Received-SPF: pass (google.com: domain of linux-ext4+bounces-1302-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <b4555d09-1408-8303-3274-9343f7944b7f@huaweicloud.com>
Date: Wed, 21 Feb 2024 17:05:56 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.5.1
Subject: Re: [PATCH] ext4: clear EXT4_GROUP_INFO_WAS_TRIMMED_BIT even mount
 with discard
To: yangerkun <yangerkun@huawei.com>, tytso@mit.edu,
 adilger.kernel@dilger.ca, Jan Kara <jack@suse.cz>
Cc: boyu.mt@taobao.com, linux-ext4@vger.kernel.org
References: <20231230070654.178638-1-yangerkun@huaweicloud.com>
 <4284c67d-eefd-d4eb-2df3-d863ca4651b4@huawei.com>
From: yangerkun <yangerkun@huaweicloud.com>
In-Reply-To: <4284c67d-eefd-d4eb-2df3-d863ca4651b4@huawei.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Gently ping...

在 2024/1/16 9:02, yangerkun 写道:
> Ping...
> 
> 在 2023/12/30 15:06, yangerkun 写道:
>> Commit 3d56b8d2c74c ("ext4: Speed up FITRIM by recording flags in
>> ext4_group_info") speed up fstrim by skipping trim trimmed group. We
>> also has the chance to clear trimmed once there exists some block free
>> for this group(mount without discard), and the next trim for this group
>> will work well too.
>>
>> For mount with discard, we will issue dicard when we free blocks, so
>> leave trimmed flag keep alive to skip useless trim trigger from
>> userspace seems reasonable. But for some case like ext4 build on
>> dm-thinpool(ext4 blocksize 4K, pool blocksize 128K), discard from ext4
>> maybe unaligned for dm thinpool, and thinpool will just finish this
>> discard(see process_discard_bio when begein equals to end) without
>> actually process discard. For this case, trim from userspace can really
>> help us to free some thinpool block.
>>
>> So convert to clear trimmed flag for all case no matter mounted with
>> discard or not.
>>
>> Fixes: 3d56b8d2c74c ("ext4: Speed up FITRIM by recording flags in 
>> ext4_group_info")
>> Signed-off-by: yangerkun <yangerkun@huaweicloud.com>
>> ---
>>   fs/ext4/mballoc.c | 10 ++++------
>>   1 file changed, 4 insertions(+), 6 deletions(-)
>>
>> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
>> index d72b5e3c92ec..69240ae775f1 100644
>> --- a/fs/ext4/mballoc.c
>> +++ b/fs/ext4/mballoc.c
>> @@ -3855,11 +3855,8 @@ static void ext4_free_data_in_buddy(struct 
>> super_block *sb,
>>       /*
>>        * Clear the trimmed flag for the group so that the next
>>        * ext4_trim_fs can trim it.
>> -     * If the volume is mounted with -o discard, online discard
>> -     * is supported and the free blocks will be trimmed online.
>>        */
>> -    if (!test_opt(sb, DISCARD))
>> -        EXT4_MB_GRP_CLEAR_TRIMMED(db);
>> +    EXT4_MB_GRP_CLEAR_TRIMMED(db);
>>       if (!db->bb_free_root.rb_node) {
>>           /* No more items in the per group rb tree
>> @@ -6481,8 +6478,9 @@ static void ext4_mb_clear_bb(handle_t *handle, 
>> struct inode *inode,
>>                        " group:%u block:%d count:%lu failed"
>>                        " with %d", block_group, bit, count,
>>                        err);
>> -        } else
>> -            EXT4_MB_GRP_CLEAR_TRIMMED(e4b.bd_info);
>> +        }
>> +
>> +        EXT4_MB_GRP_CLEAR_TRIMMED(e4b.bd_info);
>>           ext4_lock_group(sb, block_group);
>>           mb_free_blocks(inode, &e4b, bit, count_clusters);