From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
To: Ian Campbell <Ian.Campbell@citrix.com>,
Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Cc: "Wei Liu (Intern)" <wei.liu2@citrix.com>,
xen-devel@lists.xensource.com,
Ian Jackson <ian.jackson@eu.citrix.com>,
linux@eikelenboom.it
Subject: Re: [PATCH for-4.5] reset PCI devices on force removal even when QEMU returns error
Date: Tue, 02 Dec 2014 10:09:26 -0500 [thread overview]
Message-ID: <547DD626.3070600@oracle.com> (raw)
In-Reply-To: <1417527800.24320.29.camel@citrix.com>
On 12/02/2014 08:43 AM, Ian Campbell wrote:
> On Fri, 2014-11-28 at 16:53 +0000, Stefano Stabellini wrote:
>
> CCing Boris because he was fixing a similar sounding issue in the same
> area. Not sure if this is related to those patches or not.
Not really. I was trying to fix something that another Stefano's patch
from yesterday is trying to address:
http://lists.xenproject.org/archives/html/xen-devel/2014-12/msg00160.html
although I thought that removing the call to xc_domain_irq_permission()
would be the solution.
(which reminds me --- and maybe I shouldn't overload this thread ---
that this patch from IanJ needs to get in as well:
http://lists.xenproject.org/archives/html/xen-devel/2014-11/msg01342.html
)
Thanks.
-boris
>
>> On do_pci_remove when QEMU returns error, we just bail out early without
>> resetting the device. On domain shutdown we are racing with QEMU exiting
>> and most often QEMU closes the QMP connection before executing the
>> requested command.
>>
>> In these cases if force=1, it makes sense to go ahead with rest of the
>> PCI device removal, that includes resetting the device and calling
>> xc_deassign_device. Otherwise we risk not resetting the device properly
>> on domain shutdown.
> ISTR seeing a conversation along the lines of there being a better
> solution which was more 4.6 material, is that right?
>
>> Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> Acked-by: Ian Campbell <ian.campbell@citrix.com>
>
> But, I'd prefer to see a version which logs when the qemu side has
> failed but it is continuing. Probably just adding right after the
> existing if:
> if (rc)
> LOG("Something appropriate");
> would do the trick.
>
> Also this needs RM input from Konrad.
>
>> diff --git a/tools/libxl/libxl_pci.c b/tools/libxl/libxl_pci.c
>> index 316643c..0ac0b93 100644
>> --- a/tools/libxl/libxl_pci.c
>> +++ b/tools/libxl/libxl_pci.c
>> @@ -1243,7 +1245,7 @@ static int do_pci_remove(libxl__gc *gc, uint32_t domid,
>> rc = ERROR_INVAL;
>> goto out_fail;
>> }
>> - if (rc) {
>> + if (rc && !force) {
>> rc = ERROR_FAIL;
>> goto out_fail;
>> }
>
next prev parent reply other threads:[~2014-12-02 15:09 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-28 16:53 [PATCH for-4.5] reset PCI devices on force removal even when QEMU returns error Stefano Stabellini
2014-12-02 13:43 ` Ian Campbell
2014-12-02 15:09 ` Boris Ostrovsky [this message]
2014-12-12 15:13 ` Sander Eikelenboom
2014-12-12 16:50 ` Konrad Rzeszutek Wilk
2014-12-15 11:13 ` Stefano Stabellini
2014-12-17 21:07 ` Konrad Rzeszutek Wilk
2015-01-06 11:45 ` Ian Campbell
2015-01-06 14:02 ` Sander Eikelenboom
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=547DD626.3070600@oracle.com \
--to=boris.ostrovsky@oracle.com \
--cc=Ian.Campbell@citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=linux@eikelenboom.it \
--cc=stefano.stabellini@eu.citrix.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.