linux-f2fs-devel.lists.sourceforge.net archive mirror
 help / color / mirror / Atom feed
From: Chao Yu <chao@kernel.org>
To: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net
Subject: Re: [f2fs-dev] [PATCH 2/2] f2fs: keep io_flags to avoid IO split due to different op_flags in two fio holders
Date: Wed, 20 Apr 2022 10:45:25 +0800	[thread overview]
Message-ID: <47f19561-11a7-9da3-1cf9-78843e0d0ec3@kernel.org> (raw)
In-Reply-To: <Yl88F3RVRMfwyB7+@google.com>

On 2022/4/20 6:47, Jaegeuk Kim wrote:
> On 04/17, Chao Yu wrote:
>> On 2022/4/13 23:49, Jaegeuk Kim wrote:
>>> Let's attach io_flags to bio only, so that we can merge IOs given original
>>> io_flags only.
>>>
>>> Fixes: 64bf0eef0171 ("f2fs: pass the bio operation to bio_alloc_bioset")
>>
>> Nice catch.
>>
>> Wasn't this bug introduced by:
> 
> I don't think so.

Oh, correct, it looks in original patch, we only attach IO flag before
__submit_bio().

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,

> 
>>
>> commit da9953b729c1 ("f2fs: introduce sysfs/data_io_flag to attach REQ_META/FUA")
>>
>> static void __attach_data_io_flag(struct f2fs_io_info *fio)
>> {
>> ...
>>         if ((1 << fio->temp) & meta_flag)
>>                 fio->op_flags |= REQ_META;
>>         if ((1 << fio->temp) & fua_flag)
>>                 fio->op_flags |= REQ_FUA;
>> ...
>>
>> Thanks,
>>
>>> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
>>> ---
>>>    fs/f2fs/data.c | 33 +++++++++++++++++++++------------
>>>    1 file changed, 21 insertions(+), 12 deletions(-)
>>>
>>> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
>>> index acc523f893ba..fcf0daa386de 100644
>>> --- a/fs/f2fs/data.c
>>> +++ b/fs/f2fs/data.c
>>> @@ -387,11 +387,23 @@ int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr)
>>>    	return 0;
>>>    }
>>> -static void __attach_io_flag(struct f2fs_io_info *fio, unsigned int io_flag)
>>> +static unsigned int f2fs_io_flags(struct f2fs_io_info *fio)
>>>    {
>>>    	unsigned int temp_mask = (1 << NR_TEMP_TYPE) - 1;
>>> -	unsigned int fua_flag = io_flag & temp_mask;
>>> -	unsigned int meta_flag = (io_flag >> NR_TEMP_TYPE) & temp_mask;
>>> +	unsigned int fua_flag, meta_flag, io_flag;
>>> +	unsigned int op_flags = 0;
>>> +
>>> +	if (fio->op != REQ_OP_WRITE)
>>> +		return 0;
>>> +	if (fio->type == DATA)
>>> +		io_flag = fio->sbi->data_io_flag;
>>> +	else if (fio->type == NODE)
>>> +		io_flag = fio->sbi->node_io_flag;
>>> +	else
>>> +		return 0;
>>> +
>>> +	fua_flag = io_flag & temp_mask;
>>> +	meta_flag = (io_flag >> NR_TEMP_TYPE) & temp_mask;
>>>    	/*
>>>    	 * data/node io flag bits per temp:
>>> @@ -400,9 +412,10 @@ static void __attach_io_flag(struct f2fs_io_info *fio, unsigned int io_flag)
>>>    	 * Cold | Warm | Hot | Cold | Warm | Hot |
>>>    	 */
>>>    	if ((1 << fio->temp) & meta_flag)
>>> -		fio->op_flags |= REQ_META;
>>> +		op_flags |= REQ_META;
>>>    	if ((1 << fio->temp) & fua_flag)
>>> -		fio->op_flags |= REQ_FUA;
>>> +		op_flags |= REQ_FUA;
>>> +	return op_flags;
>>>    }
>>>    static struct bio *__bio_alloc(struct f2fs_io_info *fio, int npages)
>>> @@ -412,14 +425,10 @@ static struct bio *__bio_alloc(struct f2fs_io_info *fio, int npages)
>>>    	sector_t sector;
>>>    	struct bio *bio;
>>> -	if (fio->type == DATA)
>>> -		__attach_io_flag(fio, sbi->data_io_flag);
>>> -	else if (fio->type == NODE)
>>> -		__attach_io_flag(fio, sbi->node_io_flag);
>>> -
>>>    	bdev = f2fs_target_device(sbi, fio->new_blkaddr, &sector);
>>> -	bio = bio_alloc_bioset(bdev, npages, fio->op | fio->op_flags, GFP_NOIO,
>>> -			       &f2fs_bioset);
>>> +	bio = bio_alloc_bioset(bdev, npages,
>>> +				fio->op | fio->op_flags | f2fs_io_flags(fio),
>>> +				GFP_NOIO, &f2fs_bioset);
>>>    	bio->bi_iter.bi_sector = sector;
>>>    	if (is_read_io(fio->op)) {
>>>    		bio->bi_end_io = f2fs_read_end_io;


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

  reply	other threads:[~2022-04-20  2:45 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-13 15:49 [f2fs-dev] [PATCH 1/2] f2fs: remove obsolete whint_mode Jaegeuk Kim
2022-04-13 15:49 ` [f2fs-dev] [PATCH 2/2] f2fs: keep io_flags to avoid IO split due to different op_flags in two fio holders Jaegeuk Kim
2022-04-17  6:55   ` Chao Yu
2022-04-19 22:47     ` Jaegeuk Kim
2022-04-20  2:45       ` Chao Yu [this message]
2022-04-17  6:32 ` [f2fs-dev] [PATCH 1/2] f2fs: remove obsolete whint_mode 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=47f19561-11a7-9da3-1cf9-78843e0d0ec3@kernel.org \
    --to=chao@kernel.org \
    --cc=jaegeuk@kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).