linux-kernel - Re: [PATCH 07/16] GFS2: Directory handling

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]

Message-ID: <20060905084334.GA16788@elte.hu>
Date:	Tue, 5 Sep 2006 10:43:34 +0200
From:	Ingo Molnar <mingo@...e.hu>
To:	Steven Whitehouse <swhiteho@...hat.com>
Cc:	Jan Engelhardt <jengelh@...ux01.gwdg.de>,
	linux-kernel@...r.kernel.org,
	Russell Cattelan <cattelan@...hat.com>,
	David Teigland <teigland@...hat.com>, hch@...radead.org
Subject: Re: [PATCH 07/16] GFS2: Directory handling


* Steven Whitehouse <swhiteho@...hat.com> wrote:

> > >+static inline int __gfs2_dirent_find(const struct gfs2_dirent *dent,
> > >+				     const struct qstr *name, int ret)
> > >+{
> > >+	if (dent->de_inum.no_addr != 0 &&
> > >+	    be32_to_cpu(dent->de_hash) == name->hash &&
> > >+	    be16_to_cpu(dent->de_name_len) == name->len &&
> > >+	    memcmp((char *)(dent+1), name->name, name->len) == 0)
> > 
> > Nocast.
> > 
> ok

actually, sizeof(*dent) != 1, so how can a non-casted memcmp be correct 
here?

> > >+	if ((char *)cur + cur_rec_len >= bh_end) {
> > >+		if ((char *)cur + cur_rec_len > bh_end) {
> > >+			gfs2_consist_inode(dip);
> > >+			return -EIO;
> > >+		}
> > >+		return -ENOENT;
> > >+	}
> > 
> > if((char *)cur + cur_rec_len > bh_end) {
> > 	gfs2_consist_inode(dip);
> > 	return -EIO;
> > } else if((char *)cur + cur_rec_len == bh_end)
> > 	return -ENOENT;
> > 
> ok

this one is not OK! Firstly, Jan, and i mentioned this before, please 
stop using 'if(', it is highly inconsistent and against basic taste. We 
only use this construct for function calls (and macros), not for C 
statements.

Secondly, whenever we have curly braces in the first block, we tend to 
do it in the second block too, for easier parsing. I.e.:

	if ((char *)cur + cur_rec_len > bh_end) {
		gfs2_consist_inode(dip);
		return -EIO;
	} else {
		if ((char *)cur + cur_rec_len == bh_end)
			return -ENOENT;
	}

Thirdly, the original code was quite fine as-is! What's the point of 
introducing random perturbations like this? It is an open invitation for 
the introduction of bugs... So unless there is a clear style reason to 
do a change, i'd suggest to not touch the code.

	Ingo
-
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/