Re: [PATCH net-next V3 3/3] tun: rx batching

From: Jason Wang Date: Fri, 30 Dec 2016 13:20:51 +0800 > @@ -1283,10 +1314,15 @@ static ssize_t tun_get_user(struct tun_struct *tun, > struct tun_file *tfile, > skb_probe_transport_header(skb, 0); > > rxhash = skb_get_hash(skb); > + > #ifndef CONFIG_4KSTACKS

Re: [PATCH net-next V2 3/3] tun: rx batching

From: Jason Wang Date: Wed, 28 Dec 2016 16:09:31 +0800 > + spin_lock(>lock); > + qlen = skb_queue_len(queue); > + if (qlen > rx_batched) > + goto drop; > + __skb_queue_tail(queue, skb); > + if (!more || qlen + 1 > rx_batched) { > +

Re: [PATCH] net: Use kmemdup instead of kmalloc and memcpy

From: Shyam Saini Date: Sat, 24 Dec 2016 00:44:58 +0530 > when some other buffer is immediately copied into allocated region. > Replace calls to kmalloc followed by a memcpy with a direct > call to kmemdup. > > Signed-off-by: Shyam Saini

Re: [PATCH net 0/9] several fixups for virtio-net XDP

From: Jason Wang Date: Fri, 23 Dec 2016 22:37:23 +0800 > Merry Xmas and a Happy New year to all: > > This series tries to fixes several issues for virtio-net XDP which > could be categorized into several parts: > > - fix several issues during XDP linearizing > - allow

Re: [PATCH net] virtio-net: correctly enable multiqueue

From: Jason Wang Date: Tue, 13 Dec 2016 14:23:05 +0800 > Commit 4490001029012539937ff02778fe6180613fa949 ("virtio-net: enable > multiqueue by default") blindly set the affinity instead of queues > during probe which can cause a mismatch of #queues between guest and > host.

Re: [PATCH-RESEND] vhost-vsock: fix orphan connection reset

From: Peng Tao Date: Fri, 9 Dec 2016 01:10:46 +0800 > local_addr.svm_cid is host cid. We should check guest cid instead, > which is remote_addr.svm_cid. Otherwise we end up resetting all > connections to all guests. > > Cc: sta...@vger.kernel.org [4.8+] > Reviewed-by:

Re: [PATCH 1/2] vhost-vsock: remove unused vq variable

From: Peng Tao Date: Wed, 7 Dec 2016 17:52:18 +0800 > Signed-off-by: Peng Tao Applied. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org

Re: [PATCH] virtio-net: Fix DMA-from-the-stack in virtnet_set_mac_address()

From: Andy Lutomirski Date: Mon, 5 Dec 2016 18:10:58 -0800 > With CONFIG_VMAP_STACK=y, virtnet_set_mac_address() can be passed a > pointer to the stack and it will OOPS. Copy the address to the heap > to prevent the crash. > > Cc: Michael S. Tsirkin > Cc:

Re: [PATCH net-next] virtio-net: enable multiqueue by default

From: Jason Wang Date: Fri, 25 Nov 2016 12:37:26 +0800 > We use single queue even if multiqueue is enabled and let admin to > enable it through ethtool later. This is used to avoid possible > regression (small packet TCP stream transmission). But looks like an > overkill

Re: [PATCH net-next] virtio-net: enable multiqueue by default

From: "Michael S. Tsirkin" Date: Fri, 25 Nov 2016 06:43:08 +0200 > On Fri, Nov 25, 2016 at 12:37:26PM +0800, Jason Wang wrote: >> We use single queue even if multiqueue is enabled and let admin to >> enable it through ethtool later. This is used to avoid possible >> regression

Re: [PATCH] virtio-net: drop legacy features in virtio 1 mode

From: "Michael S. Tsirkin" Date: Fri, 4 Nov 2016 12:55:36 +0200 > Virtio 1.0 spec says VIRTIO_F_ANY_LAYOUT and VIRTIO_NET_F_GSO are > legacy-only feature bits. Do not negotiate them in virtio 1 mode. Note > this is a spec violation so we need to backport it to stable/downstream

Re: [PATCH v2 net-next] virtio-net: Update the mtu code to match virtio spec

From: Aaron Conole Date: Tue, 25 Oct 2016 16:12:12 -0400 > The virtio committee recently ratified a change, VIRTIO-152, which > defines the mtu field to be 'max' MTU, not simply desired MTU. > > This commit brings the virtio-net device in compliance with VIRTIO-152. > >

