[f2fs-dev] [PATCH v3] f2fs: modify f2fs_is_checkpoint_ready logic to allow more data to be written with the CP disable

[f2fs-dev] [PATCH v3] f2fs: modify f2fs_is_checkpoint_ready logic to allow more data to be written with the CP disable

[Qi Han via Linux-f2fs-devel]

When the free segment is used up during CP disable, many write or
ioctl operations will get ENOSPC error codes, even if there are
still many blocks available. We can reproduce it in the following
steps:

dd if=/dev/zero of=f2fs.img bs=1M count=65
mkfs.f2fs -f f2fs.img
mount f2fs.img f2fs_dir -o checkpoint=disable:10%
cd f2fs_dir
i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; dd \
if=/dev/random of=$file_name bs=1M count=2); i=$((i+1)); done
sync
i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; truncate \
-s 1K $file_name); i=$((i+1)); done
sync
dd if=/dev/zero of=./file bs=1M count=20

In f2fs_need_SSR() function, it is allowed to use SSR to allocate
blocks when CP is disabled, so in f2fs_is_checkpoint_ready function,
can we judge the number of invalid blocks when free segment is not
enough, and return ENOSPC only if the number of invalid blocks is
also not enough?

Signed-off-by: Qi Han <hanqi@vivo.com>
---
 fs/f2fs/segment.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index 71adb4a43bec..20b568eaa95e 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -637,12 +637,29 @@ static inline bool has_enough_free_secs(struct f2fs_sb_info *sbi,
 	return !has_not_enough_free_secs(sbi, freed, needed);
 }
 
+static inline bool has_enough_free_blks(struct f2fs_sb_info *sbi)
+{
+	unsigned int total_free_blocks = 0;
+	unsigned int avail_user_block_count;
+
+	spin_lock(&sbi->stat_lock);
+
+	avail_user_block_count = get_available_block_count(sbi, NULL, true);
+	total_free_blocks = avail_user_block_count - (unsigned int)valid_user_blocks(sbi);
+
+	spin_unlock(&sbi->stat_lock);
+
+	return total_free_blocks > 0;
+}
+
 static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
 {
 	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
 		return true;
 	if (likely(has_enough_free_secs(sbi, 0, 0)))
 		return true;
+	if (likely(has_enough_free_blks(sbi)))
+		return true;
 	return false;
 }
 
-- 
2.39.0



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Re: [f2fs-dev] [PATCH v3] f2fs: modify f2fs_is_checkpoint_ready logic to allow more data to be written with the CP disable

[Chao Yu via Linux-f2fs-devel]

On 2024/10/23 10:59, Qi Han wrote:
> When the free segment is used up during CP disable, many write or
> ioctl operations will get ENOSPC error codes, even if there are
> still many blocks available. We can reproduce it in the following
> steps:
> 
> dd if=/dev/zero of=f2fs.img bs=1M count=65
> mkfs.f2fs -f f2fs.img
> mount f2fs.img f2fs_dir -o checkpoint=disable:10%
> cd f2fs_dir
> i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; dd \
> if=/dev/random of=$file_name bs=1M count=2); i=$((i+1)); done
> sync
> i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; truncate \
> -s 1K $file_name); i=$((i+1)); done
> sync
> dd if=/dev/zero of=./file bs=1M count=20
> 
> In f2fs_need_SSR() function, it is allowed to use SSR to allocate
> blocks when CP is disabled, so in f2fs_is_checkpoint_ready function,
> can we judge the number of invalid blocks when free segment is not
> enough, and return ENOSPC only if the number of invalid blocks is
> also not enough?
> 
> Signed-off-by: Qi Han <hanqi@vivo.com>

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Re: [f2fs-dev] [PATCH v3] f2fs: modify f2fs_is_checkpoint_ready logic to allow more data to be written with the CP disable

[韩棋 via Linux-f2fs-devel]

在 2024/10/23 10:59, Qi Han 写道:
> When the free segment is used up during CP disable, many write or
> ioctl operations will get ENOSPC error codes, even if there are
> still many blocks available. We can reproduce it in the following
> steps:
>
> dd if=/dev/zero of=f2fs.img bs=1M count=65
> mkfs.f2fs -f f2fs.img
> mount f2fs.img f2fs_dir -o checkpoint=disable:10%
> cd f2fs_dir
> i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; dd \
> if=/dev/random of=$file_name bs=1M count=2); i=$((i+1)); done
> sync
> i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; truncate \
> -s 1K $file_name); i=$((i+1)); done
> sync
> dd if=/dev/zero of=./file bs=1M count=20
>
> In f2fs_need_SSR() function, it is allowed to use SSR to allocate
> blocks when CP is disabled, so in f2fs_is_checkpoint_ready function,
> can we judge the number of invalid blocks when free segment is not
> enough, and return ENOSPC only if the number of invalid blocks is
> also not enough?
>
> Signed-off-by: Qi Han <hanqi@vivo.com>
> ---
>   fs/f2fs/segment.h | 17 +++++++++++++++++
>   1 file changed, 17 insertions(+)
>
> diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
> index 71adb4a43bec..20b568eaa95e 100644
> --- a/fs/f2fs/segment.h
> +++ b/fs/f2fs/segment.h
> @@ -637,12 +637,29 @@ static inline bool has_enough_free_secs(struct f2fs_sb_info *sbi,
>   	return !has_not_enough_free_secs(sbi, freed, needed);
>   }
>   
> +static inline bool has_enough_free_blks(struct f2fs_sb_info *sbi)
> +{
> +	unsigned int total_free_blocks = 0;
> +	unsigned int avail_user_block_count;
> +
> +	spin_lock(&sbi->stat_lock);
> +
> +	avail_user_block_count = get_available_block_count(sbi, NULL, true);
> +	total_free_blocks = avail_user_block_count - (unsigned int)valid_user_blocks(sbi);
> +
> +	spin_unlock(&sbi->stat_lock);
> +
> +	return total_free_blocks > 0;
> +}
> +
>   static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
>   {
>   	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
>   		return true;
>   	if (likely(has_enough_free_secs(sbi, 0, 0)))
>   		return true;

