similar to: kernel vhost demands an interrupt from guest when the ring is full in order to enable guest to submit new packets to the queue

?On 12/17/18, 2:55 PM, "Michael S. Tsirkin" <mst at redhat.com> wrote: On Thu, Dec 13, 2018 at 11:24:28PM +0000, Steven Luong (sluong) via Virtualization wrote: > Folks, > > > > We came across a memory race condition between VPP vhost driver and the kernel > vhost. VPP is running a tap interface over vhost backend. In this case,

On Thu, Dec 13, 2018 at 11:24:28PM +0000, Steven Luong (sluong) via Virtualization wrote: > Folks, > > > > We came across a memory race condition between VPP vhost driver and the kernel > vhost. VPP is running a tap interface over vhost backend. In this case, VPP is > acting as the vhost driver mode and the kernel vhost is acting as the vhost > device mode. > >

Patches before the last are small cleanups. In the last patch I'm trying to extract / generalize an idea from Stefan Hajnoczi's review of my virtio-net driver for OVMF. Unfortunately I can't find Stefan's email on any mailing list archive (sourceforge, gmane, mail-archive etc. all have only my response), so I'll quote it here. The patch Stefan was reviewing is

Patches before the last are small cleanups. In the last patch I'm trying to extract / generalize an idea from Stefan Hajnoczi's review of my virtio-net driver for OVMF. Unfortunately I can't find Stefan's email on any mailing list archive (sourceforge, gmane, mail-archive etc. all have only my response), so I'll quote it here. The patch Stefan was reviewing is

Changes from v2: added padding between avail idx and flags, and changed virtio to only publish used index when callbacks are enabled. Here's a rewrite of the original patch with a new layout. I haven't tested it yet so no idea how this performs, but I think this addresses the cache bounce issue raised by Avi. Posting for early flames/comments. Generally, the Host end of the virtio ring

Changes from v2: added padding between avail idx and flags, and changed virtio to only publish used index when callbacks are enabled. Here's a rewrite of the original patch with a new layout. I haven't tested it yet so no idea how this performs, but I think this addresses the cache bounce issue raised by Avi. Posting for early flames/comments. Generally, the Host end of the virtio ring

Here's a rewrite of the original patch with a new layout. I haven't tested it yet so no idea how this performs, but I think this addresses the cache bounce issue raised by Avi. Posting for early flames/comments. Generally, the Host end of the virtio ring doesn't need to see where Guest is up to in consuming the ring. However, to completely understand what's going on from the

Here's a rewrite of the original patch with a new layout. I haven't tested it yet so no idea how this performs, but I think this addresses the cache bounce issue raised by Avi. Posting for early flames/comments. Generally, the Host end of the virtio ring doesn't need to see where Guest is up to in consuming the ring. However, to completely understand what's going on from the

Hi. I would like to print out a field of a vnode, but the argument to the function is of type vnode_t **. If I set "this->vpp" to arg2 in an entry function to zfs_lookup(), I''d like to retrieve the value of one of the fields of *(this->vpp) in the return function. Something like the below (but the syntax below is not right): printf("zfs_lookup:

platform x86_64-unknown-linux-gnu arch x86_64 os linux-gnu system x86_64, linux-gnu status major 2 minor 6.2 year 2008 month 02 day

A virtio driver does virtqueue_add_buf() multiple times before finally calling virtqueue_kick(); previously we only exposed the added buffers in the virtqueue_kick() call. This means we don't need a memory barrier in virtqueue_add_buf(), but it reduces concurrency as the device (ie. host) can't see the buffers until the kick. Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>

A virtio driver does virtqueue_add_buf() multiple times before finally calling virtqueue_kick(); previously we only exposed the added buffers in the virtqueue_kick() call. This means we don't need a memory barrier in virtqueue_add_buf(), but it reduces concurrency as the device (ie. host) can't see the buffers until the kick. Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>

Hi Nikolay, I poked at src/lib/virtio/virtq.lua a bit - I was surprised to find no explicit CPU memory barriers in the virtq implementation. These are typically required when using virtio on smp machines - the spec actually mention where barriers are necessary. Are the barriers implicit somehow for lua? I'd be curious to learn. Thanks, -- MST

Hi Nikolay, I poked at src/lib/virtio/virtq.lua a bit - I was surprised to find no explicit CPU memory barriers in the virtq implementation. These are typically required when using virtio on smp machines - the spec actually mention where barriers are necessary. Are the barriers implicit somehow for lua? I'd be curious to learn. Thanks, -- MST

virtio allows drivers to suppress callbacks (ie. interrupts) for efficiency (no locking, it's just an optimization). There's a similar mechanism for the host to suppress notifications coming from the guest: in that case, we ignore the suppression if the ring is completely full. It turns out that life is simpler if the host similarly ignores callback suppression when the ring is

virtio allows drivers to suppress callbacks (ie. interrupts) for efficiency (no locking, it's just an optimization). There's a similar mechanism for the host to suppress notifications coming from the guest: in that case, we ignore the suppression if the ring is completely full. It turns out that life is simpler if the host similarly ignores callback suppression when the ring is

Hello I?m need fitt growth curve with data length-age. I want to evaluate which is the function that best predicts my data, to do so I compare the Akaikes of different models. I'm now need to evaluate if changing the initial values changes the parameters and which do not allow to estimate the model. To do this I use the function nls(); and I randomize the initial values (real positive number).

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I lost my DC with all fsmo-roles. I try to "seize" the roles to another DC. It worked four out of five roles: root at SVL-V-AD1:~# samba-tool fsmo seize --role=rid Attempting transfer... Transfer unsuccessful, seizing... FSMO seize of 'rid' role successful root at SVL-V-AD1:~# samba-tool fsmo seize --role=pdc Attempting

Anyone have problems with getaddrinfo under AIX4.3? I have built 2.5.0 under linux, HP and VPP without too much in the way of problems. However, AIX4.3 version builds OK but doesn't execute getaddrinfo correctly. Any ideas? Terry Raggett ------------- -- +-------------------+----------------------------------+---------------------+ | Terry Raggett | |

> -----Original Message----- > From: Michael S. Tsirkin [mailto:mst at redhat.com] > Sent: Tuesday, April 10, 2018 9:36 PM > To: Liang, Cunming <cunming.liang at intel.com> > Cc: Paolo Bonzini <pbonzini at redhat.com>; Bie, Tiwei <tiwei.bie at intel.com>; > Jason Wang <jasowang at redhat.com>; alex.williamson at redhat.com; > ddutile at redhat.com;