
Code: Select all
-march=x86-64
Code: Select all
gcc -march=native -E -v - </dev/null 2>&1 | grep cc1

Apple M1 Maxalamahant wrote:What kind of virtualized cpu does Parallels assign to your vm?
Safest is to omit -march= or set it toCode: Select all
-march=x86-64

# gcc -march=native -E -v - </dev/null 2>&1 | grep cc1alamahant wrote:Plz check it withCode: Select all
gcc -march=native -E -v - </dev/null 2>&1 | grep cc1



Code: Select all
-mlittle-endian -mabi=lp64 -march=armv8-a+crypto+crc+lse+rcpc+rdma+dotprod+fp16fml+sb+ssbs+flagm+pauth -dumpbase -Code: Select all
-march=x86-64 

some packages can't build with -marth=native. for example - www-client/chromium.NeddySeagoon wrote:leonchik1976,
The M1 is an aarch64 CPU. That's 64 bit ARM.
Theis what gcc uses for -march=native.Code: Select all
-mlittle-endian -mabi=lp64 -march=armv8-a+crypto+crc+lse+rcpc+rdma+dotprod+fp16fml+sb+ssbs+flagm+pauth -dumpbase -
Don't useas gcc will try to output code for amd64 and fail. It will be harmless though.Code: Select all
-march=x86-64
What problems are you having. gcc should be OK.
clang needs version 15 to understand --march=native on your CPU


Code: Select all
-march=armv8-a+crypto+crc+lse+rcpc+rdma+dotprod+fp16fml+sb+ssbs+flagm+pauth
NeddySeagoon, after replacing -march=native with -march=armv8-a+crypto+crc+lse+rcpc+rdma+dotprod+fp16fml+sb+ssbs+flagm+pauth,NeddySeagoon wrote:leonchik1976,
Lets test withwhich is what gcc detected.Code: Select all
-march=armv8-a+crypto+crc+lse+rcpc+rdma+dotprod+fp16fml+sb+ssbs+flagm+pauth
That will be used by both gcc and clang.