Re: [PATCH/RFC 0/5] cpu_relax: introduce yield, remove lowlatency

From: Christian Borntraeger <borntrae...@de.ibm.com> Date: Fri, 21 Oct 2016 17:08:54 +0200 > On 10/21/2016 04:57 PM, David Miller wrote: >> From: Christian Borntraeger <borntrae...@de.ibm.com> >> Date: Fri, 21 Oct 2016 13:58:53 +0200 >> >>> For s

Re: [PATCH net-next V4 0/6] switch to use tx skb array in tun

From: Jason Wang Date: Thu, 30 Jun 2016 14:45:30 +0800 > This series tries to switch to use skb array in tun. This is used to > eliminate the spinlock contention between producer and consumer. The > conversion was straightforward: just introdce a tx skb array and use > it

Re: [PATCH net-next V2] tun: introduce tx skb ring

From: Jason Wang Date: Wed, 15 Jun 2016 16:38:17 +0800 > We used to queue tx packets in sk_receive_queue, this is less > efficient since it requires spinlocks to synchronize between producer > and consumer. > > This patch tries to address this by: > > - introduce a new

Re: [PATCH 0/6] virtio_net: use common code for virtio_net_hdr and skb GSO conversion

From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:16 +0300 > This patches introduce virtio_net_hdr_{from,to}_skb functions for > conversion of GSO information between skb and virtio_net_hdr. Looks like a nice cleanup to me, series applied, thanks Mike.

Re: [PATCH -next] virtio_net: Update the feature bit to comply with spec

From: Aaron Conole Date: Thu, 9 Jun 2016 13:41:15 -0400 > A draft version of the MTU Advice feature bit was specified as 25. This > bit is not within the allowed range for network device feature bits, and > should be changed to be feature bit 3 to fully comply with the

Re: [PATCH] macvtap: fix bugon.cocci warnings

From: Julia Lawall Date: Wed, 8 Jun 2016 18:15:29 +0200 (CEST) > Use BUG_ON instead of a if condition followed by BUG. > > Generated by: scripts/coccinelle/misc/bugon.cocci > > CC: Mike Rapoport > Signed-off-by: Julia Lawall

Re: [PATCH V3 1/2] vhost_net: stop polling socket during rx processing

From: Jason Wang Date: Wed, 1 Jun 2016 01:56:33 -0400 > We don't stop rx polling socket during rx processing, this will lead > unnecessary wakeups from under layer net devices (E.g > sock_def_readable() form tun). Rx will be slowed down in this > way. This patch avoids this

Re: [PATCH v3] virtio-net: Add initial MTU advice feature

From: Aaron Conole Date: Fri, 3 Jun 2016 16:57:12 -0400 > This commit adds the feature bit and associated mtu device entry for the > virtio network device. When a virtio device comes up, it checks the > feature bit for the VIRTIO_NET_F_MTU feature. If such feature bit is >

Re: [PATCH V3 0/2] vhost_net polling optimization

From: Jason Wang Date: Wed, 1 Jun 2016 01:56:32 -0400 > This series tries to optimize vhost_net polling at two points: > > - Stop rx polling for reduicng the unnecessary wakeups during > handle_rx(). > - Conditonally enable tx polling for reducing the unnecessary >

Re: [PATCH] VSOCK: Only check error on skb_recv_datagram when skb is NULL

From: Jorgen Hansen Date: Mon, 18 Apr 2016 23:58:52 -0700 > If skb_recv_datagram returns an skb, we should ignore the err > value returned. Otherwise, datagram receives will return EAGAIN > when they have to wait for a datagram. > > Acked-by: Adit Ranadive

Re: [PATCH V3 0/3] basic busy polling support for vhost_net

From: Jason Wang Date: Fri, 26 Feb 2016 16:42:41 +0800 > This series tries to add basic busy polling for vhost net. The idea is > simple: at the end of tx/rx processing, busy polling for new tx added > descriptor and rx receive socket for a while. The maximum number of >

Re: bonding (IEEE 802.3ad) not working with qemu/virtio

From: Nikolay Aleksandrov Date: Fri, 29 Jan 2016 22:48:26 +0100 > On 01/29/2016 10:45 PM, Jay Vosburgh wrote: >> Nikolay Aleksandrov wrote: >> >>> On 01/25/2016 05:24 PM, Bjørnar Ness wrote: As subject says, 802.3ad bonding is not

Re: [PATCH v2 24/32] sparc: define __smp_xxx

From: "Michael S. Tsirkin" Date: Thu, 31 Dec 2015 21:08:53 +0200 > This defines __smp_xxx barriers for sparc, > for use by virtualization. > > smp_xxx barriers are removed as they are > defined correctly by asm-generic/barriers.h > > Signed-off-by: Michael S. Tsirkin

