linux/drivers/vhost/net.c

1802 lines
44 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* Copyright (C) 2009 Red Hat, Inc.
* Author: Michael S. Tsirkin <mst@redhat.com>
*
* virtio-net server in host kernel.
*/
#include <linux/compat.h>
#include <linux/eventfd.h>
#include <linux/vhost.h>
#include <linux/virtio_net.h>
#include <linux/miscdevice.h>
#include <linux/module.h>
#include <linux/moduleparam.h>
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
#include <linux/mutex.h>
#include <linux/workqueue.h>
#include <linux/file.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/sched/clock.h>
#include <linux/sched/signal.h>
#include <linux/vmalloc.h>
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
#include <linux/net.h>
#include <linux/if_packet.h>
#include <linux/if_arp.h>
#include <linux/if_tun.h>
#include <linux/if_macvlan.h>
#include <linux/if_tap.h>
#include <linux/if_vlan.h>
#include <linux/skb_array.h>
#include <linux/skbuff.h>
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
#include <net/sock.h>
#include <net/xdp.h>
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
#include "vhost.h"
static int experimental_zcopytx = 0;
module_param(experimental_zcopytx, int, 0444);
MODULE_PARM_DESC(experimental_zcopytx, "Enable Zero Copy TX;"
" 1 -Enable; 0 - Disable");
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* Max number of bytes transferred before requeueing the job.
* Using this limit prevents one virtqueue from starving others. */
#define VHOST_NET_WEIGHT 0x80000
vhost-net: set packet weight of tx polling to 2 * vq size 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 length. Ping-Latencies shown below were tested between two Virtual Machines using netperf (UDP_STREAM, len=1), and then another machine pinged the client: vq size=256 Packet-Weight Ping-Latencies(millisecond) min avg max Origin 3.319 18.489 57.303 64 1.643 2.021 2.552 128 1.825 2.600 3.224 256 1.997 2.710 4.295 512 1.860 3.171 4.631 1024 2.002 4.173 9.056 2048 2.257 5.650 9.688 4096 2.093 8.508 15.943 vq size=512 Packet-Weight Ping-Latencies(millisecond) min avg max Origin 6.537 29.177 66.245 64 2.798 3.614 4.403 128 2.861 3.820 4.775 256 3.008 4.018 4.807 512 3.254 4.523 5.824 1024 3.079 5.335 7.747 2048 3.944 8.201 12.762 4096 4.158 11.057 19.985 Seems pretty consistent, a small dip at 2 VQ sizes. Ring size is a hint from device about a burst size it can tolerate. Based on benchmarks, set the weight to 2 * vq size. To evaluate this change, another tests were done using netperf(RR, TX) between two machines with Intel(R) Xeon(R) Gold 6133 CPU @ 2.50GHz, and vq size was tweaked through qemu. Results shown below does not show obvious changes. vq size=256 TCP_RR vq size=512 TCP_RR size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 1/ 1/ -7%/ -2% 1/ 1/ 0%/ -2% 1/ 4/ +1%/ 0% 1/ 4/ +1%/ 0% 1/ 8/ +1%/ -2% 1/ 8/ 0%/ +1% 64/ 1/ -6%/ 0% 64/ 1/ +7%/ +3% 64/ 4/ 0%/ +2% 64/ 4/ -1%/ +1% 64/ 8/ 0%/ 0% 64/ 8/ -1%/ -2% 256/ 1/ -3%/ -4% 256/ 1/ -4%/ -2% 256/ 4/ +3%/ +4% 256/ 4/ +1%/ +2% 256/ 8/ +2%/ 0% 256/ 8/ +1%/ -1% vq size=256 UDP_RR vq size=512 UDP_RR size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 1/ 1/ -5%/ +1% 1/ 1/ -3%/ -2% 1/ 4/ +4%/ +1% 1/ 4/ -2%/ +2% 1/ 8/ -1%/ -1% 1/ 8/ -1%/ 0% 64/ 1/ -2%/ -3% 64/ 1/ +1%/ +1% 64/ 4/ -5%/ -1% 64/ 4/ +2%/ 0% 64/ 8/ 0%/ -1% 64/ 8/ -2%/ +1% 256/ 1/ +7%/ +1% 256/ 1/ -7%/ 0% 256/ 4/ +1%/ +1% 256/ 4/ -3%/ -4% 256/ 8/ +2%/ +2% 256/ 8/ +1%/ +1% vq size=256 TCP_STREAM vq size=512 TCP_STREAM size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 64/ 1/ 0%/ -3% 64/ 1/ 0%/ 0% 64/ 4/ +3%/ -1% 64/ 4/ -2%/ +4% 64/ 8/ +9%/ -4% 64/ 8/ -1%/ +2% 256/ 1/ +1%/ -4% 256/ 1/ +1%/ +1% 256/ 4/ -1%/ -1% 256/ 4/ -3%/ 0% 256/ 8/ +7%/ +5% 256/ 8/ -3%/ 0% 512/ 1/ +1%/ 0% 512/ 1/ -1%/ -1% 512/ 4/ +1%/ -1% 512/ 4/ 0%/ 0% 512/ 8/ +7%/ -5% 512/ 8/ +6%/ -1% 1024/ 1/ 0%/ -1% 1024/ 1/ 0%/ +1% 1024/ 4/ +3%/ 0% 1024/ 4/ +1%/ 0% 1024/ 8/ +8%/ +5% 1024/ 8/ -1%/ 0% 2048/ 1/ +2%/ +2% 2048/ 1/ -1%/ 0% 2048/ 4/ +1%/ 0% 2048/ 4/ 0%/ -1% 2048/ 8/ -2%/ 0% 2048/ 8/ 5%/ -1% 4096/ 1/ -2%/ 0% 4096/ 1/ -2%/ 0% 4096/ 4/ +2%/ 0% 4096/ 4/ 0%/ 0% 4096/ 8/ +9%/ -2% 4096/ 8/ -5%/ -1% Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Haibin Zhang <haibinzhang@tencent.com> Signed-off-by: Yunfang Tai <yunfangtai@tencent.com> Signed-off-by: Lidong Chen <lidongchen@tencent.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-09 15:22:17 +08:00
/* Max number of packets transferred before requeueing the job.
* Using this limit prevents one virtqueue from starving others with small
* pkts.
*/
#define VHOST_NET_PKT_WEIGHT 256
vhost-net: set packet weight of tx polling to 2 * vq size 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 length. Ping-Latencies shown below were tested between two Virtual Machines using netperf (UDP_STREAM, len=1), and then another machine pinged the client: vq size=256 Packet-Weight Ping-Latencies(millisecond) min avg max Origin 3.319 18.489 57.303 64 1.643 2.021 2.552 128 1.825 2.600 3.224 256 1.997 2.710 4.295 512 1.860 3.171 4.631 1024 2.002 4.173 9.056 2048 2.257 5.650 9.688 4096 2.093 8.508 15.943 vq size=512 Packet-Weight Ping-Latencies(millisecond) min avg max Origin 6.537 29.177 66.245 64 2.798 3.614 4.403 128 2.861 3.820 4.775 256 3.008 4.018 4.807 512 3.254 4.523 5.824 1024 3.079 5.335 7.747 2048 3.944 8.201 12.762 4096 4.158 11.057 19.985 Seems pretty consistent, a small dip at 2 VQ sizes. Ring size is a hint from device about a burst size it can tolerate. Based on benchmarks, set the weight to 2 * vq size. To evaluate this change, another tests were done using netperf(RR, TX) between two machines with Intel(R) Xeon(R) Gold 6133 CPU @ 2.50GHz, and vq size was tweaked through qemu. Results shown below does not show obvious changes. vq size=256 TCP_RR vq size=512 TCP_RR size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 1/ 1/ -7%/ -2% 1/ 1/ 0%/ -2% 1/ 4/ +1%/ 0% 1/ 4/ +1%/ 0% 1/ 8/ +1%/ -2% 1/ 8/ 0%/ +1% 64/ 1/ -6%/ 0% 64/ 1/ +7%/ +3% 64/ 4/ 0%/ +2% 64/ 4/ -1%/ +1% 64/ 8/ 0%/ 0% 64/ 8/ -1%/ -2% 256/ 1/ -3%/ -4% 256/ 1/ -4%/ -2% 256/ 4/ +3%/ +4% 256/ 4/ +1%/ +2% 256/ 8/ +2%/ 0% 256/ 8/ +1%/ -1% vq size=256 UDP_RR vq size=512 UDP_RR size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 1/ 1/ -5%/ +1% 1/ 1/ -3%/ -2% 1/ 4/ +4%/ +1% 1/ 4/ -2%/ +2% 1/ 8/ -1%/ -1% 1/ 8/ -1%/ 0% 64/ 1/ -2%/ -3% 64/ 1/ +1%/ +1% 64/ 4/ -5%/ -1% 64/ 4/ +2%/ 0% 64/ 8/ 0%/ -1% 64/ 8/ -2%/ +1% 256/ 1/ +7%/ +1% 256/ 1/ -7%/ 0% 256/ 4/ +1%/ +1% 256/ 4/ -3%/ -4% 256/ 8/ +2%/ +2% 256/ 8/ +1%/ +1% vq size=256 TCP_STREAM vq size=512 TCP_STREAM size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 64/ 1/ 0%/ -3% 64/ 1/ 0%/ 0% 64/ 4/ +3%/ -1% 64/ 4/ -2%/ +4% 64/ 8/ +9%/ -4% 64/ 8/ -1%/ +2% 256/ 1/ +1%/ -4% 256/ 1/ +1%/ +1% 256/ 4/ -1%/ -1% 256/ 4/ -3%/ 0% 256/ 8/ +7%/ +5% 256/ 8/ -3%/ 0% 512/ 1/ +1%/ 0% 512/ 1/ -1%/ -1% 512/ 4/ +1%/ -1% 512/ 4/ 0%/ 0% 512/ 8/ +7%/ -5% 512/ 8/ +6%/ -1% 1024/ 1/ 0%/ -1% 1024/ 1/ 0%/ +1% 1024/ 4/ +3%/ 0% 1024/ 4/ +1%/ 0% 1024/ 8/ +8%/ +5% 1024/ 8/ -1%/ 0% 2048/ 1/ +2%/ +2% 2048/ 1/ -1%/ 0% 2048/ 4/ +1%/ 0% 2048/ 4/ 0%/ -1% 2048/ 8/ -2%/ 0% 2048/ 8/ 5%/ -1% 4096/ 1/ -2%/ 0% 4096/ 1/ -2%/ 0% 4096/ 4/ +2%/ 0% 4096/ 4/ 0%/ 0% 4096/ 8/ +9%/ -2% 4096/ 8/ -5%/ -1% Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Haibin Zhang <haibinzhang@tencent.com> Signed-off-by: Yunfang Tai <yunfangtai@tencent.com> Signed-off-by: Lidong Chen <lidongchen@tencent.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-09 15:22:17 +08:00
/* MAX number of TX used buffers for outstanding zerocopy */
#define VHOST_MAX_PEND 128
#define VHOST_GOODCOPY_LEN 256
/*
* For transmit, used buffer len is unused; we override it to track buffer
* status internally; used for zerocopy tx only.
