Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 19 Feb 2020 16:04:57 +0800
From:   Gao Xiang <gaoxiang25@huawei.com>
To:     Chao Yu <yuchao0@huawei.com>
CC:     <linux-erofs@lists.ozlabs.org>,
        Matthew Wilcox <willy@infradead.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Miao Xie <miaoxie@huawei.com>
Subject: Re: [PATCH v2] erofs: convert workstn to XArray
Message-ID: <20200219080456.GA126843@architecture4>
References: <20200217033042.137855-1-gaoxiang25@huawei.com>
 <58f1ff26-e1f8-96a4-fa7b-ee86f972b0aa@huawei.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <58f1ff26-e1f8-96a4-fa7b-ee86f972b0aa@huawei.com>
User-Agent: Mutt/1.9.4 (2018-02-28)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Chao,

On Wed, Feb 19, 2020 at 03:46:16PM +0800, Chao Yu wrote:
> On 2020/2/17 11:30, Gao Xiang wrote:

[]

> > -	return err;
> > +	sbi = EROFS_SB(sb);
> > +repeat:
> > +	xa_lock(&sbi->managed_pslots);
> > +	pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index,
> > +			   NULL, grp, GFP_NOFS);
> > +	if (pre) {
> 
> It looks __xa_cmpxchg() could return negative value in case of failure, e.g.
> no memory case. We'd better handle that case and old valid workgroup separately?

Thanks, that is a quite good catch!

To be honest, I'm not quite sure whether __xa_cmpxchg
could fail due to no memory here (as a quick scan, it
will do __xas_nomem loop interally.)

But anyway, it needs bailing out all potential errors
by using xa_is_err(). Will do some research and send
v3 then.

Thanks,
Gao Xiang