This patch tries to batched used ring update during RX. This is pretty
fit for the case when guest is much faster (e.g dpdk based
backend). In this case, used ring is almost empty:
- we may get serious cache line misses/contending on both used ring
and used idx.
- at most 1 packet could be dequeued at one time, batching in guest
does not make much effect.
Update used ring in a batch can help since guest won't access the used
ring until used idx was advanced for several descriptors and since we
advance used ring for every N packets, guest will only need to access
used idx for every N packet since it can cache the used idx. To have a
better interaction for both batch dequeuing and dpdk batching,
VHOST_RX_BATCH was used as the maximum number of descriptors that
could be batched.
Test were done between two machines with 2.40GHz Intel(R) Xeon(R) CPU
E5-2630 connected back to back through ixgbe. Traffic were generated
on one remote ixgbe through MoonGen and measure the RX pps through
testpmd in guest when do xdp_redirect_map from local ixgbe to
tap. RX pps were increased from 3.05 Mpps to 4.00 Mpps (about 31%
improvement).
One possible concern for this is the implications for TCP (especially
latency sensitive workload). Result[1] does not show obvious changes
for most of the netperf test (RR, TX, and RX). And we do get some
improvements for RX on some specific size.
Guest RX:
size/sessions/+thu%/+normalize%
64/ 1/ +2%/ +2%
64/ 2/ +2%/ -1%
64/ 4/ +1%/ +1%
64/ 8/ 0%/ 0%
256/ 1/ +6%/ -3%
256/ 2/ -3%/ +2%
256/ 4/ +11%/ +11%
256/ 8/ 0%/ 0%
512/ 1/ +4%/ 0%
512/ 2/ +2%/ +2%
512/ 4/ 0%/ -1%
512/ 8/ -8%/ -8%
1024/ 1/ -7%/ -17%
1024/ 2/ -8%/ -7%
1024/ 4/ +1%/ 0%
1024/ 8/ 0%/ 0%
2048/ 1/ +30%/ +14%
2048/ 2/ +46%/ +40%
2048/ 4/ 0%/ 0%
2048/ 8/ 0%/ 0%
4096/ 1/ +23%/ +22%
4096/ 2/ +26%/ +23%
4096/ 4/ 0%/ +1%
4096/ 8/ 0%/ 0%
16384/ 1/ -2%/ -3%
16384/ 2/ +1%/ -4%
16384/ 4/ -1%/ -3%
16384/ 8/ 0%/ -1%
65535/ 1/ +15%/ +7%
65535/ 2/ +4%/ +7%
65535/ 4/ 0%/ +1%
65535/ 8/ 0%/ 0%
TCP_RR:
size/sessions/+thu%/+normalize%
1/ 1/ 0%/ +1%
1/ 25/ +2%/ +1%
1/ 50/ +4%/ +1%
64/ 1/ 0%/ -4%
64/ 25/ +2%/ +1%
64/ 50/ 0%/ -1%
256/ 1/ 0%/ 0%
256/ 25/ 0%/ 0%
256/ 50/ +4%/ +2%
Guest TX:
size/sessions/+thu%/+normalize%
64/ 1/ +4%/ -2%
64/ 2/ -6%/ -5%
64/ 4/ +3%/ +6%
64/ 8/ 0%/ +3%
256/ 1/ +15%/ +16%
256/ 2/ +11%/ +12%
256/ 4/ +1%/ 0%
256/ 8/ +5%/ +5%
512/ 1/ -1%/ -6%
512/ 2/ 0%/ -8%
512/ 4/ -2%/ +4%
512/ 8/ +6%/ +9%
1024/ 1/ +3%/ +1%
1024/ 2/ +3%/ +9%
1024/ 4/ 0%/ +7%
1024/ 8/ 0%/ +7%
2048/ 1/ +8%/ +2%
2048/ 2/ +3%/ -1%
2048/ 4/ -1%/ +11%
2048/ 8/ +3%/ +9%
4096/ 1/ +8%/ +8%
4096/ 2/ 0%/ -7%
4096/ 4/ +4%/ +4%
4096/ 8/ +2%/ +5%
16384/ 1/ -3%/ +1%
16384/ 2/ -1%/ -12%
16384/ 4/ -1%/ +5%
16384/ 8/ 0%/ +1%
65535/ 1/ 0%/ -3%
65535/ 2/ +5%/ +16%
65535/ 4/ +1%/ +2%
65535/ 8/ +1%/ -1%
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch implements XDP transmission for TAP. Since we can't create
new queues for TAP during XDP set, exist ptr_ring was reused for
queuing XDP buffers. To differ xdp_buff from sk_buff, TUN_XDP_FLAG
(0x1UL) was encoded into lowest bit of xpd_buff pointer during
ptr_ring_produce, and was decoded during consuming. XDP metadata was
stored in the headroom of the packet which should work in most of
cases since driver usually reserve enough headroom. Very minor changes
were done for vhost_net: it just need to peek the length depends on
the type of pointer.
