From: Steven Whitehouse <swhiteho@redhat.com>
To: Jan Engelhardt <jengelh@linux01.gwdg.de>
Cc: linux-kernel@vger.kernel.org,
Russell Cattelan <cattelan@redhat.com>,
David Teigland <teigland@redhat.com>, Ingo Molnar <mingo@elte.hu>,
hch@infradead.org
Subject: Re: [PATCH 07/16] GFS2: Directory handling
Date: Tue, 05 Sep 2006 09:44:14 +0100 [thread overview]
Message-ID: <1157445854.3384.965.camel@quoit.chygwyn.com> (raw)
In-Reply-To: <Pine.LNX.4.61.0609041314470.21005@yvahk01.tjqt.qr>
Hi,
On Mon, 2006-09-04 at 13:35 +0200, Jan Engelhardt wrote:
> >+
> >+ return copied;
> >+fail:
> >+ return (copied) ? copied : error;
> >+}
> >+
> >+typedef int (*gfs2_dscan_t)(const struct gfs2_dirent *dent,
> >+ const struct qstr *name,
> >+ void *opaque);
>
> Collect all the typedefs above all functions (applies to all .c files).
>
ok
> >+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
> >+static struct gfs2_dirent *gfs2_dirent_scan(struct inode *inode,
[lines snipped]
> >+ offset = ret;
> >+ prev = NULL;
> >+ dent = (struct gfs2_dirent *)(buf + offset);
>
> Nocast.
>
ok
> >+ dent = (struct gfs2_dirent *)(buf + offset);
>
>
> >+ 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
> >+ tmp = (struct gfs2_dirent *)((char *)cur + cur_rec_len);
> >+
> >+ if ((char *)tmp + be16_to_cpu(tmp->de_rec_len) > bh_end) {
>
> Aah, this makes my eyes hurt. Though, it is probably a task not too short to
> think of something that would do without casts.
>
Well I've had a crack at this. I've got it down to one cast. Let me know
if you think thats ok now... I managed to factor out some common code at
the same time.
> >+ leaf->lf_depth = cpu_to_be16(depth);
> >+ leaf->lf_entries = cpu_to_be16(0);
>
> 0 is said to be portable across all CPUs, therefore
>
> leaf->lf_entries = 0;
>
> should suffice.
>
> >+ leaf->lf_next = cpu_to_be64(0);
>
ok, both now fixed.
>
> >+ for (x = sdp->sd_hash_ptrs; x--; from++) {
> >+ *to++ = *from; /* No endianess worries */
> >+ *to++ = *from;
> >+ }
>
> Add /* Hakuna Matata */ and there will never be worries :-)
>
:-) I had to use google to understand the reference :-)
> >+static int compare_dents(const void *a, const void *b)
> >+{
> >+ struct gfs2_dirent *dent_a, *dent_b;
> >+ uint32_t hash_a, hash_b;
> >+ int ret = 0;
> >+
> >+ dent_a = *(struct gfs2_dirent **)a;
>
> But in this function you can have it const! Or not?
>
Yes, and now all converted to const
[lines snipped]
> ret = memcmp((char *)(dent_a + 1),
> >+ (char *)(dent_b + 1),
> >+ len_a);
>
> Nocast.
>
ok
> >+ error = filldir(opaque, (char *)(dent + 1),
>
> If you case, then cast it directly, in this case, filldir takes "const char *"
> as 2nd type.
>
ok
> >+ larr = vmalloc((leaves + entries) * sizeof(void*));
> ^
> Space, to go in line with all the other casts.
>
fixed
> >+static inline uint32_t gfs2_disk_hash(const char *data, int len)
> >+{
> >+ return crc32_le(0xFFFFFFFF, data, len) ^ 0xFFFFFFFF;
> >+}
>
> Mind using (uint32_t)-1 or (uint32_t)~0 for that?
>
~0 is the variant I've chosen.
> >+ memcpy((char*)(dent+1), name->name, name->len);
>
> Last but not least, nocast.
>
ok.
The patch is:
http://www.kernel.org/git/?p=linux/kernel/git/steve/gfs2-2.6.git;a=commitdiff;h=2bdbc5d73961c040fdc9b30d985fab3047d697a0
Thanks for doing all this review work, btw. I'll be working on your more
recent emails today,
Steve.
next prev parent reply other threads:[~2006-09-05 8:38 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-08-31 13:34 Steven Whitehouse
2006-09-04 11:35 ` Jan Engelhardt
2006-09-05 8:44 ` Steven Whitehouse [this message]
2006-09-05 8:43 ` Ingo Molnar
2006-09-05 8:58 ` Andreas Schwab
2006-09-05 10:58 ` Ingo Molnar
2006-09-05 9:22 ` Jan Engelhardt
2006-09-05 11:19 ` Ingo Molnar
2006-09-05 11:22 ` Ingo Molnar
2006-09-05 14:02 ` Steven Whitehouse
-- strict thread matches above, loose matches on Subject: below --
2006-04-21 16:16 Steven Whitehouse
2006-04-21 16:16 ` Christoph Hellwig
2006-04-24 9:16 ` Steven Whitehouse
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1157445854.3384.965.camel@quoit.chygwyn.com \
--to=swhiteho@redhat.com \
--cc=cattelan@redhat.com \
--cc=hch@infradead.org \
--cc=jengelh@linux01.gwdg.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=teigland@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®