Re: [PATCH v5 2/3] Skip NaN mode check for soft-float

[Peter Maydell <peter.maydell@linaro.org>]

On Wed, 26 Feb 2025 at 17:03, Aleksandar Rakic
<aleksandar.rakic@htecgroup.com> wrote:
>
> From: Aleksandar Rakic <aleksandar.rakic@htecgroup.com>
>
> Skip NaN mode check for soft-float since NaN mode is irrelevant if an ELF
> binary's FPU mode is soft-float, i.e. it doesn't utilize a FPU.
>
> Cherry-picked 63492a56485f6b755fccf7ad623f7a189bfc79b6
> from https://github.com/MIPS/gnutools-qemu
>
> Signed-off-by: Faraz Shahbazker <fshahbazker@wavecomp.com>
> Signed-off-by: Aleksandar Rakic <aleksandar.rakic@htecgroup.com>
> ---
>  linux-user/mips/cpu_loop.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/linux-user/mips/cpu_loop.c b/linux-user/mips/cpu_loop.c
> index 462387a073..07c1ebe287 100644
> --- a/linux-user/mips/cpu_loop.c
> +++ b/linux-user/mips/cpu_loop.c
> @@ -304,8 +304,10 @@ void target_cpu_copy_regs(CPUArchState *env, struct target_pt_regs *regs)
>      if (env->insn_flags & ISA_NANOMIPS32) {
>          return;
>      }
> -    if (((info->elf_flags & EF_MIPS_NAN2008) != 0) !=
> -        ((env->active_fpu.fcr31 & (1 << FCR31_NAN2008)) != 0)) {
> +    if (info->fp_abi != MIPS_ABI_FP_SOFT
> +        && ((info->elf_flags & EF_MIPS_NAN2008) != 0) !=
> +           ((env->active_fpu.fcr31 & (1 << FCR31_NAN2008)) != 0))
> +      {
>          if ((env->active_fpu.fcr31_rw_bitmask &
>                (1 << FCR31_NAN2008)) == 0) {
>              fprintf(stderr, "ELF binary's NaN mode not supported by CPU\n");

Unless I'm misreading the code, the kernel's MIPS ELF loader
does not look at the fp_abi to decide how to handle the
NAN2008 bit:

https://elixir.bootlin.com/linux/v6.13.6/source/arch/mips/kernel/elf.c#L154

I think QEMU should handle this bit the same way as the kernel's
loader.

thanks
-- PMM