From: Jan Kara <jack@suse.cz>
To: Kemeng Shi <shikemeng@huaweicloud.com>
Cc: Jan Kara <jack@suse.cz>,
axboe@kernel.dk, linux-block@vger.kernel.org,
linux-kernel@vger.kernel.org, kbusch@kernel.org, mwilck@suse.com,
wuchi <wuchi.zero@gmail.com>
Subject: Re: [PATCH RESEND v2 2/5] sbitmap: remove redundant check in __sbitmap_queue_get_batch
Date: Thu, 22 Dec 2022 13:16:10 +0100 [thread overview]
Message-ID: <20221222121610.6o23vbarpiaqwkcx@quack3> (raw)
In-Reply-To: <792b0caa-0e99-94b2-60bf-90ad719c63d7@huaweicloud.com>
[Added to CC original author of the problematic commit and reviewer]
On Thu 22-12-22 19:49:12, Kemeng Shi wrote:
> Hi Jan, thanks for review.
> on 12/22/2022 7:23 PM, Jan Kara wrote:
> >> diff --git a/lib/sbitmap.c b/lib/sbitmap.c
> >> index cb5e03a2d65b..11e75f4040fb 100644
> >> --- a/lib/sbitmap.c
> >> +++ b/lib/sbitmap.c
> >> @@ -518,11 +518,9 @@ unsigned long __sbitmap_queue_get_batch(struct sbitmap_queue *sbq, int nr_tags,
> >>
> >> get_mask = ((1UL << nr_tags) - 1) << nr;
> >> val = READ_ONCE(map->word);
> >> - do {
> >> - if ((val & ~get_mask) != val)
> >> - goto next;
> >> - } while (!atomic_long_try_cmpxchg(ptr, &val,
> >> - get_mask | val));
> >> + while (!atomic_long_try_cmpxchg(ptr, &val,
> >> + get_mask | val))
> >> + ;
> >> get_mask = (get_mask & ~val) >> nr;
> >> if (get_mask) {
> >> *offset = nr + (index << sb->shift);
> >
> > So I agree this will result in correct behavior but it can change
> > performance. In the original code, we end up doing
> > atomic_long_try_cmpxchg() only for words where we have a chance of getting
> > all tags allocated. Now we just accept any word where we could allocate at
> > least one bit. Frankly the original code looks rather restrictive and also
> > the fact that we look only from the first zero bit in the word looks
> > unnecessarily restrictive so maybe I miss some details about what's
> > expected from __sbitmap_queue_get_batch(). So all in all I wanted to point
> > out this needs more scrutiny from someone understanding better expectations
> > from __sbitmap_queue_get_batch().
> In the very beginning, __sbitmap_queue_get_batch will return if we only
> get partial tags allocated. Recent commit fbb564a557809 ("lib/sbitmap: Fix
> invalid loop in __sbitmap_queue_get_batch()") thought we may reuse busying
> bits in old codes and change behavior of __sbitmap_queue_get_batch() to get
> all tags. However we will not reuse busying bits in old codes actually. So
> I try to revert this wrong fix and keep the behavior of
> __sbitmap_queue_get_batch() as it designed to be at beginning.
I see and now I agree. Please add tag:
Fixes: fbb564a557809 ("lib/sbitmap: Fix invalid loop in __sbitmap_queue_get_batch()")
to your commit. Also feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
> Besides, if we keep to get all tags,the check below is redundant.
> get_mask = (get_mask & ~ret) >> nr;
> if (get_mask) {
> ...
> }
> As we only reach here if we get all tags and the check above will always
> pass. So this check in old codes should be removed.
Yeah, I agree.
Honza
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
next prev parent reply other threads:[~2022-12-22 12:18 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-22 14:33 [PATCH RESEND v2 0/5] A few bugfix and cleanup patches for sbitmap Kemeng Shi
2022-12-22 14:33 ` [PATCH RESEND v2 1/5] sbitmap: remove unnecessary calculation of alloc_hint in __sbitmap_get_shallow Kemeng Shi
2022-12-22 11:01 ` Jan Kara
2022-12-22 14:33 ` [PATCH RESEND v2 2/5] sbitmap: remove redundant check in __sbitmap_queue_get_batch Kemeng Shi
2022-12-22 11:23 ` Jan Kara
2022-12-22 11:49 ` Kemeng Shi
2022-12-22 12:16 ` Jan Kara [this message]
2022-12-22 14:33 ` [PATCH RESEND v2 3/5] sbitmap: rewrite sbitmap_find_bit_in_index to reduce repeat code Kemeng Shi
2022-12-22 12:23 ` Jan Kara
2022-12-22 14:33 ` [PATCH RESEND v2 4/5] sbitmap: add sbitmap_find_bit to remove repeat code in __sbitmap_get/__sbitmap_get_shallow Kemeng Shi
2022-12-22 12:42 ` Jan Kara
2022-12-22 14:33 ` [PATCH RESEND v2 5/5] sbitmap: correct wake_batch recalculation to avoid potential IO hung Kemeng Shi
2022-12-22 13:41 ` Jan Kara
2022-12-26 7:50 ` Yu Kuai
2022-12-26 8:57 ` Yu Kuai
2023-01-03 2:12 ` Kemeng Shi
2023-01-16 2:15 ` Kemeng Shi
2023-01-16 9:54 ` Jan Kara
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=20221222121610.6o23vbarpiaqwkcx@quack3 \
--to=jack@suse.cz \
--cc=axboe@kernel.dk \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mwilck@suse.com \
--cc=shikemeng@huaweicloud.com \
--cc=wuchi.zero@gmail.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®