Linux RAID subsystem development
 help / color / mirror / Atom feed
From: Guoqing Jiang <guoqing.jiang@linux.dev>
To: Xiao Ni <xni@redhat.com>
Cc: "Tkaczyk, Mariusz" <mariusz.tkaczyk@intel.com>,
	Song Liu <song@kernel.org>,
	linux-raid <linux-raid@vger.kernel.org>,
	Heinz Mauelshagen <heinzm@redhat.com>,
	Nigel Croxon <ncroxon@redhat.com>
Subject: Re: The read data is wrong from raid5 when recovery happens
Date: Tue, 30 May 2023 08:53:44 +0800	[thread overview]
Message-ID: <5cc6370a-1207-9b15-87c0-b817fa6b1b95@linux.dev> (raw)
In-Reply-To: <CALTww29N5-WFKH9JG8yzcYyHy5v+5D-NR2jd=fe5fG6tAF2_-w@mail.gmail.com>



On 5/29/23 21:51, Xiao Ni wrote:
> I can reproduce this after setting /sys/block/md126/queue/iostats to
> 0. But if I comment md_account_bio, this can't be reproduced. If
> setting iostats to 0, md_account_bio only checks the bit
> QUEUE_FLAG_IO_STAT of the request queue of md126.

I don't get why the issue can't be reproduced after comment md_acount_bio
out manually, the only difference between it and disable iostats is that 
chunk
aligned read bio still record time.

> In chunk_aligned_read, it can split the original bio and returns the
> split bio if it can't do the align read. Are there problems in this
> case?

Not sure.

Thanks,
Guoqing


      reply	other threads:[~2023-05-30  0:54 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CALTww28aV5CGXQAu46Rkc=fG1jK=ARzCT8VGoVyje8kQdqEXMg@mail.gmail.com>
2023-05-26  2:08 ` Fwd: The read data is wrong from raid5 when recovery happens Xiao Ni
2023-05-26  2:17   ` Yu Kuai
2023-05-26  2:40     ` Xiao Ni
2023-05-26  2:47       ` Yu Kuai
2023-05-26  3:02         ` Xiao Ni
2023-05-26  3:56   ` d tbsky
2023-05-26  6:20     ` Xiao Ni
2024-02-14 15:15   ` Fwd: " Mateusz Kusiak
2024-02-14 17:12     ` Song Liu
     [not found]     ` <CALTww29s1WupaVRSrEX1GbD=1Bt7b5cxseDnBLARkH1uHUhtCA@mail.gmail.com>
2024-02-15 10:41       ` Mateusz Kusiak
2023-05-26  3:09 ` Guoqing Jiang
2023-05-26  6:45   ` Xiao Ni
2023-05-26  7:12     ` Guoqing Jiang
2023-05-26  7:23       ` Xiao Ni
2023-05-26  9:13         ` Mariusz Tkaczyk
2023-05-26 21:13           ` Song Liu
2023-05-27  0:56             ` Xiao Ni
2023-07-11  0:39               ` Xiao Ni
2023-07-14  1:30                 ` Yu Kuai
2023-05-29  2:25         ` Guoqing Jiang
2023-05-29  3:41           ` Xiao Ni
2023-05-29  8:33             ` Guoqing Jiang
2023-05-29  8:40               ` Xiao Ni
2023-05-30  1:36                 ` Guoqing Jiang
2023-05-30  2:02                   ` Yu Kuai
2023-05-30  2:11                     ` Xiao Ni
2023-05-30  2:23                       ` Guoqing Jiang
2023-05-30  2:30                         ` Xiao Ni
2023-05-30  2:43                           ` Guoqing Jiang
2023-06-14  8:27                           ` Kusiak, Mateusz
2023-06-14  8:46                             ` Xiao Ni
2023-05-29 13:51           ` Xiao Ni
2023-05-30  0:53             ` Guoqing Jiang [this message]

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=5cc6370a-1207-9b15-87c0-b817fa6b1b95@linux.dev \
    --to=guoqing.jiang@linux.dev \
    --cc=heinzm@redhat.com \
    --cc=linux-raid@vger.kernel.org \
    --cc=mariusz.tkaczyk@intel.com \
    --cc=ncroxon@redhat.com \
    --cc=song@kernel.org \
    --cc=xni@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