*/
/* Lower device DMA failed */
#define VHOST_DMA_FAILED_LEN ((__force __virtio32)3)
/* Lower device DMA done */
#define VHOST_DMA_DONE_LEN ((__force __virtio32)2)
/* Lower device DMA in progress */
#define VHOST_DMA_IN_PROGRESS ((__force __virtio32)1)
/* Buffer unused */
#define VHOST_DMA_CLEAR_LEN ((__force __virtio32)0)
#define VHOST_DMA_IS_DONE(len) ((__force u32)(len) >= (__force u32)VHOST_DMA_DONE_LEN)
enum {
VHOST_NET_FEATURES = VHOST_FEATURES |
(1ULL << VHOST_NET_F_VIRTIO_NET_HDR) |
2016-06-23 14:04:32 +08:00
(1ULL << VIRTIO_NET_F_MRG_RXBUF) |
(1ULL << VIRTIO_F_ACCESS_PLATFORM)
};
enum {
VHOST_NET_BACKEND_FEATURES = (1ULL << VHOST_BACKEND_F_IOTLB_MSG_V2)
};
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
enum {
VHOST_NET_VQ_RX = 0,
VHOST_NET_VQ_TX = 1,
VHOST_NET_VQ_MAX = 2,
};
struct vhost_net_ubuf_ref {
/* refcount follows semantics similar to kref:
* 0: object is released
* 1: no outstanding ubufs
* >1: outstanding ubufs
*/
atomic_t refcount;
wait_queue_head_t wait;
struct vhost_virtqueue *vq;
};
#define VHOST_NET_BATCH 64
struct vhost_net_buf {
void **queue;
int tail;
int head;
};
struct vhost_net_virtqueue {
struct vhost_virtqueue vq;
size_t vhost_hlen;
size_t sock_hlen;
/* vhost zerocopy support fields below: */
/* last used idx for outstanding DMA zerocopy buffers */
int upend_idx;
/* For TX, first used idx for DMA done zerocopy buffers
* For RX, number of batched heads
*/
int done_idx;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
/* Number of XDP frames batched */
int batched_xdp;
/* an array of userspace buffers info */
struct ubuf_info *ubuf_info;
/* Reference counting for outstanding ubufs.
* Protected by vq mutex. Writers must also take device mutex. */
struct vhost_net_ubuf_ref *ubufs;
struct ptr_ring *rx_ring;
struct vhost_net_buf rxq;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
/* Batched XDP buffs */
struct xdp_buff *xdp;
};
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
struct vhost_net {
struct vhost_dev dev;
struct vhost_net_virtqueue vqs[VHOST_NET_VQ_MAX];
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
struct vhost_poll poll[VHOST_NET_VQ_MAX];
/* Number of TX recently submitted.
* Protected by tx vq lock. */
unsigned tx_packets;
/* Number of times zerocopy TX recently failed.
* Protected by tx vq lock. */
unsigned tx_zcopy_err;
/* Flush in progress. Protected by tx vq lock. */
bool tx_flush;
/* Private page frag */
struct page_frag page_frag;
/* Refcount bias of page frag */
int refcnt_bias;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
};
static unsigned vhost_net_zcopy_mask __read_mostly;
static void *vhost_net_buf_get_ptr(struct vhost_net_buf *rxq)
{
if (rxq->tail != rxq->head)
return rxq->queue[rxq->head];
else
return NULL;
}
static int vhost_net_buf_get_size(struct vhost_net_buf *rxq)
{
return rxq->tail - rxq->head;
}
static int vhost_net_buf_is_empty(struct vhost_net_buf *rxq)
{
return rxq->tail == rxq->head;
}
static void *vhost_net_buf_consume(struct vhost_net_buf *rxq)
{
void *ret = vhost_net_buf_get_ptr(rxq);
++rxq->head;
return ret;
}
static int vhost_net_buf_produce(struct vhost_net_virtqueue *nvq)
{
struct vhost_net_buf *rxq = &nvq->rxq;
rxq->head = 0;
rxq->tail = ptr_ring_consume_batched(nvq->rx_ring, rxq->queue,
VHOST_NET_BATCH);
return rxq->tail;
}
static void vhost_net_buf_unproduce(struct vhost_net_virtqueue *nvq)
{
struct vhost_net_buf *rxq = &nvq->rxq;
if (nvq->rx_ring && !vhost_net_buf_is_empty(rxq)) {
ptr_ring_unconsume(nvq->rx_ring, rxq->queue + rxq->head,
vhost_net_buf_get_size(rxq),
tun_ptr_free);
rxq->head = rxq->tail = 0;
}
}
static int vhost_net_buf_peek_len(void *ptr)
{
if (tun_is_xdp_frame(ptr)) {
struct xdp_frame *xdpf = tun_ptr_to_xdp(ptr);
return xdpf->len;
}
return __skb_array_len_with_tag(ptr);
}
static int vhost_net_buf_peek(struct vhost_net_virtqueue *nvq)
{
struct vhost_net_buf *rxq = &nvq->rxq;
if (!vhost_net_buf_is_empty(rxq))
goto out;
if (!vhost_net_buf_produce(nvq))
return 0;
out:
return vhost_net_buf_peek_len(vhost_net_buf_get_ptr(rxq));
}
static void vhost_net_buf_init(struct vhost_net_buf *rxq)
{
rxq->head = rxq->tail = 0;
}
static void vhost_net_enable_zcopy(int vq)
{
vhost_net_zcopy_mask |= 0x1 << vq;
}
static struct vhost_net_ubuf_ref *
vhost_net_ubuf_alloc(struct vhost_virtqueue *vq, bool zcopy)
{
struct vhost_net_ubuf_ref *ubufs;
/* No zero copy backend? Nothing to count. */
if (!zcopy)
return NULL;
ubufs = kmalloc(sizeof(*ubufs), GFP_KERNEL);
if (!ubufs)
return ERR_PTR(-ENOMEM);
atomic_set(&ubufs->refcount, 1);
init_waitqueue_head(&ubufs->wait);
ubufs->vq = vq;
return ubufs;
}
static int vhost_net_ubuf_put(struct vhost_net_ubuf_ref *ubufs)
{
int r = atomic_sub_return(1, &ubufs->refcount);
if (unlikely(!r))
wake_up(&ubufs->wait);
return r;
}
static void vhost_net_ubuf_put_and_wait(struct vhost_net_ubuf_ref *ubufs)
{
vhost_net_ubuf_put(ubufs);
wait_event(ubufs->wait, !atomic_read(&ubufs->refcount));
}
static void vhost_net_ubuf_put_wait_and_free(struct vhost_net_ubuf_ref *ubufs)
{
vhost_net_ubuf_put_and_wait(ubufs);
kfree(ubufs);
}
static void vhost_net_clear_ubuf_info(struct vhost_net *n)
{
int i;
for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
kfree(n->vqs[i].ubuf_info);
n->vqs[i].ubuf_info = NULL;
}
}
static int vhost_net_set_ubuf_info(struct vhost_net *n)
{
bool zcopy;
int i;
for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
zcopy = vhost_net_zcopy_mask & (0x1 << i);
if (!zcopy)
continue;
treewide: kmalloc() -> kmalloc_array() The kmalloc() function has a 2-factor argument form, kmalloc_array(). This patch replaces cases of: kmalloc(a * b, gfp) with: kmalloc_array(a * b, gfp) as well as handling cases of: kmalloc(a * b * c, gfp) with: kmalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kmalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kmalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The tools/ directory was manually excluded, since it has its own implementation of kmalloc(). The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kmalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kmalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kmalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(char) * COUNT + COUNT , ...) | kmalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kmalloc + kmalloc_array ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kmalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kmalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kmalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kmalloc(C1 * C2 * C3, ...) | kmalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kmalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kmalloc(sizeof(THING) * C2, ...) | kmalloc(sizeof(TYPE) * C2, ...) | kmalloc(C1 * C2 * C3, ...) | kmalloc(C1 * C2, ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - (E1) * E2 + E1, E2 , ...) | - kmalloc + kmalloc_array ( - (E1) * (E2) + E1, E2 , ...) | - kmalloc + kmalloc_array ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 04:55:00 +08:00
n->vqs[i].ubuf_info =
kmalloc_array(UIO_MAXIOV,
sizeof(*n->vqs[i].ubuf_info),
GFP_KERNEL);
if (!n->vqs[i].ubuf_info)
goto err;
}
return 0;
err:
vhost_net_clear_ubuf_info(n);
return -ENOMEM;
}
static void vhost_net_vq_reset(struct vhost_net *n)
{
int i;
vhost_net_clear_ubuf_info(n);
for (i = 0; i < VHOST_NET_VQ_MAX; i++) {
n->vqs[i].done_idx = 0;
n->vqs[i].upend_idx = 0;
n->vqs[i].ubufs = NULL;
n->vqs[i].vhost_hlen = 0;
n->vqs[i].sock_hlen = 0;
vhost_net_buf_init(&n->vqs[i].rxq);
}
}
static void vhost_net_tx_packet(struct vhost_net *net)
{
++net->tx_packets;
if (net->tx_packets < 1024)
return;
net->tx_packets = 0;
net->tx_zcopy_err = 0;
}
static void vhost_net_tx_err(struct vhost_net *net)
{
++net->tx_zcopy_err;
}
static bool vhost_net_tx_select_zcopy(struct vhost_net *net)
{
/* TX flush waits for outstanding DMAs to be done.
* Don't start new DMAs.
*/
return !net->tx_flush &&
net->tx_packets / 64 >= net->tx_zcopy_err;
}
static bool vhost_sock_zcopy(struct socket *sock)
{
return unlikely(experimental_zcopytx) &&
sock_flag(sock->sk, SOCK_ZEROCOPY);
}
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
static bool vhost_sock_xdp(struct socket *sock)
{
return sock_flag(sock->sk, SOCK_XDP);
}
/* In case of DMA done not in order in lower device driver for some reason.