Hi, Chao,

After Zhiguo's reminder, I just saw your previous patch:
f2fs: fix to account dirty data in __get_secs_required(),
the current modification may still return true if the segment
is found to be insufficient when LFS and CP is closed, should
I add the LFS mode restriction here?

Thanks.

> +	if (likely(has_enough_free_blks(sbi)))
> +		return true;
>   	return false;
>   }
>   


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Re: [f2fs-dev] [PATCH v3] f2fs: modify f2fs_is_checkpoint_ready logic to allow more data to be written with the CP disable

[Chao Yu via Linux-f2fs-devel]

On 2024/10/23 15:40, 韩棋 wrote:
> 在 2024/10/23 10:59, Qi Han 写道:
>> When the free segment is used up during CP disable, many write or
>> ioctl operations will get ENOSPC error codes, even if there are
>> still many blocks available. We can reproduce it in the following
>> steps:
>>
>> dd if=/dev/zero of=f2fs.img bs=1M count=65
>> mkfs.f2fs -f f2fs.img
>> mount f2fs.img f2fs_dir -o checkpoint=disable:10%
>> cd f2fs_dir
>> i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; dd \
>> if=/dev/random of=$file_name bs=1M count=2); i=$((i+1)); done
>> sync
>> i=1 ; while [[ $i -lt 50 ]] ; do (file_name=./2M_file$i ; truncate \
>> -s 1K $file_name); i=$((i+1)); done
>> sync
>> dd if=/dev/zero of=./file bs=1M count=20
>>
>> In f2fs_need_SSR() function, it is allowed to use SSR to allocate
>> blocks when CP is disabled, so in f2fs_is_checkpoint_ready function,
>> can we judge the number of invalid blocks when free segment is not
>> enough, and return ENOSPC only if the number of invalid blocks is
>> also not enough?
>>
>> Signed-off-by: Qi Han <hanqi@vivo.com>
>> ---
>>    fs/f2fs/segment.h | 17 +++++++++++++++++
>>    1 file changed, 17 insertions(+)
>>
>> diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
>> index 71adb4a43bec..20b568eaa95e 100644
>> --- a/fs/f2fs/segment.h
>> +++ b/fs/f2fs/segment.h
>> @@ -637,12 +637,29 @@ static inline bool has_enough_free_secs(struct f2fs_sb_info *sbi,
>>    	return !has_not_enough_free_secs(sbi, freed, needed);
>>    }
>>    
>> +static inline bool has_enough_free_blks(struct f2fs_sb_info *sbi)
>> +{
>> +	unsigned int total_free_blocks = 0;
>> +	unsigned int avail_user_block_count;



>> +
>> +	spin_lock(&sbi->stat_lock);
>> +
>> +	avail_user_block_count = get_available_block_count(sbi, NULL, true);
>> +	total_free_blocks = avail_user_block_count - (unsigned int)valid_user_blocks(sbi);
>> +
>> +	spin_unlock(&sbi->stat_lock);
>> +
>> +	return total_free_blocks > 0;
>> +}
>> +
>>    static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
>>    {
>>    	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
>>    		return true;
>>    	if (likely(has_enough_free_secs(sbi, 0, 0)))
>>    		return true;
> 
> Hi, Chao,
> 
> After Zhiguo's reminder, I just saw your previous patch:
> f2fs: fix to account dirty data in __get_secs_required(),
> the current modification may still return true if the segment
> is found to be insufficient when LFS and CP is closed, should
> I add the LFS mode restriction here?

Hi Qi,

I guess so, I think we need to add a lfs_mode check condition in
has_enough_free_blks(), otherwise this patch will trigger system
panic w/ below testcase:

https://lore.kernel.org/fstests/20241015025106.3203676-1-chao@kernel.org/

Thanks,

> 
> Thanks.
> 
>> +	if (likely(has_enough_free_blks(sbi)))
>> +		return true;
>>    	return false;
>>    }
>>    
> 



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel