linux-kernel - Re: sys_readlink() with empty filename

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] [day] [month] [year] [list]

Message-ID: <20110802154746.GC29456@e102144-lin.cambridge.arm.com>
Date:	Tue, 2 Aug 2011 16:47:47 +0100
From:	Will Deacon <will.deacon@....com>
To:	"viro@...iv.linux.org.uk" <viro@...iv.linux.org.uk>
Cc:	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: sys_readlink() with empty filename

On Thu, Jul 28, 2011 at 07:21:29PM +0100, Will Deacon wrote:
> Hi Al,
> 
> In commit 65cfc672 ("readlinkat(), fchownat() and fstatat() with empty
> relative pathnames") you changed the behaviour of the readlink syscall so
> that LOOKUP_EMPTY is passed in the lookup flags. This causes the syscall to
> return -EINVAL rather than -ENOENT if the filename is empty.
> 
> Looking at the man page for readlink, it's not obvious to me what the
> correct behaviour should be. However, this change has also been noticed by
> the gnulib guys because it causes the coreutils tests to fail:
> 
> http://comments.gmane.org/gmane.comp.lib.gnulib.bugs/25791
> 
> I noticed it as a regression in LTP.
> 
> So is this behaviour unspecified by POSIX and are the tests all broken or
> should it be fixed in the kernel / C library?

Any comments on this? I'm reluctant to start changing tests if this is just
an oversight in the kernel.

Thanks,

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