefer-pop -fdelete-null-pointer-checks -fdevirtualize -fdwarf2-cfi-asm -fearly-inlining -feliminate-unused-debug-types -fexpensive-optimizations -fforward-propagate -ffunction-cse -fgcse -fgcse-lm -fgnu-runtime -fgnu-unique -fguess-branch-probability -fhoist-adjacent-loads -fident -fif-conversion -fif-conversion2 -findirect-inlining -finline -finline-atomics -finline-functions-called-once -finline-small-functions -fipa-cp -fipa-profile -fipa-pure-const -fipa-reference -fipa-sra -fira-hoist-pressure -fira-share-save-slots -fira-share-spill-slots -fivopts -fkeep-static-consts -fleading-underscore -fmath-errno -fmerge-constants -fmerge-debug-strings -fmove-loop-invariants -fomit-frame-pointer -foptimize-register-move -foptimize-sibling-calls -foptimize-strlen -fpartial-inlining -fpeephole -fpeephole2 -fprefetch-loop-arrays -free -freg-struct-return -fregmove -freorder-blocks -freorder-functions -frerun-cse-after-loop -fsched-critical-path-heuristic -fsched-dep-count-heuristic -fsched-group-heuristic -fsched-interblock -fsched-last-insn-heuristic -fsched-rank-heuristic -fsched-spec -fsched-spec-insn-heuristic -fsched-stalled-insns-dep -fschedule-insns2 -fshow-column -fshrink-wrap -fsigned-zeros -fsplit-ivs-in-unroller -fsplit-wide-types -fstrict-aliasing -fstrict-overflow -fstrict-volatile-bitfields -fsync-libcalls -fthread-jumps -ftoplevel-reorder -ftrapping-math -ftree-bit-ccp -ftree-builtin-call-dce -ftree-ccp -ftree-ch -ftree-coalesce-vars -ftree-copy-prop -ftree-copyrename -ftree-cselim -ftree-dce -ftree-dominator-opts -ftree-dse -ftree-forwprop -ftree-fre -ftree-loop-if-convert -ftree-loop-im -ftree-loop-ivcanon -ftree-loop-optimize -ftree-parallelize-loops= -ftree-phiprop -ftree-pre -ftree-pta -ftree-reassoc -ftree-scev-cprop -ftree-sink -ftree-slp-vectorize -ftree-slsr -ftree-sra -ftree-switch-conversion -ftree-tail-merge -ftree-ter -ftree-vect-loop-version -ftree-vrp -funit-at-a-time -funwind-tables -fvar-tracking -fvar-tracking-assignments -fzero-initialized-in-bss -m128bit-long-double -m64 -m80387 -maccumulate-outgoing-args -malign-stringops -mcx16 -mfancy-math-387 -mfp-ret-in-387 -mfxsr -mglibc -mieee-fp -mlong-double-80 -mmmx -mno-sse4 -mpush-args -mred-zone -msahf -msse -msse2 -msse3 -mssse3 -mtls-direct-seg-refs The following patches adds VIA nano CPU detection for gcc 4.8.x and 4.9.x. The 4.9.x patch also applies to gcc 5.2 but, for any reason, I cannot get gcc enter the CPU detection routine, as if "-march=native", "-mtune=native" & "-mcpu=native" options are discarded somewhere. However, gcc 5.2 compiles for VIA nano CPU while using "-march=native" but lacks to enable SSE3 & SSSE3 support. You'll also find the sys-devel/gcc overlay directory I created for my tests. As a test, I recompiled the whole distribution using patched gcc 4.8.4 with: emerge --verbose --empty @system emerge --verbose --empty @world (in a newly opened shell) It compiled and runs fine. gcc versions 4.8.5 & 4.9.3 where only checked from their output on a test program. Please note that those patches should fix support for all known VIA nano but won't help for VIA Eden support, which is likely to be broken too as far as I can see. I also opened a bug on gcc bug tracker, for information: <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67310> Please consider applying thoses patches.
Created attachment 409920 [details, diff] patch to properly detect VIA nano CPU for gcc 4.8.x versions
Created attachment 409922 [details, diff] patch to properly detect VIA nano CPU for gcc 4.9+ versions
Created attachment 409928 [details] Overlay for sys-devel/gcc with VIA nano CPU patches included for gcc 4.8.4, 4.8.5, 4.9.3 and 5.2
The following patch has been commited by the gcc maintainers team and allows -march=native option to be usable on any x86_64 VIA CPU and applies to gcc >= 4.9. Refer to: https://gcc.gnu.org/ml/gcc-patches/2016-06/msg00044.html Please consider adding this patch to gentoo ones, until it's commited in all gcc branches.
Created attachment 436152 [details, diff] Patch commited to gcc mainstream to fix VIA Nano issue with -march=native
i've added the backport to gcc-5.4.0: https://sources.gentoo.org/gentoo/src/patchsets/gcc/5.4.0/gentoo/71_all_gcc-5-march-native-pr67310.patch i'll close the bug once it's queued for 4.9 too
fix is in the 4.9.4 release
