Re: [Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-07 Thread Fam Zheng
On Wed, 05/07 11:06, Kevin Wolf wrote: > Am 07.05.2014 um 10:57 hat Fam Zheng geschrieben: > > On Wed, 05/07 10:20, Kevin Wolf wrote: > > > Am 07.05.2014 um 03:45 hat Fam Zheng geschrieben: > > > > On Tue, 05/06 10:32, Fam Zheng wrote: > > > > > On mounted NFS filesystem, ftruncate is much much slo

Re: [Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-07 Thread Kevin Wolf
Am 07.05.2014 um 10:57 hat Fam Zheng geschrieben: > On Wed, 05/07 10:20, Kevin Wolf wrote: > > Am 07.05.2014 um 03:45 hat Fam Zheng geschrieben: > > > On Tue, 05/06 10:32, Fam Zheng wrote: > > > > On mounted NFS filesystem, ftruncate is much much slower than doing a > > > > zero write. Changing thi

Re: [Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-07 Thread Fam Zheng
On Wed, 05/07 10:20, Kevin Wolf wrote: > Am 07.05.2014 um 03:45 hat Fam Zheng geschrieben: > > On Tue, 05/06 10:32, Fam Zheng wrote: > > > On mounted NFS filesystem, ftruncate is much much slower than doing a > > > zero write. Changing this significantly speeds up cluster allocation. > > > > > > C

Re: [Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-07 Thread Kevin Wolf
Am 07.05.2014 um 03:45 hat Fam Zheng geschrieben: > On Tue, 05/06 10:32, Fam Zheng wrote: > > On mounted NFS filesystem, ftruncate is much much slower than doing a > > zero write. Changing this significantly speeds up cluster allocation. > > > > Comparing by converting a cirros image (296M) to VMD

Re: [Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-07 Thread Stefan Hajnoczi
On Wed, May 07, 2014 at 09:45:17AM +0800, Fam Zheng wrote: > On Tue, 05/06 10:32, Fam Zheng wrote: > > @@ -1110,12 +,20 @@ static int get_cluster_offset(BlockDriverState *bs, > > } > > > > /* Avoid the L2 tables update for the images that have snapshots. > > */ > > -

Re: [Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-06 Thread Fam Zheng
On Tue, 05/06 10:32, Fam Zheng wrote: > On mounted NFS filesystem, ftruncate is much much slower than doing a > zero write. Changing this significantly speeds up cluster allocation. > > Comparing by converting a cirros image (296M) to VMDK on an NFS mount > point, over 1Gbe LAN: > > $ time qe

[Qemu-devel] [PATCH v2] vmdk: Optimize cluster allocation

2014-05-05 Thread Fam Zheng
On mounted NFS filesystem, ftruncate is much much slower than doing a zero write. Changing this significantly speeds up cluster allocation. Comparing by converting a cirros image (296M) to VMDK on an NFS mount point, over 1Gbe LAN: $ time qemu-img convert cirros-0.3.1.img /mnt/a.raw -O vmdk