search for: sys_req

Guys, Here is a little something to look at in the code. I have the following set: Run command 1: /opt/kde3/bin/ksnapshot <Super>Print Run command 2: /opt/kde3/bin/ksnapshot --current <Alt><Super><Sys_Req> The bug occurs when you choose 'Grab key combination' for <Alt><Super><Sys_Req>. When you enter the keystroke, the keystroke is clearly accepted/interpreted by the grab key combination routine, but the little rectangle that says "Please enter the new key combina...

...addresses, but virtio will break. > > > > > >> Signed-off-by: Jason Wang <jasowang at redhat.com> > > What limits amount of entries that kernel keeps around? > > It depends on guest working set I think. Looking at > http://dpdk.org/doc/guides/linux_gsg/sys_reqs.html: > > - For 2MB page size in guest, it suggests hugepages=1024 > - For 1GB page size, it suggests a hugepages=4 > > So I choose 2048 to make sure it can cover this. 4K page size is rather common, too. > > Do we want at least a mod parameter for this? > > Maybe....

...addresses, but virtio will break. > > > > > >> Signed-off-by: Jason Wang <jasowang at redhat.com> > > What limits amount of entries that kernel keeps around? > > It depends on guest working set I think. Looking at > http://dpdk.org/doc/guides/linux_gsg/sys_reqs.html: > > - For 2MB page size in guest, it suggests hugepages=1024 > - For 1GB page size, it suggests a hugepages=4 > > So I choose 2048 to make sure it can cover this. 4K page size is rather common, too. > > Do we want at least a mod parameter for this? > > Maybe....

...y. So looks ok? If not, looks not a problem to device IOTLB API itself. > > >> Signed-off-by: Jason Wang <jasowang at redhat.com> > What limits amount of entries that kernel keeps around? It depends on guest working set I think. Looking at http://dpdk.org/doc/guides/linux_gsg/sys_reqs.html: - For 2MB page size in guest, it suggests hugepages=1024 - For 1GB page size, it suggests a hugepages=4 So I choose 2048 to make sure it can cover this. > Do we want at least a mod parameter for this? Maybe. > >> --- >> drivers/vhost/net.c | 6 +- >> dr...

...his could be solved. > > > >>> >>>> Signed-off-by: Jason Wang <jasowang at redhat.com> >>> What limits amount of entries that kernel keeps around? >> It depends on guest working set I think. Looking at >> http://dpdk.org/doc/guides/linux_gsg/sys_reqs.html: >> >> - For 2MB page size in guest, it suggests hugepages=1024 >> - For 1GB page size, it suggests a hugepages=4 >> >> So I choose 2048 to make sure it can cover this. > 4K page size is rather common, too. I assume hugepages is used widely, and there's a...

On Fri, Mar 25, 2016 at 10:34:34AM +0800, Jason Wang wrote: > This patch tries to implement an device IOTLB for vhost. This could be > used with for co-operation with userspace(qemu) implementation of DMA > remapping. > > The idea is simple. When vhost meets an IOTLB miss, it will request > the assistance of userspace to do the translation, this is done > through: > >

On Fri, Mar 25, 2016 at 10:34:34AM +0800, Jason Wang wrote: > This patch tries to implement an device IOTLB for vhost. This could be > used with for co-operation with userspace(qemu) implementation of DMA > remapping. > > The idea is simple. When vhost meets an IOTLB miss, it will request > the assistance of userspace to do the translation, this is done > through: > >