Tests were done on two Intel E5-2630 2.40GHz machines connected back
to back through two 82599ES. Traffic were generated/received through
MoonGen/testpmd(rxonly). It reports ~20% improvements when
xdp_redirect_map is doing redirection from ixgbe to TAP (from 2.50Mpps
to 3.05Mpps)
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch switches to use ptr_ring instead of skb_array. This will be
used to enqueue different types of pointers by encoding type into
lower bits.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Matthew found a roughly 40% tcp throughput regression with commit
c67df11f(vhost_net: try batch dequing from skb array) as discussed
in the following thread:
https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html
Eventually we figured out that it was a skb leak in handle_rx()
when sending packets to the VM. This usually happens when a guest
can not drain out vq as fast as vhost fills in, afterwards it sets
off the traffic jam and leaks skb(s) which occurs as no headcount
to send on the vq from vhost side.
This can be avoided by making sure we have got enough headcount
before actually consuming a skb from the batched rx array while
transmitting, which is simply done by moving checking the zero
headcount a bit ahead.
Signed-off-by: Wei Xu <wexu@redhat.com>
Reported-by: Matthew Rosato <mjrosato@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We always poll tx for socket, this is sub optimal since this will
slightly increase the waitqueue traversing time and more important,
vhost could not benefit from commit 9e641bdcfa ("net-tun:
restructure tun_do_read for better sleep/wakeup efficiency") even if
we've stopped rx polling during handle_rx(), tx poll were still left
in the waitqueue.
Pktgen from a remote host to VM over mlx4 on two 2.00GHz Xeon E5-2650
shows 11.7% improvements on rx PPS. (from 1.28Mpps to 1.44Mpps)
Cc: Wei Xu <wexu@redhat.com>
Cc: Matthew Rosato <mjrosato@linux.vnet.ibm.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Vhost-net has a hard limit on the number of zerocopy skbs in flight.
When reached, transmission stalls. Stalls cause latency, as well as
head-of-line blocking of other flows that do not use zerocopy.
Instead of stalling, revert to copy-based transmission.
Tested by sending two udp flows from guest to host, one with payload
of VHOST_GOODCOPY_LEN, the other too small for zerocopy (1B). The
large flow is redirected to a netem instance with 1MBps rate limit
and deep 1000 entry queue.
modprobe ifb
ip link set dev ifb0 up
tc qdisc add dev ifb0 root netem limit 1000 rate 1MBit
tc qdisc add dev tap0 ingress
tc filter add dev tap0 parent ffff: protocol ip \
u32 match ip dport 8000 0xffff \
action mirred egress redirect dev ifb0
Before the delay, both flows process around 80K pps. With the delay,
before this patch, both process around 400. After this patch, the
large flow is still rate limited, while the small reverts to its
original rate. See also discussion in the first link, below.
Without rate limiting, {1, 10, 100}x TCP_STREAM tests continued to
send at 100% zerocopy.
