[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CALjAwxgtuEv=d1TfvH3EdXqZs=_6F7TinyVhbPdUmTsdK_5ghw@mail.gmail.com>
Date: Mon, 2 Apr 2018 18:32:08 +0100
From: Sitsofe Wheeler <sitsofe@...il.com>
To: Jens Axboe <axboe@...nel.dk>
Cc: linux-kernel@...r.kernel.org, linux-block@...r.kernel.org,
Tejun Heo <tj@...nel.org>,
Stefan Haberland <sth@...ux.vnet.ibm.com>,
Jan Hoeppner <hoeppner@...ux.vnet.ibm.com>,
Bart Van Assche <Bart.VanAssche@....com>
Subject: Re: [BISECTED][REGRESSION] Hang while booting EeePC 900
On 2 April 2018 at 15:33, Jens Axboe <axboe@...nel.dk> wrote:
> On 4/1/18 8:35 AM, Sitsofe Wheeler wrote:
>>
>> While trying to boot an EeePC 900 on the latest git mainline kernel a
>> hang is encountered while systemd is starting services but this did
>> not happen with 4.15. A bisection narrowed it down to the commit
>> 358f70da49d77c43f2ca11b5da584213b2add29c ("blk-mq: make
>> blk_abort_request() trigger timeout path"). Here's the bisection log:
>>
>> git bisect bad 358f70da49d77c43f2ca11b5da584213b2add29c
>> # first bad commit: [358f70da49d77c43f2ca11b5da584213b2add29c] blk-mq:
>> make blk_abort_request() trigger timeout path
>
> Are you using scsi-mq?
Yes:
$ cat /sys/module/scsi_mod/parameters/use_blk_mq
Y
> Would probably also be handy to see a dmesg log from a 4.15 boot,
> and from a failed 4.16 boot.
Attached. See the dmesg-g67818d25738b-ok.txt (parent of
358f70da49d77c43f2ca11b5da584213b2add29c) for the dmesg on an OK
commit. See dmesg-g358f70da49d7-regression.txt for a dmesg of a boot
that hangs (matching commit 358f70da49d77c43f2ca11b5da584213b2add29c).
--
Sitsofe | http://sucs.org/~sits/
View attachment "dmesg-g358f70da49d7-regression.txt" of type "text/plain" (53605 bytes)
View attachment "dmesg-g67818d25738b-ok.txt" of type "text/plain" (53708 bytes)
Powered by blists - more mailing lists