From: Chao Yu <chao@kernel.org>
To: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: linux-f2fs-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/6] f2fs: compress: fix to guarantee persisting compressed blocks by CP
Date: Thu, 28 Dec 2023 10:44:09 +0800 [thread overview]
Message-ID: <aae654e7-8a7e-478d-9f5a-65807a0e0343@kernel.org> (raw)
In-Reply-To: <ZYyrddJK8KJNpKXz@google.com>
On 2023/12/28 6:55, Jaegeuk Kim wrote:
> On 12/27, Chao Yu wrote:
>> On 2023/12/27 5:02, Jaegeuk Kim wrote:
>>> On 12/20, Chao Yu wrote:
>>>> If data block in compressed cluster is not persisted with metadata
>>>> during checkpoint, after SPOR, the data may be corrupted, let's
>>>> guarantee to write compressed page by checkpoint.
>>>>
>>>> Fixes: 4c8ff7095bef ("f2fs: support data compression")
>>>> Signed-off-by: Chao Yu <chao@kernel.org>
>>>> ---
>>>> fs/f2fs/compress.c | 3 ++-
>>>> fs/f2fs/data.c | 12 +++++++++---
>>>> fs/f2fs/f2fs.h | 3 ++-
>>>> 3 files changed, 13 insertions(+), 5 deletions(-)
>>>>
>>>> diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c
>>>> index 5b076329e9bf..1122db8cc0b0 100644
>>>> --- a/fs/f2fs/compress.c
>>>> +++ b/fs/f2fs/compress.c
>>>> @@ -1442,6 +1442,7 @@ void f2fs_compress_write_end_io(struct bio *bio, struct page *page)
>>>> struct f2fs_sb_info *sbi = bio->bi_private;
>>>> struct compress_io_ctx *cic =
>>>> (struct compress_io_ctx *)page_private(page);
>>>> + enum count_type type = WB_DATA_TYPE(page);
>>>> int i;
>>>> if (unlikely(bio->bi_status))
>>>> @@ -1449,7 +1450,7 @@ void f2fs_compress_write_end_io(struct bio *bio, struct page *page)
>>>> f2fs_compress_free_page(page);
>>>> - dec_page_count(sbi, F2FS_WB_DATA);
>>>> + dec_page_count(sbi, type);
>>>> if (atomic_dec_return(&cic->pending_pages))
>>>> return;
>>>> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
>>>> index d28c97282e68..6c72a6e86ba8 100644
>>>> --- a/fs/f2fs/data.c
>>>> +++ b/fs/f2fs/data.c
>>>> @@ -48,7 +48,7 @@ void f2fs_destroy_bioset(void)
>>>> bioset_exit(&f2fs_bioset);
>>>> }
>>>> -static bool __is_cp_guaranteed(struct page *page)
>>>> +bool f2fs_is_cp_guaranteed(struct page *page)
>>>> {
>>>> struct address_space *mapping = page->mapping;
>>>> struct inode *inode;
>>>> @@ -66,7 +66,7 @@ static bool __is_cp_guaranteed(struct page *page)
>>>> return true;
>>>> if (f2fs_is_compressed_page(page))
>>>> - return false;
>>>> + return true;
>>>> if ((S_ISREG(inode->i_mode) && IS_NOQUOTA(inode)) ||
>>>> page_private_gcing(page))
>>>> return true;
>>>> @@ -1007,6 +1007,7 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio)
>>>> enum page_type btype = PAGE_TYPE_OF_BIO(fio->type);
>>>> struct f2fs_bio_info *io = sbi->write_io[btype] + fio->temp;
>>>> struct page *bio_page;
>>>> + enum count_type type;
>>>> f2fs_bug_on(sbi, is_read_io(fio->op));
>>>> @@ -1046,7 +1047,12 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio)
>>>> /* set submitted = true as a return value */
>>>> fio->submitted = 1;
>>>> - inc_page_count(sbi, WB_DATA_TYPE(bio_page));
>>>> + type = WB_DATA_TYPE(bio_page);
>>>> + /* override count type if page is compressed one */
>>>> + if (fio->compressed_page)
>>>> + type = WB_DATA_TYPE(fio->compressed_page);
>>>
>>> Doesn't bio_page already point fio->compressed_page?
