[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [Xen-devel] guest cannot boot up successfully using qem
From: |
Liu, SongtaoX |
Subject: |
Re: [Qemu-devel] [Xen-devel] guest cannot boot up successfully using qemu.git in Xen |
Date: |
Wed, 24 Apr 2013 08:52:44 +0000 |
latest commit:bb71623811686ce3c34ce724f073f5c5dd95f51b in qemu.git works for
this issue.
guest could boot up normally.
Thanks.
Regards
Carl(Songtao)
-----Original Message-----
From: Stefano Stabellini [mailto:address@hidden
Sent: Monday, April 22, 2013 5:50 PM
To: Liu, SongtaoX
Cc: address@hidden; address@hidden; Xu, YongweiX; Ren, Yongjie; Paolo Bonzini;
address@hidden; Sander Eikelenboom; Gerd Hoffmann
Subject: Re: [Xen-devel] guest cannot boot up successfully using qemu.git in Xen
On Mon, 22 Apr 2013, Liu, SongtaoX wrote:
> I encountered a problem that the guest cannot boot up successfully using
> qemu.git in Xen.
> after creating the guest by "xl cr xlexample.hvm", "xl list" shows the
> guest's vcpu has no status, and "xl vnc $dom_id" cannot connect to the guest.
>
> The qemu.git source is from git://git.qemu.org/qemu.git
>
> The qemu compile steps:
> ./configure --enable-xen --target-list=x86_64-softmmu
> --extra-cflags="-I/xen-unstable.hg/tools/include
> -I/xen-unstable.hg/tools/libxc -I/xen-unstable.hg/tools/xenstore"
> --extra-ldflags="-L/xen-unstable.hg/tools/libxc
> -L/xen-unstable.hg/tools/xenstore"
> make
> make install
>
> The xlexample.hvm includes the two lines:
> device_model_override= '/usr/local/bin/qemu-system-x86_64'
> device_model_version= 'qemu-xen'
>
> Using "git bisect", the commit: 0ca5aa4f4c4a8bcc73988dd52a536241d35e5223
> introduced the problem.
>
> commit 0ca5aa4f4c4a8bcc73988dd52a536241d35e5223
> Author: Paolo Bonzini <address@hidden>
> Date: Wed Apr 10 15:23:27 2013 +0200
>
> qemu-char: another io_add_watch_poll fix
>
> After attaching the source, we have to remove the reference we hold
> to it, because we do not hold anymore a pointer to the source.
>
> If we do not do this, removing the source will not finalize it and
> will not drop the "real" I/O watch source.
>
> This showed up when backporting the new flow control patches to older
> versions of QEMU that still used select. The whole select then failed
> with EBADF (poll instead will reporting POLLNVAL on a single pollfd)
> and QEMU froze.
>
> Signed-off-by: Paolo Bonzini <address@hidden>
> Message-id: address@hidden
> Signed-off-by: Anthony Liguori <address@hidden>
Sander already reported this issue:
http://marc.info/?l=qemu-devel&m=136614022215765&w=2
Some patches were posted to fix the issue, it should just be a matter of time
before they are committed.