Re: [PATCH] virtio-net: Remove inclusion of pci.h

2019-04-06 Thread David Miller
From: Yuval Shaia Date: Wed, 3 Apr 2019 11:20:45 +0300 > This header is not in use - remove it. > > Signed-off-by: Yuval Shaia Applied to net-next ___ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundatio

Re: [PATCH] virtio-net: Fix some minor formatting errors

2019-04-06 Thread David Miller
From: Yuval Shaia Date: Wed, 3 Apr 2019 12:10:13 +0300 > Signed-off-by: Yuval Shaia Applied to net-next ___ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [PATCH net v8] failover: allow name change on IFF_UP slave interfaces

2019-04-10 Thread David Miller
From: Si-Wei Liu Date: Mon, 8 Apr 2019 19:45:27 -0400 > When a netdev appears through hot plug then gets enslaved by a failover > master that is already up and running, the slave will be opened > right away after getting enslaved. Today there's a race that userspace > (udev) may fail to rename t

Re: [PATCH net] vhost: reject zero size iova range

2019-04-10 Thread David Miller
From: Jason Wang Date: Tue, 9 Apr 2019 12:10:25 +0800 > We used to accept zero size iova range which will lead a infinite loop > in translate_desc(). Fixing this by failing the request in this case. > > Reported-by: syzbot+d21e6e297322a900c...@syzkaller.appspotmail.com > Fixes: 6b1e6cc7 ("vhost

Re: [PATCH v2 2/8] vsock/virtio: free packets during the socket release

2019-05-10 Thread David Miller
From: Stefano Garzarella Date: Fri, 10 May 2019 14:58:37 +0200 > @@ -827,12 +827,20 @@ static bool virtio_transport_close(struct vsock_sock > *vsk) > > void virtio_transport_release(struct vsock_sock *vsk) > { > + struct virtio_vsock_sock *vvs = vsk->trans; > + struct virtio_vsock_bu

Re: [PATCH net] vhost: don't use kmap() to log dirty pages

2019-05-13 Thread David Miller
From: Jason Wang Date: Mon, 13 May 2019 01:27:45 -0400 > Vhost log dirty pages directly to a userspace bitmap through GUP and > kmap_atomic() since kernel doesn't have a set_bit_to_user() > helper. This will cause issues for the arch that has virtually tagged > caches. The way to fix is to keep u

Re: [PATCH v3] vsock/virtio: free packets during the socket release

2019-05-17 Thread David Miller
From: Stefano Garzarella Date: Fri, 17 May 2019 16:45:43 +0200 > When the socket is released, we should free all packets > queued in the per-socket list in order to avoid a memory > leak. > > Signed-off-by: Stefano Garzarella Applied and queued up for -stable. _

Re: [PATCH RESEND] vsock/virtio: Initialize core virtio vsock before registering the driver

2019-05-18 Thread David Miller
From: "Jorge E. Moreira" Date: Thu, 16 May 2019 13:51:07 -0700 > Avoid a race in which static variables in net/vmw_vsock/af_vsock.c are > accessed (while handling interrupts) before they are initialized. ... > Fixes: 22b5c0b63f32 ("vsock/virtio: fix kernel panic after device hot-unplug") > Cc: S

Re: [PATCH V2 0/4] Prevent vhost kthread from hogging CPU

2019-05-18 Thread David Miller
From: Jason Wang Date: Fri, 17 May 2019 00:29:48 -0400 > Hi: > > This series try to prevent a guest triggerable CPU hogging through > vhost kthread. This is done by introducing and checking the weight > after each requrest. The patch has been tested with reproducer of > vsock and virtio-net. Onl

Re: [PATCH 1/4] vsock/virtio: fix locking around 'the_virtio_vsock'

2019-05-29 Thread David Miller
From: Stefano Garzarella Date: Tue, 28 May 2019 12:56:20 +0200 > @@ -68,7 +68,13 @@ struct virtio_vsock { > > static struct virtio_vsock *virtio_vsock_get(void) > { > - return the_virtio_vsock; > + struct virtio_vsock *vsock; > + > + mutex_lock(&the_virtio_vsock_mutex); > + vs

Re: [PATCH net-next 0/6] vhost: accelerate metadata access

2019-05-30 Thread David Miller
From: Jason Wang Date: Fri, 24 May 2019 04:12:12 -0400 > This series tries to access virtqueue metadata through kernel virtual > address instead of copy_user() friends since they had too much > overheads like checks, spec barriers or even hardware feature > toggling like SMAP. This is done throug

Re: [PATCH net-next 0/6] vhost: accelerate metadata access

2019-05-30 Thread David Miller
From: "Michael S. Tsirkin" Date: Thu, 30 May 2019 14:13:28 -0400 > On Thu, May 30, 2019 at 11:07:30AM -0700, David Miller wrote: >> From: Jason Wang >> Date: Fri, 24 May 2019 04:12:12 -0400 >> >> > This series tries to access virtqueue metadata throug

Re: [PATCH v3 2/5] vsock/virtio: fix locking for fwd_cnt and buf_alloc

2019-06-02 Thread David Miller
From: Stefano Garzarella Date: Fri, 31 May 2019 15:39:51 +0200 > @@ -434,7 +434,9 @@ void virtio_transport_set_buffer_size(struct vsock_sock > *vsk, u64 val) > if (val > vvs->buf_size_max) > vvs->buf_size_max = val; > vvs->buf_size = val; > + spin_lock_bh(&vvs->rx_l

