Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Paolo Bonzini
Il 28/06/2013 18:17, Alex Bligh ha scritto: > > But as Stefano says: > > --On 28 June 2013 11:56:29 +0100 Stefano Stabellini > wrote: > >> This is PV only, at least for the moment: emulated disks always use >> writeback caching. >> From the performance point of view, making this change for IDE

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Alex Bligh
--On 28 June 2013 11:44:41 +0100 Ian Jackson wrote: Looks useful. Are you planning to do this for both emulated and pv disks? Emulated disks don't have the same problem because they don't try to use O_DIRECT on pages shared with the guest via the Xen grant table mechanism. I should have

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Stefano Stabellini
On Fri, 28 Jun 2013, Paolo Bonzini wrote: > Il 27/06/2013 20:16, Stefano Stabellini ha scritto: > > Support backend option "direct-io-safe". This is documented as > > follows in the Xen backend specification: > > > > * direct-io-safe > > * Values: 0/1 (boolean) > > * Default

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Stefano Stabellini
On Fri, 28 Jun 2013, Alex Bligh wrote: > Stefano, > > --On 27 June 2013 19:16:30 +0100 Stefano Stabellini > wrote: > > > * Therefore this option gives the backend permission to use > > * O_DIRECT, notwithstanding that bug. > > Looks useful. Are you planning to do this for both emula

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Ian Jackson
Paolo Bonzini writes ("Re: [PATCH v2] xen_disk: support "direct-io-safe" backend option"): > Il 27/06/2013 20:16, Stefano Stabellini ha scritto: > > The original proposal for a "cache" backend option has been dropped > > because it was believed too wide, especially considering that at the > > mome

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Ian Jackson
Alex Bligh writes ("Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option"): > Stefano, > --On 27 June 2013 19:16:30 +0100 Stefano Stabellini > wrote: > > > * Therefore this option gives the backend permission to use >

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Alex Bligh
Stefano, --On 27 June 2013 19:16:30 +0100 Stefano Stabellini wrote: * Therefore this option gives the backend permission to use * O_DIRECT, notwithstanding that bug. Looks useful. Are you planning to do this for both emulated and pv disks? -- Alex Bligh

Re: [Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-28 Thread Paolo Bonzini
Il 27/06/2013 20:16, Stefano Stabellini ha scritto: > Support backend option "direct-io-safe". This is documented as > follows in the Xen backend specification: > > * direct-io-safe > * Values: 0/1 (boolean) > * Default Value: 0 > * > * The underlying storage is not a

[Qemu-devel] [PATCH v2] xen_disk: support "direct-io-safe" backend option

2013-06-27 Thread Stefano Stabellini
Support backend option "direct-io-safe". This is documented as follows in the Xen backend specification: * direct-io-safe * Values: 0/1 (boolean) * Default Value: 0 * * The underlying storage is not affected by the direct IO memory * lifetime bug. See: *