Re: [PATCH v2 07/32] sparc: reuse asm-generic/barrier.h

From: "Michael S. Tsirkin" Date: Thu, 31 Dec 2015 21:06:38 +0200 > On sparc 64 bit dma_rmb, dma_wmb, smp_store_mb, smp_mb, smp_rmb, > smp_wmb, read_barrier_depends and smp_read_barrier_depends match the > asm-generic variants exactly. Drop the local definitions and pull in >

Re: [PATCH 07/34] sparc: reuse asm-generic/barrier.h

I find it a little bit irritating when I receive a patch directly addressed to me to review, but I don't see the "00/34" cover letter. I want to see what this series is about, and what the motiviation for this change is. I'm also highly frustrated, in general, with mix-and-match CC: lists for

Re: [PATCH 00/34] arch: barrier cleanup + __smp_xxx barriers for virt

From: "Michael S. Tsirkin" Date: Wed, 30 Dec 2015 14:58:19 +0200 > -. Patch 1 documents the __smp APIs, and explains why they are >useful for virt If virt is doing things like interacting with descriptors that are shared with a (potentially SMP) host, why don't we just

Re: [PATCH 07/34] sparc: reuse asm-generic/barrier.h

From: "Michael S. Tsirkin" Date: Wed, 30 Dec 2015 21:55:33 +0200 > The problem I had in the past is that vger rejects email with > too many CC addresses. I can't think of a patch set where a > 1024 character CC: list is legitimate anyways, even if all interested parties are

Re: [PATCH 00/34] arch: barrier cleanup + __smp_xxx barriers for virt

From: "Michael S. Tsirkin" <m...@redhat.com> Date: Wed, 30 Dec 2015 23:36:29 +0200 > On Wed, Dec 30, 2015 at 03:46:46PM -0500, David Miller wrote: >> From: "Michael S. Tsirkin" <m...@redhat.com> >> Date: Wed, 30 Dec 2015 14:58:19 +0200 >> &

Re: [PATCH 0/6] VSOCK: revert virtio-vsock until device spec is finalized

From: Stefan Hajnoczi Date: Tue, 8 Dec 2015 19:57:30 +0800 > Please revert for now. Please don't revert it piece by piece like this. Instead, send me one big revert commit that undoes the whole thing. There is even a merge commit that you can use to create that revert

Re: [PATCH] vhost: vsock: select CONFIG_VHOST

From: "Michael S. Tsirkin" Date: Tue, 8 Dec 2015 18:09:44 +0200 > On Tue, Dec 08, 2015 at 04:46:08PM +0100, Arnd Bergmann wrote: >> When building the new vsock code without vhost, we get a build error: >> >> drivers/built-in.o: In function `vhost_vsock_flush': >>

Re: [PATCH v2] Revert "Merge branch 'vsock-virtio'"

From: Stefan Hajnoczi Date: Wed, 9 Dec 2015 10:51:12 +0800 > This reverts commit 0d76d6e8b2507983a2cae4c09880798079007421 and merge > commit c402293bd76fbc93e52ef8c0947ab81eea3ae019, reversing changes made > to c89359a42e2a49656451569c382eed63e781153c. > > The virtio-vsock

Re: [PATCH] VSOCK: fix returnvar.cocci warnings

From: Julia Lawall Date: Sun, 6 Dec 2015 06:56:23 +0100 (CET) > Remove unneeded variable used to store return value. > > Generated by: scripts/coccinelle/misc/returnvar.cocci > > CC: Asias He > Signed-off-by: Fengguang Wu >

Re: [PATCH v2 0/5] Add virtio transport for AF_VSOCK

From: Stefan Hajnoczi Date: Wed, 2 Dec 2015 14:43:58 +0800 > This patch series adds a virtio transport for AF_VSOCK (net/vmw_vsock/). Series applied, thanks. ___ Virtualization mailing list

Re: [PATCH v2] vhost: replace % with & on data path

From: "Michael S. Tsirkin" Date: Mon, 30 Nov 2015 11:15:23 +0200 > We know vring num is a power of 2, so use & > to mask the high bits. > > Signed-off-by: Michael S. Tsirkin Acked-by: David S. Miller

Re: [PATCH] vhost: replace % with & on data path

From: "Michael S. Tsirkin" Date: Mon, 30 Nov 2015 10:34:07 +0200 > We know vring num is a power of 2, so use & > to mask the high bits. > > Signed-off-by: Michael S. Tsirkin > --- > drivers/vhost/vhost.c | 8 +--- > 1 file changed, 5 insertions(+), 3

Re: [PATCH net-next 0/3] basic busy polling support for vhost_net

From: Jason Wang Date: Wed, 25 Nov 2015 15:11:26 +0800 > This series tries to add basic busy polling for vhost net. The idea is > simple: at the end of tx/rx processing, busy polling for new tx added > descriptor and rx receive socket for a while. The maximum number of >

