linux-kernel - Re: [PATCH 1/1] block: Check the queue limit before bio submitting

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <3f30258d7f728757892b21637b7f8ba686269917.camel@mediatek.com>
Date:   Fri, 3 Nov 2023 09:05:16 +0000
From:   Ed Tsai (蔡宗軒) <Ed.Tsai@...iatek.com>
To:     "hch@...radead.org" <hch@...radead.org>
CC:     Will Shiu (許恭瑜) <Will.Shiu@...iatek.com>,
        Peter Wang (王信友) 
        <peter.wang@...iatek.com>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Alice Chao (趙珮均) 
        <Alice.Chao@...iatek.com>,
        "linux-mediatek@...ts.infradead.org" 
        <linux-mediatek@...ts.infradead.org>,
        wsd_upstream <wsd_upstream@...iatek.com>,
        "axboe@...nel.dk" <axboe@...nel.dk>,
        Casper Li (李中榮) <casper.li@...iatek.com>,
        Chun-Hung Wu (巫駿宏) 
        <Chun-hung.Wu@...iatek.com>,
        Powen Kao (高伯文) <Powen.Kao@...iatek.com>,
        Naomi Chu (朱詠田) <Naomi.Chu@...iatek.com>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        Stanley Chu (朱原陞) 
        <stanley.chu@...iatek.com>,
        "matthias.bgg@...il.com" <matthias.bgg@...il.com>,
        "angelogioacchino.delregno@...labora.com" 
        <angelogioacchino.delregno@...labora.com>
Subject: Re: [PATCH 1/1] block: Check the queue limit before bio submitting

On Fri, 2023-11-03 at 01:15 -0700, Christoph Hellwig wrote:
> You need to look into the splitting code to see why the splits are
> suboptimal for your device.  We can't limit the upper bio size as
> we have code relying on it not having arbitrary limits.
> 

The splitting code is working correctly based on the queue limit.
However, since 5.1, full bios are no longer a fixed size and often do
not align with the queue limit. This results in the block layer
splitting some small fragments when submitting consecutive full bios. I
don't think this is specific to a particular device.

While I understand your point about not limiting the bio size, in the
past, the 1MB limit prevented the generation of fragmented bios during
the splitting.