* upend_idx is used to track end of used idx, done_idx is used to track head
* of used idx. Once lower device DMA done contiguously, we will signal KVM
* guest used idx.
*/
static void vhost_zerocopy_signal_used(struct vhost_net *net,
struct vhost_virtqueue *vq)
{
struct vhost_net_virtqueue *nvq =
container_of(vq, struct vhost_net_virtqueue, vq);
int i, add;
int j = 0;
for (i = nvq->done_idx; i != nvq->upend_idx; i = (i + 1) % UIO_MAXIOV) {
if (vq->heads[i].len == VHOST_DMA_FAILED_LEN)
vhost_net_tx_err(net);
if (VHOST_DMA_IS_DONE(vq->heads[i].len)) {
vq->heads[i].len = VHOST_DMA_CLEAR_LEN;
++j;
} else
break;
}
while (j) {
add = min(UIO_MAXIOV - nvq->done_idx, j);
vhost_add_used_and_signal_n(vq->dev, vq,
&vq->heads[nvq->done_idx], add);
nvq->done_idx = (nvq->done_idx + add) % UIO_MAXIOV;
j -= add;
}
}
static void vhost_zerocopy_callback(struct ubuf_info *ubuf, bool success)
{
struct vhost_net_ubuf_ref *ubufs = ubuf->ctx;
struct vhost_virtqueue *vq = ubufs->vq;
int cnt;
rcu_read_lock_bh();
/* set len to mark this desc buffers done DMA */
vq->heads[ubuf->desc].len = success ?
VHOST_DMA_DONE_LEN : VHOST_DMA_FAILED_LEN;
cnt = vhost_net_ubuf_put(ubufs);
/*
* Trigger polling thread if guest stopped submitting new buffers:
* in this case, the refcount after decrement will eventually reach 1.
* We also trigger polling periodically after each 16 packets
* (the value 16 here is more or less arbitrary, it's tuned to trigger
* less than 10% of times).
*/
if (cnt <= 1 || !(cnt % 16))
vhost_poll_queue(&vq->poll);
rcu_read_unlock_bh();
}
static inline unsigned long busy_clock(void)
{
return local_clock() >> 10;
}
vhost_net: Avoid tx vring kicks during busyloop Under heavy load vhost busypoll may run without suppressing notification. For example tx zerocopy callback can push tx work while handle_tx() is running, then busyloop exits due to vhost_has_work() condition and enables notification but immediately reenters handle_tx() because the pushed work was tx. In this case handle_tx() tries to disable notification again, but when using event_idx it by design cannot. Then busyloop will run without suppressing notification. Another example is the case where handle_tx() tries to enable notification but avail idx is advanced so disables it again. This case also leads to the same situation with event_idx. The problem is that once we enter this situation busyloop does not work under heavy load for considerable amount of time, because notification is likely to happen during busyloop and handle_tx() immediately enables notification after notification happens. Specifically busyloop detects notification by vhost_has_work() and then handle_tx() calls vhost_enable_notify(). Because the detected work was the tx work, it enters handle_tx(), and enters busyloop without suppression again. This is likely to be repeated, so with event_idx we are almost not able to suppress notification in this case. To fix this, poll the work instead of enabling notification when busypoll is interrupted by something. IMHO vhost_has_work() is kind of interruption rather than a signal to completely cancel the busypoll, so let's run busypoll after the necessary work is done. Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 15:31:32 +08:00
static bool vhost_can_busy_poll(unsigned long endtime)
{
vhost_net: Avoid tx vring kicks during busyloop Under heavy load vhost busypoll may run without suppressing notification. For example tx zerocopy callback can push tx work while handle_tx() is running, then busyloop exits due to vhost_has_work() condition and enables notification but immediately reenters handle_tx() because the pushed work was tx. In this case handle_tx() tries to disable notification again, but when using event_idx it by design cannot. Then busyloop will run without suppressing notification. Another example is the case where handle_tx() tries to enable notification but avail idx is advanced so disables it again. This case also leads to the same situation with event_idx. The problem is that once we enter this situation busyloop does not work under heavy load for considerable amount of time, because notification is likely to happen during busyloop and handle_tx() immediately enables notification after notification happens. Specifically busyloop detects notification by vhost_has_work() and then handle_tx() calls vhost_enable_notify(). Because the detected work was the tx work, it enters handle_tx(), and enters busyloop without suppression again. This is likely to be repeated, so with event_idx we are almost not able to suppress notification in this case. To fix this, poll the work instead of enabling notification when busypoll is interrupted by something. IMHO vhost_has_work() is kind of interruption rather than a signal to completely cancel the busypoll, so let's run busypoll after the necessary work is done. Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 15:31:32 +08:00
return likely(!need_resched() && !time_after(busy_clock(), endtime) &&
!signal_pending(current));
}
static void vhost_net_disable_vq(struct vhost_net *n,
struct vhost_virtqueue *vq)
{
struct vhost_net_virtqueue *nvq =
container_of(vq, struct vhost_net_virtqueue, vq);
struct vhost_poll *poll = n->poll + (nvq - n->vqs);
if (!vhost_vq_get_backend(vq))
return;
vhost_poll_stop(poll);
}
static int vhost_net_enable_vq(struct vhost_net *n,
struct vhost_virtqueue *vq)
{
struct vhost_net_virtqueue *nvq =
container_of(vq, struct vhost_net_virtqueue, vq);
struct vhost_poll *poll = n->poll + (nvq - n->vqs);
struct socket *sock;
sock = vhost_vq_get_backend(vq);
if (!sock)
return 0;
return vhost_poll_start(poll, sock->file);
}
static void vhost_net_signal_used(struct vhost_net_virtqueue *nvq)
{
struct vhost_virtqueue *vq = &nvq->vq;
struct vhost_dev *dev = vq->dev;
if (!nvq->done_idx)
return;
vhost_add_used_and_signal_n(dev, vq, vq->heads, nvq->done_idx);
nvq->done_idx = 0;
}
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
static void vhost_tx_batch(struct vhost_net *net,
struct vhost_net_virtqueue *nvq,
struct socket *sock,
struct msghdr *msghdr)
{
struct tun_msg_ctl ctl = {
.type = TUN_MSG_PTR,
.num = nvq->batched_xdp,
.ptr = nvq->xdp,
};
int err;
if (nvq->batched_xdp == 0)
goto signal_used;
msghdr->msg_control = &ctl;
err = sock->ops->sendmsg(sock, msghdr, 0);
if (unlikely(err < 0)) {
vq_err(&nvq->vq, "Fail to batch sending packets\n");
return;
}
signal_used:
vhost_net_signal_used(nvq);
nvq->batched_xdp = 0;
}
static int sock_has_rx_data(struct socket *sock)
{
if (unlikely(!sock))
return 0;
if (sock->ops->peek_len)
return sock->ops->peek_len(sock);
return skb_queue_empty(&sock->sk->sk_receive_queue);
}
static void vhost_net_busy_poll_try_queue(struct vhost_net *net,
struct vhost_virtqueue *vq)
{
if (!vhost_vq_avail_empty(&net->dev, vq)) {
vhost_poll_queue(&vq->poll);
} else if (unlikely(vhost_enable_notify(&net->dev, vq))) {
vhost_disable_notify(&net->dev, vq);
vhost_poll_queue(&vq->poll);
}
}
static void vhost_net_busy_poll(struct vhost_net *net,
struct vhost_virtqueue *rvq,
struct vhost_virtqueue *tvq,
bool *busyloop_intr,
bool poll_rx)
{
unsigned long busyloop_timeout;
unsigned long endtime;
struct socket *sock;
struct vhost_virtqueue *vq = poll_rx ? tvq : rvq;
/* Try to hold the vq mutex of the paired virtqueue. We can't
* use mutex_lock() here since we could not guarantee a
* consistenet lock ordering.
