From: Chao Yu <chao@kernel.org>
To: Yangtao Li <frank.li@vivo.com>, jaegeuk@kernel.org
Cc: linux-kernel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net
Subject: Re: [f2fs-dev] [PATCH] f2fs: merge f2fs_show_injection_info() into time_to_inject()
Date: Sat, 17 Dec 2022 09:14:16 +0800 [thread overview]
Message-ID: <0377292c-ba43-06cd-2638-8e41e2e75f3a@kernel.org> (raw)
In-Reply-To: <20221216193248.71333-1-frank.li@vivo.com>
On 2022/12/17 3:32, Yangtao Li wrote:
> Hi Chao,
>
>> After moving f2fs_show_injection_info() core functionality into time_to_inject(),
>> __builtin_return_address(0) result changes from return address of caller of
>> f2fs_show_injection_info() to return address of time_to_inject().
>
> It seems you are wrong, and the original patch didn't change the logic here.
> Because time_to_inject is an inline function, __builtin_return_address(0)
> has not return address of time_to_inject().
Yeah, but it depends on compile option, right? If user use -O0 or -fno-inline
option, inline function will not be expanded...
How about using deterministic implementation?
Thanks,
>
> My test:
>
> (1). w/ below patch, we found that the value of __builtin_return_address(0) in
> __time_to_inject() is the same as that in f2fs_show_injection_info().
>
> # mount -t f2fs -o fault_type=0xffff,fault_injection=1 /mnt/9p/f2fs.img /mnt/f2fs
> [ 19.739661] loop0: detected capacity change from 0 to 2097152
> [ 19.749759] new F2FS-fs (loop0) : inject kmalloc in f2fs_kmalloc of f2fs_fill_super+0x7eb/0x1780
> [ 19.750313] raw F2FS-fs (loop0) : inject kmalloc in f2fs_kmalloc of f2fs_fill_super+0x7eb/0x1780
> mount: mounting /dev/loop0 on /mnt/f2fs failed: Cannot allocate memory
>
> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> index 4035dab1f570..1e5030633f4d 100644
> --- a/fs/f2fs/f2fs.h
> +++ b/fs/f2fs/f2fs.h
> @@ -1876,11 +1876,13 @@ struct f2fs_sb_info {
>
> #ifdef CONFIG_F2FS_FAULT_INJECTION
> #define f2fs_show_injection_info(sbi, type) \
> - printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n", \
> + printk_ratelimited("%sraw F2FS-fs (%s) : inject %s in %s of %pS\n", \
> KERN_INFO, sbi->sb->s_id, \
> f2fs_fault_name[type], \
> __func__, __builtin_return_address(0))
> -static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
> +
> +#define time_to_inject(sbi, type) __time_to_inject(sbi, type, __func__)
> +static inline bool __time_to_inject(struct f2fs_sb_info *sbi, int type, const char *func_name)
> {
> struct f2fs_fault_info *ffi = &F2FS_OPTION(sbi).fault_info;
>
> @@ -1893,6 +1895,10 @@ static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
> atomic_inc(&ffi->inject_ops);
> if (atomic_read(&ffi->inject_ops) >= ffi->inject_rate) {
> atomic_set(&ffi->inject_ops, 0);
> + printk_ratelimited("%s new F2FS-fs (%s) : inject %s in %s of %pS\n",
> + KERN_INFO, sbi->sb->s_id,
> + f2fs_fault_name[type],
> + func_name, __builtin_return_address(0));
> return true;
> }
> return false;
>
> (2). w/ below patch(remove inline form __time_to_inject()), we found that the value of
> __builtin_return_address(0) in __time_to_inject() is different from that in
> f2fs_show_injection_info().
>
> # mount -t f2fs -o fault_type=0xffff,fault_injection=1 /mnt/9p/f2fs.img /mnt/f2fs
> [ 81.019451] loop0: detected capacity change from 0 to 2097152
> [ 81.031058] new F2FS-fs (loop0) : inject kmalloc in f2fs_kmalloc of f2fs_init_write_merge_io+0x35/0x1c0
> [ 81.031745] raw F2FS-fs (loop0) : inject kmalloc in f2fs_kmalloc of f2fs_fill_super+0x7eb/0x1710
> mount: mounting /dev/loop0 on /mnt/f2fs failed: Cannot allocate memory
>
> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> index 4035dab1f570..f15001b5d73b 100644
> --- a/fs/f2fs/f2fs.h
> +++ b/fs/f2fs/f2fs.h
> @@ -1876,11 +1876,13 @@ struct f2fs_sb_info {
>
> #ifdef CONFIG_F2FS_FAULT_INJECTION
> #define f2fs_show_injection_info(sbi, type) \
> - printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n", \
> + printk_ratelimited("%sraw F2FS-fs (%s) : inject %s in %s of %pS\n", \
> KERN_INFO, sbi->sb->s_id, \
> f2fs_fault_name[type], \
> __func__, __builtin_return_address(0))
> -static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
> +
> +#define time_to_inject(sbi, type) __time_to_inject(sbi, type, __func__)
> +static bool __time_to_inject(struct f2fs_sb_info *sbi, int type, const char *func_name)
> {
> struct f2fs_fault_info *ffi = &F2FS_OPTION(sbi).fault_info;
>
> @@ -1893,6 +1895,10 @@ static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
> atomic_inc(&ffi->inject_ops);
> if (atomic_read(&ffi->inject_ops) >= ffi->inject_rate) {
> atomic_set(&ffi->inject_ops, 0);
> + printk_ratelimited("%s new F2FS-fs (%s) : inject %s in %s of %pS\n",
> + KERN_INFO, sbi->sb->s_id,
> + f2fs_fault_name[type],
> + func_name, __builtin_return_address(0));
> return true;
> }
> return false;
>
> Thx,
> Yangtao
_______________________________________________
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:[~2022-12-17 1:14 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-15 12:20 [f2fs-dev] [PATCH] f2fs: merge f2fs_show_injection_info() into time_to_inject() Yangtao Li via Linux-f2fs-devel
2022-12-16 2:51 ` Chao Yu
2022-12-16 3:48 ` Yangtao Li via Linux-f2fs-devel
2022-12-16 10:47 ` Chao Yu
2022-12-16 19:32 ` Yangtao Li via Linux-f2fs-devel
2022-12-17 1:14 ` Chao Yu [this message]
2022-12-17 4:51 ` Yangtao Li via Linux-f2fs-devel
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=0377292c-ba43-06cd-2638-8e41e2e75f3a@kernel.org \
--to=chao@kernel.org \
--cc=frank.li@vivo.com \
--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).