diff mbox

f2fs: add bug_on when f2fs_gc even fails to get one victim

Message ID 1507729864-118702-1-git-send-email-yunlong.song@huawei.com (mailing list archive)
State New, archived
Headers show

Commit Message

Yunlong Song Oct. 11, 2017, 1:51 p.m. UTC 
This can help us to debug on some corner case.

Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
---
 fs/f2fs/gc.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

Comments

Jaegeuk Kim Oct. 12, 2017, 11:23 p.m. UTC | #1 
Hi Yunlong,

On 10/11, Yunlong Song wrote:
> This can help us to debug on some corner case.
> 
> Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
> ---
>  fs/f2fs/gc.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
> index 197ebf4..960503e 100644
> --- a/fs/f2fs/gc.c
> +++ b/fs/f2fs/gc.c
> @@ -986,6 +986,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>  		.ilist = LIST_HEAD_INIT(gc_list.ilist),
>  		.iroot = RADIX_TREE_INIT(GFP_NOFS),
>  	};
> +	bool need_gc = false;
>  
>  	trace_f2fs_gc_begin(sbi->sb, sync, background,
>  				get_pages(sbi, F2FS_DIRTY_NODES),
> @@ -1018,8 +1019,10 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>  			if (ret)
>  				goto stop;
>  		}
> -		if (has_not_enough_free_secs(sbi, 0, 0))
> +		if (has_not_enough_free_secs(sbi, 0, 0)) {
>  			gc_type = FG_GC;
> +			need_gc = true;
> +		}
>  	}
>  
>  	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
> @@ -1028,6 +1031,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>  		goto stop;
>  	}
>  	if (!__get_victim(sbi, &segno, gc_type)) {
> +		f2fs_bug_on(sbi, !total_freed && need_gc);

How about this, if this is for debugging purpose?

		if (!total_freed && need_gc)
			f2fs_msg(...);

It's not a good way to define a value for f2fs_bug_on(), since it is only
activated by CHECK_FS config.

Thanks,

>  		ret = -ENODATA;
>  		goto stop;
>  	}
> -- 
> 1.8.5.2
Yunlong Song Oct. 13, 2017, 1:09 a.m. UTC | #2 
Hi, Jay,
     I think it should not happen when need_gc == true but total_freed 
==0, so I add it as bug_on to let it panic at once. And even CHECK_FS is 
closed, this can also printk WARNING message for notice.

On 2017/10/13 7:23, Jaegeuk Kim wrote:
> Hi Yunlong,
>
> On 10/11, Yunlong Song wrote:
>> This can help us to debug on some corner case.
>>
>> Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
>> ---
>>   fs/f2fs/gc.c | 6 +++++-
>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
>> index 197ebf4..960503e 100644
>> --- a/fs/f2fs/gc.c
>> +++ b/fs/f2fs/gc.c
>> @@ -986,6 +986,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>   		.ilist = LIST_HEAD_INIT(gc_list.ilist),
>>   		.iroot = RADIX_TREE_INIT(GFP_NOFS),
>>   	};
>> +	bool need_gc = false;
>>   
>>   	trace_f2fs_gc_begin(sbi->sb, sync, background,
>>   				get_pages(sbi, F2FS_DIRTY_NODES),
>> @@ -1018,8 +1019,10 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>   			if (ret)
>>   				goto stop;
>>   		}
>> -		if (has_not_enough_free_secs(sbi, 0, 0))
>> +		if (has_not_enough_free_secs(sbi, 0, 0)) {
>>   			gc_type = FG_GC;
>> +			need_gc = true;
>> +		}
>>   	}
>>   
>>   	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
>> @@ -1028,6 +1031,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>   		goto stop;
>>   	}
>>   	if (!__get_victim(sbi, &segno, gc_type)) {
>> +		f2fs_bug_on(sbi, !total_freed && need_gc);
> How about this, if this is for debugging purpose?
>
> 		if (!total_freed && need_gc)
> 			f2fs_msg(...);
>
> It's not a good way to define a value for f2fs_bug_on(), since it is only
> activated by CHECK_FS config.
>
> Thanks,
>
>>   		ret = -ENODATA;
>>   		goto stop;
>>   	}
>> -- 
>> 1.8.5.2
> .
>
Jaegeuk Kim Oct. 13, 2017, 2:29 a.m. UTC | #3 
On 10/13, Yunlong Song wrote:
> Hi, Jay,
>     I think it should not happen when need_gc == true but total_freed ==0,
> so I add it as bug_on to let it panic at once. And even CHECK_FS is closed,
> this can also printk WARNING message for notice.

