[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] device_add error handling regressed in v1.5.0
From: |
Markus Armbruster |
Subject: |
Re: [Qemu-devel] device_add error handling regressed in v1.5.0 |
Date: |
Mon, 20 Jan 2014 14:10:18 +0100 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.2 (gnu/linux) |
Markus Armbruster <address@hidden> writes:
> Paolo Bonzini <address@hidden> writes:
>
>> Il 17/01/2014 11:23, Markus Armbruster ha scritto:
>>> Watch this:
>>>
>>> $ upstream-qemu -nodefaults -S -display none -monitor stdio
>>> QEMU 1.7.50 monitor - type 'help' for more information
>>> (qemu) device_add e1000,netdev=xxx
>>> Property 'e1000.netdev' can't find value 'xxx'
>>> (qemu) info qtree
> [...]
>>> Segmentation fault (core dumped)
>>>
>>> Even though device_add failed, it still created a node in the qtree!
>>>
>>> Same issue observed with scsi-hd. Looks like a qdev problem, not a
>>> device problem.
>>>
>>> git-bisect blames this one:
>>>
>>> e0a83fc2c1582dc8d4453849852ebe6c258b7c3a is the first bad commit
>>> commit e0a83fc2c1582dc8d4453849852ebe6c258b7c3a
>>> Author: Paolo Bonzini <address@hidden>
>>> Date: Tue Apr 2 15:50:00 2013 +0200
>>>
>>> qom: do nothing on unparent of object without parent
>>>
>>> Otherwise, device_unparent will fail to get a canonical path of
>>> the object.
>>>
>>> Signed-off-by: Paolo Bonzini <address@hidden>
>>> Message-id: address@hidden
>>> Signed-off-by: Anthony Liguori <address@hidden>
>>>
>>
>> I think Amos had a fix for this.
>
> Amos, do you?
Found it: "[PATCH v2] qdev: add the device to the QOM tree before using
it to set a link" seems to fix it.