*/
if (!mutex_trylock(&vq->mutex))
return;
vhost_disable_notify(&net->dev, vq);
sock = vhost_vq_get_backend(rvq);
busyloop_timeout = poll_rx ? rvq->busyloop_timeout:
tvq->busyloop_timeout;
preempt_disable();
endtime = busy_clock() + busyloop_timeout;
while (vhost_can_busy_poll(endtime)) {
if (vhost_has_work(&net->dev)) {
*busyloop_intr = true;
break;
}
if ((sock_has_rx_data(sock) &&
!vhost_vq_avail_empty(&net->dev, rvq)) ||
!vhost_vq_avail_empty(&net->dev, tvq))
break;
cpu_relax();
}
preempt_enable();
if (poll_rx || sock_has_rx_data(sock))
vhost_net_busy_poll_try_queue(net, vq);
else if (!poll_rx) /* On tx here, sock has no rx data. */
vhost_enable_notify(&net->dev, rvq);
mutex_unlock(&vq->mutex);
}
static int vhost_net_tx_get_vq_desc(struct vhost_net *net,
struct vhost_net_virtqueue *tnvq,
vhost_net: Avoid tx vring kicks during busyloop Under heavy load vhost busypoll may run without suppressing notification. For example tx zerocopy callback can push tx work while handle_tx() is running, then busyloop exits due to vhost_has_work() condition and enables notification but immediately reenters handle_tx() because the pushed work was tx. In this case handle_tx() tries to disable notification again, but when using event_idx it by design cannot. Then busyloop will run without suppressing notification. Another example is the case where handle_tx() tries to enable notification but avail idx is advanced so disables it again. This case also leads to the same situation with event_idx. The problem is that once we enter this situation busyloop does not work under heavy load for considerable amount of time, because notification is likely to happen during busyloop and handle_tx() immediately enables notification after notification happens. Specifically busyloop detects notification by vhost_has_work() and then handle_tx() calls vhost_enable_notify(). Because the detected work was the tx work, it enters handle_tx(), and enters busyloop without suppression again. This is likely to be repeated, so with event_idx we are almost not able to suppress notification in this case. To fix this, poll the work instead of enabling notification when busypoll is interrupted by something. IMHO vhost_has_work() is kind of interruption rather than a signal to completely cancel the busypoll, so let's run busypoll after the necessary work is done. Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 15:31:32 +08:00
unsigned int *out_num, unsigned int *in_num,
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
struct msghdr *msghdr, bool *busyloop_intr)
{
struct vhost_net_virtqueue *rnvq = &net->vqs[VHOST_NET_VQ_RX];
struct vhost_virtqueue *rvq = &rnvq->vq;
struct vhost_virtqueue *tvq = &tnvq->vq;
int r = vhost_get_vq_desc(tvq, tvq->iov, ARRAY_SIZE(tvq->iov),
2016-06-23 14:04:32 +08:00
out_num, in_num, NULL, NULL);
if (r == tvq->num && tvq->busyloop_timeout) {
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
/* Flush batched packets first */
if (!vhost_sock_zcopy(vhost_vq_get_backend(tvq)))
vhost_tx_batch(net, tnvq,
vhost_vq_get_backend(tvq),
msghdr);
vhost_net_busy_poll(net, rvq, tvq, busyloop_intr, false);
r = vhost_get_vq_desc(tvq, tvq->iov, ARRAY_SIZE(tvq->iov),
2016-06-23 14:04:32 +08:00
out_num, in_num, NULL, NULL);
}
return r;
}
static bool vhost_exceeds_maxpend(struct vhost_net *net)
{
struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_TX];
struct vhost_virtqueue *vq = &nvq->vq;
vhost_net: do not stall on zerocopy depletion 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>
2017-10-07 01:22:31 +08:00
return (nvq->upend_idx + UIO_MAXIOV - nvq->done_idx) % UIO_MAXIOV >
min_t(unsigned int, VHOST_MAX_PEND, vq->num >> 2);
}
static size_t init_iov_iter(struct vhost_virtqueue *vq, struct iov_iter *iter,
size_t hdr_size, int out)
{
/* Skip header. TODO: support TSO. */
size_t len = iov_length(vq->iov, out);
iov_iter_init(iter, WRITE, vq->iov, out, len);
iov_iter_advance(iter, hdr_size);
return iov_iter_count(iter);
}
static int get_tx_bufs(struct vhost_net *net,
struct vhost_net_virtqueue *nvq,
struct msghdr *msg,
unsigned int *out, unsigned int *in,
size_t *len, bool *busyloop_intr)
{
struct vhost_virtqueue *vq = &nvq->vq;
int ret;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
ret = vhost_net_tx_get_vq_desc(net, nvq, out, in, msg, busyloop_intr);
if (ret < 0 || ret == vq->num)
return ret;
if (*in) {
vq_err(vq, "Unexpected descriptor format for TX: out %d, int %d\n",
*out, *in);
return -EFAULT;
}
/* Sanity check */
*len = init_iov_iter(vq, &msg->msg_iter, nvq->vhost_hlen, *out);
if (*len == 0) {
vq_err(vq, "Unexpected header len for TX: %zd expected %zd\n",
*len, nvq->vhost_hlen);
return -EFAULT;
}
return ret;
}
static bool tx_can_batch(struct vhost_virtqueue *vq, size_t total_len)
{
return total_len < VHOST_NET_WEIGHT &&
!vhost_vq_avail_empty(vq->dev, vq);
}
#define SKB_FRAG_PAGE_ORDER get_order(32768)
static bool vhost_net_page_frag_refill(struct vhost_net *net, unsigned int sz,
struct page_frag *pfrag, gfp_t gfp)
{
if (pfrag->page) {
if (pfrag->offset + sz <= pfrag->size)
return true;
__page_frag_cache_drain(pfrag->page, net->refcnt_bias);
}
pfrag->offset = 0;
net->refcnt_bias = 0;
if (SKB_FRAG_PAGE_ORDER) {
/* Avoid direct reclaim but allow kswapd to wake */
pfrag->page = alloc_pages((gfp & ~__GFP_DIRECT_RECLAIM) |
__GFP_COMP | __GFP_NOWARN |
__GFP_NORETRY,
SKB_FRAG_PAGE_ORDER);
if (likely(pfrag->page)) {
pfrag->size = PAGE_SIZE << SKB_FRAG_PAGE_ORDER;
goto done;
}
}
pfrag->page = alloc_page(gfp);
if (likely(pfrag->page)) {
pfrag->size = PAGE_SIZE;
goto done;
}
return false;
done:
net->refcnt_bias = USHRT_MAX;
page_ref_add(pfrag->page, USHRT_MAX - 1);
return true;
}
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
#define VHOST_NET_RX_PAD (NET_IP_ALIGN + NET_SKB_PAD)
static int vhost_net_build_xdp(struct vhost_net_virtqueue *nvq,
struct iov_iter *from)
{
struct vhost_virtqueue *vq = &nvq->vq;
struct vhost_net *net = container_of(vq->dev, struct vhost_net,
dev);
struct socket *sock = vhost_vq_get_backend(vq);
struct page_frag *alloc_frag = &net->page_frag;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
struct virtio_net_hdr *gso;
struct xdp_buff *xdp = &nvq->xdp[nvq->batched_xdp];
struct tun_xdp_hdr *hdr;
size_t len = iov_iter_count(from);
int headroom = vhost_sock_xdp(sock) ? XDP_PACKET_HEADROOM : 0;
int buflen = SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
int pad = SKB_DATA_ALIGN(VHOST_NET_RX_PAD + headroom + nvq->sock_hlen);
int sock_hlen = nvq->sock_hlen;
void *buf;
int copied;
if (unlikely(len < nvq->sock_hlen))
return -EFAULT;
if (SKB_DATA_ALIGN(len + pad) +
SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) > PAGE_SIZE)
return -ENOSPC;
buflen += SKB_DATA_ALIGN(len + pad);
alloc_frag->offset = ALIGN((u64)alloc_frag->offset, SMP_CACHE_BYTES);
if (unlikely(!vhost_net_page_frag_refill(net, buflen,
alloc_frag, GFP_KERNEL)))
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
return -ENOMEM;
buf = (char *)page_address(alloc_frag->page) + alloc_frag->offset;
copied = copy_page_from_iter(alloc_frag->page,
alloc_frag->offset +
offsetof(struct tun_xdp_hdr, gso),
sock_hlen, from);
if (copied != sock_hlen)
return -EFAULT;
hdr = buf;
gso = &hdr->gso;
if ((gso->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) &&
vhost16_to_cpu(vq, gso->csum_start) +
vhost16_to_cpu(vq, gso->csum_offset) + 2 >
vhost16_to_cpu(vq, gso->hdr_len)) {
gso->hdr_len = cpu_to_vhost16(vq,
vhost16_to_cpu(vq, gso->csum_start) +
vhost16_to_cpu(vq, gso->csum_offset) + 2);
if (vhost16_to_cpu(vq, gso->hdr_len) > len)
return -EINVAL;
}
len -= sock_hlen;
copied = copy_page_from_iter(alloc_frag->page,
alloc_frag->offset + pad,
len, from);
if (copied != len)
return -EFAULT;
xdp->data_hard_start = buf;
xdp->data = buf + pad;
xdp->data_end = xdp->data + len;
hdr->buflen = buflen;
xdp->frame_sz = buflen;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
--net->refcnt_bias;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
alloc_frag->offset += buflen;
++nvq->batched_xdp;
return 0;
}
static void handle_tx_copy(struct vhost_net *net, struct socket *sock)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_TX];
struct vhost_virtqueue *vq = &nvq->vq;
unsigned out, in;
int head;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
struct msghdr msg = {
.msg_name = NULL,
.msg_namelen = 0,
.msg_control = NULL,
.msg_controllen = 0,
.msg_flags = MSG_DONTWAIT,
};
size_t len, total_len = 0;
int err;
vhost-net: set packet weight of tx polling to 2 * vq size 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 length. Ping-Latencies shown below were tested between two Virtual Machines using netperf (UDP_STREAM, len=1), and then another machine pinged the client: vq size=256 Packet-Weight Ping-Latencies(millisecond) min avg max Origin 3.319 18.489 57.303 64 1.643 2.021 2.552 128 1.825 2.600 3.224 256 1.997 2.710 4.295 512 1.860 3.171 4.631 1024 2.002 4.173 9.056 2048 2.257 5.650 9.688 4096 2.093 8.508 15.943 vq size=512 Packet-Weight Ping-Latencies(millisecond) min avg max Origin 6.537 29.177 66.245 64 2.798 3.614 4.403 128 2.861 3.820 4.775 256 3.008 4.018 4.807 512 3.254 4.523 5.824 1024 3.079 5.335 7.747 2048 3.944 8.201 12.762 4096 4.158 11.057 19.985 Seems pretty consistent, a small dip at 2 VQ sizes. Ring size is a hint from device about a burst size it can tolerate. Based on benchmarks, set the weight to 2 * vq size. To evaluate this change, another tests were done using netperf(RR, TX) between two machines with Intel(R) Xeon(R) Gold 6133 CPU @ 2.50GHz, and vq size was tweaked through qemu. Results shown below does not show obvious changes. vq size=256 TCP_RR vq size=512 TCP_RR size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 1/ 1/ -7%/ -2% 1/ 1/ 0%/ -2% 1/ 4/ +1%/ 0% 1/ 4/ +1%/ 0% 1/ 8/ +1%/ -2% 1/ 8/ 0%/ +1% 64/ 1/ -6%/ 0% 64/ 1/ +7%/ +3% 64/ 4/ 0%/ +2% 64/ 4/ -1%/ +1% 64/ 8/ 0%/ 0% 64/ 8/ -1%/ -2% 256/ 1/ -3%/ -4% 256/ 1/ -4%/ -2% 256/ 4/ +3%/ +4% 256/ 4/ +1%/ +2% 256/ 8/ +2%/ 0% 256/ 8/ +1%/ -1% vq size=256 UDP_RR vq size=512 UDP_RR size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 1/ 1/ -5%/ +1% 1/ 1/ -3%/ -2% 1/ 4/ +4%/ +1% 1/ 4/ -2%/ +2% 1/ 8/ -1%/ -1% 1/ 8/ -1%/ 0% 64/ 1/ -2%/ -3% 64/ 1/ +1%/ +1% 64/ 4/ -5%/ -1% 64/ 4/ +2%/ 0% 64/ 8/ 0%/ -1% 64/ 8/ -2%/ +1% 256/ 1/ +7%/ +1% 256/ 1/ -7%/ 0% 256/ 4/ +1%/ +1% 256/ 4/ -3%/ -4% 256/ 8/ +2%/ +2% 256/ 8/ +1%/ +1% vq size=256 TCP_STREAM vq size=512 TCP_STREAM size/sessions/+thu%/+normalize% size/sessions/+thu%/+normalize% 64/ 1/ 0%/ -3% 64/ 1/ 0%/ 0% 64/ 4/ +3%/ -1% 64/ 4/ -2%/ +4% 64/ 8/ +9%/ -4% 64/ 8/ -1%/ +2% 256/ 1/ +1%/ -4% 256/ 1/ +1%/ +1% 256/ 4/ -1%/ -1% 256/ 4/ -3%/ 0% 256/ 8/ +7%/ +5% 256/ 8/ -3%/ 0% 512/ 1/ +1%/ 0% 512/ 1/ -1%/ -1% 512/ 4/ +1%/ -1% 512/ 4/ 0%/ 0% 512/ 8/ +7%/ -5% 512/ 8/ +6%/ -1% 1024/ 1/ 0%/ -1% 1024/ 1/ 0%/ +1% 1024/ 4/ +3%/ 0% 1024/ 4/ +1%/ 0% 1024/ 8/ +8%/ +5% 1024/ 8/ -1%/ 0% 2048/ 1/ +2%/ +2% 2048/ 1/ -1%/ 0% 2048/ 4/ +1%/ 0% 2048/ 4/ 0%/ -1% 2048/ 8/ -2%/ 0% 2048/ 8/ 5%/ -1% 4096/ 1/ -2%/ 0% 4096/ 1/ -2%/ 0% 4096/ 4/ +2%/ 0% 4096/ 4/ 0%/ 0% 4096/ 8/ +9%/ -2% 4096/ 8/ -5%/ -1% Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Haibin Zhang <haibinzhang@tencent.com> Signed-off-by: Yunfang Tai <yunfangtai@tencent.com> Signed-off-by: Lidong Chen <lidongchen@tencent.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-09 15:22:17 +08:00
int sent_pkts = 0;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
bool sock_can_batch = (sock->sk->sk_sndbuf == INT_MAX);
do {
bool busyloop_intr = false;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
if (nvq->done_idx == VHOST_NET_BATCH)
vhost_tx_batch(net, nvq, sock, &msg);
head = get_tx_bufs(net, nvq, &msg, &out, &in, &len,
&busyloop_intr);
/* On error, stop handling until the next kick. */
if (unlikely(head < 0))
break;
/* Nothing new? Wait for eventfd to tell us they refilled. */
if (head == vq->num) {
if (unlikely(busyloop_intr)) {
vhost_poll_queue(&vq->poll);
} else if (unlikely(vhost_enable_notify(&net->dev,
vq))) {
vhost_disable_notify(&net->dev, vq);
continue;
}
break;
}
2016-06-23 14:04:32 +08:00
total_len += len;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
/* For simplicity, TX batching is only enabled if
* sndbuf is unlimited.
