Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 28 Oct 2022 16:24:52 +0530
From:   Ojaswin Mujoo <ojaswin@linux.ibm.com>
To:     Guoqing Jiang <guoqing.jiang@linux.dev>
Cc:     Jason Yan <yanaijie@huawei.com>, tytso@mit.edu,
        adilger.kernel@dilger.ca, linux-ext4@vger.kernel.org
Subject: Re: [PATCH] ext4: make ext4_mb_initialize_context return void
Message-ID: <Y1u05l3L/gb/cSYg@li-bb2b2a4c-3307-11b2-a85c-8fa5c3a69313.ibm.com>
References: <20221027032435.27374-1-guoqing.jiang@linux.dev>
 <bf9dba6f-50c6-5ba8-31e3-b60de18105f1@huawei.com>
 <23c58b71-8dbc-b314-de53-31e2593f94d4@linux.dev>
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <23c58b71-8dbc-b314-de53-31e2593f94d4@linux.dev>
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
Precedence: bulk

On Thu, Oct 27, 2022 at 04:12:45PM +0800, Guoqing Jiang wrote:
> 
> 
> On 10/27/22 2:29 PM, Jason Yan wrote:
> > 
> > On 2022/10/27 11:24, Guoqing Jiang wrote:
> > > Change the return type to void since it always return 0, and no need
> > > to do the checking in ext4_mb_new_blocks.
> > > 
> > > Signed-off-by: Guoqing Jiang <guoqing.jiang@linux.dev>
> > > ---
> > > ? fs/ext4/mballoc.c | 10 ++--------
> > > ? 1 file changed, 2 insertions(+), 8 deletions(-)
> > > 
> > > diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> > > index 9dad93059945..5b2ae37a8b80 100644
> > > --- a/fs/ext4/mballoc.c
> > > +++ b/fs/ext4/mballoc.c
> > > @@ -5204,7 +5204,7 @@ static void ext4_mb_group_or_file(struct
> > > ext4_allocation_context *ac)
> > > ????? mutex_lock(&ac->ac_lg->lg_mutex);
> > > ? }
> > > ? -static noinline_for_stack int
> > > +static noinline_for_stack void
> > > ? ext4_mb_initialize_context(struct ext4_allocation_context *ac,
> > > ????????????????? struct ext4_allocation_request *ar)
> > > ? {
> > > @@ -5253,8 +5253,6 @@ ext4_mb_initialize_context(struct
> > > ext4_allocation_context *ac,
> > > ????????????? (unsigned) ar->lleft, (unsigned) ar->pleft,
> > > ????????????? (unsigned) ar->lright, (unsigned) ar->pright,
> > > ????????????? inode_is_open_for_write(ar->inode) ? "" : "non-");
> > > -??? return 0;
> > > -
> > > ? }
> > > ? ? static noinline_for_stack void
> > > @@ -5591,11 +5589,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
> > > ????????? goto out;
> > > ????? }
> > > ? -??? *errp = ext4_mb_initialize_context(ac, ar);
> > > -??? if (*errp) {
> > > -??????? ar->len = 0;
> > > -??????? goto out;
> > > -??? }
> > > +??? ext4_mb_initialize_context(ac, ar);
> > 
> > This changed the logic here slightly. *errp will not be intialized with
> > zero after this change. So we need to carefully check whether this will
> > cause any issues.
> 
> Yes, thanks for reminder. I think "*errp" is always set later with below.
> 
> https://elixir.bootlin.com/linux/v6.1-rc2/source/fs/ext4/mballoc.c#L5606
> https://elixir.bootlin.com/linux/v6.1-rc2/source/fs/ext4/mballoc.c#L5611
> https://elixir.bootlin.com/linux/v6.1-rc2/source/fs/ext4/mballoc.c#L5629
> https://elixir.bootlin.com/linux/v6.1-rc2/source/fs/ext4/mballoc.c#L5646
Hi Guoqing,

I agree, it seems to be intialized correctly later in the code. The
flow is something like:

  ext4_fsblk_t ext4_mb_new_blocks(...)
  {
      ...
      ext4_mb_initialize_context(ac, ar);
      ...
      if (!ext4_mb_use_preallocated(ac)) {
          *errp = ext4_mb_pa_alloc(ac);  // *errp init to 0 on success
          ...
      }

      if (likely(ac->ac_status == AC_STATUS_FOUND)) {
          // *errp init to 0 on success
          *errp = ext4_mb_mark_diskspace_used(ac, handle, reserv_clstrs);
          ...
      } else {
          ...
          *errp = -ENOSPC;
      }
      ...
  }

So it seems like this cleanup won't alter the behavior. Feel free to
add:

Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>

Regards,
ojaswin