The limit in vhost_exceeds_maxpend must be carefully chosen. With
vq->num >> 1, the flows remain correlated. This value happens to
correspond to VHOST_MAX_PENDING for vq->num == 256. Allow smaller
fractions and ensure correctness also for much smaller values of
vq->num, by testing the min() of both explicitly. See also the
discussion in the second link below.
Changes
v1 -> v2
- replaced min with typed min_t
- avoid unnecessary whitespace change
Link:http://lkml.kernel.org/r/CAF=yD-+Wk9sc9dXMUq1+x_hh=3ThTXa6BnZkygP3tgVpjbp93g@mail.gmail.com
Link:http://lkml.kernel.org/r/20170819064129.27272-1-den@klaipeden.com
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We check tx avail through vhost_enable_notify() in the past which is
wrong since it only checks whether or not guest has filled more
available buffer since last avail idx synchronization which was just
done by vhost_vq_avail_empty() before. What we really want is checking
pending buffers in the avail ring. Fix this by calling
vhost_vq_avail_empty() instead.
This issue could be noticed by doing netperf TCP_RR benchmark as
client from guest (but not host). With this fix, TCP_RR from guest to
localhost restores from 1375.91 trans per sec to 55235.28 trans per
sec on my laptop (Intel(R) Core(TM) i7-5600U CPU @ 2.60GHz).
Fixes: 0308813724 ("vhost_net: basic polling support")
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
v2: added the change in drivers/vhost/net.c as spotted
by Willem.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Prepare the datapath for refcounted ubuf_info. Clone ubuf_info with
skb_zerocopy_clone() wherever needed due to skb split, merge, resize
or clone.
Split skb_orphan_frags into two variants. The split, merge, .. paths
support reference counted zerocopy buffers, so do not do a deep copy.
Add skb_orphan_frags_rx for paths that may loop packets to receive
sockets. That is not allowed, as it may cause unbounded latency.
Deep copy all zerocopy copy buffers, ref-counted or not, in this path.
The exact locations to modify were chosen by exhaustively searching
through all code that might modify skb_frag references and/or the
the SKBTX_DEV_ZEROCOPY tx_flags bit.
The changes err on the safe side, in two ways.
(1) legacy ubuf_info paths virtio and tap are not modified. They keep
a 1:1 ubuf_info to sk_buff relationship. Calls to skb_orphan_frags
still call skb_copy_ubufs and thus copy frags in this case.
(2) not all copies deep in the stack are addressed yet. skb_shift,
skb_split and skb_try_coalesce can be refined to avoid copying.
These are not in the hot path and this patch is hairy enough as
is, so that is left for future refinement.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__GFP_REPEAT was designed to allow retry-but-eventually-fail semantic to
the page allocator. This has been true but only for allocations
requests larger than PAGE_ALLOC_COSTLY_ORDER. It has been always
ignored for smaller sizes. This is a bit unfortunate because there is
no way to express the same semantic for those requests and they are
considered too important to fail so they might end up looping in the
page allocator for ever, similarly to GFP_NOFAIL requests.
Now that the whole tree has been cleaned up and accidental or misled
usage of __GFP_REPEAT flag has been removed for !costly requests we can
give the original flag a better name and more importantly a more useful
semantic. Let's rename it to __GFP_RETRY_MAYFAIL which tells the user
that the allocator would try really hard but there is no promise of a
success. This will work independent of the order and overrides the
default allocator behavior. Page allocator users have several levels of
guarantee vs. cost options (take GFP_KERNEL as an example)
- GFP_KERNEL & ~__GFP_RECLAIM - optimistic allocation without _any_
attempt to free memory at all. The most light weight mode which even
doesn't kick the background reclaim. Should be used carefully because
it might deplete the memory and the next user might hit the more
aggressive reclaim
- GFP_KERNEL & ~__GFP_DIRECT_RECLAIM (or GFP_NOWAIT)- optimistic
allocation without any attempt to free memory from the current
context but can wake kswapd to reclaim memory if the zone is below
the low watermark. Can be used from either atomic contexts or when
the request is a performance optimization and there is another
fallback for a slow path.
