On Mon, Jul 19, 2021 at 08:58:44PM +, John Johnson wrote:
>
>
> > On Jul 19, 2021, at 12:59 PM, Peter Xu wrote:
> >
> >
> > Here vfio_user_send_recv() looks tricky to me - it releases the bql within a
> > memory update procedure, and IMHO it needs some serious justification on
> > why it
> On Jul 19, 2021, at 12:59 PM, Peter Xu wrote:
>
>
> Here vfio_user_send_recv() looks tricky to me - it releases the bql within a
> memory update procedure, and IMHO it needs some serious justification on why
> it
> can do so. For example, what if memory layout changed when waiting for the
c: Paolo Bonzini ; John Levon
> > ; John G Johnson ;
> > Markus Armbruster ; QEMU Devel Mailing List
> >
> > Subject: Re: Question on memory commit during MR finalize()
> >
> > Omg I don't know how I missed that, of course I'll ignore SIGUSR1 and
> > rete
1, 16:58
> To: Thanos Makatos
> Cc: Paolo Bonzini; Markus Armbruster; QEMU Devel Mailing List; John Levon;
> John G Johnson
> Subject: Re: Question on memory commit during MR finalize()
>
>
> Hi, Thanos,
>
> On Mon, Jul 19, 2021 at 02:38:52PM +, Thanos Makatos wrote:
&g
on memory commit during MR finalize()
Hi, Thanos,
On Mon, Jul 19, 2021 at 02:38:52PM +, Thanos Makatos wrote:
> I can trivially trigger an assertion with a build where I merged the recent
> vfio-user patches
> (https://urldefense.proofpoint.com/v2/url?u=https-3A__patchew.org_Q
Hi, Thanos,
On Mon, Jul 19, 2021 at 02:38:52PM +, Thanos Makatos wrote:
> I can trivially trigger an assertion with a build where I merged the recent
> vfio-user patches
> (https://patchew.org/QEMU/cover.1626675354.git.elena.ufimts...@oracle.com/)
> to master and then merging the result
> -Original Message-
> From: Peter Xu
> Sent: 16 July 2021 15:19
> To: Thanos Makatos
> Cc: Paolo Bonzini ; Markus Armbruster
> ; QEMU Devel Mailing List de...@nongnu.org>; John Levon ; John G
> Johnson
> Subject: Re: Question on memory commit during MR fin
u.org>; John Levon ; John G
> > Johnson
> > Subject: Re: Question on memory commit during MR finalize()
> >
> > On Thu, Jul 15, 2021 at 02:27:48PM +, Thanos Makatos wrote:
> > > Hi Peter,
> >
> > Hi, Thanos,
> >
> > > We're hitting t
> -Original Message-
> From: Peter Xu
> Sent: 15 July 2021 19:35
> To: Thanos Makatos
> Cc: Paolo Bonzini ; Markus Armbruster
> ; QEMU Devel Mailing List de...@nongnu.org>; John Levon ; John G
> Johnson
> Subject: Re: Question on memory commit during MR fin
On Thu, Jul 15, 2021 at 02:27:48PM +, Thanos Makatos wrote:
> Hi Peter,
Hi, Thanos,
> We're hitting this issue using a QEMU branch where JJ is using vfio-user as
> the transport for multiprocess-qemu
> (https://github.com/oracle/qemu/issues/9). We can reproduce it fairly
> reliably by
Hi Peter,
> -Original Message-
> From: Qemu-devel bounces+thanos.makatos=nutanix@nongnu.org> On Behalf Of Peter
> Xu
> Sent: 21 April 2020 11:44
> To: Paolo Bonzini
> Cc: Markus Armbruster ; QEMU Devel Mailing List
>
> Subject: Re: Question on memor
On Tue, Apr 21, 2020 at 11:43:36AM +0200, Paolo Bonzini wrote:
> On 21/04/20 01:31, Peter Xu wrote:
> >>
> >> However, instead of memory_region_transaction_commit,
> >> memory_region_finalize probably should do something like
> >>
> >> --memory_region_transaction_depth;
> >> assert
On 21/04/20 01:31, Peter Xu wrote:
>>
>> However, instead of memory_region_transaction_commit,
>> memory_region_finalize probably should do something like
>>
>> --memory_region_transaction_depth;
>> assert (memory_region_transaction_depth ||
>> (!memory_region_update_pending &&
>>
On Mon, Apr 20, 2020 at 11:44:11PM +0200, Paolo Bonzini wrote:
> On 20/04/20 23:00, Peter Xu wrote:
> >
> > I'm still uncertain how the dirty ring branch can easily trigger this,
> > however
> > the backtrace looks really odd to me in that we're going to do memory commit
> > and even sending KVM
On 20/04/20 23:00, Peter Xu wrote:
>
> I'm still uncertain how the dirty ring branch can easily trigger this, however
> the backtrace looks really odd to me in that we're going to do memory commit
> and even sending KVM ioctls during finalize(), especially in the RCU thread...
> I never expected
15 matches
Mail list logo