*/
if (sock_can_batch) {
err = vhost_net_build_xdp(nvq, &msg.msg_iter);
if (!err) {
goto done;
} else if (unlikely(err != -ENOSPC)) {
vhost_tx_batch(net, nvq, sock, &msg);
vhost_discard_vq_desc(vq, 1);
vhost_net_enable_vq(net, vq);
break;
}
/* We can't build XDP buff, go for single
* packet path but let's flush batched
* packets.
*/
vhost_tx_batch(net, nvq, sock, &msg);
msg.msg_control = NULL;
} else {
if (tx_can_batch(vq, total_len))
msg.msg_flags |= MSG_MORE;
else
msg.msg_flags &= ~MSG_MORE;
}
/* TODO: Check specific error and bomb out unless ENOBUFS? */
err = sock->ops->sendmsg(sock, &msg, len);
if (unlikely(err < 0)) {
vhost_discard_vq_desc(vq, 1);
vhost_net_enable_vq(net, vq);
break;
}
if (err != len)
pr_debug("Truncated TX packet: len %d != %zd\n",
err, len);
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
done:
vq->heads[nvq->done_idx].id = cpu_to_vhost32(vq, head);
vq->heads[nvq->done_idx].len = 0;
++nvq->done_idx;
} while (likely(!vhost_exceeds_weight(vq, ++sent_pkts, total_len)));
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
vhost_tx_batch(net, nvq, sock, &msg);
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock)
{
struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_TX];
struct vhost_virtqueue *vq = &nvq->vq;
unsigned out, in;
int head;
struct msghdr msg = {
.msg_name = NULL,
.msg_namelen = 0,
.msg_control = NULL,
.msg_controllen = 0,
.msg_flags = MSG_DONTWAIT,
};
struct tun_msg_ctl ctl;
size_t len, total_len = 0;
int err;
treewide: Remove uninitialized_var() usage Using uninitialized_var() is dangerous as it papers over real bugs[1] (or can in the future), and suppresses unrelated compiler warnings (e.g. "unused variable"). If the compiler thinks it is uninitialized, either simply initialize the variable or make compiler changes. In preparation for removing[2] the[3] macro[4], remove all remaining needless uses with the following script: git grep '\buninitialized_var\b' | cut -d: -f1 | sort -u | \ xargs perl -pi -e \ 's/\buninitialized_var\(([^\)]+)\)/\1/g; s:\s*/\* (GCC be quiet|to make compiler happy) \*/$::g;' drivers/video/fbdev/riva/riva_hw.c was manually tweaked to avoid pathological white-space. No outstanding warnings were found building allmodconfig with GCC 9.3.0 for x86_64, i386, arm64, arm, powerpc, powerpc64le, s390x, mips, sparc64, alpha, and m68k. [1] https://lore.kernel.org/lkml/20200603174714.192027-1-glider@google.com/ [2] https://lore.kernel.org/lkml/CA+55aFw+Vbj0i=1TGqCR5vQkCzWJ0QxK6CernOU6eedsudAixw@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CA+55aFwgbgqhbp1fkxvRKEpzyR5J8n1vKT1VZdz9knmPuXhOeg@mail.gmail.com/ [4] https://lore.kernel.org/lkml/CA+55aFz2500WfbKXAx8s67wrm9=yVJu65TpLgN_ybYNv0VEOKA@mail.gmail.com/ Reviewed-by: Leon Romanovsky <leonro@mellanox.com> # drivers/infiniband and mlx4/mlx5 Acked-by: Jason Gunthorpe <jgg@mellanox.com> # IB Acked-by: Kalle Valo <kvalo@codeaurora.org> # wireless drivers Reviewed-by: Chao Yu <yuchao0@huawei.com> # erofs Signed-off-by: Kees Cook <keescook@chromium.org>
2020-06-04 04:09:38 +08:00
struct vhost_net_ubuf_ref *ubufs;
bool zcopy_used;
int sent_pkts = 0;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
do {
vhost_net: Avoid tx vring kicks during busyloop Under heavy load vhost busypoll may run without suppressing notification. For example tx zerocopy callback can push tx work while handle_tx() is running, then busyloop exits due to vhost_has_work() condition and enables notification but immediately reenters handle_tx() because the pushed work was tx. In this case handle_tx() tries to disable notification again, but when using event_idx it by design cannot. Then busyloop will run without suppressing notification. Another example is the case where handle_tx() tries to enable notification but avail idx is advanced so disables it again. This case also leads to the same situation with event_idx. The problem is that once we enter this situation busyloop does not work under heavy load for considerable amount of time, because notification is likely to happen during busyloop and handle_tx() immediately enables notification after notification happens. Specifically busyloop detects notification by vhost_has_work() and then handle_tx() calls vhost_enable_notify(). Because the detected work was the tx work, it enters handle_tx(), and enters busyloop without suppression again. This is likely to be repeated, so with event_idx we are almost not able to suppress notification in this case. To fix this, poll the work instead of enabling notification when busypoll is interrupted by something. IMHO vhost_has_work() is kind of interruption rather than a signal to completely cancel the busypoll, so let's run busypoll after the necessary work is done. Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 15:31:32 +08:00
bool busyloop_intr;
/* Release DMAs done buffers first */
vhost_zerocopy_signal_used(net, vq);
vhost_net: Avoid tx vring kicks during busyloop Under heavy load vhost busypoll may run without suppressing notification. For example tx zerocopy callback can push tx work while handle_tx() is running, then busyloop exits due to vhost_has_work() condition and enables notification but immediately reenters handle_tx() because the pushed work was tx. In this case handle_tx() tries to disable notification again, but when using event_idx it by design cannot. Then busyloop will run without suppressing notification. Another example is the case where handle_tx() tries to enable notification but avail idx is advanced so disables it again. This case also leads to the same situation with event_idx. The problem is that once we enter this situation busyloop does not work under heavy load for considerable amount of time, because notification is likely to happen during busyloop and handle_tx() immediately enables notification after notification happens. Specifically busyloop detects notification by vhost_has_work() and then handle_tx() calls vhost_enable_notify(). Because the detected work was the tx work, it enters handle_tx(), and enters busyloop without suppression again. This is likely to be repeated, so with event_idx we are almost not able to suppress notification in this case. To fix this, poll the work instead of enabling notification when busypoll is interrupted by something. IMHO vhost_has_work() is kind of interruption rather than a signal to completely cancel the busypoll, so let's run busypoll after the necessary work is done. Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 15:31:32 +08:00
busyloop_intr = false;
head = get_tx_bufs(net, nvq, &msg, &out, &in, &len,
&busyloop_intr);
/* On error, stop handling until the next kick. */
if (unlikely(head < 0))
break;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* Nothing new? Wait for eventfd to tell us they refilled. */
if (head == vq->num) {
vhost_net: Avoid tx vring kicks during busyloop Under heavy load vhost busypoll may run without suppressing notification. For example tx zerocopy callback can push tx work while handle_tx() is running, then busyloop exits due to vhost_has_work() condition and enables notification but immediately reenters handle_tx() because the pushed work was tx. In this case handle_tx() tries to disable notification again, but when using event_idx it by design cannot. Then busyloop will run without suppressing notification. Another example is the case where handle_tx() tries to enable notification but avail idx is advanced so disables it again. This case also leads to the same situation with event_idx. The problem is that once we enter this situation busyloop does not work under heavy load for considerable amount of time, because notification is likely to happen during busyloop and handle_tx() immediately enables notification after notification happens. Specifically busyloop detects notification by vhost_has_work() and then handle_tx() calls vhost_enable_notify(). Because the detected work was the tx work, it enters handle_tx(), and enters busyloop without suppression again. This is likely to be repeated, so with event_idx we are almost not able to suppress notification in this case. To fix this, poll the work instead of enabling notification when busypoll is interrupted by something. IMHO vhost_has_work() is kind of interruption rather than a signal to completely cancel the busypoll, so let's run busypoll after the necessary work is done. Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 15:31:32 +08:00
if (unlikely(busyloop_intr)) {
vhost_poll_queue(&vq->poll);
} else if (unlikely(vhost_enable_notify(&net->dev, vq))) {
vhost_disable_notify(&net->dev, vq);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