- (GFP_KERNEL|__GFP_HIGH) & ~__GFP_DIRECT_RECLAIM (aka GFP_ATOMIC) -
non sleeping allocation with an expensive fallback so it can access
some portion of memory reserves. Usually used from interrupt/bh
context with an expensive slow path fallback.
- GFP_KERNEL - both background and direct reclaim are allowed and the
_default_ page allocator behavior is used. That means that !costly
allocation requests are basically nofail but there is no guarantee of
that behavior so failures have to be checked properly by callers
(e.g. OOM killer victim is allowed to fail currently).
- GFP_KERNEL | __GFP_NORETRY - overrides the default allocator behavior
and all allocation requests fail early rather than cause disruptive
reclaim (one round of reclaim in this implementation). The OOM killer
is not invoked.
- GFP_KERNEL | __GFP_RETRY_MAYFAIL - overrides the default allocator
behavior and all allocation requests try really hard. The request
will fail if the reclaim cannot make any progress. The OOM killer
won't be triggered.
- GFP_KERNEL | __GFP_NOFAIL - overrides the default allocator behavior
and all allocation requests will loop endlessly until they succeed.
This might be really dangerous especially for larger orders.
Existing users of __GFP_REPEAT are changed to __GFP_RETRY_MAYFAIL
because they already had their semantic. No new users are added.
__alloc_pages_slowpath is changed to bail out for __GFP_RETRY_MAYFAIL if
there is no progress and we have already passed the OOM point.
This means that all the reclaim opportunities have been exhausted except
the most disruptive one (the OOM killer) and a user defined fallback
behavior is more sensible than keep retrying in the page allocator.
[akpm@linux-foundation.org: fix arch/sparc/kernel/mdesc.c]
[mhocko@suse.com: semantic fix]
Link: http://lkml.kernel.org/r/20170626123847.GM11534@dhcp22.suse.cz
[mhocko@kernel.org: address other thing spotted by Vlastimil]
Link: http://lkml.kernel.org/r/20170626124233.GN11534@dhcp22.suse.cz
Link: http://lkml.kernel.org/r/20170623085345.11304-3-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Alex Belits <alex.belits@cavium.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: David Daney <david.daney@cavium.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: NeilBrown <neilb@suse.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We used to dequeue one skb during recvmsg() from skb_array, this could
be inefficient because of the bad cache utilization and spinlock
touching for each packet. This patch tries to batch them by calling
batch dequeuing helpers explicitly on the exported skb array and pass
the skb back through msg_control for underlayer socket to finish the
userspace copying. Batch dequeuing is also the requirement for more
batching improvement on receive path.
Tests were done by pktgen on tap with XDP1 in guest. Host is Intel(R)
Xeon(R) CPU E5-2650 0 @ 2.00GHz.
rx batch | pps
0 2.25Mpps
1 2.33Mpps (+3.56%)
4 2.33Mpps (+3.56%)
16 2.35Mpps (+4.44%)
64 2.42Mpps (+7.56%) <- Default rx batching
128 2.40Mpps (+6.67%)
256 2.38Mpps (+5.78%)
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
vhost code uses __GFP_REPEAT when allocating vhost_virtqueue resp.
vhost_vsock because it would really like to prefer kmalloc to the
vmalloc fallback - see 23cc5a991c ("vhost-net: extend device
allocation to vmalloc") for more context. Michael Tsirkin has also
noted:
"__GFP_REPEAT overhead is during allocation time. Using vmalloc means
all accesses are slowed down. Allocation is not on data path, accesses
are."
The similar applies to other vhost_kvzalloc users.
