linux-kernel - Re: [PATCH v2 09/10] selftests/nolibc: test return value of read() in test

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:   Tue, 1 Aug 2023 08:59:17 +0200
From:   Willy Tarreau <w@....eu>
To:     Thomas Weißschuh <linux@...ssschuh.net>
Cc:     Shuah Khan <shuah@...nel.org>, linux-kselftest@...r.kernel.org,
        linux-kernel@...r.kernel.org, Yuan Tan <tanyuan@...ylab.org>,
        Zhangjin Wu <falcon@...ylab.org>
Subject: Re: [PATCH v2 09/10] selftests/nolibc: test return value of read()
 in test_vfprintf

On Tue, Aug 01, 2023 at 07:30:16AM +0200, Thomas Weißschuh wrote:
> If read() fails and returns -1 buf would be accessed out of bounds.
> 
> Signed-off-by: Thomas Weißschuh <linux@...ssschuh.net>
> ---
>  tools/testing/selftests/nolibc/nolibc-test.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/tools/testing/selftests/nolibc/nolibc-test.c b/tools/testing/selftests/nolibc/nolibc-test.c
> index 82714051c72f..a334f8450a34 100644
> --- a/tools/testing/selftests/nolibc/nolibc-test.c
> +++ b/tools/testing/selftests/nolibc/nolibc-test.c
> @@ -1031,6 +1031,12 @@ static int expect_vfprintf(int llen, int c, const char *expected, const char *fm
>  	lseek(fd, 0, SEEK_SET);
>  
>  	r = read(fd, buf, sizeof(buf) - 1);
> +	if (r == -1) {
> +		llen += printf(" read() = %s", errorname(errno));
> +		result(llen, FAIL);
> +		return 1;
> +	}
> +
>  	buf[r] = '\0';

In fact given the nature of this file (test if we properly implemented
our syscalls), I think that a more conservative approach is deserved
because if we messed up on read() we can have anything on return and we
don't want to trust that. As such I would suggest that we declare r as
ssize_t and verify that it's neither negative nor larger than
sizeof(buf)-1, which becomes:

        if ((size_t)r >= sizeof(buf)) {
            ... fail ...
        }

You'll also have to turn w to ssize_t then due to the test later BTW.

Willy