continue;
}
break;
}
zcopy_used = len >= VHOST_GOODCOPY_LEN
&& !vhost_exceeds_maxpend(net)
&& vhost_net_tx_select_zcopy(net);
/* use msg_control to pass vhost zerocopy ubuf info to skb */
if (zcopy_used) {
struct ubuf_info *ubuf;
ubuf = nvq->ubuf_info + nvq->upend_idx;
vq->heads[nvq->upend_idx].id = cpu_to_vhost32(vq, head);
vq->heads[nvq->upend_idx].len = VHOST_DMA_IN_PROGRESS;
ubuf->callback = vhost_zerocopy_callback;
ubuf->ctx = nvq->ubufs;
ubuf->desc = nvq->upend_idx;
refcount_set(&ubuf->refcnt, 1);
msg.msg_control = &ctl;
ctl.type = TUN_MSG_UBUF;
ctl.ptr = ubuf;
msg.msg_controllen = sizeof(ctl);
ubufs = nvq->ubufs;
atomic_inc(&ubufs->refcount);
nvq->upend_idx = (nvq->upend_idx + 1) % UIO_MAXIOV;
} else {
vhost_net: clear msg.control for non-zerocopy case during tx When we decide not use zero-copy, msg.control should be set to NULL otherwise macvtap/tap may set zerocopy callbacks which may decrease the kref of ubufs wrongly. Bug were introduced by commit cedb9bdce099206290a2bdd02ce47a7b253b6a84 (vhost-net: skip head management if no outstanding). This solves the following warnings: WARNING: at include/linux/kref.h:47 handle_tx+0x477/0x4b0 [vhost_net]() Modules linked in: vhost_net macvtap macvlan tun nfsd exportfs bridge stp llc openvswitch kvm_amd kvm bnx2 megaraid_sas [last unloaded: tun] CPU: 5 PID: 8670 Comm: vhost-8668 Not tainted 3.10.0-rc2+ #1566 Hardware name: Dell Inc. PowerEdge R715/00XHKG, BIOS 1.5.2 04/19/2011 ffffffffa0198323 ffff88007c9ebd08 ffffffff81796b73 ffff88007c9ebd48 ffffffff8103d66b 000000007b773e20 ffff8800779f0000 ffff8800779f43f0 ffff8800779f8418 000000000000015c 0000000000000062 ffff88007c9ebd58 Call Trace: [<ffffffff81796b73>] dump_stack+0x19/0x1e [<ffffffff8103d66b>] warn_slowpath_common+0x6b/0xa0 [<ffffffff8103d6b5>] warn_slowpath_null+0x15/0x20 [<ffffffffa0197627>] handle_tx+0x477/0x4b0 [vhost_net] [<ffffffffa0197690>] handle_tx_kick+0x10/0x20 [vhost_net] [<ffffffffa019541e>] vhost_worker+0xfe/0x1a0 [vhost_net] [<ffffffffa0195320>] ? vhost_attach_cgroups_work+0x30/0x30 [vhost_net] [<ffffffffa0195320>] ? vhost_attach_cgroups_work+0x30/0x30 [vhost_net] [<ffffffff81061f46>] kthread+0xc6/0xd0 [<ffffffff81061e80>] ? kthread_freezable_should_stop+0x70/0x70 [<ffffffff817a1aec>] ret_from_fork+0x7c/0xb0 [<ffffffff81061e80>] ? kthread_freezable_should_stop+0x70/0x70 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>
2013-06-05 15:40:46 +08:00
msg.msg_control = NULL;
ubufs = NULL;
}
total_len += len;
if (tx_can_batch(vq, total_len) &&
likely(!vhost_exceeds_maxpend(net))) {
msg.msg_flags |= MSG_MORE;
} else {
msg.msg_flags &= ~MSG_MORE;
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* TODO: Check specific error and bomb out unless ENOBUFS? */
err = sock->ops->sendmsg(sock, &msg, len);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (unlikely(err < 0)) {
if (zcopy_used) {
vhost_net_ubuf_put(ubufs);
nvq->upend_idx = ((unsigned)nvq->upend_idx - 1)
% UIO_MAXIOV;
}
vhost_discard_vq_desc(vq, 1);
vhost_net_enable_vq(net, vq);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
break;
}
if (err != len)
pr_debug("Truncated TX packet: "
" len %d != %zd\n", err, len);
if (!zcopy_used)
vhost_add_used_and_signal(&net->dev, vq, head, 0);
else
vhost_zerocopy_signal_used(net, vq);
vhost_net_tx_packet(net);
} while (likely(!vhost_exceeds_weight(vq, ++sent_pkts, total_len)));
}
/* Expects to be always run from workqueue - which acts as
* read-size critical section for our kind of RCU. */
static void handle_tx(struct vhost_net *net)
{
struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_TX];
struct vhost_virtqueue *vq = &nvq->vq;
struct socket *sock;
mutex_lock_nested(&vq->mutex, VHOST_NET_VQ_TX);
sock = vhost_vq_get_backend(vq);
if (!sock)
goto out;
if (!vq_meta_prefetch(vq))
goto out;
vhost_disable_notify(&net->dev, vq);
vhost_net_disable_vq(net, vq);
if (vhost_sock_zcopy(sock))
handle_tx_zerocopy(net, sock);
else
handle_tx_copy(net, sock);
out:
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_unlock(&vq->mutex);
}
static int peek_head_len(struct vhost_net_virtqueue *rvq, struct sock *sk)
{
struct sk_buff *head;
int len = 0;
unsigned long flags;
if (rvq->rx_ring)
return vhost_net_buf_peek(rvq);
spin_lock_irqsave(&sk->sk_receive_queue.lock, flags);
head = skb_peek(&sk->sk_receive_queue);
if (likely(head)) {
len = head->len;
if (skb_vlan_tag_present(head))
len += VLAN_HLEN;
}
spin_unlock_irqrestore(&sk->sk_receive_queue.lock, flags);
return len;
}
static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk,
bool *busyloop_intr)
{
struct vhost_net_virtqueue *rnvq = &net->vqs[VHOST_NET_VQ_RX];
struct vhost_net_virtqueue *tnvq = &net->vqs[VHOST_NET_VQ_TX];
struct vhost_virtqueue *rvq = &rnvq->vq;
struct vhost_virtqueue *tvq = &tnvq->vq;
int len = peek_head_len(rnvq, sk);
if (!len && rvq->busyloop_timeout) {
/* Flush batched heads first */
vhost_net_signal_used(rnvq);
/* Both tx vq and rx socket were polled here */
vhost_net_busy_poll(net, rvq, tvq, busyloop_intr, true);
len = peek_head_len(rnvq, sk);
}
return len;
}
/* This is a multi-buffer version of vhost_get_desc, that works if
* vq has read descriptors only.
* @vq - the relevant virtqueue
* @datalen - data length we'll be reading
* @iovcount - returned count of io vectors we fill
* @log - vhost log
* @log_num - log offset
* @quota - headcount quota, 1 for big buffer
* returns number of buffer heads allocated, negative on error
*/
static int get_rx_bufs(struct vhost_virtqueue *vq,
struct vring_used_elem *heads,
int datalen,
unsigned *iovcount,
struct vhost_log *log,
unsigned *log_num,
unsigned int quota)
{
unsigned int out, in;
int seg = 0;
int headcount = 0;
unsigned d;
int r, nlogs = 0;
/* len is always initialized before use since we are always called with
* datalen > 0.
*/
treewide: Remove uninitialized_var() usage Using uninitialized_var() is dangerous as it papers over real bugs[1] (or can in the future), and suppresses unrelated compiler warnings (e.g. "unused variable"). If the compiler thinks it is uninitialized, either simply initialize the variable or make compiler changes. In preparation for removing[2] the[3] macro[4], remove all remaining needless uses with the following script: git grep '\buninitialized_var\b' | cut -d: -f1 | sort -u | \ xargs perl -pi -e \ 's/\buninitialized_var\(([^\)]+)\)/\1/g; s:\s*/\* (GCC be quiet|to make compiler happy) \*/$::g;' drivers/video/fbdev/riva/riva_hw.c was manually tweaked to avoid pathological white-space. No outstanding warnings were found building allmodconfig with GCC 9.3.0 for x86_64, i386, arm64, arm, powerpc, powerpc64le, s390x, mips, sparc64, alpha, and m68k. [1] https://lore.kernel.org/lkml/20200603174714.192027-1-glider@google.com/ [2] https://lore.kernel.org/lkml/CA+55aFw+Vbj0i=1TGqCR5vQkCzWJ0QxK6CernOU6eedsudAixw@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CA+55aFwgbgqhbp1fkxvRKEpzyR5J8n1vKT1VZdz9knmPuXhOeg@mail.gmail.com/ [4] https://lore.kernel.org/lkml/CA+55aFz2500WfbKXAx8s67wrm9=yVJu65TpLgN_ybYNv0VEOKA@mail.gmail.com/ Reviewed-by: Leon Romanovsky <leonro@mellanox.com> # drivers/infiniband and mlx4/mlx5 Acked-by: Jason Gunthorpe <jgg@mellanox.com> # IB Acked-by: Kalle Valo <kvalo@codeaurora.org> # wireless drivers Reviewed-by: Chao Yu <yuchao0@huawei.com> # erofs Signed-off-by: Kees Cook <keescook@chromium.org>
2020-06-04 04:09:38 +08:00
u32 len;
while (datalen > 0 && headcount < quota) {
if (unlikely(seg >= UIO_MAXIOV)) {
r = -ENOBUFS;
goto err;
}
r = vhost_get_vq_desc(vq, vq->iov + seg,
ARRAY_SIZE(vq->iov) - seg, &out,
&in, log, log_num);
if (unlikely(r < 0))
goto err;
d = r;
if (d == vq->num) {
r = 0;
goto err;
}
if (unlikely(out || in <= 0)) {
vq_err(vq, "unexpected descriptor format for RX: "
"out %d, in %d\n", out, in);
r = -EINVAL;
goto err;
}
if (unlikely(log)) {
nlogs += *log_num;
log += *log_num;
}
heads[headcount].id = cpu_to_vhost32(vq, d);
len = iov_length(vq->iov + seg, in);
heads[headcount].len = cpu_to_vhost32(vq, len);
datalen -= len;
++headcount;
seg += in;
}
heads[headcount - 1].len = cpu_to_vhost32(vq, len + datalen);
*iovcount = seg;
if (unlikely(log))
*log_num = nlogs;
/* Detect overrun */
if (unlikely(datalen > 0)) {
r = UIO_MAXIOV + 1;
goto err;
}
return headcount;
err:
vhost_discard_vq_desc(vq, headcount);
return r;
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* Expects to be always run from workqueue - which acts as
* read-size critical section for our kind of RCU. */
static void handle_rx(struct vhost_net *net)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_RX];
struct vhost_virtqueue *vq = &nvq->vq;