Let's teach kvmalloc_node to handle __GFP_REPEAT properly. There are
two things to be careful about. First we should prevent from the OOM
killer and so have to involve __GFP_NORETRY by default and secondly
override __GFP_REPEAT for !costly order requests as the __GFP_REPEAT is
ignored for !costly orders.
Supporting __GFP_REPEAT like semantic for !costly request is possible it
would require changes in the page allocator. This is out of scope of
this patch.
This patch shouldn't introduce any functional change.
Link: http://lkml.kernel.org/r/20170306103032.2540-3-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix up affected files that include this signal functionality via sched.h.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We are going to split <linux/sched/clock.h> out of <linux/sched.h>, which
will have to be picked up from other headers and .c files.
Create a trivial placeholder <linux/sched/clock.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.
Include the new header in the files that are going to need it.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Renaming tap related APIs, data structures and macros in tap.c from macvtap_.* to tap_.*
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch tries to utilize tuntap rx batching by peeking the tx
virtqueue during transmission, if there's more available buffers in
the virtqueue, set MSG_MORE flag for a hint for backend (e.g tuntap)
to batch the packets.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With the s390 special case of a yielding cpu_relax() implementation gone,
we can now remove all users of cpu_relax_lowlatency() and replace them
with cpu_relax().
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Noam Camus <noamc@ezchip.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: virtualization@lists.linux-foundation.org
Cc: xen-devel@lists.xenproject.org
Link: http://lkml.kernel.org/r/1477386195-32736-5-git-send-email-borntraeger@de.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
- New vsock device support in host and guest
- Platform IOMMU support in host and guest,
including compatibility quirks for legacy systems.
- Misc fixes and cleanups.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJXofvbAAoJECgfDbjSjVRpUTIH/iEoK9h636tBayXy0PXkPby0
6fMaRFy6H1HgEttgDhJE8Pqg/ba3qaW9Em0fHyFq7Mp2waFHAZ8hAT8phC6TAK3c
CIBnfzyyuI8u3N9SnNOfelPVcwCBfuALuuTsXB/rwKbYQEVv+U5Rdt3Vyx9+lXkj
P005klz7PfqxFhQrrnj4Eh7VawtHwmMuLH8YoWpCZpM71dHPo6eL+3ftKwhH2boo
qK86uVprwba03Pewpm13vQnotemfVfUUkjXd4EJpG3dx7E0KZosuj0ZG9OV8mPGQ
Cl2gBdUhocdJgeUnAHmf6tumYi9KFlYfy6xLy44YMmN7FL3E9nQjaKZp25UKfiM=
=ztIm
-----END PGP SIGNATURE-----
Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Pull virtio/vhost updates from Michael Tsirkin:
- new vsock device support in host and guest
- platform IOMMU support in host and guest, including compatibility
quirks for legacy systems.
- misc fixes and cleanups.
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
VSOCK: Use kvfree()
vhost: split out vringh Kconfig
vhost: detect 32 bit integer wrap around
vhost: new device IOTLB API
vhost: drop vringh dependency
vhost: convert pre sorted vhost memory array to interval tree
vhost: introduce vhost memory accessors
VSOCK: Add Makefile and Kconfig
VSOCK: Introduce vhost_vsock.ko
VSOCK: Introduce virtio_transport.ko
VSOCK: Introduce virtio_vsock_common.ko
VSOCK: defer sock removal to transports
VSOCK: transport-specific vsock_transport functions
vhost: drop vringh dependency
vop: pull in vhost Kconfig
virtio: new feature to detect IOMMU device quirk
balloon: check the number of available pages in leak balloon
vhost: lockless enqueuing
vhost: simplify work flushing
This patch tries to implement an device IOTLB for vhost. This could be
used with userspace(qemu) implementation of DMA remapping
to emulate an IOMMU for the guest.
The idea is simple, cache the translation in a software device IOTLB
(which is implemented as an interval tree) in vhost and use vhost_net
file descriptor for reporting IOTLB miss and IOTLB
update/invalidation. When vhost meets an IOTLB miss, the fault
address, size and access can be read from the file. After userspace
finishes the translation, it writes the translated address to the
vhost_net file to update the device IOTLB.