Ah, got it. Merged. :)
Thanks,

> 
> On 2017/10/13 7:23, Jaegeuk Kim wrote:
> > Hi Yunlong,
> > 
> > On 10/11, Yunlong Song wrote:
> > > This can help us to debug on some corner case.
> > > 
> > > Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
> > > ---
> > >   fs/f2fs/gc.c | 6 +++++-
> > >   1 file changed, 5 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
> > > index 197ebf4..960503e 100644
> > > --- a/fs/f2fs/gc.c
> > > +++ b/fs/f2fs/gc.c
> > > @@ -986,6 +986,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
> > >   		.ilist = LIST_HEAD_INIT(gc_list.ilist),
> > >   		.iroot = RADIX_TREE_INIT(GFP_NOFS),
> > >   	};
> > > +	bool need_gc = false;
> > >   	trace_f2fs_gc_begin(sbi->sb, sync, background,
> > >   				get_pages(sbi, F2FS_DIRTY_NODES),
> > > @@ -1018,8 +1019,10 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
> > >   			if (ret)
> > >   				goto stop;
> > >   		}
> > > -		if (has_not_enough_free_secs(sbi, 0, 0))
> > > +		if (has_not_enough_free_secs(sbi, 0, 0)) {
> > >   			gc_type = FG_GC;
> > > +			need_gc = true;
> > > +		}
> > >   	}
> > >   	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
> > > @@ -1028,6 +1031,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
> > >   		goto stop;
> > >   	}
> > >   	if (!__get_victim(sbi, &segno, gc_type)) {
> > > +		f2fs_bug_on(sbi, !total_freed && need_gc);
> > How about this, if this is for debugging purpose?
> > 
> > 		if (!total_freed && need_gc)
> > 			f2fs_msg(...);
> > 
> > It's not a good way to define a value for f2fs_bug_on(), since it is only
> > activated by CHECK_FS config.
> > 
> > Thanks,
> > 
> > >   		ret = -ENODATA;
> > >   		goto stop;
> > >   	}
> > > -- 
> > > 1.8.5.2
> > .
> > 
> 
> -- 
> Thanks,
> Yunlong Song
>
Chao Yu Oct. 13, 2017, 11:09 a.m. UTC | #4 
On 2017/10/11 21:51, Yunlong Song wrote:
> This can help us to debug on some corner case.
> 
> Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
> ---
>  fs/f2fs/gc.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
> index 197ebf4..960503e 100644
> --- a/fs/f2fs/gc.c
> +++ b/fs/f2fs/gc.c
> @@ -986,6 +986,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>  		.ilist = LIST_HEAD_INIT(gc_list.ilist),
>  		.iroot = RADIX_TREE_INIT(GFP_NOFS),
>  	};
> +	bool need_gc = false;

How about changing variable name to need_fggc for better readability?

Thanks,

>  
>  	trace_f2fs_gc_begin(sbi->sb, sync, background,
>  				get_pages(sbi, F2FS_DIRTY_NODES),
> @@ -1018,8 +1019,10 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>  			if (ret)
>  				goto stop;
>  		}
> -		if (has_not_enough_free_secs(sbi, 0, 0))
> +		if (has_not_enough_free_secs(sbi, 0, 0)) {
>  			gc_type = FG_GC;
> +			need_gc = true;
> +		}
>  	}
>  
>  	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
> @@ -1028,6 +1031,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>  		goto stop;
>  	}
>  	if (!__get_victim(sbi, &segno, gc_type)) {
> +		f2fs_bug_on(sbi, !total_freed && need_gc);
>  		ret = -ENODATA;
>  		goto stop;
>  	}
>
Yunlong Song Oct. 13, 2017, 11:20 a.m. UTC | #5 
Yep, both are OK, since I do not care it is foreground or background. 
Here it just needs to do gc.