Re: [PATCH net-next] vsock: correct removal of socket from the list

2019-06-14 Thread David Miller
From: Sunil Muthuswamy Date: Thu, 13 Jun 2019 03:52:27 + > The current vsock code for removal of socket from the list is both > subject to race and inefficient. It takes the lock, checks whether > the socket is in the list, drops the lock and if the socket was on the > list, deletes it from t

Re: [PATCH net-next] vhost_net: disable zerocopy by default

2019-06-17 Thread David Miller
From: Jason Wang Date: Mon, 17 Jun 2019 05:20:54 -0400 > Vhost_net was known to suffer from HOL[1] issues which is not easy to > fix. Several downstream disable the feature by default. What's more, > the datapath was split and datacopy path got the support of batching > and XDP support recently w

Re: [PATCH v3 0/3] vsock/virtio: several fixes in the .probe() and .remove()

2019-07-08 Thread David Miller
From: Stefano Garzarella Date: Fri, 5 Jul 2019 13:04:51 +0200 > During the review of "[PATCH] vsock/virtio: Initialize core virtio vsock > before registering the driver", Stefan pointed out some possible issues > in the .probe() and .remove() callbacks of the virtio-vsock driver. ... Series ap

Re: [PATCH v5 12/29] compat_ioctl: move drivers to compat_ptr_ioctl

2019-07-30 Thread David Miller
From: Arnd Bergmann Date: Tue, 30 Jul 2019 21:50:28 +0200 > Each of these drivers has a copy of the same trivial helper function to > convert the pointer argument and then call the native ioctl handler. > > We now have a generic implementation of that, so use it. > > Acked-by: Greg Kroah-Hartma

Re: [PATCH V4 0/9] Fixes for metadata accelreation

2019-08-08 Thread David Miller
From: Jason Wang Date: Wed, 7 Aug 2019 03:06:08 -0400 > This series try to fix several issues introduced by meta data > accelreation series. Please review. ... My impression is that patch #7 will be changed to use spinlocks so there will be a v5. ___

Re: [PATCH V5 0/9] Fixes for vhost metadata acceleration

2019-08-11 Thread David Miller
From: Jason Wang Date: Mon, 12 Aug 2019 10:44:51 +0800 > On 2019/8/11 上午1:52, Michael S. Tsirkin wrote: >> At this point how about we revert >> 7f466032dc9e5a61217f22ea34b2df932786bbfc >> for this release, and then re-apply a corrected version >> for the next one? > > If possible, consider we've

Re: [PATCH net-next] vsock/virtio: a better comment on credit update

2019-09-05 Thread David Miller
From: "Michael S. Tsirkin" Date: Tue, 3 Sep 2019 03:38:16 -0400 > The comment we have is just repeating what the code does. > Include the *reason* for the condition instead. > > Cc: Stefano Garzarella > Signed-off-by: Michael S. Tsirkin Applied. ___

Re: [PATCH 0/2] Revert and rework on the metadata accelreation

2019-09-06 Thread David Miller
From: Jason Wang Date: Fri, 6 Sep 2019 18:02:35 +0800 > On 2019/9/5 下午9:59, Jason Gunthorpe wrote: >> I think you should apply the revert this cycle and rebase the other >> patch for next.. >> >> Jason > > Yes, the plan is to revert in this release cycle. Then you should reset patch #1 all by i

Re: [PATCH net 1/2] vhost: use mutex_lock_nested() in vhost_dev_lock_vqs()

2018-01-24 Thread David Miller
From: Jason Wang Date: Tue, 23 Jan 2018 17:27:25 +0800 > We used to call mutex_lock() in vhost_dev_lock_vqs() which tries to > hold mutexes of all virtqueues. This may confuse lockdep to report a > possible deadlock because of trying to hold locks belong to same > class. Switch to use mutex_lock_

Re: [PATCH net 1/2] vhost: use mutex_lock_nested() in vhost_dev_lock_vqs()

2018-01-24 Thread David Miller
From: "Michael S. Tsirkin" Date: Wed, 24 Jan 2018 23:46:19 +0200 > On Wed, Jan 24, 2018 at 04:38:30PM -0500, David Miller wrote: >> From: Jason Wang >> Date: Tue, 23 Jan 2018 17:27:25 +0800 >> >> > We used to call mutex_lock() in vhost_dev_lock_vqs()

Re: [PATCH net] vhost_net: stop device during reset owner

2018-01-29 Thread David Miller
From: Jason Wang Date: Thu, 25 Jan 2018 22:03:52 +0800 > We don't stop device before reset owner, this means we could try to > serve any virtqueue kick before reset dev->worker. This will result a > warn since the work was pending at llist during owner resetting. Fix > this by stopping device dur

Re: [PATCH net] virtio-net: disable NAPI only when enabled during XDP set

2018-02-28 Thread David Miller
From: Jason Wang Date: Wed, 28 Feb 2018 18:20:04 +0800 > We try to disable NAPI to prevent a single XDP TX queue being used by > multiple cpus. But we don't check if device is up (NAPI is enabled), > this could result stall because of infinite wait in > napi_disable(). Fixing this by checking dev

Re: [PATCH net V2] virtio-net: re enable XDP_REDIRECT for mergeable buffer

2018-03-04 Thread David Miller
From: Jason Wang Date: Fri, 2 Mar 2018 17:29:14 +0800 > XDP_REDIRECT support for mergeable buffer was removed since commit > 7324f5399b06 ("virtio_net: disable XDP_REDIRECT in receive_mergeable() > case"). This is because we don't reserve enough tailroom for struct > skb_shared_info which breaks

Re: [PATCH net V2] virtio-net: re enable XDP_REDIRECT for mergeable buffer

2018-03-04 Thread David Miller
From: Jason Wang Date: Mon, 5 Mar 2018 10:43:41 +0800 > > > On 2018年03月05日 07:38, David Miller wrote: >> From: Jason Wang >> Date: Fri, 2 Mar 2018 17:29:14 +0800 >> >>> XDP_REDIRECT support for mergeable buffer was removed since commit >>> 7324f

Re: [PATCH] drivers: vhost: vsock: fixed a brace coding style issue

2018-03-09 Thread David Miller
From: Vaibhav Murkute Date: Fri, 9 Mar 2018 08:26:03 +0530 > Fixed a coding style issue. > > Signed-off-by: Vaibhav Murkute Applied to net-next, thanks. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfou

Re: [PATCH net 0/3] Several fixes for vhost_net ptr_ring usage

2018-03-09 Thread David Miller
From: Jason Wang Date: Fri, 9 Mar 2018 14:50:31 +0800 > This small series try to fix several bugs of ptr_ring usage in > vhost_net. Please review. Series applied, thanks Jason. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org

Re: [PATCH net] vhost_net: add missing lock nesting notation

2018-03-26 Thread David Miller
From: Jason Wang Date: Mon, 26 Mar 2018 16:10:23 +0800 > We try to hold TX virtqueue mutex in vhost_net_rx_peek_head_len() > after RX virtqueue mutex is held in handle_rx(). This requires an > appropriate lock nesting notation to calm down deadlock detector. > > Fixes: 0308813724606 ("vhost_net:

Re: [PATCH net V2] vhost: correctly remove wait queue during poll failure

2018-03-27 Thread David Miller
From: Jason Wang Date: Tue, 27 Mar 2018 20:50:52 +0800 > We tried to remove vq poll from wait queue, but do not check whether > or not it was in a list before. This will lead double free. Fixing > this by switching to use vhost_poll_stop() which zeros poll->wqh after > removing poll from waitqueu

Re: [PATCH net] vhost: validate log when IOTLB is enabled

2018-03-29 Thread David Miller
From: "Michael S. Tsirkin" Date: Thu, 29 Mar 2018 15:48:22 +0300 > On Thu, Mar 29, 2018 at 04:00:04PM +0800, Jason Wang wrote: >> Vq log_base is the userspace address of bitmap which has nothing to do >> with IOTLB. So it needs to be validated unconditionally otherwise we >> may try use 0 as log_

Re: [PATCH net-next 0/5] virtio-net: Add SCTP checksum offload support

2018-04-02 Thread David Miller
From: Vladislav Yasevich Date: Mon, 2 Apr 2018 09:40:01 -0400 > Now that we have SCTP offload capabilities in the kernel, we can add > them to virtio as well. First step is SCTP checksum. Vlad, the net-next tree is closed, please resubmit this when the merge window is over and the net-next tre

Re: [RFC PATCH 2/3] netdev: kernel-only IFF_HIDDEN netdevice

2018-04-04 Thread David Miller
From: David Ahern Date: Wed, 4 Apr 2018 11:21:54 -0600 > It is a netdev so there is no reason to have a separate ip command to > inspect it. 'ip link' is the right place. I agree on this. What I really don't understand still is the use case... really. So there are control netdevs, what exactly

Re: [RFC PATCH 2/3] netdev: kernel-only IFF_HIDDEN netdevice

2018-04-04 Thread David Miller
From: David Ahern Date: Wed, 4 Apr 2018 11:37:52 -0600 > Networking vendors have out of tree kernel modules. Those modules use a > netdev (call it a master netdev, a control netdev, cpu port, whatever) > to pull packets from the ASIC and deliver to virtual netdevices > representing physical ports

Re: [RFC PATCH 2/3] netdev: kernel-only IFF_HIDDEN netdevice

2018-04-08 Thread David Miller
From: Siwei Liu Date: Fri, 6 Apr 2018 19:32:05 -0700 > And I assume everyone here understands the use case for live > migration (in the context of providing cloud service) is very > different, and we have to hide the netdevs. If not, I'm more than > happy to clarify. I think you still need to cl

Re: [PATCH] vhost-net: set packet weight of tx polling to 2 * vq size

2018-04-08 Thread David Miller
From: haibinzhang(张海斌) Date: Fri, 6 Apr 2018 08:22:37 + > handle_tx will delay rx for tens or even hundreds of milliseconds when tx busy > polling udp packets with small length(e.g. 1byte udp payload), because setting > VHOST_NET_WEIGHT takes into account only sent-bytes but no single packet

Re: [PATCH RESEND v2] vhost-net: set packet weight of tx polling to 2 * vq size

2018-04-09 Thread David Miller
From: haibinzhang(张海斌) Date: Mon, 9 Apr 2018 07:22:17 + > handle_tx will delay rx for tens or even hundreds of milliseconds when tx busy > polling udp packets with small length(e.g. 1byte udp payload), because setting > VHOST_NET_WEIGHT takes into account only sent-bytes but no single packet

Re: [PATCH v2 0/2] vhost: fix vhost_vq_access_ok() log check

2018-04-10 Thread David Miller
From: Jason Wang Date: Tue, 10 Apr 2018 14:40:10 +0800 > On 2018年04月10日 13:26, Stefan Hajnoczi wrote: >> v2: >> * Rewrote the conditional to make the vq access check clearer [Linus] >> * Added Patch 2 to make the return type consistent and harder to misuse >> * [Linus] >> >> The first patch

Re: [PATCH v3 0/2] vhost: fix vhost_vq_access_ok() log check

2018-04-11 Thread David Miller
From: "Michael S. Tsirkin" Date: Wed, 11 Apr 2018 16:24:02 +0300 > On Wed, Apr 11, 2018 at 10:35:39AM +0800, Stefan Hajnoczi wrote: >> v3: >> * Rebased onto net/master and resolved conflict [DaveM] >> >> v2: >> * Rewrote the conditional to make the vq access check clearer [Linus] >> * Added P

Re: [PATCH net] virtio-net: add missing virtqueue kick when flushing packets

2018-04-13 Thread David Miller
From: Jason Wang Date: Fri, 13 Apr 2018 14:58:25 +0800 > We tends to batch submitting packets during XDP_TX. This requires to > kick virtqueue after a batch, we tried to do it through > xdp_do_flush_map() which only makes sense for devmap not XDP_TX. So > explicitly kick the virtqueue in this cas

Re: [PATCH] net: don't use kvzalloc for DMA memory

2018-04-18 Thread David Miller
From: Mikulas Patocka Date: Wed, 18 Apr 2018 12:44:25 -0400 (EDT) > The structure net_device is followed by arbitrary driver-specific data > (accessible with the function netdev_priv). And for virtio-net, these > driver-specific data must be in DMA memory. And we are saying that this assumptio

Re: [PATCH] net: don't use kvzalloc for DMA memory

2018-04-18 Thread David Miller
From: Eric Dumazet Date: Wed, 18 Apr 2018 09:51:25 -0700 > I suggest that virtio_net clearly identifies which part needs a specific > allocation > and does its itself, instead of abusing the netdev_priv storage. > > Ie use a pointer to a block of memory, allocated by virtio_net, for > virtio_n

Re: [PATCH v7 net-next 4/4] netvsc: refactor notifier/event handling code to use the failover framework

2018-04-20 Thread David Miller
From: Stephen Hemminger Date: Fri, 20 Apr 2018 08:28:02 -0700 > On Thu, 19 Apr 2018 18:42:04 -0700 > Sridhar Samudrala wrote: > >> Use the registration/notification framework supported by the generic >> failover infrastructure. >> >> Signed-off-by: Sridhar Samudrala > > Do what you want to o

Re: [PATCH v7 net-next 4/4] netvsc: refactor notifier/event handling code to use the failover framework

2018-04-20 Thread David Miller
From: "Michael S. Tsirkin" Date: Fri, 20 Apr 2018 18:43:54 +0300 > On Fri, Apr 20, 2018 at 08:28:02AM -0700, Stephen Hemminger wrote: >> Plus, DPDK is now dependent on existing model. > > DPDK does the kernel bypass thing, doesn't it? Why does the kernel care? +1 ___

Re: [PATCH] vhost_net: use packet weight for rx handler, too

2018-04-24 Thread David Miller
From: Paolo Abeni Date: Tue, 24 Apr 2018 10:34:36 +0200 > Similar to commit a2ac99905f1e ("vhost-net: set packet weight of > tx polling to 2 * vq size"), we need a packet-based limit for > handler_rx, too - elsewhere, under rx flood with small packets, > tx can be delayed for a very long time, ev

Re: [PATCH] vhost: make msg padding explicit

2018-04-29 Thread David Miller
From: "Michael S. Tsirkin" Date: Fri, 27 Apr 2018 19:02:05 +0300 > There's a 32 bit hole just after type. It's best to > give it a name, this way compiler is forced to initialize > it with rest of the structure. > > Reported-by: Kevin Easton > Signed-off-by: Michael S. Tsirkin Who applied thi

Re: [PATCH] vhost: make msg padding explicit

2018-05-01 Thread David Miller
From: "Michael S. Tsirkin" Date: Fri, 27 Apr 2018 19:02:05 +0300 > There's a 32 bit hole just after type. It's best to > give it a name, this way compiler is forced to initialize > it with rest of the structure. > > Reported-by: Kevin Easton > Signed-off-by: Michael S. Tsirkin Michael, will y

Re: [PATCH] vhost: make msg padding explicit

2018-05-01 Thread David Miller
From: "Michael S. Tsirkin" Date: Tue, 1 May 2018 20:19:19 +0300 > On Tue, May 01, 2018 at 11:28:22AM -0400, David Miller wrote: >> From: "Michael S. Tsirkin" >> Date: Fri, 27 Apr 2018 19:02:05 +0300 >> >> > There's a 32 bit hole just

Re: [PATCH] vhost: make msg padding explicit

2018-05-02 Thread David Miller
From: "Michael S. Tsirkin" Date: Wed, 2 May 2018 16:36:37 +0300 > Ouch. True - and in particular the 32 bit ABI on 64 bit kernels doesn't > work at all. Hmm. It's relatively new and maybe there aren't any 32 bit > users yet. Thoughts? If it's been in a released kernel version, we really aren't

Re: [PATCH] Revert "vhost: make msg padding explicit"

2018-05-02 Thread David Miller
From: "Michael S. Tsirkin" Date: Wed, 2 May 2018 17:19:05 +0300 > This reverts commit 93c0d549c4c5a7382ad70de6b86610b7aae57406. > > Unfortunately the padding will break 32 bit userspace. > Ouch. Need to add some compat code, revert for now. > > Signed-off-by: Michael S. Tsirkin Applied, thank

Re: [PATCH net V2 0/4] Fix several issues of virtio-net mergeable XDP

2018-05-23 Thread David Miller
From: Jason Wang Date: Tue, 22 May 2018 11:44:27 +0800 > Please review the patches that tries to fix sevreal issues of > virtio-net mergeable XDP. > > Changes from V1: > - check against 1 before decreasing instead of resetting to 1 > - typoe fixes Series applied and queued up for -stable. _

Re: [PATCH net] vhost: synchronize IOTLB message with dev cleanup

2018-05-24 Thread David Miller
From: Jason Wang Date: Tue, 22 May 2018 19:58:57 +0800 > DaeRyong Jeong reports a race between vhost_dev_cleanup() and > vhost_process_iotlb_msg(): > > Thread interleaving: > CPU0 (vhost_process_iotlb_msg)CPU1 (vhost_dev_cleanup) > (In the case of both VHOST_IOTLB_UPDATE

Re: [PATCH net] vhost: synchronize IOTLB message with dev cleanup

2018-05-24 Thread David Miller
From: Jason Wang Date: Tue, 22 May 2018 19:58:57 +0800 > DaeRyong Jeong reports a race between vhost_dev_cleanup() and > vhost_process_iotlb_msg(): > > Thread interleaving: > CPU0 (vhost_process_iotlb_msg)CPU1 (vhost_dev_cleanup) > (In the case of both VHOST_IOTLB_UPDATE

Re: [PATCH net-next v12 0/5] Enable virtio_net to act as a standby for a passthru device

2018-05-28 Thread David Miller
From: Sridhar Samudrala Date: Thu, 24 May 2018 09:55:12 -0700 > The main motivation for this patch is to enable cloud service providers > to provide an accelerated datapath to virtio-net enabled VMs in a > transparent manner with no/minimal guest userspace changes. This also > enables hypervisor

Re: [PATCH net] vhost_net: flush batched heads before trying to busy polling

2018-05-30 Thread David Miller
From: Jason Wang Date: Tue, 29 May 2018 14:18:19 +0800 > After commit e2b3b35eb989 ("vhost_net: batch used ring update in rx"), > we tend to batch updating used heads. But it doesn't flush batched > heads before trying to do busy polling, this will cause vhost to wait > for guest TX which waits f

Re: [PATCH net-next] virtio_net: fix error return code in virtnet_probe()

2018-05-31 Thread David Miller
From: Wei Yongjun Date: Thu, 31 May 2018 02:05:07 + > Fix to return a negative error code from the failover create fail error > handling case instead of 0, as done elsewhere in this function. > > Fixes: ba5e4426e80e ("virtio_net: Extend virtio to use VF datapath when > available") > Signed-

Re: [PATCH net] vhost_net: remove VHOST_NET_F_VIRTIO_NET_HDR support

2018-06-10 Thread David Miller
From: Jason Wang Date: Fri, 8 Jun 2018 11:50:42 +0800 > This feature bit is duplicated with VIRTIO_F_ANY_LAYOUT, this means if > a userpsace want to enable VRITIO_F_ANY_LAYOUT, > VHOST_NET_F_VIRTIO_NET_HDR will be implied too. This is wrong and will > break networking. Fixing this by safely remo

Re: [PATCH net] vhost_net: validate sock before trying to put its fd

2018-06-22 Thread David Miller
From: Jason Wang Date: Thu, 21 Jun 2018 13:11:31 +0800 > Sock will be NULL if we pass -1 to vhost_net_set_backend(), but when > we meet errors during ubuf allocation, the code does not check for > NULL before calling sockfd_put(), this will lead NULL > dereferencing. Fixing by checking sock point

Re: [PATCH v2 net-next 0/4] vhost_net: Avoid vq kicks during busyloop

2018-07-04 Thread David Miller
From: Toshiaki Makita Date: Tue, 3 Jul 2018 16:31:30 +0900 > Under heavy load vhost tx busypoll tend not to suppress vq kicks, which > causes poor guest tx performance. The detailed scenario is described in > commitlog of patch 2. > Rx seems not to have that serious problem, but for consistency

Re: [PATCH net-next v2 0/5] virtio: support packed ring

2018-07-12 Thread David Miller
From: Tiwei Bie Date: Wed, 11 Jul 2018 10:27:06 +0800 > Hello everyone, > > This patch set implements packed ring support in virtio driver. > > Some functional tests have been done with Jason's > packed ring implementation in vhost: > > https://lkml.org/lkml/2018/7/3/33 > > Both of ping and n

Re: [PATCH net-next V2 0/8] Packed virtqueue support for vhost

2018-07-17 Thread David Miller
From: "Michael S. Tsirkin" Date: Mon, 16 Jul 2018 15:49:04 +0300 > I'm not sure I understand this approach. Packed ring is just an > optimization. What value is there in merging it if it does not help > speed? So it seems that both Tiwei's and Jason's packed patch sets are kind of in limbo due

Re: [PATCH net-next 0/9] TX used ring batched updating for vhost

2018-07-21 Thread David Miller
From: Jason Wang Date: Fri, 20 Jul 2018 08:15:12 +0800 > This series implement batch updating of used ring for TX. This help to > reduce the cache contention on used ring. The idea is first split > datacopy path from zerocopy, and do only batching for datacopy. This > is because zercopy had alrea

Re: [PATCH net-next 0/9] TX used ring batched updating for vhost

2018-07-22 Thread David Miller
From: "Michael S. Tsirkin" Date: Sun, 22 Jul 2018 17:37:05 +0300 > On Fri, Jul 20, 2018 at 08:15:12AM +0800, Jason Wang wrote: >> Hi: >> >> This series implement batch updating of used ring for TX. This help to >> reduce the cache contention on used ring. The idea is first split >> datacopy path

Re: [PATCH net-next 0/6] virtio_net: Add ethtool stat items

2018-07-24 Thread David Miller
From: Toshiaki Makita Date: Mon, 23 Jul 2018 23:36:03 +0900 > From: Toshiaki Makita > > Add some ethtool stat items useful for performance analysis. > > Signed-off-by: Toshiaki Makita Michael and Jason, any objections to these new stats? ___ Virtua

Re: [PATCH net-next 0/6] virtio_net: Add ethtool stat items

2018-07-25 Thread David Miller
From: "Michael S. Tsirkin" Date: Wed, 25 Jul 2018 12:40:12 +0300 > On Mon, Jul 23, 2018 at 11:36:03PM +0900, Toshiaki Makita wrote: >> From: Toshiaki Makita >> >> Add some ethtool stat items useful for performance analysis. >> >> Signed-off-by: Toshiaki Makita > > Series: > > Acked-by: Mich

Re: [PATCH net-next v6 0/4] net: vhost: improve performance when enable busyloop

2018-07-25 Thread David Miller
From: xiangxia.m@gmail.com Date: Sat, 21 Jul 2018 11:03:58 -0700 > From: Tonghao Zhang > > This patches improve the guest receive performance. > On the handle_tx side, we poll the sock receive queue > at the same time. handle_rx do that in the same way. > > For more performance report, see

Re: [PATCH net-next 1/2] virtio-net: correctly update XDP_TX counters

2018-07-31 Thread David Miller
From: Jason Wang Date: Tue, 31 Jul 2018 17:43:38 +0800 > Commit 5b8f3c8d30a6 ("virtio_net: Add XDP related stats") tries to > count TX XDP stats in virtnet_receive(). This will cause several > issues: > > - virtnet_xdp_sq() was called without checking whether or not XDP is > set. This may caus

Re: [PATCH net-next 2/2] virtio-net: get rid of unnecessary container of rq stats

2018-07-31 Thread David Miller
From: Jason Wang Date: Tue, 31 Jul 2018 17:43:39 +0800 > We don't maintain tx counters in rx stats any more. There's no need > for an extra container of rq stats. > > Cc: Toshiaki Makita > Signed-off-by: Jason Wang Applied. ___ Virtualization mailin

Re: [PATCH net-next] vhost: switch to use new message format

2018-08-04 Thread David Miller
From: Jason Wang Date: Fri, 3 Aug 2018 15:04:51 +0800 > So fixing this by introducing a new message type with an explicit > 32bit reserved field after type like: > > struct vhost_msg_v2 { > int type; > __u32 reserved; Please use fixed sized types consistently. Use 's32' instead of

Re: [PATCH net-next] virtio-net: mark expected switch fall-throughs

2018-08-05 Thread David Miller
From: "Gustavo A. R. Silva" Date: Sat, 4 Aug 2018 21:42:05 -0500 > In preparation to enabling -Wimplicit-fallthrough, mark switch cases > where we are expecting to fall through. > > Addresses-Coverity-ID: 1402059 ("Missing break in switch") > Addresses-Coverity-ID: 1402060 ("Missing break in swi

Re: [PATCH net-next V2] vhost: switch to use new message format

2018-08-06 Thread David Miller
From: Jason Wang Date: Mon, 6 Aug 2018 11:17:47 +0800 > We use to have message like: > > struct vhost_msg { > int type; > union { > struct vhost_iotlb_msg iotlb; > __u8 padding[64]; > }; > }; > > Unfortunately, there will be a hole of 32bit in 64bi

Re: [PATCH net-next] net: allow to call netif_reset_xps_queues() under cpu_read_lock

2018-08-07 Thread David Miller
From: Andrei Vagin Date: Mon, 6 Aug 2018 21:14:54 -0700 > This patch adds an ability to call __netif_set_xps_queue under > cpu_read_lock(). Please don't add conditional locking using a boolean argument. Simply wrap calls to __netif_set_xps_queue() with cpu read lock held by the caller or simil

Re: [PATCH net] vhost: reset metadata cache when initializing new IOTLB

2018-08-08 Thread David Miller
From: Jason Wang Date: Wed, 8 Aug 2018 11:43:04 +0800 > We need to reset metadata cache during new IOTLB initialization, > otherwise the stale pointers to previous IOTLB may be still accessed > which will lead a use after free. > > Reported-by: syzbot+c51e6736a1bf614b3...@syzkaller.appspotmail.

