linux-kernel - Re: [PATCH] isofs: fix undefined behavior in iso

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 9 Jul 2023 12:41:25 +0100
From:   Matthew Wilcox <willy@...radead.org>
To:     Linke Li <lilinke99@...mail.com>
Cc:     linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
        Jan Kara <jack@...e.cz>, Linke Li <lilinke99@...il.com>
Subject: Re: [PATCH] isofs: fix undefined behavior in iso_date()

On Sun, Jul 09, 2023 at 02:42:55PM +0800, Linke Li wrote:
> From: Linke Li <lilinke99@...il.com>
> 
> Fix undefined behavior in the code by properly handling the left shift operaion.
> Instead of left-shifting a negative value, explicitly cast -1 to an unsigned int
> before the shift. This ensures well defined behavior and resolves any potential
> issues.

This certainly fixes the problem, but wouldn't it be easier to get the
compiler to do the work for us?

#include <stdio.h>

int f(unsigned char *p)
{
	return (signed char)p[0];
}

int main(void)
{
	unsigned char x = 0xa5;

	printf("%d\n", f(&x));

	return 0;
}

prints -91.