When device IOTLB is enabled by setting VIRTIO_F_IOMMU_PLATFORM all vq
addresses set by ioctl are treated as iova instead of virtual address and
the accessing can only be done through IOTLB instead of direct userspace
memory access. Before each round or vq processing, all vq metadata is
prefetched in device IOTLB to make sure no translation fault happens
during vq processing.
In most cases, virtqueues are contiguous even in virtual address space.
The IOTLB translation for virtqueue itself may make it a little
slower. We might add fast path cache on top of this patch.
Signed-off-by: Jason Wang <jasowang@redhat.com>
[mst: use virtio feature bit: VHOST_F_DEVICE_IOTLB -> VIRTIO_F_IOMMU_PLATFORM ]
[mst: fix build warnings ]
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
[ weiyj.lk: missing unlock on error ]
Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com>
Current pre-sorted memory region array has some limitations for future
device IOTLB conversion:
1) need extra work for adding and removing a single region, and it's
expected to be slow because of sorting or memory re-allocation.
2) need extra work of removing a large range which may intersect
several regions with different size.
3) need trick for a replacement policy like LRU
To overcome the above shortcomings, this patch convert it to interval
tree which can easily address the above issue with almost no extra
work.
The patch could be used for:
- Extend the current API and only let the userspace to send diffs of
memory table.
- Simplify Device IOTLB implementation.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
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:
- switch from sk_receive_queue to a skb_array, and resize it when
tx_queue_len was changed.
- introduce a new proto_ops peek_len which was used for peeking the
skb length.
- implement a tun version of peek_len for vhost_net to use and convert
vhost_net to use peek_len if possible.
Pktgen test shows about 15.3% improvement on guest receiving pps for small
buffers:
Before: ~1300000pps
After : ~1500000pps
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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 by stop polling socket during rx
processing. A small drawback is that this introduces some overheads in
light load case because of the extra start/stop polling, but single
netperf TCP_RR does not notice any change. In a super heavy load case,
e.g using pktgen to inject packet to guest, we get about ~8.8%
improvement on pps:
before: ~1240000 pkt/s
after: ~1350000 pkt/s
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch tries to poll for new added tx buffer or socket receive
queue for a while at the end of tx/rx processing. The maximum time
spent on polling were specified through a new kind of vring ioctl.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Looking at how callers use this, maybe we should just rename init_used
to vhost_vq_init_access. The _used suffix was a hint that we
access the vq used ring. But maybe what callers care about is
that it must be called after access_ok.
Also, this function manipulates the vq->is_le field which isn't related
to the vq used ring.
This patch simply renames vhost_init_used() to vhost_vq_init_access() as
suggested by Michael.
No behaviour change.
Signed-off-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Conflicts:
drivers/net/ethernet/rocker/rocker.c
The rocker commit was two overlapping changes, one to rename
the ->vport member to ->pport, and another making the bitmask
expression use '1ULL' instead of plain '1'.
Signed-off-by: David S. Miller <davem@davemloft.net>
After TIPC doesn't depend on iocb argument in its internal
implementations of sendmsg() and recvmsg() hooks defined in proto
structure, no any user is using iocb argument in them at all now.
Then we can drop the redundant iocb argument completely from kinds of
implementations of both sendmsg() and recvmsg() in the entire
networking stack.
Cc: Christoph Hellwig <hch@lst.de>
Suggested-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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 <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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 updated by copy_to_iter,
sometimes-by iov_iter_advance.
Rearrange code to make this obvious.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit ba7438aed9 ("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 recvmsg(). This work well but was broken after
commit 6e03f896b5 ("Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net") which tries
to advance sizeof(struct virtio_net_hdr_mrg_rxbuf). It will fill the
number of buffers at the wrong place. This patch fixes this.