Re: [PATCH] vhost: move is_le setup to the backend

From: Greg Kurz Date: Fri, 30 Oct 2015 12:42:35 +0100 > The vq->is_le field is used to fix endianness when accessing the vring via > the cpu_to_vhost16() and vhost16_to_cpu() helpers in the following cases: > > 1) host is big endian and device is modern virtio > > 2)

Re: [PATCH] VSOCK: define VSOCK_SS_LISTEN once only

From: Stefan Hajnoczi Date: Thu, 29 Oct 2015 11:57:42 + > The SS_LISTEN socket state is defined by both af_vsock.c and > vmci_transport.c. This is risky since the value could be changed in one > file and the other would be out of sync. > > Rename from SS_LISTEN to

Re: [PATCH] vhost: fix performance on LE hosts

From: "Michael S. Tsirkin" Date: Tue, 27 Oct 2015 11:37:39 +0200 > commit 2751c9882b947292fcfb084c4f604e01724af804 ("vhost: cross-endian > support for legacy devices") introduced a minor regression: even with > cross-endian disabled, and even on LE host, vhost_is_little_endian

Re: [PATCH] vhost: fix performance on LE hosts

From: "Michael S. Tsirkin" Date: Tue, 27 Oct 2015 11:37:39 +0200 > commit 2751c9882b947292fcfb084c4f604e01724af804 ("vhost: cross-endian > support for legacy devices") introduced a minor regression: even with > cross-endian disabled, and even on LE host, vhost_is_little_endian

Re: [PATCH] VSOCK: sock_put wasn't safe to call in interrupt context

From: Jorgen Hansen Date: Wed, 21 Oct 2015 04:53:56 -0700 > In the vsock vmci_transport driver, sock_put wasn't safe to call > in interrupt context, since that may call the vsock destructor > which in turn calls several functions that should only be called > from process

Re: [PATCH] Fix AF_PACKET ABI breakage in 4.2

From: David Woodhouse Date: Wed, 23 Sep 2015 15:44:30 +0100 > +#define VIO_LE virtio_legacy_is_little_endian() When you define a shorthand macro, the defines to a function call, make the macro have parenthesis too. ___

Re: [PATCH net-next] virtio-net: avoid unnecessary sg initialzation

From: Jason Wang jasow...@redhat.com Date: Thu, 27 Aug 2015 14:53:06 +0800 Usually an skb does not have up to MAX_SKB_FRAGS frags. So no need to initialize the unuse part of sg. This patch initialize the sg based on the real number it will used: - during xmit, it could be inferred from

Re: [PATCH net V2] virtio-net: drop NETIF_F_FRAGLIST

From: Jason Wang jasow...@redhat.com Date: Wed, 5 Aug 2015 10:34:04 +0800 virtio declares support for NETIF_F_FRAGLIST, but assumes that there are at most MAX_SKB_FRAGS + 2 fragments which isn't always true with a fraglist. A longer fraglist in the skb will make the call to skb_to_sgvec

Re: [PATCH net V2] virtio-net: drop NETIF_F_FRAGLIST

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 6 Aug 2015 16:13:11 +0300 On Wed, Aug 05, 2015 at 10:34:04AM +0800, Jason Wang wrote: virtio declares support for NETIF_F_FRAGLIST, but assumes that there are at most MAX_SKB_FRAGS + 2 fragments which isn't always true with a fraglist. A

Re: [PATCH net-next] virtio_net: add gro capability

From: Eric Dumazet eric.duma...@gmail.com Date: Fri, 31 Jul 2015 18:25:17 +0200 From: Eric Dumazet eduma...@google.com Straightforward patch to add GRO processing to virtio_net. napi_complete_done() usage allows more aggressive aggregation, opted-in by setting

Re: [PATCH net-next] virtio_net: add gro capability

From: Eric Dumazet eric.duma...@gmail.com Date: Fri, 31 Jul 2015 18:25:17 +0200 From: Eric Dumazet eduma...@google.com Straightforward patch to add GRO processing to virtio_net. ... Signed-off-by: Eric Dumazet eduma...@google.com Tested-by: Rick Jones rick.jon...@hp.com Cc: Michael S.

