[Qemu-discuss] Very poor compilation performance in armhf chroot vs native

[Aboul Fad]

Hello,

I am not entirely sure if this belongs here, if not it would be greatly
appreciated to redirect me to the best mailing list.

I have setup an ARM chroot on a Xenial system, compiled and installed
qemu-arm-static from source v4.1.0.
I compiled using gcc a simple hello world on the native system (Xenial) and
the same in the ARM chroot using qemu-arm-static.

The performance degradation is large (~15x slower) and am unsure if this is
expected because qemu-arm-static is emulating instructions in user space or
what ?

Thank you for your help.

Results:

> root@ubumini:/tmp# uname -m
> armv7l
> root@ubumini:/tmp# time gcc hello.c -o hello
> real  0m1.032s
> user  0m0.976s
> sys 0m0.052s
> root@ubumini:/tmp# exit
> user@ubumini:/tmp$ uname -m
> x86_64
> user@ubumini:/tmp$ time gcc hello.c -o hello
> real  0m0.073s
> user  0m0.044s
> sys 0m0.012s
>
>