Fixes 6e03f896b5
("Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net")
Cc: David S. Miller <davem@davemloft.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/vxlan.c
drivers/vhost/net.c
include/linux/if_vlan.h
net/core/dev.c
The net/core/dev.c conflict was the overlap of one commit marking an
existing function static whilst another was adding a new function.
In the include/linux/if_vlan.h case, the type used for a local
variable was changed in 'net', whereas the function got rewritten
to fix a stacked vlan bug in 'net-next'.
In drivers/vhost/net.c, Al Viro's iov_iter conversions in 'net-next'
overlapped with an endainness fix for VHOST 1.0 in 'net'.
In drivers/net/vxlan.c, vxlan_find_vni() added a 'flags' parameter
in 'net-next' whereas in 'net' there was a bug fix to pass in the
correct network namespace pointer in calls to this function.
Signed-off-by: David S. Miller <davem@davemloft.net>
In virtio 1.0 mode, when mergeable buffers are enabled on a big-endian
host, num_buffers wasn't byte-swapped correctly, so large incoming
packets got corrupted.
To fix, fill it in within hdr - this also makes sure it gets
the correct type.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/xen-netfront.c
Minor overlapping changes in xen-netfront.c, mostly to do
with some buffer management changes alongside the split
of stats into TX and RX.
Signed-off-by: David S. Miller <davem@davemloft.net>
The same macros are used for rx as well. So rename it.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 8b38694a2d
vhost/net: virtio 1.0 byte swap
had this chunk:
- heads[headcount - 1].len += datalen;
+ heads[headcount - 1].len = cpu_to_vhost32(vq, len - datalen);
This adds datalen with the wrong sign, causing guest panics.
Fixes: 8b38694a2d
Reported-by: Alex Williamson <alex.williamson@redhat.com>
Suggested-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Pull networking updates from David Miller:
1) New offloading infrastructure and example 'rocker' driver for
offloading of switching and routing to hardware.
This work was done by a large group of dedicated individuals, not
limited to: Scott Feldman, Jiri Pirko, Thomas Graf, John Fastabend,
Jamal Hadi Salim, Andy Gospodarek, Florian Fainelli, Roopa Prabhu
2) Start making the networking operate on IOV iterators instead of
modifying iov objects in-situ during transfers. Thanks to Al Viro
and Herbert Xu.
3) A set of new netlink interfaces for the TIPC stack, from Richard
Alpe.
4) Remove unnecessary looping during ipv6 routing lookups, from Martin
KaFai Lau.
5) Add PAUSE frame generation support to gianfar driver, from Matei
Pavaluca.
6) Allow for larger reordering levels in TCP, which are easily
achievable in the real world right now, from Eric Dumazet.
7) Add a variable of napi_schedule that doesn't need to disable cpu
interrupts, from Eric Dumazet.
8) Use a doubly linked list to optimize neigh_parms_release(), from
Nicolas Dichtel.
9) Various enhancements to the kernel BPF verifier, and allow eBPF
programs to actually be attached to sockets. From Alexei
Starovoitov.
10) Support TSO/LSO in sunvnet driver, from David L Stevens.
11) Allow controlling ECN usage via routing metrics, from Florian
Westphal.
12) Remote checksum offload, from Tom Herbert.
13) Add split-header receive, BQL, and xmit_more support to amd-xgbe
driver, from Thomas Lendacky.
14) Add MPLS support to openvswitch, from Simon Horman.
15) Support wildcard tunnel endpoints in ipv6 tunnels, from Steffen
Klassert.
16) Do gro flushes on a per-device basis using a timer, from Eric
Dumazet. This tries to resolve the conflicting goals between the
desired handling of bulk vs. RPC-like traffic.
17) Allow userspace to ask for the CPU upon what a packet was
received/steered, via SO_INCOMING_CPU. From Eric Dumazet.
18) Limit GSO packets to half the current congestion window, from Eric
Dumazet.
19) Add a generic helper so that all drivers set their RSS keys in a
consistent way, from Eric Dumazet.
