linux-kernel - Re: [PATCH RFC md-6.16 v3 15/19] md/md-llbitmap: implement APIs to dirty bits and clear bits

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <03f64fc7-4e57-2f32-bffc-04836a9df790@huaweicloud.com>
Date: Tue, 13 May 2025 14:32:04 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: Yu Kuai <yukuai1@...weicloud.com>, Christoph Hellwig <hch@....de>
Cc: xni@...hat.com, colyli@...nel.org, agk@...hat.com, snitzer@...nel.org,
 mpatocka@...hat.com, song@...nel.org, linux-kernel@...r.kernel.org,
 dm-devel@...ts.linux.dev, linux-raid@...r.kernel.org, yi.zhang@...wei.com,
 yangerkun@...wei.com, johnny.chenyi@...wei.com,
 "yukuai (C)" <yukuai3@...wei.com>
Subject: Re: [PATCH RFC md-6.16 v3 15/19] md/md-llbitmap: implement APIs to
 dirty bits and clear bits

Hi,

在 2025/05/12 21:36, Yu Kuai 写道:
> Hi,
> 
> 在 2025/05/12 21:30, Christoph Hellwig 写道:
>> On Mon, May 12, 2025 at 03:26:41PM +0200, Christoph Hellwig wrote:
>>>> 1) bitmap bio must be done before this bio can be issued;
>>>> 2) bitmap bio will be added to current->bio_list, and wait for this bio
>>>> to be issued;
>>>>
>>>> Do you have a better sulution to this problem?
>>>
>>> A bew block layer API that bypasses bio_list maybe?  I.e. export
>>> __submit_bio with a better name and a kerneldoc detailing the narrow
>>> use case.
>>
>> That won't work as we'd miss a lot of checks, cgroup handling, etc.
>>
>> But maybe a flag to skip the recursion avoidance?
> 
> I think this can work, and this can also improve performance. I'll look
> into this.

So, I did a quick test with old internal bitmap and make sure following
patch can work.

However, for bitmap file case, bio is issued from submit_bh(), I'll have
to change buffer_head code and I'm not sure if we want to do that.

Thanks,
Kuai

diff --git a/block/blk-core.c b/block/blk-core.c
index b862c66018f2..66ced5769694 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -745,7 +745,7 @@ void submit_bio_noacct_nocheck(struct bio *bio)
          * to collect a list of requests submited by a ->submit_bio 
method while
          * it is active, and then process them after it returned.
          */
-       if (current->bio_list)
+       if (current->bio_list && !bio_flagged(bio, BIO_STACKED_META))
                 bio_list_add(&current->bio_list[0], bio);
         else if (!bdev_test_flag(bio->bi_bdev, BD_HAS_SUBMIT_BIO))
                 __submit_bio_noacct_mq(bio);
diff --git a/drivers/md/md-bitmap.c b/drivers/md/md-bitmap.c
index 431a3ab2e449..e0cb210a4ea4 100644
--- a/drivers/md/md-bitmap.c
+++ b/drivers/md/md-bitmap.c
@@ -1257,35 +1257,6 @@ static void __bitmap_unplug(struct bitmap *bitmap)
                 md_bitmap_file_kick(bitmap);
  }

-struct bitmap_unplug_work {
-       struct work_struct work;
-       struct bitmap *bitmap;
-       struct completion *done;
-};
-
-static void md_bitmap_unplug_fn(struct work_struct *work)
-{
-       struct bitmap_unplug_work *unplug_work =
-               container_of(work, struct bitmap_unplug_work, work);
-
-       __bitmap_unplug(unplug_work->bitmap);
-       complete(unplug_work->done);
-}
-
-static void bitmap_unplug_async(struct bitmap *bitmap)
-{
-       DECLARE_COMPLETION_ONSTACK(done);
-       struct bitmap_unplug_work unplug_work;
-
-       INIT_WORK_ONSTACK(&unplug_work.work, md_bitmap_unplug_fn);
-       unplug_work.bitmap = bitmap;
-       unplug_work.done = &done;
-
-       queue_work(md_bitmap_wq, &unplug_work.work);
-       wait_for_completion(&done);
-       destroy_work_on_stack(&unplug_work.work);
-}
-
  static void bitmap_unplug(struct mddev *mddev, bool sync)
  {
         struct bitmap *bitmap = mddev->bitmap;
@@ -1293,10 +1264,7 @@ static void bitmap_unplug(struct mddev *mddev, 
bool sync)
         if (!bitmap)
                 return;

-       if (sync)
-               __bitmap_unplug(bitmap);
-       else
-               bitmap_unplug_async(bitmap);
+       __bitmap_unplug(bitmap);
  }

  static void md_bitmap_set_memory_bits(struct bitmap *bitmap, sector_t 
offset, int needed);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 32b997dfe6f4..179eabd6e038 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1050,6 +1050,7 @@ void md_super_write(struct mddev *mddev, struct 
md_rdev *rdev,
         __bio_add_page(bio, page, size, 0);
         bio->bi_private = rdev;
         bio->bi_end_io = super_written;
+       bio_set_flag(bio, BIO_STACKED_META);

         if (test_bit(MD_FAILFAST_SUPPORTED, &mddev->flags) &&
             test_bit(FailFast, &rdev->flags) &&
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index f38425338c3f..88164cdae6aa 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -300,6 +300,7 @@ enum {
         BIO_REMAPPED,
         BIO_ZONE_WRITE_PLUGGING, /* bio handled through zone write 
plugging */
         BIO_EMULATES_ZONE_APPEND, /* bio emulates a zone append 
operation */
+       BIO_STACKED_META,       /* bio is metadata from stacked device */
         BIO_FLAG_LAST
  };

> 
> Thanks,
> Kuai
> 
>>
>> .
>>
> 
> .
>