From: Chao Yu <yuchao0@huawei.com>
To: "yuchangchun (C)" <yuchangchun1@huawei.com>,
Qilong Zhang <zhangqilong3@huawei.com>, <jaegeuk@kernel.org>
Cc: linux-f2fs-devel@lists.sourceforge.net
Subject: Re: [f2fs-dev] [PATCH v4] f2fs: add f2fs_gc exception handle in f2fs_ioc_gc_range
Date: Sun, 28 Jun 2020 17:24:20 +0800 [thread overview]
Message-ID: <a645df9e-5ad7-27e0-08a6-d5064f857c8f@huawei.com> (raw)
In-Reply-To: <7c68a31c-7388-cffa-97db-91154b572456@huawei.com>
On 2020/6/28 16:06, yuchangchun (C) wrote:
> 在 2020/6/28 14:39, Qilong Zhang 写道:
>> When f2fs_ioc_gc_range performs multiple segments gc ops, the return
>> value of f2fs_ioc_gc_range is determined by the last segment gc ops.
>> If its ops failed, the f2fs_ioc_gc_range will be considered to be failed
>> despite some of previous segments gc ops succeeded. Therefore, so we
>> fix: Redefine the return value of getting victim ops and add exception
>> handle for f2fs_gc. In particular, 1).if target has no valid block, it
>> will go on. 2).if target sectoion has valid block(s), but it is current
>> section, we will reminder the caller.
>>
>> Signed-off-by: Qilong Zhang <zhangqilong3@huawei.com>
>> Reviewed-by: Chao Yu <yuchao0@huawei.com>
>> ---
>> fs/f2fs/file.c | 5 +++++
>> fs/f2fs/gc.c | 19 +++++++++++++------
>> fs/f2fs/segment.c | 4 ++--
>> 3 files changed, 20 insertions(+), 8 deletions(-)
>>
>> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
>> index 3268f8dd59bb..fe006477a4e4 100644
>> --- a/fs/f2fs/file.c
>> +++ b/fs/f2fs/file.c
>> @@ -2527,6 +2527,11 @@ static int f2fs_ioc_gc_range(struct file *filp, unsigned long arg)
>> }
>>
>> ret = f2fs_gc(sbi, range.sync, true, GET_SEGNO(sbi, range.start));
>> + if (ret) {
>> + if (ret == -EBUSY)
>> + ret = -EAGAIN;
>> + goto out;
>> + }
>> range.start += BLKS_PER_SEC(sbi);
>> if (range.start <= end)
>> goto do_more;
>> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
>> index 5b95d5a146eb..71cb4a02aeee 100644
>> --- a/fs/f2fs/gc.c
>> +++ b/fs/f2fs/gc.c
>> @@ -321,6 +321,7 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
>> unsigned int secno, last_victim;
>> unsigned int last_segment;
>> unsigned int nsearched = 0;
>> + int ret;
>>
>> mutex_lock(&dirty_i->seglist_lock);
>> last_segment = MAIN_SECS(sbi) * sbi->segs_per_sec;
>> @@ -332,12 +333,18 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
>> p.min_cost = get_max_cost(sbi, &p);
>>
>> if (*result != NULL_SEGNO) {
>> - if (get_valid_blocks(sbi, *result, false) &&
>> - !sec_usage_check(sbi, GET_SEC_FROM_SEG(sbi, *result)))
>> + ret = 0;
>> + if (!get_valid_blocks(sbi, *result, false))
>> + goto out;
> Should it return specific errno to stop GC if this victim segment has no
> valid blocks in accordance with original code.
Agreed, could be:
if (!get_valid_blocks(sbi, *result, false)) {
ret = -ENODATA;
goto out;
}
Thanks,
>
> Thanks,
>
>> +
>> + if (sec_usage_check(sbi, GET_SEC_FROM_SEG(sbi, *result)))
>> + ret = -EBUSY;
>> + else
>> p.min_segno = *result;
>> goto out;
>> }
>>
>> + ret = -ENODATA;
>> if (p.max_search == 0)
>> goto out;
>>
>> @@ -440,6 +447,7 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
>> else
>> set_bit(secno, dirty_i->victim_secmap);
>> }
>> + ret = 0;
>>
>> }
>> out:
>> @@ -449,7 +457,7 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
>> prefree_segments(sbi), free_segments(sbi));
>> mutex_unlock(&dirty_i->seglist_lock);
>>
>> - return (p.min_segno == NULL_SEGNO) ? 0 : 1;
>> + return ret;
>> }
>>
>> static const struct victim_selection default_v_ops = {
>> @@ -1333,10 +1341,9 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>> ret = -EINVAL;
>> goto stop;
>> }
>> - if (!__get_victim(sbi, &segno, gc_type)) {
>> - ret = -ENODATA;
>> + ret = __get_victim(sbi, &segno, gc_type);
>> + if (ret)
>> goto stop;
>> - }
>>
>> seg_freed = do_garbage_collect(sbi, segno, &gc_list, gc_type);
>> if (gc_type == FG_GC && seg_freed == sbi->segs_per_sec)
>> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
>> index 196f31503511..b9fd93761b0a 100644
>> --- a/fs/f2fs/segment.c
>> +++ b/fs/f2fs/segment.c
>> @@ -2605,7 +2605,7 @@ static int get_ssr_segment(struct f2fs_sb_info *sbi, int type)
>> bool reversed = false;
>>
>> /* f2fs_need_SSR() already forces to do this */
>> - if (v_ops->get_victim(sbi, &segno, BG_GC, type, SSR)) {
>> + if (!v_ops->get_victim(sbi, &segno, BG_GC, type, SSR)) {
>> curseg->next_segno = segno;
>> return 1;
>> }
>> @@ -2632,7 +2632,7 @@ static int get_ssr_segment(struct f2fs_sb_info *sbi, int type)
>> for (; cnt-- > 0; reversed ? i-- : i++) {
>> if (i == type)
>> continue;
>> - if (v_ops->get_victim(sbi, &segno, BG_GC, i, SSR)) {
>> + if (!v_ops->get_victim(sbi, &segno, BG_GC, i, SSR)) {
>> curseg->next_segno = segno;
>> return 1;
>> }
>
>
>
> .
>
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
prev parent reply other threads:[~2020-06-28 9:24 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-28 6:39 [f2fs-dev] [PATCH v4] f2fs: add f2fs_gc exception handle in f2fs_ioc_gc_range Qilong Zhang
2020-06-28 8:06 ` yuchangchun (C)
2020-06-28 9:24 ` 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=a645df9e-5ad7-27e0-08a6-d5064f857c8f@huawei.com \
--to=yuchao0@huawei.com \
--cc=jaegeuk@kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=yuchangchun1@huawei.com \
--cc=zhangqilong3@huawei.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;
as well as URLs for NNTP newsgroup(s).