Re: [PATCH net-next] virtio_net: remove duplicated include from virtio_net.c

2018-08-13 Thread David Miller
From: YueHaibing Date: Mon, 13 Aug 2018 14:13:15 +0800 > Remove duplicated include linux/netdevice.h > > Signed-off-by: YueHaibing Applied. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/ma

Re: [PATCH net] vhost: correctly check the iova range when waking virtqueue

2018-08-25 Thread David Miller
From: Jason Wang Date: Fri, 24 Aug 2018 16:53:13 +0800 > We don't wakeup the virtqueue if the first byte of pending iova range > is the last byte of the range we just got updated. This will lead a > virtqueue to wait for IOTLB updating forever. Fixing by correct the > check and wake up the virtqu

Re: [PATCH] vhost: fix VHOST_GET_BACKEND_FEATURES ioctl request definition

2018-09-03 Thread David Miller
From: "Michael S. Tsirkin" Date: Mon, 3 Sep 2018 23:11:11 -0400 > On Tue, Sep 04, 2018 at 11:08:40AM +0800, Jason Wang wrote: >> >> >> On 2018年09月04日 10:22, Michael S. Tsirkin wrote: >> > On Mon, Sep 03, 2018 at 08:59:13PM +0300, Gleb Fotengauer-Malinovskiy >> > wrote: >> > > The _IOC_READ fla

Re: [PATCH net-next v9 0/6] net: vhost: improve performance when enable busyloop

2018-09-09 Thread David Miller
From: xiangxia.m@gmail.com Date: Sun, 9 Sep 2018 04:51:21 -0700 > From: Tonghao Zhang > > This patches improve the guest receive performance. > On the handle_tx side, we poll the sock receive queue > at the same time. handle_rx do that in the same way. > > For more performance report, see

Re: [PATCH net-next V2 00/11] vhost_net TX batching

2018-09-13 Thread David Miller
From: Jason Wang Date: Wed, 12 Sep 2018 11:16:58 +0800 > This series tries to batch submitting packets to underlayer socket > through msg_control during sendmsg(). This is done by: ... Series applied, thanks Jason. ___ Virtualization mailing list Virt

Re: [PATCH net-next V2 00/11] vhost_net TX batching

2018-09-13 Thread David Miller
From: "Michael S. Tsirkin" Date: Thu, 13 Sep 2018 14:02:13 -0400 > On Thu, Sep 13, 2018 at 09:28:19AM -0700, David Miller wrote: >> From: Jason Wang >> Date: Wed, 12 Sep 2018 11:16:58 +0800 >> >> > This series tries to batch submitting packets to un

Re: [PATCH net-next V2] virtio_net: ethtool tx napi configuration

2018-09-13 Thread David Miller
From: Jason Wang Date: Thu, 13 Sep 2018 13:35:45 +0800 > Toggle tx napi through a bit in tx-frames. This is not what the code implements as the interface any more. Please fix the commit message to match the code. Thanks. ___ Virtualization mailing li

Re: [REBASE PATCH net-next v9 0/4] net: vhost: improve performance when enable busyloop

2018-09-26 Thread David Miller
From: xiangxia.m@gmail.com Date: Tue, 25 Sep 2018 05:36:48 -0700 > From: Tonghao Zhang > > This patches improve the guest receive performance. > On the handle_tx side, we poll the sock receive queue > at the same time. handle_rx do that in the same way. > > For more performance report, see

Re: [PATCH net-next V3] virtio_net: ethtool tx napi configuration

2018-10-10 Thread David Miller
From: Jason Wang Date: Tue, 9 Oct 2018 10:06:26 +0800 > Implement ethtool .set_coalesce (-C) and .get_coalesce (-c) handlers. > Interrupt moderation is currently not supported, so these accept and > display the default settings of 0 usec and 1 frame. > > Toggle tx napi through setting tx-frames

Re: [PATCH v3] virtio_net: avoid using netif_tx_disable() for serializing tx routine

2018-10-17 Thread David Miller
From: Ake Koomsin Date: Wed, 17 Oct 2018 19:44:12 +0900 > Commit 713a98d90c5e ("virtio-net: serialize tx routine during reset") > introduces netif_tx_disable() after netif_device_detach() in order to > avoid use-after-free of tx queues. However, there are two issues. > > 1) Its operation is redu

Re: [PATCH] virtio_net: add local_bh_disable() around u64_stats_update_begin

2018-10-18 Thread David Miller
From: Sebastian Andrzej Siewior Date: Thu, 18 Oct 2018 10:43:13 +0200 > on 32bit, lockdep notices that virtnet_open() and refill_work() invoke > try_fill_recv() from process context while virtnet_receive() invokes the > same function from BH context. The problem that the seqcounter within > u64_s

Re: [PATCH net] vhost: Fix Spectre V1 vulnerability

2018-10-31 Thread David Miller
From: Jason Wang Date: Tue, 30 Oct 2018 14:10:49 +0800 > The idx in vhost_vring_ioctl() was controlled by userspace, hence a > potential exploitation of the Spectre variant 1 vulnerability. > > Fixing this by sanitizing idx before using it to index d->vqs. > > Cc: Michael S. Tsirkin > Cc: Josh

Re: [PATCH net-next 1/2] vhost_net: mitigate page reference counting during page frag refill

2018-11-17 Thread David Miller
From: Jason Wang Date: Thu, 15 Nov 2018 17:43:09 +0800 > We do a get_page() which involves a atomic operation. This patch tries > to mitigate a per packet atomic operation by maintaining a reference > bias which is initially USHRT_MAX. Each time a page is got, instead of > calling get_page() we d

Re: [PATCH net-next 2/2] tuntap: free XDP dropped packets in a batch

2018-11-17 Thread David Miller
From: Jason Wang Date: Thu, 15 Nov 2018 17:43:10 +0800 > Thanks to the batched XDP buffs through msg_control. Instead of > calling put_page() for each page which involves a atomic operation, > let's batch them by record the last page that needs to be freed and > its refcnt count and free them in