Re: [PATCH] virtio_net: don't require ANY_LAYOUT with VERSION_1

From: Michael S. Tsirkin m...@redhat.com Date: Wed, 15 Jul 2015 15:26:19 +0300 ANY_LAYOUT is a compatibility feature. It's implied for VERSION_1 devices, and non-transitional devices might not offer it. Change code to behave accordingly. Signed-off-by: Michael S. Tsirkin m...@redhat.com

Re: [PATCH v6 2/8] tun: add tun_is_little_endian() helper

From: Michael S. Tsirkin m...@redhat.com Date: Mon, 1 Jun 2015 12:29:43 +0200 On Fri, Apr 24, 2015 at 02:24:38PM +0200, Greg Kurz wrote: Signed-off-by: Greg Kurz gk...@linux.vnet.ibm.com Dave, could you please ack merging this through the virtio tree? Acked-by: David S. Miller

Re: [PATCH v6 3/8] macvtap: introduce macvtap_is_little_endian() helper

From: Michael S. Tsirkin m...@redhat.com Date: Mon, 1 Jun 2015 12:30:16 +0200 On Fri, Apr 24, 2015 at 02:24:48PM +0200, Greg Kurz wrote: Signed-off-by: Greg Kurz gk...@linux.vnet.ibm.com Dave, could you pls ack merging this through the virtio tree? Acked-by: David S. Miller

Re: [PATCH net-next] virtio: document queue state logic

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 2 Apr 2015 13:05:47 +0200 commit d631b94e7a15277858ec5f88d674d93080506999 virtio: change comment in transmit started clarifying the logic behind queue state management, but introduced an inaccuracy: TX_BUSY does not cause a BUG

Re: [PATCH net-next] netdevice: document NETDEV_TX_BUSY deprecation.

/msg322350.html Inspired-by: David Miller da...@davemloft.net Signed-off-by: Rusty Russell ru...@rustcorp.com.au Applied. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo

Re: [PATCH net-next] virtio: change comment in transmit

From: Stephen Hemminger step...@networkplumber.org Date: Tue, 24 Mar 2015 16:22:07 -0700 The original comment was not really informative or funny as well as sexist. Replace it with a better explanation of why the driver does stop and what the impacts are. Signed-off-by: Stephen Hemminger

Re: [PATCH net-next] virtio: change comment in transmit

From: Rusty Russell ru...@rustcorp.com.au Date: Wed, 25 Mar 2015 14:29:38 +1030 I note that there's still no comment saying don't do this in netdevice.h; I gather returning NETDEV_TX_BUSY is still considered a Bad Thing? You're not supposed to do it still, that's right. If the driver returns

Re: [PATCH net] virtio-net: correctly delete napi hash

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 12 Mar 2015 08:25:34 +0100 On Thu, Mar 12, 2015 at 01:57:44PM +0800, Jason Wang wrote: We don't delete napi from hash list during module exit. This will cause the following panic when doing module load and unload: ... This patch fixes this

Re: [PATCH] vhost: cleanup iterator update logic

From: Michael S. Tsirkin m...@redhat.com Date: Tue, 24 Feb 2015 17:31:31 +0100 Recent iterator-related changes in vhost made it harder to follow the logic fixing up the header. In fact, the fixup always happens at the same offset: sizeof(virtio_net_hdr): sometimes the fixup iterator is

Re: [PATCH] vhost: drop hard-coded num_buffers size

From: Michael S. Tsirkin m...@redhat.com Date: Tue, 24 Feb 2015 17:31:10 +0100 The 2 that we use for copy_to_iter comes from sizeof(u16), it used to be that way before the iov iter update. Fix it up, making it obvious the size of stack access is right. Signed-off-by: Michael S. Tsirkin

Re: [PATCH] vhost: drop hard-coded num_buffers size

From: Michael S. Tsirkin m...@redhat.com Date: Tue, 24 Feb 2015 17:31:10 +0100 The 2 that we use for copy_to_iter comes from sizeof(u16), it used to be that way before the iov iter update. Fix it up, making it obvious the size of stack access is right. Signed-off-by: Michael S. Tsirkin

Re: [PATCH net] vhost_net: fix wrong iter offset when setting number of buffers

From: Jason Wang jasow...@redhat.com Date: Sun, 15 Feb 2015 16:35:17 +0800 In commit ba7438aed924 (vhost: don't bother copying iovecs in handle_rx(), kill memcpy_toiovecend()), we advance iov iter fixup sizeof(struct virtio_net_hdr) bytes and fill the number of buffers after doing the socket

Re: [PATCH v3 net 1/3] ipv6: Select fragment id during UFO segmentation if not set.

