Re: [PATCH] bad strlcpy conversion breaks toshiba_acpi

From: Petr Vandrovec (vandrove@vc.cvut.cz)
Date: Fri Jul 25 2003 - 11:15:10 EST


On Fri, Jul 25, 2003 at 10:46:38AM -0400, John Belmonte wrote:
> Please revert the following item from Ben Collins' "drivers/* strlcpy
> conversions" patch dated 2003-May-26.
>
> The strlcpy function requires a zero-terminated string, which is not a
> valid assumption for the code in question. I suggest that Ben review
> all such modifications he made to the kernel for similar errors. It's
> quite annoying to have someone add bugs to your driver while you're not
> looking. Either notify the maintainer of your patch or don't make mistakes.

Nope. Kernel strlcpy implementation is crap and I do not believe that there
is single place in the kernel which can live with current implementation.

Take a look at ftp://ftp.openbsd.org/pub/OpenBSD/src/lib/libc/string/strlcpy.c
or at http://www.courtesan.com/todd/papers/strlcpy.html - it copies
at most size-1 characters. Nothing about characters beyond specified size
in the article.

Kernel should use strnlen() to get string length, if coding loop like
OpenBSD does is unacceptable.
                                                Best regards,
                                                        Petr Vandrovec

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



This archive was generated by hypermail 2b29 : Thu Jul 31 2003 - 22:00:26 EST