Re: [PATCH net-next v3 00/13] virtio: support packed ring

2018-11-21 Thread David Miller
From: "Michael S. Tsirkin" Date: Wed, 21 Nov 2018 07:20:27 -0500 > Dave, given the holiday, attempts to wrap up the 1.1 spec and the > patchset size I would very much appreciate a bit more time for > review. Say until Nov 28? Ok. ___ Virtualization mai

Re: [PATCH net 1/2] virtio-net: disable guest csum during XDP set

2018-11-24 Thread David Miller
From: Jason Wang Date: Thu, 22 Nov 2018 14:36:30 +0800 > We don't disable VIRTIO_NET_F_GUEST_CSUM if XDP was set. This means we > can receive partial csumed packets with metadata kept in the > vnet_hdr. This may have several side effects: > > - It could be overridden by header adjustment, thus i

Re: [PATCH net 2/2] virtio-net: fail XDP set if guest csum is negotiated

2018-11-24 Thread David Miller
From: Jason Wang Date: Thu, 22 Nov 2018 14:36:31 +0800 > We don't support partial csumed packet since its metadata will be lost > or incorrect during XDP processing. So fail the XDP set if guest_csum > feature is negotiated. > > Fixes: f600b6905015 ("virtio_net: Add XDP support") > Reported-by:

Re: [PATCH net-next v3 00/13] virtio: support packed ring

2018-11-26 Thread David Miller
From: "Michael S. Tsirkin" Date: Tue, 27 Nov 2018 01:08:08 -0500 > On Wed, Nov 21, 2018 at 06:03:17PM +0800, Tiwei Bie wrote: >> Hi, >> >> This patch set implements packed ring support in virtio driver. >> >> A performance test between pktgen (pktgen_sample03_burst_single_flow.sh) >> and DPDK v

Re: [PATCH net] virtio-net: keep vnet header zeroed after processing XDP

2018-11-30 Thread David Miller
From: Jason Wang Date: Thu, 29 Nov 2018 13:53:16 +0800 > We copy vnet header unconditionally in page_to_skb() this is wrong > since XDP may modify the packet data. So let's keep a zeroed vnet > header for not confusing the conversion between vnet header and skb > metadata. > > In the future, we

Re: [PATCH v2] vhost: fix IOTLB locking

2018-12-03 Thread David Miller
From: Jean-Philippe Brucker Date: Fri, 30 Nov 2018 16:05:53 + > Commit 78139c94dc8c ("net: vhost: lock the vqs one by one") moved the vq > lock to improve scalability, but introduced a possible deadlock in > vhost-iotlb. vhost_iotlb_notify_vq() now takes vq->mutex while holding > the device's

Re: [PATCH net 0/4] Fix various issue of vhost

2018-12-11 Thread David Miller
From: Jason Wang Date: Mon, 10 Dec 2018 17:44:50 +0800 > This series tries to fix various issues of vhost: > > - Patch 1 adds a missing write barrier between used idx updating and > logging. > - Patch 2-3 brings back the protection of device IOTLB through vq > mutex, this fixes possible use

Re: [PATCH net V3 0/3] Fix various issue of vhost

2018-12-12 Thread David Miller
From: Jason Wang Date: Thu, 13 Dec 2018 10:53:36 +0800 > This series tries to fix various issues of vhost: > > - Patch 1 adds a missing write barrier between used idx updating and > logging. > - Patch 2-3 brings back the protection of device IOTLB through vq > mutex, this fixes possible use

Re: [PATCH v2 1/5] VSOCK: support fill mergeable rx buffer in guest

2018-12-14 Thread David Miller
From: jiangyiwen Date: Wed, 12 Dec 2018 17:28:16 +0800 > +static int fill_mergeable_rx_buff(struct virtio_vsock *vsock, > + struct virtqueue *vq) > +{ > + struct page_frag *alloc_frag = &vsock->alloc_frag; > + struct scatterlist sg; > + /* Currently we don't use ewma len,

Re: [PATCH v2 2/5] VSOCK: support fill data to mergeable rx buffer in host

2018-12-15 Thread David Miller
From: jiangyiwen Date: Wed, 12 Dec 2018 17:29:31 +0800 > diff --git a/include/uapi/linux/virtio_vsock.h > b/include/uapi/linux/virtio_vsock.h > index 1d57ed3..2292f30 100644 > --- a/include/uapi/linux/virtio_vsock.h > +++ b/include/uapi/linux/virtio_vsock.h > @@ -63,6 +63,11 @@ struct virtio_vso

Re: [PATCH net V2 0/4] Fix various issue of vhost

2018-12-15 Thread David Miller
From: Jason Wang Date: Wed, 12 Dec 2018 18:08:15 +0800 > This series tries to fix various issues of vhost: > > - Patch 1 adds a missing write barrier between used idx updating and > logging. > - Patch 2-3 brings back the protection of device IOTLB through vq > mutex, this fixes possible use

Re: [PATCH v2 2/5] VSOCK: support fill data to mergeable rx buffer in host

2018-12-15 Thread David Miller
From: jiangyiwen Date: Thu, 13 Dec 2018 11:11:48 +0800 > I hope Host can fill fewer bytes into rx virtqueue, so > I keep structure virtio_vsock_mrg_rxbuf_hdr one byte > alignment. The question is if this actully matters. Do you know? If the obejct this is embeeded inside of is at least 2 byte

<    1   2   3   4   5   6   >