Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-10-30 Thread Konrad Rzeszutek Wilk
On Tue, Oct 30, 2012 at 07:33:16PM +0100, Roger Pau Monné wrote: > On 30/10/12 18:01, Konrad Rzeszutek Wilk wrote: > > On Wed, Oct 24, 2012 at 06:58:45PM +0200, Roger Pau Monne wrote: > >> This patch implements persistent grants for the xen-blk{front,back} > >> mechanism. The effect of this change

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-10-29 Thread Konrad Rzeszutek Wilk
On Wed, Oct 24, 2012 at 06:58:45PM +0200, Roger Pau Monne wrote: > This patch implements persistent grants for the xen-blk{front,back} > mechanism. The effect of this change is to reduce the number of unmap > operations performed, since they cause a (costly) TLB shootdown. This > allows the I/O per

[PATCH v2] Persistent grant maps for xen blk drivers

2012-10-24 Thread Roger Pau Monne
This patch implements persistent grants for the xen-blk{front,back} mechanism. The effect of this change is to reduce the number of unmap operations performed, since they cause a (costly) TLB shootdown. This allows the I/O performance to scale better when a large number of VMs are performing I/O.

Re: [Xen-devel] [PATCH v2] Persistent grant maps for xen blk drivers

2012-10-09 Thread Roger Pau Monné
On 24/09/12 13:36, Jan Beulich wrote: On 21.09.12 at 17:52, Oliver Chick wrote: >> Changes since v1: >> >> * Maximum number of persistent grants per device now 64, rather than >>256, as this is the actual maxmimum request in a (1 page) ring. > > As said previously, I don't see why this

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-27 Thread Oliver Chick
Hi Konrad, I have applied my patch to your #linux-next, and compiled. I don't experience the same problem that you do - both PV and HVM boot and work. Do you have anything special in your setup that might help me reproduce the problem? On Fri, 2012-09-21 at 19:41 +0100, Konrad Rzeszutek Wilk wr

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-24 Thread Andres Lagar-Cavilla
On Sep 24, 2012, at 11:06 AM, Konrad Rzeszutek Wilk wrote: > On Mon, Sep 24, 2012 at 10:38:48AM -0400, Andres Lagar-Cavilla wrote: >> On Sep 21, 2012, at 4:46 PM, Konrad Rzeszutek Wilk wrote: >> >>> On Fri, Sep 21, 2012 at 02:56:22PM -0400, Konrad Rzeszutek Wilk wrote: > *: With a PVHVM guest

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-24 Thread Konrad Rzeszutek Wilk
On Mon, Sep 24, 2012 at 10:38:48AM -0400, Andres Lagar-Cavilla wrote: > On Sep 21, 2012, at 4:46 PM, Konrad Rzeszutek Wilk wrote: > > > On Fri, Sep 21, 2012 at 02:56:22PM -0400, Konrad Rzeszutek Wilk wrote: > >>> *: With a PVHVM guest I get > >>> > >>> [ 261.927218] privcmd_fault: vma=88002a

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-24 Thread Andres Lagar-Cavilla
On Sep 21, 2012, at 4:46 PM, Konrad Rzeszutek Wilk wrote: > On Fri, Sep 21, 2012 at 02:56:22PM -0400, Konrad Rzeszutek Wilk wrote: >>> *: With a PVHVM guest I get >>> >>> [ 261.927218] privcmd_fault: vma=88002a31dce8 >>> 7f4edc095000-7f4edc195000, pgoff=c8, uv=7f4edc15d000 >>> >>> thou

Re: [Xen-devel] [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-24 Thread Ian Campbell
On Mon, 2012-09-24 at 12:36 +0100, Jan Beulich wrote: > >>> On 21.09.12 at 17:52, Oliver Chick wrote: > > Changes since v1: > > > > * Maximum number of persistent grants per device now 64, rather than > >256, as this is the actual maxmimum request in a (1 page) ring. > > As said previously,

Re: [Xen-devel] [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-24 Thread Jan Beulich
>>> On 21.09.12 at 17:52, Oliver Chick wrote: > Changes since v1: > > * Maximum number of persistent grants per device now 64, rather than >256, as this is the actual maxmimum request in a (1 page) ring. As said previously, I don't see why this needs to have a separate #define at all - it's

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-21 Thread Konrad Rzeszutek Wilk
On Fri, Sep 21, 2012 at 02:56:22PM -0400, Konrad Rzeszutek Wilk wrote: > > *: With a PVHVM guest I get > > > > [ 261.927218] privcmd_fault: vma=88002a31dce8 > > 7f4edc095000-7f4edc195000, pgoff=c8, uv=7f4edc15d000 > > > > thought if I applied your patch on top of v3.6-rc6 I didn't see t

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-21 Thread Konrad Rzeszutek Wilk
> *: With a PVHVM guest I get > > [ 261.927218] privcmd_fault: vma=88002a31dce8 7f4edc095000-7f4edc195000, > pgoff=c8, uv=7f4edc15d000 > > thought if I applied your patch on top of v3.6-rc6 I didn't see the > privcmd_fault but > I did see the guest crash. And that is due to c571898ffc

Re: [PATCH v2] Persistent grant maps for xen blk drivers

2012-09-21 Thread Konrad Rzeszutek Wilk
On Fri, Sep 21, 2012 at 04:52:47PM +0100, Oliver Chick wrote: > This patch implements persistent grants for the xen-blk{front,back} > mechanism. The effect of this change is to reduce the number of unmap > operations performed, since they cause a (costly) TLB shootdown. This > allows the I/O perfor

[PATCH v2] Persistent grant maps for xen blk drivers

2012-09-21 Thread Oliver Chick
This patch implements persistent grants for the xen-blk{front,back} mechanism. The effect of this change is to reduce the number of unmap operations performed, since they cause a (costly) TLB shootdown. This allows the I/O performance to scale better when a large number of VMs are performing I/O.