Re: [PATCH] riscv: Return -EFAULT if copy_to_user() failed in signal.c

[Ben Dooks <ben.dooks@codethink.co.uk>]

On 02/03/2021 07:28, Tiezhu Yang wrote:
> copy_to_user() returns the amount left to copy, it should return -EFAULT
> if copy to user failed.

This looks technically correct, but the caller (only one)
will check for non-zero and will covert that to -EFAULT
in setup_rt_frame().

I expect if this change is done, it also needs to be done
for the callers too and there's a few others than assume
!=0 is an error.

I think it would be easier to define save_fp_state() to
return non-zero on error and note it does not return an
error code. It may be worth exiting the functio nif
the first __copy_to_user fails?

Note: setup_rt_frame -> setup_sigcontext -> save_fp_frame

> 
> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
> ---
>   arch/riscv/kernel/signal.c | 6 +++++-
>   1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/riscv/kernel/signal.c b/arch/riscv/kernel/signal.c
> index 65942b3..2238fc5 100644
> --- a/arch/riscv/kernel/signal.c
> +++ b/arch/riscv/kernel/signal.c
> @@ -67,7 +67,7 @@ static long save_fp_state(struct pt_regs *regs,
>   	fstate_save(current, regs);
>   	err = __copy_to_user(state, &current->thread.fstate, sizeof(*state));
>   	if (unlikely(err))
> -		return err;
> +		return -EFAULT;
>   
>   	/* We support no other extension state at this time. */
>   	for (i = 0; i < ARRAY_SIZE(sc_fpregs->q.reserved); i++) {
> @@ -140,8 +140,12 @@ static long setup_sigcontext(struct rt_sigframe __user *frame,
>   {
>   	struct sigcontext __user *sc = &frame->uc.uc_mcontext;
>   	long err;
> +
>   	/* sc_regs is structured the same as the start of pt_regs */
>   	err = __copy_to_user(&sc->sc_regs, regs, sizeof(sc->sc_regs));
> +	if (unlikely(err))
> +		return -EFAULT;
> +
>   	/* Save the floating-point state. */
>   	if (has_fpu)
>   		err |= save_fp_state(regs, &sc->sc_fpregs);
> 


-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius

https://www.codethink.co.uk/privacy.html