public inbox for linux-btrfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Qu Wenruo <quwenruo.btrfs@gmx.com>
To: Christoph Hellwig <hch@infradead.org>, Qu Wenruo <wqu@suse.com>
Cc: linux-btrfs@vger.kernel.org
Subject: Re: [PATCH 05/13] btrfs: add btrfs_read_repair_ctrl to record corrupted sectors
Date: Wed, 4 May 2022 09:12:50 +0800	[thread overview]
Message-ID: <dac4707a-04f4-f143-342b-cd69e0ffcd80@gmx.com> (raw)
In-Reply-To: <YnFE62oGR5C/8UN2@infradead.org>



On 2022/5/3 23:06, Christoph Hellwig wrote:
> On Tue, May 03, 2022 at 02:49:49PM +0800, Qu Wenruo wrote:
>> - For the extra bitmaps
>>    We use btrfs_bio::read_repair_{cur|prev}_bitmap.
>>    This will add extra 16 bytes to btrfs_bio.
>>
>>    This is unavoidable, or we need to allocate extra memory at endio
>>    function which can be dead lock prone.
>>
>>    Furthermore, pre-allocating bitmap has a hidden benefit, if we're
>>    submitting a large read and failed to allocated enough memory for
>>    bitmap, we fail the bio, and VFS layer will automatically retry
>>    with smaller read range, giving us a higher chance to succeed in
>>    next try.
>
> This is a really bad idea.  Now every read needs to pay the quite
> large price for corner case of a read repair.  As I mentioned last time
> I think a mempool with a few entries that any read repair can dip into
> is a much better choice here.

The problem is, can mempool provide a variable length entry?

Thanks,
Qu

  reply	other threads:[~2022-05-04  1:13 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-03  6:49 [PATCH 00/13] btrfs: make read repair work in synchronous mode Qu Wenruo
2022-05-03  6:49 ` [PATCH 01/13] btrfs: introduce a pure data checksum checking helper Qu Wenruo
2022-05-03 15:03   ` Christoph Hellwig
2022-05-03  6:49 ` [PATCH 02/13] btrfs: quit early if the fs has no RAID56 support for raid56 related checks Qu Wenruo
2022-05-03  6:49 ` [PATCH 03/13] btrfs: save the original bi_iter into btrfs_bio for buffered read Qu Wenruo
2022-05-03  6:49 ` [PATCH 04/13] btrfs: remove duplicated parameters from submit_data_read_repair() Qu Wenruo
2022-05-03  6:49 ` [PATCH 05/13] btrfs: add btrfs_read_repair_ctrl to record corrupted sectors Qu Wenruo
2022-05-03 15:06   ` Christoph Hellwig
2022-05-04  1:12     ` Qu Wenruo [this message]
2022-05-04 14:05       ` Christoph Hellwig
2022-05-04 22:40         ` Qu Wenruo
2022-05-12 17:16           ` David Sterba
2022-05-13 10:33             ` Christoph Hellwig
2022-05-13 10:53               ` Qu Wenruo
2022-05-13 10:57                 ` Christoph Hellwig
2022-05-13 11:21                   ` Qu Wenruo
2022-05-13 11:23                     ` Christoph Hellwig
2022-05-17 13:32                       ` Qu Wenruo
2022-05-03  6:49 ` [PATCH 06/13] btrfs: add a helper to queue a corrupted sector for read repair Qu Wenruo
2022-05-03 15:07   ` Christoph Hellwig
2022-05-04  1:13     ` Qu Wenruo
2022-05-04 14:06       ` Christoph Hellwig
2022-05-12 17:20         ` David Sterba
2022-05-03  6:49 ` [PATCH 07/13] btrfs: introduce a helper to repair from one mirror Qu Wenruo
2022-05-03  6:49 ` [PATCH 08/13] btrfs: allow btrfs read repair to submit writes in asynchronous mode Qu Wenruo
2022-05-03  6:49 ` [PATCH 09/13] btrfs: handle RAID56 read repair differently Qu Wenruo
2022-05-03  6:49 ` [PATCH 10/13] btrfs: switch buffered read to the new read repair routine Qu Wenruo
2022-05-03  6:49 ` [PATCH 11/13] btrfs: switch direct IO routine to use btrfs_read_repair_ctrl Qu Wenruo
2022-05-03  6:49 ` [PATCH 12/13] btrfs: remove io_failure_record infrastructure completely Qu Wenruo
2022-05-03  6:49 ` [PATCH 13/13] btrfs: remove btrfs_inode::io_failure_tree Qu Wenruo
2022-05-03 15:07   ` Christoph Hellwig
2022-05-12 17:08 ` [PATCH 00/13] btrfs: make read repair work in synchronous mode David Sterba
2022-05-12 23:01   ` Qu Wenruo

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=dac4707a-04f4-f143-342b-cd69e0ffcd80@gmx.com \
    --to=quwenruo.btrfs@gmx.com \
    --cc=hch@infradead.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=wqu@suse.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