"march=native" flag not working in libgccjit

"march=native" flag not working in libgccjit

[Li, Yicheng]

[-- Attachment #1: Type: text/plain, Size: 1003 bytes --]

Hi,

In libjitgcc, I created a function and set flags “march=native” with gcc_jit_context_add_command_line_option( ctxt, “-march=native”.
However, the error is below:

libgccjit.so: error: bad value ‘native’ for ‘-march=’ switch

libgccjit.so: note: valid arguments to ‘-march=’ switch are: nocona core2 nehalem corei7 westmere sandybridge corei7-avx ivybridge core-avx-i haswell core-avx2 broadwell skylake skylake-avx512 cannonlake icelake-client rocketlake icelake-server cascadelake tigerlake cooperlake sapphirerapids emeraldrapids alderlake raptorlake meteorlake graniterapids bonnell atom silvermont slm goldmont goldmont-plus tremont sierraforest grandridge knl knm x86-64 x86-64-v2 x86-64-v3 x86-64-v4 eden-x2 nano nano-1000 nano-2000 nano-3000 nano-x2 eden-x4 nano-x4 lujiazui k8 k8-sse3 opteron opteron-sse3 athlon64 athlon64-sse3 athlon-fx amdfam10 barcelona bdver1 bdver2 bdver3 bdver4 znver1 znver2 znver3 znver4 btver1 btver2 native


Thank you
Yicheng Li

Re: "march=native" flag not working in libgccjit

[David Malcolm]

On Mon, 2023-05-15 at 20:07 +0000, Li, Yicheng via Jit wrote:
> Hi,
> 
> In libjitgcc, I created a function and set flags “march=native” with
> gcc_jit_context_add_command_line_option( ctxt, “-march=native”.
> However, the error is below:
> 
> libgccjit.so: error: bad value ‘native’ for ‘-march=’ switch
> 
> libgccjit.so: note: valid arguments to ‘-march=’ switch are: nocona
> core2 nehalem corei7 westmere sandybridge corei7-avx ivybridge core-
> avx-i haswell core-avx2 broadwell skylake skylake-avx512 cannonlake
> icelake-client rocketlake icelake-server cascadelake tigerlake
> cooperlake sapphirerapids emeraldrapids alderlake raptorlake
> meteorlake graniterapids bonnell atom silvermont slm goldmont
> goldmont-plus tremont sierraforest grandridge knl knm x86-64 x86-64-
> v2 x86-64-v3 x86-64-v4 eden-x2 nano nano-1000 nano-2000 nano-3000
> nano-x2 eden-x4 nano-x4 lujiazui k8 k8-sse3 opteron opteron-sse3
> athlon64 athlon64-sse3 athlon-fx amdfam10 barcelona bdver1 bdver2
> bdver3 bdver4 znver1 znver2 znver3 znver4 btver1 btver2 native

Sorry about this; this is
  https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66594
and
  https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107230
Unfortunately it's tricky to implement this within libgccjit due to the
differences between it and the normal command-line way of invoking gcc.

If someone wants to try fixing this, that would be great.

Dave