From: Vladislav Yasevich vyasev...@gmail.com Date: Tue, 3 Feb 2015 16:36:15 -0500 diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index 97f41a3..54520a0 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -9,6 +9,24 @@ #include net/addrconf.h #include

Re: [PATCH v3 0/3] Restore UFO support to virtio_net devices

Vlad, this still fails the same way. [davem@dokdo net]$ make -s -j8 kernel/Makefile:132: *** No X.509 certificates found *** kernel/Makefile:132: *** No X.509 certificates found *** net/built-in.o: In function `udp6_ufo_fragment': udp_offload.c:(.text+0x103514): undefined reference to

Re: [PATCH v3 0/3] Restore UFO support to virtio_net devices

From: David Miller da...@davemloft.net Date: Mon, 02 Feb 2015 13:11:58 -0800 (PST) Vlad, this still fails the same way. ... Please build allmodconfig, that is the exact build I use to test your and everyone else's changes. Oh, also, there is trailing whitespace in one of the comments added

Re: [PATCH v2 0/3] Restore UFO support to virtio_net devices

From: Vladislav Yasevich vyasev...@gmail.com Date: Fri, 30 Jan 2015 14:27:24 -0500 commit 3d0ad09412ffe00c9afa201d01effdb6023d09b4 Author: Ben Hutchings b...@decadent.org.uk Date: Thu Oct 30 18:27:12 2014 + drivers/net: Disable UFO through virtio Turned off UFO support to

Re: [PATCH v2 0/3] Restore UFO support to virtio_net devices

From: David Miller da...@davemloft.net Date: Sun, 01 Feb 2015 22:19:35 -0800 (PST) Series applied and queued up for -stable. I have to revert, this breaks the build. net/built-in.o: In function `udp6_ufo_fragment': udp_offload.c:(.text+0x103380): undefined reference to `ipv6_select_ident

Re: [PATCH 1/3] ipv6: Select fragment id during UFO/GSO segmentation if not set.

From: Ben Hutchings b...@decadent.org.uk Date: Tue, 27 Jan 2015 02:47:54 + On Mon, 2015-01-26 at 09:37 -0500, Vladislav Yasevich wrote: If the IPv6 fragment id has not been set and we perform fragmentation due to UFO, select a new fragment id. When we store the fragment id into

Re: [PATCH net-next V2] virtio-net: don't do header check for dodgy gso packets

From: Jason Wang jasow...@redhat.com Date: Wed, 24 Dec 2014 11:03:52 +0800 There's no need to do header check for virtio-net since: - Host sets dodgy for all gso packets from guest and check the header. - Host should be prepared for all kinds of evil packets from guest, since malicious

Re: [PATCH net-next] vhost: remove unnecessary forward declarations in vhost.h

From: Jason Wang jasow...@redhat.com Date: Thu, 27 Nov 2014 14:41:21 +0800 Signed-off-by: Jason Wang jasow...@redhat.com I don't think generic vhost patches should go via my tree. If you disagree, let me know why, thanks :) ___ Virtualization mailing

Re: [PATCH net V5] virtio-net: validate features during probe

From: Jason Wang jasow...@redhat.com Date: Thu, 20 Nov 2014 17:03:05 +0800 We currently trigger BUG when VIRTIO_NET_F_CTRL_VQ is not set but one of features depending on it is. That's not a friendly way to report errors to hypervisors. Let's check, and fail probe instead. Cc: Rusty

Re: [PATCH V2 net] virtio-net: validate features during probe

From: Jason Wang jasow...@redhat.com Date: Wed, 19 Nov 2014 17:21:46 +0800 More compact, looks good. Thanks I am assuming there is therefore a V3 of this patch forthcoming. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org

Re: TUN_F_UFO change breaks live migration

From: Michael S. Tsirkin m...@redhat.com Date: Tue, 11 Nov 2014 17:57:50 +0200 Basically userspace assumed that features will only ever be added, never removed, so this change is breaking it. I essentially agree. We can't just toss feature bits like this which have been present for so long.

Re: [PATCH net] Revert drivers/net: Disable UFO through virtio in macvtap and tun

From: Ben Hutchings b...@decadent.org.uk Date: Tue, 11 Nov 2014 17:12:58 + This reverts commit 88e0e0e5aa722b193c8758c8b45d041de5316924 for the tap drivers, but leaves UFO disabled in virtio_net. libvirt at least assumes that tap features will never be dropped in new kernel versions,

Re: [PATCH] virtio_net: fix use after free

From: Michael S. Tsirkin m...@redhat.com Date: Wed, 15 Oct 2014 16:23:28 +0300 You used __netif_subqueue_stopped but that seems to use a slightly more expensive test_bit internally. More expensive in what sense? It should be roughly the same as x y sans the volatile. Anyways I'm ambivalent

Re: [PATCH net-next RFC 0/3] virtio-net: Conditionally enable tx interrupt

From: Jason Wang jasow...@redhat.com Date: Sat, 11 Oct 2014 15:16:43 +0800 We free old transmitted packets in ndo_start_xmit() currently, so any packet must be orphaned also there. This was used to reduce the overhead of tx interrupt to achieve better performance. But this may not work for

