Re: [Xen-devel] ITS emulation race conditions

2017-04-11 Thread Julien Grall
On 11/04/17 10:24, Julien Grall wrote: Hi Stefano, On 04/11/2017 12:01 AM, Stefano Stabellini wrote: On Mon, 10 Apr 2017, Andre Przywara wrote: Hi, On 10/04/17 00:12, André Przywara wrote: Hi, I wanted to run some ideas on how to prevent the race conditions we are facing with the ITS

Re: [Xen-devel] ITS emulation race conditions

2017-04-11 Thread Julien Grall
Hi Stefano, On 04/11/2017 12:01 AM, Stefano Stabellini wrote: On Mon, 10 Apr 2017, Andre Przywara wrote: Hi, On 10/04/17 00:12, André Przywara wrote: Hi, I wanted to run some ideas on how to prevent the race conditions we are facing with the ITS emulation and removing devices and/or LPIs. I

Re: [Xen-devel] ITS emulation race conditions

2017-04-10 Thread Stefano Stabellini
On Mon, 10 Apr 2017, Andre Przywara wrote: > Hi, > > On 10/04/17 00:12, André Przywara wrote: > > Hi, > > > > I wanted to run some ideas on how to prevent the race conditions we are > > facing with the ITS emulation and removing devices and/or LPIs. > > I think Stefano's idea of tagging a

Re: [Xen-devel] ITS emulation race conditions

2017-04-10 Thread Andre Przywara
Hi, On 10/04/17 00:12, André Przywara wrote: > Hi, > > I wanted to run some ideas on how to prevent the race conditions we are > facing with the ITS emulation and removing devices and/or LPIs. > I think Stefano's idea of tagging a discarded LPI is the key, but still > some details are left to be

[Xen-devel] ITS emulation race conditions

2017-04-09 Thread André Przywara
Hi, I wanted to run some ideas on how to prevent the race conditions we are facing with the ITS emulation and removing devices and/or LPIs. I think Stefano's idea of tagging a discarded LPI is the key, but still some details are left to be solved. I think we are dealing with two issues: 1) A

[Xen-devel] ITS emulation race conditions

2017-04-09 Thread Andre Przywara
Hi, I wanted to run some ideas on how to prevent the race conditions we are facing with the ITS emulation and removing devices and/or LPIs. I think Stefano's idea of tagging a discarded LPI is the key, but still some details are left to be solved. I think we are dealing with two issues: 1) A