Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 2 Aug 2023 11:41:06 +0200
From:   Christoph Hellwig <hch@lst.de>
To:     Dusty Mabe <dusty@dustymabe.com>
Cc:     Minchan Kim <minchan@kernel.org>,
        Sergey Senozhatsky <senozhatsky@chromium.org>,
        Jens Axboe <axboe@kernel.dk>, linux-kernel@vger.kernel.org,
        linux-block@vger.kernel.org, Christoph Hellwig <hch@lst.de>,
        Andrew Morton <akpm@linux-foundation.org>, marmijo@redhat.com
Subject: Re: XFS metadata CRC errors on zram block device on ppc64le
 architecture
Message-ID: <20230802094106.GA28187@lst.de>
Reply-To: wq@lst.de
References: <b2d40565-7868-ba15-4bb1-fca6f0df076b@dustymabe.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <b2d40565-7868-ba15-4bb1-fca6f0df076b@dustymabe.com>
User-Agent: Mutt/1.5.17 (2007-11-01)
Precedence: bulk

On Tue, Aug 01, 2023 at 11:31:37PM -0400, Dusty Mabe wrote:
> We ran a kernel bisect and narrowed it down to offending commit af8b04c6:
> 
> ```
> [root@ibm-p8-kvm-03-guest-02 linux]# git bisect good
> af8b04c63708fa730c0257084fab91fb2a9cecc4 is the first bad commit
> commit af8b04c63708fa730c0257084fab91fb2a9cecc4
> Author: Christoph Hellwig <hch@lst.de>
> Date:   Tue Apr 11 19:14:46 2023 +0200
> 
>     zram: simplify bvec iteration in __zram_make_request
>     
>     bio_for_each_segment synthetize bvecs that never cross page boundaries, so
>     don't duplicate that work in an inner loop.

> Any ideas on how to fix the problem?

So the interesting cases are:

  - ppc64 usually uses 64k page sizes
  - ppc64 is somewhat cache incoherent (compared to say x86)

Let me think of this a bit more.