From: Pavel Begunkov <[email protected]>
To: Dave Chinner <[email protected]>
Cc: [email protected], Jens Axboe <[email protected]>,
Christoph Hellwig <[email protected]>,
Matthew Wilcox <[email protected]>,
Ming Lei <[email protected]>,
Johannes Weiner <[email protected]>,
Alexander Viro <[email protected]>,
"Darrick J . Wong" <[email protected]>,
"Martin K . Petersen" <[email protected]>,
Jonathan Corbet <[email protected]>,
[email protected], [email protected],
[email protected], [email protected],
[email protected], [email protected],
[email protected]
Subject: Re: [PATCH v1 4/6] block/psi: remove PSI annotations from direct IO
Date: Tue, 15 Dec 2020 01:03:45 +0000 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
On 15/12/2020 00:56, Dave Chinner wrote:
> On Tue, Dec 15, 2020 at 12:20:23AM +0000, Pavel Begunkov wrote:
>> As reported, we must not do pressure stall information accounting for
>> direct IO, because otherwise it tells that it's thrashing a page when
>> actually doing IO on hot data.
>>
>> Apparently, bio_iov_iter_get_pages() is used only by paths doing direct
>> IO, so just make it avoid setting BIO_WORKINGSET, it also saves us CPU
>> cycles on doing that. For fs/direct-io.c just clear the flag before
>> submit_bio(), it's not of much concern performance-wise.
>>
>> Reported-by: Christoph Hellwig <[email protected]>
>> Suggested-by: Christoph Hellwig <[email protected]>
>> Suggested-by: Johannes Weiner <[email protected]>
>> Signed-off-by: Pavel Begunkov <[email protected]>
>> ---
>> block/bio.c | 25 ++++++++++++++++---------
>> fs/direct-io.c | 2 ++
>> 2 files changed, 18 insertions(+), 9 deletions(-)
> .....
>> @@ -1099,6 +1103,9 @@ static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter)
>> * fit into the bio, or are requested in @iter, whatever is smaller. If
>> * MM encounters an error pinning the requested pages, it stops. Error
>> * is returned only if 0 pages could be pinned.
>> + *
>> + * It also doesn't set BIO_WORKINGSET, so is intended for direct IO. If used
>> + * otherwise the caller is responsible to do that to keep PSI happy.
>> */
>> int bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter)
>> {
>> diff --git a/fs/direct-io.c b/fs/direct-io.c
>> index d53fa92a1ab6..914a7f600ecd 100644
>> --- a/fs/direct-io.c
>> +++ b/fs/direct-io.c
>> @@ -426,6 +426,8 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio)
>> unsigned long flags;
>>
>> bio->bi_private = dio;
>> + /* PSI is only for paging IO */
>> + bio_clear_flag(bio, BIO_WORKINGSET);
>
> Why only do this for the old direct IO path? Why isn't this
> necessary for the iomap DIO path?
It's in the description. In short, block and iomap dio use
bio_iov_iter_get_pages(), which with this patch doesn't use
[__]bio_add_page() and so doesn't set the flag.
--
Pavel Begunkov
next prev parent reply other threads:[~2020-12-15 1:08 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-15 0:20 [PATCH v1 0/6] no-copy bvec Pavel Begunkov
2020-12-15 0:20 ` [PATCH v1 1/6] target/file: allocate the bvec array as part of struct target_core_file_cmd Pavel Begunkov
2020-12-15 0:20 ` [PATCH v1 2/6] iov_iter: optimise bvec iov_iter_advance() Pavel Begunkov
2020-12-15 9:37 ` David Laight
2020-12-15 11:23 ` Pavel Begunkov
2020-12-15 13:54 ` David Laight
2020-12-15 13:56 ` Pavel Begunkov
2020-12-22 14:03 ` Christoph Hellwig
2020-12-15 0:20 ` [PATCH v1 3/6] bio: deduplicate adding a page into bio Pavel Begunkov
2020-12-22 14:04 ` Christoph Hellwig
2020-12-15 0:20 ` [PATCH v1 4/6] block/psi: remove PSI annotations from direct IO Pavel Begunkov
2020-12-15 0:56 ` Dave Chinner
2020-12-15 1:03 ` Pavel Begunkov [this message]
2020-12-15 1:33 ` Dave Chinner
2020-12-15 11:41 ` Pavel Begunkov
2020-12-22 14:07 ` Christoph Hellwig
2020-12-15 0:20 ` [PATCH v1 5/6] bio: add a helper calculating nr segments to alloc Pavel Begunkov
2020-12-15 1:00 ` Dave Chinner
2020-12-15 1:07 ` Pavel Begunkov
2020-12-15 1:09 ` Dave Chinner
2020-12-22 14:07 ` Christoph Hellwig
2020-12-15 0:20 ` [PATCH v1 6/6] block/iomap: don't copy bvec for direct IO Pavel Begunkov
2020-12-15 1:09 ` Dave Chinner
2020-12-15 1:15 ` Pavel Begunkov
2020-12-22 14:15 ` Christoph Hellwig
2020-12-15 1:41 ` [PATCH v1 0/6] no-copy bvec Ming Lei
2020-12-15 11:14 ` Pavel Begunkov
2020-12-15 12:03 ` Ming Lei
2020-12-15 14:05 ` Pavel Begunkov
2020-12-22 14:11 ` Christoph Hellwig
2020-12-23 12:52 ` Pavel Begunkov
2020-12-23 15:51 ` Christoph Hellwig
2020-12-23 16:04 ` James Bottomley
2020-12-23 20:23 ` Douglas Gilbert
2020-12-23 20:32 ` Pavel Begunkov
2020-12-24 6:41 ` Christoph Hellwig
2020-12-24 16:45 ` Douglas Gilbert
2020-12-24 17:30 ` James Bottomley
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 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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