All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chao Yu via Linux-f2fs-devel <linux-f2fs-devel@lists.sourceforge.net>
To: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net
Subject: Re: [f2fs-dev] [PATCH] f2fs: dump more information when checkpoint was blocked for long time
Date: Wed, 30 Jul 2025 16:40:07 +0800	[thread overview]
Message-ID: <1735962b-043d-48b5-b593-d1705ddd4a97@kernel.org> (raw)
In-Reply-To: <aIjjoBJqYjdNv63m@google.com>

On 7/29/25 23:07, Jaegeuk Kim wrote:
> On 07/29, Chao Yu wrote:
>> generic/299 w/ mode=lfs will cause long time latency, let's dump more
>> information once we hit case.
>>
>> CP merge:
>>   - Queued :    0
>>   - Issued :    1
>>   - Total :    1
>>   - Cur time : 7565(ms)
>>   - Peak time : 7565(ms)
>>
>> F2FS-fs (vdc): checkpoint was blocked for 7565 ms, affecting 1 tasks
>> CPU: 8 UID: 0 PID: 1614 Comm: f2fs_ckpt-253:3 Tainted: G           O        6.16.0-rc3+ #406 PREEMPT(voluntary)
>> Tainted: [O]=OOT_MODULE
>> Call Trace:
>>  dump_stack_lvl+0x6e/0xa0
>>  __checkpoint_and_complete_reqs+0x1a6/0x1d0
>>  issue_checkpoint_thread+0x4b/0x140
>>  kthread+0x10d/0x250
>>  ret_from_fork+0x164/0x190
>>  ret_from_fork_asm+0x1a/0x30
> 
> Can we add more information for debugging this?

I guess we can add more stats in write_checkpoint() to see which
stage we cost most of the latency.

Thanks,

> 
>>
>> Cc: Jan Prusakowski <jprusakowski@google.com>
>> Signed-off-by: Chao Yu <chao@kernel.org>
>> ---
>>  fs/f2fs/checkpoint.c | 9 ++++++++-
>>  fs/f2fs/f2fs.h       | 3 +++
>>  2 files changed, 11 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
>> index db3831f7f2f5..b0dcaa8dc40d 100644
>> --- a/fs/f2fs/checkpoint.c
>> +++ b/fs/f2fs/checkpoint.c
>> @@ -1788,8 +1788,15 @@ static void __checkpoint_and_complete_reqs(struct f2fs_sb_info *sbi)
>>  
>>  	spin_lock(&cprc->stat_lock);
>>  	cprc->cur_time = (unsigned int)div64_u64(sum_diff, count);
>> -	if (cprc->peak_time < cprc->cur_time)
>> +	if (cprc->peak_time < cprc->cur_time) {
>>  		cprc->peak_time = cprc->cur_time;
>> +
>> +		if (unlikely(cprc->peak_time >= CP_LONG_LATENCY_THRESHOLD)) {
>> +			f2fs_warn_ratelimited(sbi, "checkpoint was blocked for %u ms, affecting %llu tasks",
>> +					cprc->peak_time, count);
>> +			dump_stack();
>> +		}
>> +	}
>>  	spin_unlock(&cprc->stat_lock);
>>  }
>>  
>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
>> index 46d23c2c086c..3130ca6a4770 100644
>> --- a/fs/f2fs/f2fs.h
>> +++ b/fs/f2fs/f2fs.h
>> @@ -350,6 +350,9 @@ struct ckpt_req_control {
>>  	unsigned int peak_time;		/* peak wait time in msec until now */
>>  };
>>  
>> +/* a time threshold that checkpoint was blocked for, unit: ms */
>> +#define CP_LONG_LATENCY_THRESHOLD	5000
>> +
>>  /* for the bitmap indicate blocks to be discarded */
>>  struct discard_entry {
>>  	struct list_head list;	/* list head */
>> -- 
>> 2.49.0



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

WARNING: multiple messages have this Message-ID (diff)
From: Chao Yu <chao@kernel.org>
To: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: chao@kernel.org, linux-f2fs-devel@lists.sourceforge.net,
	linux-kernel@vger.kernel.org,
	Jan Prusakowski <jprusakowski@google.com>
Subject: Re: [PATCH] f2fs: dump more information when checkpoint was blocked for long time
Date: Wed, 30 Jul 2025 16:40:07 +0800	[thread overview]
Message-ID: <1735962b-043d-48b5-b593-d1705ddd4a97@kernel.org> (raw)
In-Reply-To: <aIjjoBJqYjdNv63m@google.com>

