[snip]
Thanks, Paul.
Well, I agree the former approach would be simpler. But I still doubt
if this is more reasonable. :)
IIUC, one of the reasons for struct domain to have a rangeset list(and
a spinlock - rangesets_lock), is because there are iomem_caps and
irq_caps for each domain. These 2 ra
oject.org; Andrew Cooper;
> >> jbeul...@suse.com; Kevin Tian; zhiyuan...@intel.com
> >> Subject: Re: [Xen-devel] [PATCH] Refactor ioreq server for better
> >> performance
> >>
> >> Thanks you, Paul.
> >>
> >
> > No problem :-)
> &
] Refactor ioreq server for better
performance
Thanks you, Paul.
No problem :-)
[snip]
I think assigning a name to the rangeset and having a debug-key dump is
useful. Can you not duplicate that in your new implementation?
Well, I can add some dump routines, e.g
> -Original Message-
> From: Yu, Zhang [mailto:yu.c.zh...@linux.intel.com]
> Sent: 30 June 2015 08:11
> To: Paul Durrant; xen-de...@lists.xenproject.org; Andrew Cooper;
> jbeul...@suse.com; Kevin Tian; zhiyuan...@intel.com
> Subject: Re: [Xen-devel] [PATCH] Refactor ioreq
Thanks you, Paul.
On 6/29/2015 8:12 PM, Paul Durrant wrote:
-Original Message-
From: Yu Zhang [mailto:yu.c.zh...@linux.intel.com]
Sent: 26 June 2015 11:30
To: xen-de...@lists.xenproject.org; Paul Durrant; Andrew Cooper;
jbeul...@suse.com; Kevin Tian; zhiyuan...@intel.com
Subject: [PATCH]
> -Original Message-
> From: Yu Zhang [mailto:yu.c.zh...@linux.intel.com]
> Sent: 26 June 2015 11:30
> To: xen-de...@lists.xenproject.org; Paul Durrant; Andrew Cooper;
> jbeul...@suse.com; Kevin Tian; zhiyuan...@intel.com
> Subject: [PATCH] Refactor ioreq server for better performance
>
>
XenGT leverages ioreq server to track and forward the accesses to
GPU IO resources, e.g. the PPGTT(per-process graphic translation
tables). Currently, ioreq server uses rangeset to track the BDF/
PIO/MMIO ranges to be emulated. To select an ioreq server, the
rangeset is searched to see if the IO