Re: [PATCH 08/16] virtio_net: drop config_enable

From: Michael S. Tsirkin m...@redhat.com Date: Sun, 5 Oct 2014 19:07:13 +0300 Now that virtio core ensures config changes don't arrive during probing, drop config_enable flag in virtio net. On removal, flush is now sufficient to guarantee that no change work is queued. This help simplify

Re: [PATCH] VMXNET3: Check for map error in vmxnet3_set_mc

From: Andy King ack...@vmware.com Date: Tue, 2 Sep 2014 13:13:44 -0700 We should check if the map of the table actually succeeds, and also free resources accordingly. Version bumped to 1.2.1.0 Acked-by: Shelley Gong shelleyg...@vmware.com Acked-by: Bhavesh Davda bhav...@vmware.com

Re: [PATCH 0/3] virtio: simplify virtio_ring.

From: Rusty Russell ru...@rustcorp.com.au Date: Wed, 3 Sep 2014 13:59:13 +0930 I resurrected these patches after prompting from Andy Lutomirski's recent patches. I put them on the back-burner because vring_bench had a 15% slowdown on my laptop: pktgen testing revealed a speedup, if

Re: [PATCH] vhost: Add polling mode

From: Michael S. Tsirkin m...@redhat.com Date: Sun, 10 Aug 2014 21:45:59 +0200 On Sun, Aug 10, 2014 at 11:30:35AM +0300, Razya Ladelsky wrote: ... And, did your tests actually produce 100% load on both host CPUs? ... Michael, please do not quote an entire patch just to ask a one line

Re: [PATCH v3 net-next 0/2] rx busy polling support for virtio-net

From: Jason Wang jasow...@redhat.com Date: Wed, 23 Jul 2014 16:33:53 +0800 This series introduces the support for rx busy polling support. This was useful for reduing the latency for a kvm guest. Instead of introducing new states and spinlocks, this series re-uses NAPI state to synchonrize

Re: [PATCH 0/25] Replace DEFINE_PCI_DEVICE_TABLE macro use

From: Benoit Taine benoit.ta...@lip6.fr Date: Fri, 18 Jul 2014 17:26:47 +0200 We should prefer `const struct pci_device_id` over `DEFINE_PCI_DEVICE_TABLE` to meet kernel coding style guidelines. This issue was reported by checkpatch. A simplified version of the semantic patch that makes

Re: [PATCH net-next V2 0/3] rx busy polling support for virtio-net

From: Jason Wang jasow...@redhat.com Date: Wed, 16 Jul 2014 14:21:44 +0800 Hi all: This series introduces the support for rx busy polling support. This was useful for reduing the latency for a kvm guest. Patch 1-2 introduces helpers which is used for rx busy polling. Patch 3 implement the

Re: [PULL net-next] vhost enhancements for 3.16

From: Michael S. Tsirkin m...@redhat.com Date: Mon, 2 Jun 2014 23:55:15 +0300 The following changes since commit 96b2e73c5471542cb9c622c4360716684f8797ed: Revert net/mlx4_en: Use affinity hint (2014-06-02 00:18:48 -0700) are available in the git repository at:

Re: [PATCH] [resend] net: get rid of SET_ETHTOOL_OPS

From: Wilfried Klaebe w-l...@lebenslange-mailadresse.de Date: Sun, 11 May 2014 00:12:32 + net: get rid of SET_ETHTOOL_OPS Dave Miller mentioned he'd like to see SET_ETHTOOL_OPS gone. This does that. Mostly done via coccinelle script: @@ struct ethtool_ops *ops; struct net_device

Re: [PATCH] vsock: Make transport the proto owner

From: Andy King ack...@vmware.com Date: Thu, 1 May 2014 15:20:43 -0700 Right now the core vsock module is the owner of the proto family. This means there's nothing preventing the transport module from unloading if there are open sockets, which results in a panic. Fix that by allowing the

Re: [PATCH] virtio_net: zero is an invald queue_pairs number

From: Amos Kong ak...@redhat.com Date: Fri, 18 Apr 2014 13:45:41 +0800 Execute ethtool -L eth0 combined 0 in guest, if multiqueue is enabled, virtnet_send_command() will return -EINVAL error, there is a validation in QEMU. But if multiqueue is disabled, virtnet_set_queues() will just

Re: [PATCHv2 net] vhost: fix total length when packets are too short

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 27 Mar 2014 12:00:26 +0200 When mergeable buffers are disabled, and the incoming packet is too large for the rx buffer, get_rx_bufs returns success. This was intentional in order for make recvmsg truncate the packet and then handle_rx