On 7/29/25 23:07, Jaegeuk Kim wrote:
> On 07/29, Chao Yu wrote:
>> generic/299 w/ mode=lfs will cause long time latency, let's dump more
>> information once we hit case.
>>
>> CP merge:
>>   - Queued :    0
>>   - Issued :    1
>>   - Total :    1
>>   - Cur time : 7565(ms)
>>   - Peak time : 7565(ms)
>>
>> F2FS-fs (vdc): checkpoint was blocked for 7565 ms, affecting 1 tasks
>> CPU: 8 UID: 0 PID: 1614 Comm: f2fs_ckpt-253:3 Tainted: G           O        6.16.0-rc3+ #406 PREEMPT(voluntary)
>> Tainted: [O]=OOT_MODULE
>> Call Trace:
>>  dump_stack_lvl+0x6e/0xa0
>>  __checkpoint_and_complete_reqs+0x1a6/0x1d0
>>  issue_checkpoint_thread+0x4b/0x140
>>  kthread+0x10d/0x250
>>  ret_from_fork+0x164/0x190
>>  ret_from_fork_asm+0x1a/0x30
> 
> Can we add more information for debugging this?

I guess we can add more stats in write_checkpoint() to see which
stage we cost most of the latency.

Thanks,

> 
>>
>> Cc: Jan Prusakowski <jprusakowski@google.com>
>> Signed-off-by: Chao Yu <chao@kernel.org>
>> ---
>>  fs/f2fs/checkpoint.c | 9 ++++++++-
>>  fs/f2fs/f2fs.h       | 3 +++
>>  2 files changed, 11 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
>> index db3831f7f2f5..b0dcaa8dc40d 100644
>> --- a/fs/f2fs/checkpoint.c
>> +++ b/fs/f2fs/checkpoint.c
>> @@ -1788,8 +1788,15 @@ static void __checkpoint_and_complete_reqs(struct f2fs_sb_info *sbi)
>>  
>>  	spin_lock(&cprc->stat_lock);
>>  	cprc->cur_time = (unsigned int)div64_u64(sum_diff, count);
>> -	if (cprc->peak_time < cprc->cur_time)
>> +	if (cprc->peak_time < cprc->cur_time) {
>>  		cprc->peak_time = cprc->cur_time;
>> +
>> +		if (unlikely(cprc->peak_time >= CP_LONG_LATENCY_THRESHOLD)) {
>> +			f2fs_warn_ratelimited(sbi, "checkpoint was blocked for %u ms, affecting %llu tasks",
>> +					cprc->peak_time, count);
>> +			dump_stack();
>> +		}
>> +	}
>>  	spin_unlock(&cprc->stat_lock);
>>  }
>>  
>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
>> index 46d23c2c086c..3130ca6a4770 100644
>> --- a/fs/f2fs/f2fs.h
>> +++ b/fs/f2fs/f2fs.h
>> @@ -350,6 +350,9 @@ struct ckpt_req_control {
>>  	unsigned int peak_time;		/* peak wait time in msec until now */
>>  };
>>  
>> +/* a time threshold that checkpoint was blocked for, unit: ms */
>> +#define CP_LONG_LATENCY_THRESHOLD	5000
>> +
>>  /* for the bitmap indicate blocks to be discarded */
>>  struct discard_entry {
>>  	struct list_head list;	/* list head */
>> -- 
>> 2.49.0


  parent reply	other threads:[~2025-07-30  8:40 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-29  6:33 [f2fs-dev] [PATCH] f2fs: dump more information when checkpoint was blocked for long time Chao Yu via Linux-f2fs-devel
2025-07-29  6:33 ` Chao Yu
2025-07-29 15:07 ` [f2fs-dev] " Jaegeuk Kim via Linux-f2fs-devel
2025-07-29 15:07   ` Jaegeuk Kim
2025-07-30  8:00   ` [f2fs-dev] " Chao Yu via Linux-f2fs-devel
2025-07-30  8:00     ` Chao Yu
2025-07-30  8:40   ` Chao Yu via Linux-f2fs-devel [this message]
2025-07-30  8:40     ` Chao Yu

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=1735962b-043d-48b5-b593-d1705ddd4a97@kernel.org \
    --to=linux-f2fs-devel@lists.sourceforge.net \
    --cc=chao@kernel.org \
    --cc=jaegeuk@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.