When device IOTLB is enabled, all address translations were stored in
interval tree. O(lgN) searching time could be slow for virtqueue
metadata (avail, used and descriptors) since they were accessed much
often than other addresses. So this patch introduces an O(1) array
which points to the interval tree nodes that store the translations of
vq metadata. Those array were update during vq IOTLB prefetching and
were reset during each invalidation and tlb update. Each time we want
to access vq metadata, this small array were queried before interval
tree. This would be sufficient for static mappings but not dynamic
mappings, we could do optimizations on top.
Test were done with l2fwd in guest (2M hugepage):
noiommu | before | after
tx 1.32Mpps | 1.06Mpps(82%) | 1.30Mpps(98%)
rx 2.33Mpps | 1.46Mpps(63%) | 2.29Mpps(98%)
We can almost reach the same performance as noiommu mode.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
When event index was enabled, we need to fetch used event from
userspace memory each time. This userspace fetch (with memory
barrier) could be saved sometime when 1) caching used event and 2)
if used event is ahead of new and old to new updating does not cross
it, we're sure there's no need to notify guest.
This will be useful for heavy tx load e.g guest pktgen test with Linux
driver shows ~3.5% improvement.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
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 use spinlock to synchronize the work list now which may cause
unnecessary contentions. So this patch switch to use llist to remove
this contention. Pktgen tests shows about 5% improvement:
Before:
~1300000 pps
After:
~1370000 pps
Signed-off-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
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>
This patch introduces a helper which will return true if we're sure
that the available ring is empty for a specific vq. When we're not
sure, e.g vq access failure, return false instead. This could be used
for busy polling code to exit the busy loop.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This path introduces a helper which can give a hint for whether or not
there's a work queued in the work list. This could be used for busy
polling code to exit the busy loop.
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>
commit 2751c9882b ("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 is
checking is_le flag so there's always a branch.
To fix, simply check virtio_legacy_is_little_endian first.
Cc: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch brings cross-endian support to vhost when used to implement
legacy virtio devices. Since it is a relatively rare situation, the
feature availability is controlled by a kernel config option (not set
by default).
The vq->is_le boolean field is added to cache the endianness to be
used for ring accesses. It defaults to native endian, as expected
by legacy virtio devices. When the ring gets active, we force little
endian if the device is modern. When the ring is deactivated, we
revert to the native endian default.
If cross-endian was compiled in, a vq->user_be boolean field is added
so that userspace may request a specific endianness. This field is
used to override the default when activating the ring of a legacy
device. It has no effect on modern devices.
Signed-off-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
The current memory accessors logic is:
- little endian if little_endian
- native endian (i.e. no byteswap) if !little_endian
If we want to fully support cross-endian vhost, we also need to be
able to convert to big endian.
Instead of changing the little_endian argument to some 3-value enum, this
patch changes the logic to:
- little endian if little_endian
- big endian if !little_endian
The native endian case is handled by all users with a trivial helper. This
patch doesn't change any functionality, nor it does add overhead.
Signed-off-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Add guest memory access wrappers to handle virtio endianness
conversions.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
We need to use bit 32 for virtio 1.0.
Make vhost_has_feature bool to avoid discarding high bits.
Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Cc: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jason Wang <jasowang@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>
Since vhost_dev_init() forever return 0, some branches are never run,
therefore need to be removed.
Signed-off-by: Zhi Yong Wu <wuzhy@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now, vq->private_data is always accessed under vq mutex. No need to play
the vhost rcu trick.
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Currently, vhost-net and vhost-scsi are sharing the vhost core code.
However, vhost-scsi shares the code by including the vhost.c file
directly.
Making vhost a separate module makes it is easier to share code with
other vhost devices.
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
If device has an owner, we shouldn't touch ubuf_info
since it might be in use.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is supposed to be removed when hdr is moved into vhost_net_virtqueue.
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
vhost.h should not depend on device specific marcos like
VHOST_NET_F_VIRTIO_NET_HDR and VIRTIO_NET_F_MRG_RXBUF.
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
RESET_OWNER ioctl would leave the fd in a bad state if
memory allocation failed: device is stopped
but owner is not reset. Make state changes
after allocating memory, such that a failed
ioctl has no effect.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
On top of 'vhost: Allow device specific fields per vq', we can move device
specific fields to device virt queue from vhost virt queue.
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This is useful for any device who wants device specific fields per vq.
For example, tcm_vhost wants a per vq field to track requests which are
in flight on the vq. Also, on top of this we can add patches to move
things like ubufs from vhost.h out to net.c.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Currently, the polling errors were ignored, which can lead following issues:
- vhost remove itself unconditionally from waitqueue when stopping the poll,
this may crash the kernel since the previous attempt of starting may fail to
add itself to the waitqueue
- userspace may think the backend were successfully set even when the polling
failed.
Solve this by:
- check poll->wqh before trying to remove from waitqueue
- report polling errors in vhost_poll_start(), tx_poll_start(), the return value
will be checked and returned when userspace want to set the backend
After this fix, there still could be a polling failure after backend is set, it
will addressed by the next patch.
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>
vring changes already do a flush internally where appropriate, so we do
not need a second flush.