Re: [PATCH net] vhost: validate vhost_get_vq_desc return value

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 27 Mar 2014 12:53:37 +0200 vhost fails to validate negative error code from vhost_get_vq_desc causing a crash: we are using -EFAULT which is 0xfff2 as vector size, which exceeds the allocated size. The code in question was introduced

Re: [PATCH net V2] vhost: net: switch to use data copy if pending DMAs exceed the limit

From: Jason Wang jasow...@redhat.com Date: Fri, 7 Mar 2014 13:28:27 +0800 This is because the delay added by htb may lead the delay the finish of DMAs and cause the pending DMAs for tap0 exceeds the limit (VHOST_MAX_PEND). In this case vhost stop handling tx request until htb send some

Re: [PATCH net V2] virtio-net: alloc big buffers also when guest can receive UFO

From: Jason Wang jasow...@redhat.com Date: Fri, 21 Feb 2014 13:08:04 +0800 We should alloc big buffers also when guest can receive UFO pakcets to let the big packets be fit into guest rx buffer. Fixes 5c5167515d80f78f6bb538492c423adcae31ad65 (virtio-net: Allow UFO feature to be set and

Re: [PATCH net v2] vhost: fix a theoretical race in device cleanup

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 13 Feb 2014 11:45:11 +0200 vhost_zerocopy_callback accesses VQ right after it drops a ubuf reference. In theory, this could race with device removal which waits on the ubuf kref, and crash on use after free. Do all accesses within rcu

Re: [PATCH net v2] vhost: fix ref cnt checking deadlock

From: Michael S. Tsirkin m...@redhat.com Date: Thu, 13 Feb 2014 11:42:05 +0200 vhost checked the counter within the refcnt before decrementing. It really wanted to know that it is the one that has the last reference, as a way to batch freeing resources a bit more efficiently. Note: we only

Re: [PATCH net 1/3] kref: add kref_sub_return

From: Greg KH gre...@linuxfoundation.org Date: Thu, 13 Feb 2014 16:03:20 -0800 So how about just open coding a kref for this structure, as it wants something that doesn't fit into the kref model, and should be pretty simple to do (you can ensure you get the locking right, unlike almost all

Re: [PATCH net 1/3] kref: add kref_sub_return

From: Greg KH gre...@linuxfoundation.org Date: Wed, 12 Feb 2014 17:39:02 -0800 Yes, that's horrible as well, but as was already pointed out in this thread, you can't rely on that value to really be 1 after reading it due to the way krefs work, what happened if someone else just grabbed it?

Re: [PATCH net 1/3] kref: add kref_sub_return

From: David Miller da...@davemloft.net Date: Wed, 12 Feb 2014 23:05:06 -0500 (EST) From: Greg KH gre...@linuxfoundation.org Date: Wed, 12 Feb 2014 17:39:02 -0800 Yes, that's horrible as well, but as was already pointed out in this thread, you can't rely on that value to really be 1 after

Re: [PATCH net-next] virtio-net: fix build error when CONFIG_AVERAGE is not enabled

From: Michael Dalton mwdal...@google.com Date: Fri, 17 Jan 2014 01:27:08 -0800 Commit ab7db91705e9 (virtio-net: auto-tune mergeable rx buffer size for improved performance) introduced a virtio-net dependency on EWMA. The inclusion of EWMA is controlled by CONFIG_AVERAGE. Fix build error when

Re: [PATCH net-next v4 1/6] net: allow 0 order atomic page alloc in skb_page_frag_refill

From: David Miller da...@davemloft.net Date: Thu, 16 Jan 2014 15:28:00 -0800 (PST) All 6 patches applied. Actually, I reverted, please resubmit this series with the following build warning corrected: net/core/net-sysfs.c: In function ‘rx_queue_add_kobject’: net/core/net-sysfs.c:767:21: warning

Re: [PATCH net-next] virtio-net: drop rq-max and rq-num

From: Jason Wang jasow...@redhat.com Date: Thu, 16 Jan 2014 14:45:24 +0800 It looks like there's no need for those two fields: - Unless there's a failure for the first refill try, rq-max should be always equal to the vring size. - rq-num is only used to determine the condition that we

Re: [PATCH net-next v5 0/6] virtio-net: mergeable rx buffer size auto-tuning

This series does not apply cleanly to net-next. ___ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [PATCH net-next v6 0/6] virtio-net: mergeable rx buffer size auto-tuning

From: Michael Dalton mwdal...@google.com Date: Thu, 16 Jan 2014 22:23:24 -0800 The virtio-net device currently uses aligned MTU-sized mergeable receive packet buffers. Network throughput for workloads with large average packet size can be improved by posting larger receive packet buffers.

<    1   2   3   4   5   >