linux-kernel - Re: [PATCH v3 1/2] md: add helper rdev_needs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <dc6e25cd-17b6-4a32-8ef9-162069646a25@huaweicloud.com>
Date: Thu, 14 Aug 2025 19:09:49 +0800
From: Zheng Qixing <zhengqixing@...weicloud.com>
To: Li Nan <linan666@...weicloud.com>, song@...nel.org, yukuai3@...wei.com
Cc: linux-raid@...r.kernel.org, linux-kernel@...r.kernel.org,
 pmenzel@...gen.mpg.de, yi.zhang@...wei.com, yangerkun@...wei.com,
 houtao1@...wei.com, zhengqixing@...wei.com
Subject: Re: [PATCH v3 1/2] md: add helper rdev_needs_recovery()

Hi,


在 2025/8/14 15:09, Li Nan 写道:
>
>
> 在 2025/8/14 9:57, Zheng Qixing 写道:
>> From: Zheng Qixing <zhengqixing@...wei.com>
>>
>> Add a helper for checking if an rdev needs recovery.
>>
>> Signed-off-by: Zheng Qixing <zhengqixing@...wei.com>
>> Reviewed-by: Paul Menzel <pmenzel@...gen.mpg.de>
>> Reviewed-by: Yu Kuai <yukuai3@...wei.com>
>> ---
>>   drivers/md/md.c | 18 ++++++++++--------
>>   1 file changed, 10 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/md/md.c b/drivers/md/md.c
>> index ac85ec73a409..4663e172864e 100644
>> --- a/drivers/md/md.c
>> +++ b/drivers/md/md.c
>> @@ -4835,6 +4835,14 @@ metadata_store(struct mddev *mddev, const char 
>> *buf, size_t len)
>>   static struct md_sysfs_entry md_metadata =
>>   __ATTR_PREALLOC(metadata_version, S_IRUGO|S_IWUSR, metadata_show, 
>> metadata_store);
>>   +static bool rdev_needs_recovery(struct md_rdev *rdev, sector_t 
>> sectors)
>> +{
>> +    return !test_bit(Journal, &rdev->flags) &&
>> +           !test_bit(Faulty, &rdev->flags) &&
>> +           !test_bit(In_sync, &rdev->flags) &&
>> +           rdev->recovery_offset < sectors;
>> +}
>> +
>
> Every caller is already checking 'rdev->raid_disk >= 0'. Should we 
> move it
> into rdev_needs_recovery()?
>

Good point, thanks.


Qixing


>>   enum sync_action md_sync_action(struct mddev *mddev)
>>   {
>>       unsigned long recovery = mddev->recovery;
>> @@ -8969,10 +8977,7 @@ static sector_t md_sync_position(struct mddev 
>> *mddev, enum sync_action action)
>>           rcu_read_lock();
>>           rdev_for_each_rcu(rdev, mddev)
>>               if (rdev->raid_disk >= 0 &&
>> -                !test_bit(Journal, &rdev->flags) &&
>> -                !test_bit(Faulty, &rdev->flags) &&
>> -                !test_bit(In_sync, &rdev->flags) &&
>> -                rdev->recovery_offset < start)
>> +                rdev_needs_recovery(rdev, start))
>>                   start = rdev->recovery_offset;
>>           rcu_read_unlock();
>>   @@ -9333,10 +9338,7 @@ void md_do_sync(struct md_thread *thread)
>>                   rdev_for_each_rcu(rdev, mddev)
>>                       if (rdev->raid_disk >= 0 &&
>>                           mddev->delta_disks >= 0 &&
>> -                        !test_bit(Journal, &rdev->flags) &&
>> -                        !test_bit(Faulty, &rdev->flags) &&
>> -                        !test_bit(In_sync, &rdev->flags) &&
>> -                        rdev->recovery_offset < mddev->curr_resync)
>> +                        rdev_needs_recovery(rdev, mddev->curr_resync))
>>                           rdev->recovery_offset = mddev->curr_resync;
>>                   rcu_read_unlock();
>>               }
>