treewide: Remove uninitialized_var() usage Using uninitialized_var() is dangerous as it papers over real bugs[1] (or can in the future), and suppresses unrelated compiler warnings (e.g. "unused variable"). If the compiler thinks it is uninitialized, either simply initialize the variable or make compiler changes. In preparation for removing[2] the[3] macro[4], remove all remaining needless uses with the following script: git grep '\buninitialized_var\b' | cut -d: -f1 | sort -u | \ xargs perl -pi -e \ 's/\buninitialized_var\(([^\)]+)\)/\1/g; s:\s*/\* (GCC be quiet|to make compiler happy) \*/$::g;' drivers/video/fbdev/riva/riva_hw.c was manually tweaked to avoid pathological white-space. No outstanding warnings were found building allmodconfig with GCC 9.3.0 for x86_64, i386, arm64, arm, powerpc, powerpc64le, s390x, mips, sparc64, alpha, and m68k. [1] https://lore.kernel.org/lkml/20200603174714.192027-1-glider@google.com/ [2] https://lore.kernel.org/lkml/CA+55aFw+Vbj0i=1TGqCR5vQkCzWJ0QxK6CernOU6eedsudAixw@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CA+55aFwgbgqhbp1fkxvRKEpzyR5J8n1vKT1VZdz9knmPuXhOeg@mail.gmail.com/ [4] https://lore.kernel.org/lkml/CA+55aFz2500WfbKXAx8s67wrm9=yVJu65TpLgN_ybYNv0VEOKA@mail.gmail.com/ Reviewed-by: Leon Romanovsky <leonro@mellanox.com> # drivers/infiniband and mlx4/mlx5 Acked-by: Jason Gunthorpe <jgg@mellanox.com> # IB Acked-by: Kalle Valo <kvalo@codeaurora.org> # wireless drivers Reviewed-by: Chao Yu <yuchao0@huawei.com> # erofs Signed-off-by: Kees Cook <keescook@chromium.org>
2020-06-04 04:09:38 +08:00
unsigned in, log;
struct vhost_log *vq_log;
struct msghdr msg = {
.msg_name = NULL,
.msg_namelen = 0,
.msg_control = NULL, /* FIXME: get and handle RX aux data. */
.msg_controllen = 0,
.msg_flags = MSG_DONTWAIT,
};
struct virtio_net_hdr hdr = {
.flags = 0,
.gso_type = VIRTIO_NET_HDR_GSO_NONE
};
size_t total_len = 0;
int err, mergeable;
s16 headcount;
size_t vhost_hlen, sock_hlen;
size_t vhost_len, sock_len;
bool busyloop_intr = false;
struct socket *sock;
struct iov_iter fixup;
__virtio16 num_buffers;
int recv_pkts = 0;
mutex_lock_nested(&vq->mutex, VHOST_NET_VQ_RX);
sock = vhost_vq_get_backend(vq);
if (!sock)
goto out;
2016-06-23 14:04:32 +08:00
if (!vq_meta_prefetch(vq))
2016-06-23 14:04:32 +08:00
goto out;
vhost_disable_notify(&net->dev, vq);
vhost_net_disable_vq(net, vq);
vhost_hlen = nvq->vhost_hlen;
sock_hlen = nvq->sock_hlen;
vq_log = unlikely(vhost_has_feature(vq, VHOST_F_LOG_ALL)) ?
vq->log : NULL;
mergeable = vhost_has_feature(vq, VIRTIO_NET_F_MRG_RXBUF);
do {
sock_len = vhost_net_rx_peek_head_len(net, sock->sk,
&busyloop_intr);
if (!sock_len)
break;
sock_len += sock_hlen;
vhost_len = sock_len + vhost_hlen;
headcount = get_rx_bufs(vq, vq->heads + nvq->done_idx,
vhost_len, &in, vq_log, &log,
likely(mergeable) ? UIO_MAXIOV : 1);
/* On error, stop handling until the next kick. */
if (unlikely(headcount < 0))
goto out;
/* OK, now we need to know about added descriptors. */
if (!headcount) {
if (unlikely(busyloop_intr)) {
vhost_poll_queue(&vq->poll);
} else if (unlikely(vhost_enable_notify(&net->dev, vq))) {
/* They have slipped one in as we were
* doing that: check again. */
vhost_disable_notify(&net->dev, vq);
continue;
}
/* Nothing new? Wait for eventfd to tell us
* they refilled. */
goto out;
}
busyloop_intr = false;
if (nvq->rx_ring)
msg.msg_control = vhost_net_buf_consume(&nvq->rxq);
/* On overrun, truncate and discard */
if (unlikely(headcount > UIO_MAXIOV)) {
iov_iter_init(&msg.msg_iter, READ, vq->iov, 1, 1);
err = sock->ops->recvmsg(sock, &msg,
1, MSG_DONTWAIT | MSG_TRUNC);
pr_debug("Discarded rx packet: len %zd\n", sock_len);
continue;
}
/* We don't need to be notified again. */
iov_iter_init(&msg.msg_iter, READ, vq->iov, in, vhost_len);
fixup = msg.msg_iter;
if (unlikely((vhost_hlen))) {
/* We will supply the header ourselves
* TODO: support TSO.
*/
iov_iter_advance(&msg.msg_iter, vhost_hlen);
}
err = sock->ops->recvmsg(sock, &msg,
sock_len, MSG_DONTWAIT | MSG_TRUNC);
/* Userspace might have consumed the packet meanwhile:
* it's not supposed to do this usually, but might be hard
* to prevent. Discard data we got (if any) and keep going. */
if (unlikely(err != sock_len)) {
pr_debug("Discarded rx packet: "
" len %d, expected %zd\n", err, sock_len);
vhost_discard_vq_desc(vq, headcount);
continue;
}
/* Supply virtio_net_hdr if VHOST_NET_F_VIRTIO_NET_HDR */
if (unlikely(vhost_hlen)) {
if (copy_to_iter(&hdr, sizeof(hdr),
&fixup) != sizeof(hdr)) {
vq_err(vq, "Unable to write vnet_hdr "
"at addr %p\n", vq->iov->iov_base);
goto out;
}
} else {
/* Header came from socket; we'll need to patch
* ->num_buffers over if VIRTIO_NET_F_MRG_RXBUF
*/
iov_iter_advance(&fixup, sizeof(hdr));
}
/* TODO: Should check and handle checksum. */
num_buffers = cpu_to_vhost16(vq, headcount);
if (likely(mergeable) &&
copy_to_iter(&num_buffers, sizeof num_buffers,
&fixup) != sizeof num_buffers) {
vq_err(vq, "Failed num_buffers write");
vhost_discard_vq_desc(vq, headcount);
goto out;
}
nvq->done_idx += headcount;
if (nvq->done_idx > VHOST_NET_BATCH)
vhost_net_signal_used(nvq);
if (unlikely(vq_log))
vhost_log_write(vq, vq_log, log, vhost_len,
vq->iov, in);
total_len += vhost_len;
} while (likely(!vhost_exceeds_weight(vq, ++recv_pkts, total_len)));
if (unlikely(busyloop_intr))
vhost_poll_queue(&vq->poll);
else if (!sock_len)
vhost_net_enable_vq(net, vq);
out:
vhost_net_signal_used(nvq);
mutex_unlock(&vq->mutex);
}
static void handle_tx_kick(struct vhost_work *work)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue,
poll.work);
struct vhost_net *net = container_of(vq->dev, struct vhost_net, dev);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
handle_tx(net);
}
static void handle_rx_kick(struct vhost_work *work)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue,
poll.work);
struct vhost_net *net = container_of(vq->dev, struct vhost_net, dev);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
handle_rx(net);
}
static void handle_tx_net(struct vhost_work *work)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct vhost_net *net = container_of(work, struct vhost_net,
poll[VHOST_NET_VQ_TX].work);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
handle_tx(net);
}
static void handle_rx_net(struct vhost_work *work)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct vhost_net *net = container_of(work, struct vhost_net,
poll[VHOST_NET_VQ_RX].work);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
handle_rx(net);
}
static int vhost_net_open(struct inode *inode, struct file *f)
{
struct vhost_net *n;
struct vhost_dev *dev;
struct vhost_virtqueue **vqs;
void **queue;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
struct xdp_buff *xdp;
int i;
mm, tree wide: replace __GFP_REPEAT by __GFP_RETRY_MAYFAIL with more useful semantic __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>
2017-07-13 05:36:45 +08:00
n = kvmalloc(sizeof *n, GFP_KERNEL | __GFP_RETRY_MAYFAIL);
if (!n)
return -ENOMEM;
treewide: kmalloc() -> kmalloc_array() The kmalloc() function has a 2-factor argument form, kmalloc_array(). This patch replaces cases of: kmalloc(a * b, gfp) with: kmalloc_array(a * b, gfp) as well as handling cases of: kmalloc(a * b * c, gfp) with: kmalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kmalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kmalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The tools/ directory was manually excluded, since it has its own implementation of kmalloc(). The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kmalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kmalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kmalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(char) * COUNT + COUNT , ...) | kmalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kmalloc + kmalloc_array ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kmalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kmalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kmalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kmalloc(C1 * C2 * C3, ...) | kmalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kmalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kmalloc(sizeof(THING) * C2, ...) | kmalloc(sizeof(TYPE) * C2, ...) | kmalloc(C1 * C2 * C3, ...) | kmalloc(C1 * C2, ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - (E1) * E2 + E1, E2 , ...) | - kmalloc + kmalloc_array ( - (E1) * (E2) + E1, E2 , ...) | - kmalloc + kmalloc_array ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 04:55:00 +08:00
vqs = kmalloc_array(VHOST_NET_VQ_MAX, sizeof(*vqs), GFP_KERNEL);
if (!vqs) {
kvfree(n);
return -ENOMEM;
}
queue = kmalloc_array(VHOST_NET_BATCH, sizeof(void *),
GFP_KERNEL);