It's currently not very expensive but a follow-up patch makes flush more
heavy-weight, so remove the extra flush here to avoid regressing
performance if call or kick fds are changed on data path.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zerocopy handling code is vhost-net specific.
Move it from vhost.c/vhost.h out to net.c
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This will be used to disable zerocopy when error rate
is high.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Better document macros for DMA tracking. Add an
explicit one for DMA in progress instead of
relying on user supplying len != 1.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Even if skb is marked for zero copy, net core might still decide
to copy it later which is somewhat slower than a copy in user context:
besides copying the data we need to pin/unpin the pages.
Add a parameter reporting such cases through zero copy callback:
if this happens a lot, device can take this into account
and switch to copying in user context.
This patch updates all users but ignores the passed value for now:
it will be used by follow-up patches.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The vhost work queue allows processing to be done in vhost worker thread
context, which uses the owner process mm. Access to the vring and guest
memory is typically only possible from vhost worker context so it is
useful to allow work to be queued directly by users.
Currently vhost_net only uses the poll wrappers which do not expose the
work queue functions. However, for tcm_vhost (vhost_scsi) it will be
necessary to queue custom work.
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Cc: Zhi Yong Wu <wuzhy@cn.ibm.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In order for other vhost devices to use the VHOST_FEATURES bits the
vhost-net specific bits need to be moved to their own VHOST_NET_FEATURES
constant.
(Asias: Update drivers/vhost/test.c to use VHOST_NET_FEATURES)
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Cc: Zhi Yong Wu <wuzhy@cn.ibm.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Asias He <asias@redhat.com>
Signed-off-by: Nicholas A. Bellinger <nab@risingtidesystems.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The skb struct ubuf_info callback gets passed struct ubuf_info
itself, not the arg value as the field name and the function signature
seem to imply. Rename the arg field to ctx to match usage,
add documentation and change the callback argument type
to make usage clear and to have compiler check correctness.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We shouldn't hold any locks on release path. Pass a flag to
vhost_dev_cleanup to use the lockdep info correctly.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Sasha Levin <levinsasha928@gmail.com>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move the used ring initialization after backend was set. This
makes it possible to disable the backend and tweak the used ring,
then restart. This will also make it possible to log the used ring
write correctly.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
>From: Shirley Ma <mashirle@us.ibm.com>
This adds experimental zero copy support in vhost-net,
disabled by default. To enable, set
experimental_zcopytx module option to 1.
This patch maintains the outstanding userspace buffers in the
sequence it is delivered to vhost. The outstanding userspace buffers
will be marked as done once the lower device buffers DMA has finished.
This is monitored through last reference of kfree_skb callback. Two
buffer indices are used for this purpose.
The vhost-net device passes the userspace buffers info to lower device
skb through message control. DMA done status check and guest
notification are handled by handle_tx: in the worst case is all buffers
in the vq are in pending/done status, so we need to notify guest to
release DMA done buffers first before we get any new buffers from the
vq.
One known problem is that if the guest stops submitting
buffers, buffers might never get used until some
further action, e.g. device reset. This does not
seem to affect linux guests.
Signed-off-by: Shirley <xma@us.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Support the new event index feature. When acked,
utilize it to reduce the # of interrupts sent to the guest.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
When built with rcu checks enabled, vhost triggers
bogus warnings as vhost features are read without
dev->mutex sometimes, and private pointer is read
with our kind of rcu where work serves as a
read side critical section.
Fixing it properly is not trivial.
Disable the warnings by stubbing out the checks for now.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1699 commits)
bnx2/bnx2x: Unsupported Ethtool operations should return -EINVAL.
vlan: Calling vlan_hwaccel_do_receive() is always valid.
tproxy: use the interface primary IP address as a default value for --on-ip
tproxy: added IPv6 support to the socket match
cxgb3: function namespace cleanup
tproxy: added IPv6 support to the TPROXY target
tproxy: added IPv6 socket lookup function to nf_tproxy_core
be2net: Changes to use only priority codes allowed by f/w
tproxy: allow non-local binds of IPv6 sockets if IP_TRANSPARENT is enabled
tproxy: added tproxy sockopt interface in the IPV6 layer
tproxy: added udp6_lib_lookup function
tproxy: added const specifiers to udp lookup functions
tproxy: split off ipv6 defragmentation to a separate module
l2tp: small cleanup
nf_nat: restrict ICMP translation for embedded header
can: mcp251x: fix generation of error frames
can: mcp251x: fix endless loop in interrupt handler if CANINTF_MERRF is set
can-raw: add msg_flags to distinguish local traffic
9p: client code cleanup
rds: make local functions/variables static
...
Fix up conflicts in net/core/dev.c, drivers/net/pcmcia/smc91c92_cs.c and
drivers/net/wireless/ath/ath9k/debug.c as per David
Qemu supports up to UIO_MAXIOV s/g so we have to match that because guest
drivers may rely on this.
Allocate indirect and log arrays dynamically to avoid using too much contigious
memory and make the length of hdr array to match the header length since each
iovec entry has a least one byte.
Test with copying large files w/ and w/o migration in both linux and windows
guests.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Also add rcu_dereference_protected() for code paths where locks are held.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
