From: Liu Yuan <namei.unix@gmail.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: Kevin Wolf <kwolf@redhat.com>,
MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>,
sheepdog@lists.wpkg.org, qemu-devel@nongnu.org,
Stefan Hajnoczi <stefanha@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v2 2/2] sheepdog: implement .bdrv_co_is_allocated()
Date: Mon, 22 Apr 2013 20:10:58 +0800 [thread overview]
Message-ID: <517528D2.1040409@gmail.com> (raw)
In-Reply-To: <20130422120007.GB21317@stefanha-thinkpad.redhat.com>
On 04/22/2013 08:00 PM, Stefan Hajnoczi wrote:
> On Mon, Apr 22, 2013 at 02:59:10PM +0800, Liu Yuan wrote:
>> +static coroutine_fn int
>> +sd_co_is_allocated(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
>> + int *pnum)
>> +{
>> + BDRVSheepdogState *s = bs->opaque;
>> + SheepdogInode *inode = &s->inode;
>> + unsigned long start = sector_num * BDRV_SECTOR_SIZE / SD_DATA_OBJ_SIZE, idx,
>> + end = DIV_ROUND_UP((sector_num + nb_sectors) *
>> + BDRV_SECTOR_SIZE, SD_DATA_OBJ_SIZE);
>
> Please put the variable declarations on separate lines, it's very easy
> to miss "idx".
Okay.
>
>> + int ret = 1;
>> +
>> + for (idx = start; idx <= end; idx++) {
>
> Should this be idx < end? Otherwise you are checking one beyond the
> last SD_DATA_OBJ_SIZE.
No. the end is index of last object, not index of last object + 1.
>
>> + if (inode->data_vdi_id[idx] == 0) {
>> + break;
>> + }
>> + }
>> + if (idx == start) {
>> + /* Get te longest length of unallocated sectors */
>
> s/te/the/
>
>> + ret = 0;
>> + for (idx = start + 1; idx <= end; idx++) {
>> + if (inode->data_vdi_id[idx] != 0) {
>> + break;
>> + }
>> + }
>> + }
>
> Here is a more concise way of implementing these two loops:
>
> int ret = !!inode->data_vdi_id[idx];
> for (idx = start + 1; idx < end; idx++) {
> if (!!inode->data_vdi_id[idx] != ret) {
> break;
> }
> }
>
> I like this better because it avoids code duplication, but it's a
> question of style. Feel free to stick to your approach if you like.
The trick of your code looks fantastic to me and I like your idea to
reduce the duplicated code as much as possible but the sacrifice of code
readability for the resulted code is somewhat too high, so I think not
worth of it and I'll stick to my stupid but more clear version in V3.
Thanks,
Yuan
next prev parent reply other threads:[~2013-04-22 12:11 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-22 6:59 [Qemu-devel] [PATCH v2 0/2] implement .bdrv_co_is_allocated Liu Yuan
2013-04-22 6:59 ` [Qemu-devel] [PATCH v2 1/2] sheepdog: use BDRV_SECTOR_SIZE Liu Yuan
2013-04-22 6:59 ` [Qemu-devel] [PATCH v2 2/2] sheepdog: implement .bdrv_co_is_allocated() Liu Yuan
2013-04-22 12:00 ` Stefan Hajnoczi
2013-04-22 12:10 ` Liu Yuan [this message]
2013-04-22 15:03 ` Stefan Hajnoczi
2013-04-22 15:18 ` Liu Yuan
2013-04-22 20:46 ` Stefan Hajnoczi
2013-04-23 5:55 ` Liu Yuan
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=517528D2.1040409@gmail.com \
--to=namei.unix@gmail.com \
--cc=kwolf@redhat.com \
--cc=morita.kazutaka@lab.ntt.co.jp \
--cc=qemu-devel@nongnu.org \
--cc=sheepdog@lists.wpkg.org \
--cc=stefanha@gmail.com \
--cc=stefanha@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;
as well as URLs for NNTP newsgroup(s).