>>
>> Please check below codes, bio_page will point to fio->encrypted_page if
>> both software encryption feature and compression feature are on, for this
>> case, we still need to account F2FS_WB_CP_DATA.
>
> So, it seems you want to make F2FS_WB_CP_DATA regardless of conditions. Then,
> how about making this explictly instead of implicit condition check of the page?
>
> #define WB_DATA_TYPE(p, f) (f || __is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA)
>
> inc_page_count(sbi, WB_DATA_TYPE(bio_page, bio_page == fio->compressed_page));
Do you mean inc_page_count(sbi, WB_DATA_TYPE(bio_page, fio->compressed_page));?
If we use inc_page_count(sbi, WB_DATA_TYPE(bio_page, bio_page == fio->compressed_page));
if bio_page points to fio->encrypted_page, but fio->compressed_page is valid, then
WB_DATA_TYPE() will return F2FS_WB_DATA, it doesn't as expect.
Or am I missing something?
Thanks,
>
> dec_page_count(sbi, WB_DATA_TYPE(page, f2fs_is_compressed_page(page)));
>
>>
>> if (fio->encrypted_page)
>> bio_page = fio->encrypted_page;
>> else if (fio->compressed_page)
>> bio_page = fio->compressed_page;
>> else
>> bio_page = fio->page;
>>
>> Thanks,
>>
>>>
>>>> +
>>>> + inc_page_count(sbi, type);
>>>> if (io->bio &&
>>>> (!io_is_mergeable(sbi, io->bio, io, fio, io->last_block_in_bio,
>>>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
>>>> index 76e9a8682e38..bcb3940ab5ba 100644
>>>> --- a/fs/f2fs/f2fs.h
>>>> +++ b/fs/f2fs/f2fs.h
>>>> @@ -1092,7 +1092,7 @@ struct f2fs_sm_info {
>>>> * f2fs monitors the number of several block types such as on-writeback,
>>>> * dirty dentry blocks, dirty node blocks, and dirty meta blocks.
>>>> */
>>>> -#define WB_DATA_TYPE(p) (__is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA)
>>>> +#define WB_DATA_TYPE(p) (f2fs_is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA)
>>>> enum count_type {
>>>> F2FS_DIRTY_DENTS,
>>>> F2FS_DIRTY_DATA,
>>>> @@ -3824,6 +3824,7 @@ void f2fs_init_ckpt_req_control(struct f2fs_sb_info *sbi);
>>>> */
>>>> int __init f2fs_init_bioset(void);
>>>> void f2fs_destroy_bioset(void);
>>>> +bool f2fs_is_cp_guaranteed(struct page *page);
>>>> int f2fs_init_bio_entry_cache(void);
>>>> void f2fs_destroy_bio_entry_cache(void);
>>>> void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, struct bio *bio,
>>>> --
>>>> 2.40.1
prev parent reply other threads:[~2023-12-28 2:44 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-20 13:59 [PATCH 1/6] f2fs: compress: fix to guarantee persisting compressed blocks by CP Chao Yu
2023-12-20 13:59 ` [PATCH 2/6] f2fs: compress: fix to cover normal cluster write with cp_rwsem Chao Yu
2023-12-20 13:59 ` [PATCH 3/6] f2fs: compress: fix to check unreleased compressed cluster Chao Yu
2023-12-20 13:59 ` [PATCH 4/6] f2fs: compress: fix to avoid inconsistent bewteen i_blocks and dnode Chao Yu
2023-12-20 13:59 ` [PATCH 5/6] f2fs: fix to remove unnecessary f2fs_bug_on() to avoid panic Chao Yu
2023-12-20 13:59 ` [PATCH 6/6] f2fs: introduce FAULT_INCONSISTENCE Chao Yu
2023-12-26 21:02 ` [PATCH 1/6] f2fs: compress: fix to guarantee persisting compressed blocks by CP Jaegeuk Kim
2023-12-27 1:04 ` Chao Yu
2023-12-27 22:55 ` Jaegeuk Kim
2023-12-28 2:44 ` Chao Yu [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=aae654e7-8a7e-478d-9f5a-65807a0e0343@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