Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH] iomap: Fix the write_count in iomap_add_to_ioend().
To:     Christoph Hellwig <hch@infradead.org>,
        Anju T Sudhakar <anju@linux.vnet.ibm.com>,
        linux-block@vger.kernel.org
Cc:     darrick.wong@oracle.com, linux-xfs@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        willy@infradead.org
References: <20200819102841.481461-1-anju@linux.vnet.ibm.com>
 <20200821060710.GC31091@infradead.org>
From:   Ritesh Harjani <riteshh@linux.ibm.com>
Date:   Fri, 21 Aug 2020 14:23:55 +0530
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.8.0
MIME-Version: 1.0
In-Reply-To: <20200821060710.GC31091@infradead.org>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Message-Id: <20200821085356.93CF64C040@d06av22.portsmouth.uk.ibm.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 8/21/20 11:37 AM, Christoph Hellwig wrote:
> On Wed, Aug 19, 2020 at 03:58:41PM +0530, Anju T Sudhakar wrote:
>> From: Ritesh Harjani <riteshh@linux.ibm.com>
>>
>> __bio_try_merge_page() may return same_page = 1 and merged = 0.
>> This could happen when bio->bi_iter.bi_size + len > UINT_MAX.
>> Handle this case in iomap_add_to_ioend() by incrementing write_count.
>> This scenario mostly happens where we have too much dirty data accumulated.
>>
>> w/o the patch we hit below kernel warning,
> 
> I think this is better fixed in the block layer rather than working
> around the problem in the callers.  Something like this:
> 
> diff --git a/block/bio.c b/block/bio.c
> index c63ba04bd62967..ef321cd1072e4e 100644
> --- a/block/bio.c
> +++ b/block/bio.c
> @@ -879,8 +879,10 @@ bool __bio_try_merge_page(struct bio *bio, struct page *page,
>   		struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt - 1];
>   
>   		if (page_is_mergeable(bv, page, len, off, same_page)) {
> -			if (bio->bi_iter.bi_size > UINT_MAX - len)
> +			if (bio->bi_iter.bi_size > UINT_MAX - len) {
> +				*same_page = false;
>   				return false;
> +			}
>   			bv->bv_len += len;
>   			bio->bi_iter.bi_size += len;
>   			return true;
> 

Ya, we had think of that. But what we then thought was, maybe the
API does return the right thing. Meaning, what API says is, same_page is
true, but the page couldn't be merged hence it returned ret = false.
With that thought, we fixed this in the caller.

But agree with you that with ret = false, there is no meaning of
same_page being true. Ok, so let linux-block comment on whether
above also looks good. If yes, I can spin out an official patch with
all details.

-ritesh