Received-SPF: pass (google.com: domain of linux-kernel+bounces-82615-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Subject: Re: [PATCH md-6.9 03/10] md/raid1: fix choose next idle in
 read_balance()
To: Xiao Ni <xni@redhat.com>, Yu Kuai <yukuai1@huaweicloud.com>
Cc: paul.e.luse@linux.intel.com, song@kernel.org, neilb@suse.com,
 shli@fb.com, linux-raid@vger.kernel.org, linux-kernel@vger.kernel.org,
 yi.zhang@huawei.com, yangerkun@huawei.com, "yukuai (C)" <yukuai3@huawei.com>
References: <20240222075806.1816400-1-yukuai1@huaweicloud.com>
 <20240222075806.1816400-4-yukuai1@huaweicloud.com>
 <CALTww28PJPdqRkSEarwATG5GmkuMmEtT0La5s-9c9r5UPy4siA@mail.gmail.com>
From: Yu Kuai <yukuai1@huaweicloud.com>
Message-ID: <6114e6f5-87a6-216e-027d-cbb0a7e8e429@huaweicloud.com>
Date: Tue, 27 Feb 2024 10:38:01 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101
 Thunderbird/60.8.0
Precedence: bulk
MIME-Version: 1.0
In-Reply-To: <CALTww28PJPdqRkSEarwATG5GmkuMmEtT0La5s-9c9r5UPy4siA@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit

Hi,

在 2024/02/27 10:23, Xiao Ni 写道:
> On Thu, Feb 22, 2024 at 4:04 PM Yu Kuai <yukuai1@huaweicloud.com> wrote:
>>
>> From: Yu Kuai <yukuai3@huawei.com>
>>
>> Commit 12cee5a8a29e ("md/raid1: prevent merging too large request") add
>> the case choose next idle in read_balance():
>>
>> read_balance:
>>   for_each_rdev
>>    if(next_seq_sect == this_sector || disk == 0)
>>    -> sequential reads
>>     best_disk = disk;
>>     if (...)
>>      choose_next_idle = 1
>>      continue;
>>
>>   for_each_rdev
>>   -> iterate next rdev
>>    if (pending == 0)
>>     best_disk = disk;
>>     -> choose the next idle disk
>>     break;
>>
>>    if (choose_next_idle)
>>     -> keep using this rdev if there are no other idle disk
>>     contine
>>
>> However, commit 2e52d449bcec ("md/raid1: add failfast handling for reads.")
>> remove the code:
>>
>> -               /* If device is idle, use it */
>> -               if (pending == 0) {
>> -                       best_disk = disk;
>> -                       break;
>> -               }
>>
>> Hence choose next idle will never work now, fix this problem by
>> following:
>>
>> 1) don't set best_disk in this case, read_balance() will choose the best
>>     disk after iterating all the disks;
>> 2) add 'pending' so that other idle disk will be chosen;
>> 3) set 'dist' to 0 so that if there is no other idle disk, and all disks
>>     are rotational, this disk will still be chosen;
>>
>> Fixes: 2e52d449bcec ("md/raid1: add failfast handling for reads.")
>> Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com>
>> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com>
>> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
>> ---
>>   drivers/md/raid1.c | 21 ++++++++++++---------
>>   1 file changed, 12 insertions(+), 9 deletions(-)
>>
>> diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
>> index c60ea58ae8c5..d0bc67e6d068 100644
>> --- a/drivers/md/raid1.c
>> +++ b/drivers/md/raid1.c
>> @@ -604,7 +604,6 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
>>          unsigned int min_pending;
>>          struct md_rdev *rdev;
>>          int choose_first;
>> -       int choose_next_idle;
>>
>>          /*
>>           * Check if we can balance. We can balance on the whole
>> @@ -619,7 +618,6 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
>>          best_pending_disk = -1;
>>          min_pending = UINT_MAX;
>>          best_good_sectors = 0;
>> -       choose_next_idle = 0;
>>          clear_bit(R1BIO_FailFast, &r1_bio->state);
>>
>>          if ((conf->mddev->recovery_cp < this_sector + sectors) ||
>> @@ -712,7 +710,6 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
>>                          int opt_iosize = bdev_io_opt(rdev->bdev) >> 9;
>>                          struct raid1_info *mirror = &conf->mirrors[disk];
>>
>> -                       best_disk = disk;
>>                          /*
>>                           * If buffered sequential IO size exceeds optimal
>>                           * iosize, check if there is idle disk. If yes, choose
>> @@ -731,15 +728,21 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
>>                              mirror->next_seq_sect > opt_iosize &&
>>                              mirror->next_seq_sect - opt_iosize >=
>>                              mirror->seq_start) {
>> -                               choose_next_idle = 1;
>> -                               continue;
>> +                               /*
>> +                                * Add 'pending' to avoid choosing this disk if
>> +                                * there is other idle disk.
>> +                                * Set 'dist' to 0, so that if there is no other
>> +                                * idle disk and all disks are rotational, this
>> +                                * disk will still be chosen.
>> +                                */
>> +                               pending++;
>> +                               dist = 0;
>> +                       } else {
>> +                               best_disk = disk;
>> +                               break;
>>                          }
>> -                       break;
>>                  }
> 
> Hi Kuai
> 
> I noticed something. In patch 12cee5a8a29e, it sets best_disk if it's
> a sequential read. If there are no other idle disks, it will read from
> the sequential disk. With this patch, it reads from the
> best_pending_disk even min_pending is not 0. It looks like a wrong
> behaviour?

Yes, nice catch, I didn't notice this yet... So there is a hidden
logical, sequential IO priority is higher than minimal 'pending'
selection, it's only less than 'choose_next_idle' where idle disk
exist.

Looks like if we want to keep this behaviour, we can add a 'sequential
disk':

if (is_sequential())
  if (!should_choose_next())
   return disk;
  ctl.sequential_disk = disk;

..

if (ctl.min_pending != 0 && ctl.sequential_disk != -1)
  return ctl.sequential_disk;

Thanks,
Kuai

> 
> Best Regards
> Xiao
>>
>> -               if (choose_next_idle)
>> -                       continue;
>> -
>>                  if (min_pending > pending) {
>>                          min_pending = pending;
>>                          best_pending_disk = disk;
>> --
>> 2.39.2
>>
>>
> 
> .
>