On 2017/10/13 19:09, Chao Yu wrote:
> On 2017/10/11 21:51, Yunlong Song wrote:
>> This can help us to debug on some corner case.
>>
>> Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
>> ---
>>   fs/f2fs/gc.c | 6 +++++-
>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
>> index 197ebf4..960503e 100644
>> --- a/fs/f2fs/gc.c
>> +++ b/fs/f2fs/gc.c
>> @@ -986,6 +986,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>   		.ilist = LIST_HEAD_INIT(gc_list.ilist),
>>   		.iroot = RADIX_TREE_INIT(GFP_NOFS),
>>   	};
>> +	bool need_gc = false;
> How about changing variable name to need_fggc for better readability?
>
> Thanks,
>
>>   
>>   	trace_f2fs_gc_begin(sbi->sb, sync, background,
>>   				get_pages(sbi, F2FS_DIRTY_NODES),
>> @@ -1018,8 +1019,10 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>   			if (ret)
>>   				goto stop;
>>   		}
>> -		if (has_not_enough_free_secs(sbi, 0, 0))
>> +		if (has_not_enough_free_secs(sbi, 0, 0)) {
>>   			gc_type = FG_GC;
>> +			need_gc = true;
>> +		}
>>   	}
>>   
>>   	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
>> @@ -1028,6 +1031,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>   		goto stop;
>>   	}
>>   	if (!__get_victim(sbi, &segno, gc_type)) {
>> +		f2fs_bug_on(sbi, !total_freed && need_gc);
>>   		ret = -ENODATA;
>>   		goto stop;
>>   	}
>>
> .
>
Chao Yu Oct. 13, 2017, 12:24 p.m. UTC | #6 
On 2017/10/13 19:20, Yunlong Song wrote:
> Yep, both are OK, since I do not care it is foreground or background. Here it just needs to do gc.

But you only set need_gc as true for foreground GC case.

Thanks,

> 
> On 2017/10/13 19:09, Chao Yu wrote:
>> On 2017/10/11 21:51, Yunlong Song wrote:
>>> This can help us to debug on some corner case.
>>>
>>> Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
>>> ---
>>>   fs/f2fs/gc.c | 6 +++++-
>>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
>>> index 197ebf4..960503e 100644
>>> --- a/fs/f2fs/gc.c
>>> +++ b/fs/f2fs/gc.c
>>> @@ -986,6 +986,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>>           .ilist = LIST_HEAD_INIT(gc_list.ilist),
>>>           .iroot = RADIX_TREE_INIT(GFP_NOFS),
>>>       };
>>> +    bool need_gc = false;
>> How about changing variable name to need_fggc for better readability?
>>
>> Thanks,
>>
>>>         trace_f2fs_gc_begin(sbi->sb, sync, background,
>>>                   get_pages(sbi, F2FS_DIRTY_NODES),
>>> @@ -1018,8 +1019,10 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>>               if (ret)
>>>                   goto stop;
>>>           }
>>> -        if (has_not_enough_free_secs(sbi, 0, 0))
>>> +        if (has_not_enough_free_secs(sbi, 0, 0)) {
>>>               gc_type = FG_GC;
>>> +            need_gc = true;
>>> +        }
>>>       }
>>>         /* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
>>> @@ -1028,6 +1031,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
>>>           goto stop;
>>>       }
>>>       if (!__get_victim(sbi, &segno, gc_type)) {
>>> +        f2fs_bug_on(sbi, !total_freed && need_gc);
>>>           ret = -ENODATA;
>>>           goto stop;
>>>       }
>>>
>> .
>>
>
diff mbox

Patch

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 197ebf4..960503e 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -986,6 +986,7 @@  int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 		.ilist = LIST_HEAD_INIT(gc_list.ilist),
 		.iroot = RADIX_TREE_INIT(GFP_NOFS),
 	};
+	bool need_gc = false;
 
 	trace_f2fs_gc_begin(sbi->sb, sync, background,
 				get_pages(sbi, F2FS_DIRTY_NODES),
@@ -1018,8 +1019,10 @@  int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 			if (ret)
 				goto stop;
 		}
-		if (has_not_enough_free_secs(sbi, 0, 0))
+		if (has_not_enough_free_secs(sbi, 0, 0)) {
 			gc_type = FG_GC;
+			need_gc = true;
+		}
 	}
 
 	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
@@ -1028,6 +1031,7 @@  int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 		goto stop;
 	}
 	if (!__get_victim(sbi, &segno, gc_type)) {
+		f2fs_bug_on(sbi, !total_freed && need_gc);
 		ret = -ENODATA;
 		goto stop;
 	}