Re: [PATCH 1/2] Introduce copy_user_handle_tail routine

[Andi Kleen]

> +unsigned long
> +copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest)
> +{
> +	char c;
> +	unsigned zero_len;
> +
> +	for (; len; --len) {
> +		if (__get_user_nocheck(c, from++, sizeof(char)))
> +			break;
> +		if (__put_user_nocheck(c, to++, sizeof(char)))

get/put user are macros and it's normally not a good idea to use ++ in macro
arguments because they might expand multiple times.

sizeof(char) is always 1

Also hopefully there's no sign extension anywhere with the signed char

Overall you could write it much simpler with a rep ; movs I think,
like traditional linux did.

> +			break;
> +	}
> +
> +	for (c = 0, zero_len = len; zerorest && zero_len; --zero_len)
> +		if (__put_user_nocheck(c, to++, sizeof(char)))
> +			break;

Similar problem with ++

If zerorest is ever 0 then retesting it on every iteration seems somewhat dumb.

I think a simple memset would be actually ok, i don't think we ever zero
anything that faults. That would be obviously racy anyways. If the zero
are supposed to override something  then a racing user thread could always
catch it.

-Andi
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/