20) Add xmit_more support to enic driver, from Govindarajulu
Varadarajan.
21) Add VLAN packet scheduler action, from Jiri Pirko.
22) Support configurable RSS hash functions via ethtool, from Eyal
Perry.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1820 commits)
Fix race condition between vxlan_sock_add and vxlan_sock_release
net/macb: fix compilation warning for print_hex_dump() called with skb->mac_header
net/mlx4: Add support for A0 steering
net/mlx4: Refactor QUERY_PORT
net/mlx4_core: Add explicit error message when rule doesn't meet configuration
net/mlx4: Add A0 hybrid steering
net/mlx4: Add mlx4_bitmap zone allocator
net/mlx4: Add a check if there are too many reserved QPs
net/mlx4: Change QP allocation scheme
net/mlx4_core: Use tasklet for user-space CQ completion events
net/mlx4_core: Mask out host side virtualization features for guests
net/mlx4_en: Set csum level for encapsulated packets
be2net: Export tunnel offloads only when a VxLAN tunnel is created
gianfar: Fix dma check map error when DMA_API_DEBUG is enabled
cxgb4/csiostor: Don't use MASTER_MUST for fw_hello call
net: fec: only enable mdio interrupt before phy device link up
net: fec: clear all interrupt events to support i.MX6SX
net: fec: reset fep link status in suspend function
net: sock: fix access via invalid file descriptor
net: introduce helper macro for_each_cmsghdr
...
Note that the code _using_ ->msg_iter at that point will be very
unhappy with anything other than unshifted iovec-backed iov_iter.
We still need to convert users to proper primitives.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
I had to add an explicit tag to suppress compiler warning:
gcc isn't smart enough to notice that
len is always initialized since function is called with size > 0.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
vhost/net keeps a copy of the used ring in host memory but (ab)uses
the length field for internal house-keeping. This works because the
length in the used ring for tx is always 0. In order to suppress sparse
warnings, we force native endianness here.
Note that these values are never exposed to guests.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
Commit 23cc5a991c ("vhost-net: extend device allocation to vmalloc")
added another open-coded version of kvfree (which is available since
v3.15-rc5), nuke it.
Signed-off-by: Romain Francoise <romain@orebokech.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
commit 2ae76693b8bcabf370b981cd00c36cd41d33fabc
vhost: replace rcu with mutex
replaced rcu sync for memory accesses with VQ mutex locl/unlock.
This is correct since all accesses are under VQ mutex, but incomplete:
we still do useless rcu lock/unlock operations, someone might copy this
code into some other context where this won't be right.
This use of RCU is also non standard and hard to understand.
Let's copy the pointer to each VQ structure, this way
the access rules become straight-forward, and there's
no need for RCU anymore.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Refactor code to make sure features are only accessed
under VQ mutex. This makes everything simpler, no need
for RCU here anymore.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Michael Mueller provided a patch to reduce the size of
vhost-net structure as some allocations could fail under
memory pressure/fragmentation. We are still left with
high order allocations though.
This patch is handling the problem at the core level, allowing
vhost structures to use vmalloc() if kmalloc() failed.
As vmalloc() adds overhead on a critical network path, add __GFP_REPEAT
to kzalloc() flags to do this fallback only when really needed.
People are still looking at cleaner ways to handle the problem
at the API level, probably passing in multiple iovecs.
This hack seems consistent with approaches
taken since then by drivers/vhost/scsi.c and net/core/dev.c
Based on patch by Romain Francoise.
Cc: Michael Mueller <mimu@linux.vnet.ibm.com>
Signed-off-by: Romain Francoise <romain@orebokech.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
vhost fails to validate negative error code
from vhost_get_vq_desc causing
a crash: we are using -EFAULT which is 0xfffffff2
as vector size, which exceeds the allocated size.
The code in question was introduced in commit
8dd014adfe
vhost-net: mergeable buffers support
CVE-2014-0055
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
