From: Chao Yu <chao@kernel.org>
To: Wu Bo <bo.wu@vivo.com>
Cc: wubo.oduw@gmail.com, jaegeuk@kernel.org,
linux-kernel@vger.kernel.org,
linux-f2fs-devel@lists.sourceforge.net
Subject: Re: [f2fs-dev] [PATCH 1/1] f2fs: fix args passed to trace_f2fs_lookup_end
Date: Mon, 29 May 2023 18:18:42 +0800 [thread overview]
Message-ID: <0c15c662-903c-3504-d502-72f0a3b70cbf@kernel.org> (raw)
In-Reply-To: <20230529041337.36741-1-bo.wu@vivo.com>
On 2023/5/29 12:13, Wu Bo wrote:
> On Sat, May 27, 2023 at 09:01:41AM +0800, Chao Yu wrote:
>> On 2023/5/27 1:21, Jaegeuk Kim wrote:
>>> On 05/24, Wu Bo wrote:
>>>> The NULL return of 'd_splice_alias' dosen't mean error.
>>>>
>>>> Signed-off-by: Wu Bo <bo.wu@vivo.com>
>>>> ---
>>>> fs/f2fs/namei.c | 2 +-
>>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
>>>> index 77a71276ecb1..e5a3e39ce90c 100644
>>>> --- a/fs/f2fs/namei.c
>>>> +++ b/fs/f2fs/namei.c
>>>> @@ -577,7 +577,7 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
>>>> #endif
>>>> new = c(inode, dentry);
>>>> err = PTR_ERR_OR_ZERO(new);
>>>> - trace_f2fs_lookup_end(dir, dentry, ino, !new ? -ENOENT : err);
>>>> + trace_f2fs_lookup_end(dir, new ? new : dentry, ino, err);
>>>
>>> Shouldn't give an error pointer to the dentry field.
>>>
>>> How about just giving the err?
>>>
>>> - err = PTR_ERR_OR_ZERO(new);
>>> - trace_f2fs_lookup_end(dir, dentry, ino, !new ? -ENOENT : err);
>>> + trace_f2fs_lookup_end(dir, dentry, ino, PTR_ERR_OR_ZERO(new));
>>
>> static inline int __must_check PTR_ERR_OR_ZERO(__force const void *ptr)
>> {
>> if (IS_ERR(ptr))
>> return PTR_ERR(ptr);
>> else
>> return 0;
>> }
>>
>> For below two cases, PTR_ERR_OR_ZERO(new) will return zero:
>> a) f2fs_lookup found existed dentry
>> b) f2fs_lookup didn't find existed dentry (-ENOENT case)
>>
>> So in below commit, I passed -ENOENT to tracepoint for case b), so we can
>> distinguish result of f2fs_lookup in tracepoint, actually, -ENOENT is expected
>> value when we create a new file/directory.
>>
>> Commit 84597b1f9b05 ("f2fs: fix wrong value of tracepoint parameter")
> I can see this commit is try to distinguish the dentry not existed case.
> But a normal case which dentry is exactly found will also go through
> 'd_splice_alias', and its return is also NULL. This makes the tracepoint always
> print 'err:-2' like the following:
> ls-11676 [004] .... 329281.943118: f2fs_lookup_end: dev = (254,39), pino = 4451, name:Alarms, ino:7093, err:-2
> ls-11676 [004] .... 329281.943145: f2fs_lookup_end: dev = (254,39), pino = 4451, name:Notifications, ino:7094, err:-2
> ls-11676 [004] .... 329281.943172: f2fs_lookup_end: dev = (254,39), pino = 4451, name:Pictures, ino:7095, err:-2
> Even these lookup are acctually successful, this is a bit strange.
Ah, I misunderstand return value's meaning of .lookup.
So, how about this? it only update err if d_splice_alias() returns a negative
value?
if (IS_ERR(new))
err = PTR_ERR(new);
trace_f2fs_lookup_end(dir, new ? new : dentry, ino, err);
Thanks,
>>
>>>
>>>
>>>> return new;
>>>> out_iput:
>>>> iput(inode);
>>>> --
>>>> 2.35.3
>>
>>
>> _______________________________________________
>> Linux-f2fs-devel mailing list
>> Linux-f2fs-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
next prev parent reply other threads:[~2023-05-29 10:19 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-24 10:08 [f2fs-dev] [PATCH 1/1] f2fs: fix args passed to trace_f2fs_lookup_end Wu Bo via Linux-f2fs-devel
2023-05-26 17:21 ` Jaegeuk Kim
2023-05-27 1:01 ` Chao Yu
2023-05-29 4:13 ` Wu Bo via Linux-f2fs-devel
2023-05-29 10:18 ` Chao Yu [this message]
2023-05-29 12:41 ` Wu Bo
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=0c15c662-903c-3504-d502-72f0a3b70cbf@kernel.org \
--to=chao@kernel.org \
--cc=bo.wu@vivo.com \
--cc=jaegeuk@kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=wubo.oduw@gmail.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).