if (!queue) {
kfree(vqs);
kvfree(n);
return -ENOMEM;
}
n->vqs[VHOST_NET_VQ_RX].rxq.queue = queue;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
xdp = kmalloc_array(VHOST_NET_BATCH, sizeof(*xdp), GFP_KERNEL);
if (!xdp) {
kfree(vqs);
kvfree(n);
kfree(queue);
return -ENOMEM;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
}
n->vqs[VHOST_NET_VQ_TX].xdp = xdp;
dev = &n->dev;
vqs[VHOST_NET_VQ_TX] = &n->vqs[VHOST_NET_VQ_TX].vq;
vqs[VHOST_NET_VQ_RX] = &n->vqs[VHOST_NET_VQ_RX].vq;
n->vqs[VHOST_NET_VQ_TX].vq.handle_kick = handle_tx_kick;
n->vqs[VHOST_NET_VQ_RX].vq.handle_kick = handle_rx_kick;
for (i = 0; i < VHOST_NET_VQ_MAX; i++) {
n->vqs[i].ubufs = NULL;
n->vqs[i].ubuf_info = NULL;
n->vqs[i].upend_idx = 0;
n->vqs[i].done_idx = 0;
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
n->vqs[i].batched_xdp = 0;
n->vqs[i].vhost_hlen = 0;
n->vqs[i].sock_hlen = 0;
vhost_net: initialize rx_ring in vhost_net_open() KMSAN reported a use of uninit memory in vhost_net_buf_unproduce() while trying to access n->vqs[VHOST_NET_VQ_TX].rx_ring: ================================================================== BUG: KMSAN: use of uninitialized memory in vhost_net_buf_unproduce+0x7bb/0x9a0 drivers/vho et.c:170 CPU: 0 PID: 3021 Comm: syz-fuzzer Not tainted 4.16.0-rc4+ #3853 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x185/0x1d0 lib/dump_stack.c:53 kmsan_report+0x142/0x1f0 mm/kmsan/kmsan.c:1093 __msan_warning_32+0x6c/0xb0 mm/kmsan/kmsan_instr.c:676 vhost_net_buf_unproduce+0x7bb/0x9a0 drivers/vhost/net.c:170 vhost_net_stop_vq drivers/vhost/net.c:974 [inline] vhost_net_stop+0x146/0x380 drivers/vhost/net.c:982 vhost_net_release+0xb1/0x4f0 drivers/vhost/net.c:1015 __fput+0x49f/0xa00 fs/file_table.c:209 ____fput+0x37/0x40 fs/file_table.c:243 task_work_run+0x243/0x2c0 kernel/task_work.c:113 tracehook_notify_resume include/linux/tracehook.h:191 [inline] exit_to_usermode_loop arch/x86/entry/common.c:166 [inline] prepare_exit_to_usermode+0x349/0x3b0 arch/x86/entry/common.c:196 syscall_return_slowpath+0xf3/0x6d0 arch/x86/entry/common.c:265 do_syscall_64+0x34d/0x450 arch/x86/entry/common.c:292 ... origin: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:303 [inline] kmsan_internal_poison_shadow+0xb8/0x1b0 mm/kmsan/kmsan.c:213 kmsan_kmalloc_large+0x6f/0xd0 mm/kmsan/kmsan.c:392 kmalloc_large_node_hook mm/slub.c:1366 [inline] kmalloc_large_node mm/slub.c:3808 [inline] __kmalloc_node+0x100e/0x1290 mm/slub.c:3818 kmalloc_node include/linux/slab.h:554 [inline] kvmalloc_node+0x1a5/0x2e0 mm/util.c:419 kvmalloc include/linux/mm.h:541 [inline] vhost_net_open+0x64/0x5f0 drivers/vhost/net.c:921 misc_open+0x7b5/0x8b0 drivers/char/misc.c:154 chrdev_open+0xc28/0xd90 fs/char_dev.c:417 do_dentry_open+0xccb/0x1430 fs/open.c:752 vfs_open+0x272/0x2e0 fs/open.c:866 do_last fs/namei.c:3378 [inline] path_openat+0x49ad/0x6580 fs/namei.c:3519 do_filp_open+0x267/0x640 fs/namei.c:3553 do_sys_open+0x6ad/0x9c0 fs/open.c:1059 SYSC_openat+0xc7/0xe0 fs/open.c:1086 SyS_openat+0x63/0x90 fs/open.c:1080 do_syscall_64+0x2f1/0x450 arch/x86/entry/common.c:287 ================================================================== Fixes: c67df11f6e480 ("vhost_net: try batch dequing from skb array") Signed-off-by: Alexander Potapenko <glider@google.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>
2018-03-09 14:50:32 +08:00
n->vqs[i].rx_ring = NULL;
vhost_net_buf_init(&n->vqs[i].rxq);
}
vhost: fix OOB in get_rx_bufs() After batched used ring updating was introduced in commit e2b3b35eb989 ("vhost_net: batch used ring update in rx"). We tend to batch heads in vq->heads for more than one packet. But the quota passed to get_rx_bufs() was not correctly limited, which can result a OOB write in vq->heads. headcount = get_rx_bufs(vq, vq->heads + nvq->done_idx, vhost_len, &in, vq_log, &log, likely(mergeable) ? UIO_MAXIOV : 1); UIO_MAXIOV was still used which is wrong since we could have batched used in vq->heads, this will cause OOB if the next buffer needs more than 960 (1024 (UIO_MAXIOV) - 64 (VHOST_NET_BATCH)) heads after we've batched 64 (VHOST_NET_BATCH) heads: Acked-by: Stefan Hajnoczi <stefanha@redhat.com> ============================================================================= BUG kmalloc-8k (Tainted: G B ): Redzone overwritten ----------------------------------------------------------------------------- INFO: 0x00000000fd93b7a2-0x00000000f0713384. First byte 0xa9 instead of 0xcc INFO: Allocated in alloc_pd+0x22/0x60 age=3933677 cpu=2 pid=2674 kmem_cache_alloc_trace+0xbb/0x140 alloc_pd+0x22/0x60 gen8_ppgtt_create+0x11d/0x5f0 i915_ppgtt_create+0x16/0x80 i915_gem_create_context+0x248/0x390 i915_gem_context_create_ioctl+0x4b/0xe0 drm_ioctl_kernel+0xa5/0xf0 drm_ioctl+0x2ed/0x3a0 do_vfs_ioctl+0x9f/0x620 ksys_ioctl+0x6b/0x80 __x64_sys_ioctl+0x11/0x20 do_syscall_64+0x43/0xf0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 INFO: Slab 0x00000000d13e87af objects=3 used=3 fp=0x (null) flags=0x200000000010201 INFO: Object 0x0000000003278802 @offset=17064 fp=0x00000000e2e6652b Fixing this by allocating UIO_MAXIOV + VHOST_NET_BATCH iovs for vhost-net. This is done through set the limitation through vhost_dev_init(), then set_owner can allocate the number of iov in a per device manner. This fixes CVE-2018-16880. Fixes: e2b3b35eb989 ("vhost_net: batch used ring update in rx") Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-28 15:05:05 +08:00
vhost_dev_init(dev, vqs, VHOST_NET_VQ_MAX,
UIO_MAXIOV + VHOST_NET_BATCH,
VHOST_NET_PKT_WEIGHT, VHOST_NET_WEIGHT, true,
NULL);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
vhost_poll_init(n->poll + VHOST_NET_VQ_TX, handle_tx_net, EPOLLOUT, dev);
vhost_poll_init(n->poll + VHOST_NET_VQ_RX, handle_rx_net, EPOLLIN, dev);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
f->private_data = n;
n->page_frag.page = NULL;
n->refcnt_bias = 0;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
return 0;
}
static struct socket *vhost_net_stop_vq(struct vhost_net *n,
struct vhost_virtqueue *vq)
{
struct socket *sock;
struct vhost_net_virtqueue *nvq =
container_of(vq, struct vhost_net_virtqueue, vq);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_lock(&vq->mutex);
sock = vhost_vq_get_backend(vq);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
vhost_net_disable_vq(n, vq);
vhost_vq_set_backend(vq, NULL);
vhost_net_buf_unproduce(nvq);
nvq->rx_ring = NULL;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_unlock(&vq->mutex);
return sock;
}
static void vhost_net_stop(struct vhost_net *n, struct socket **tx_sock,
struct socket **rx_sock)
{
*tx_sock = vhost_net_stop_vq(n, &n->vqs[VHOST_NET_VQ_TX].vq);
*rx_sock = vhost_net_stop_vq(n, &n->vqs[VHOST_NET_VQ_RX].vq);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
static void vhost_net_flush_vq(struct vhost_net *n, int index)
{
vhost_poll_flush(n->poll + index);
vhost_poll_flush(&n->vqs[index].vq.poll);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
static void vhost_net_flush(struct vhost_net *n)
{
vhost_net_flush_vq(n, VHOST_NET_VQ_TX);
vhost_net_flush_vq(n, VHOST_NET_VQ_RX);
if (n->vqs[VHOST_NET_VQ_TX].ubufs) {
mutex_lock(&n->vqs[VHOST_NET_VQ_TX].vq.mutex);
n->tx_flush = true;
mutex_unlock(&n->vqs[VHOST_NET_VQ_TX].vq.mutex);
/* Wait for all lower device DMAs done. */
vhost_net_ubuf_put_and_wait(n->vqs[VHOST_NET_VQ_TX].ubufs);
mutex_lock(&n->vqs[VHOST_NET_VQ_TX].vq.mutex);
n->tx_flush = false;
atomic_set(&n->vqs[VHOST_NET_VQ_TX].ubufs->refcount, 1);
mutex_unlock(&n->vqs[VHOST_NET_VQ_TX].vq.mutex);
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
static int vhost_net_release(struct inode *inode, struct file *f)
{
struct vhost_net *n = f->private_data;
struct socket *tx_sock;
struct socket *rx_sock;
vhost_net_stop(n, &tx_sock, &rx_sock);
vhost_net_flush(n);
vhost_dev_stop(&n->dev);
vhost_dev_cleanup(&n->dev);
vhost_net_vq_reset(n);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (tx_sock)
sockfd_put(tx_sock);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (rx_sock)
sockfd_put(rx_sock);
/* Make sure no callbacks are outstanding */
synchronize_rcu();
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* We do an extra flush before freeing memory,
* since jobs can re-queue themselves. */
vhost_net_flush(n);
kfree(n->vqs[VHOST_NET_VQ_RX].rxq.queue);
vhost_net: batch submitting XDP buffers to underlayer sockets This patch implements XDP batching for vhost_net. The idea is first to try to do userspace copy and build XDP buff directly in vhost. Instead of submitting the packet immediately, vhost_net will batch them in an array and submit every 64 (VHOST_NET_BATCH) packets to the under layer sockets through msg_control of sendmsg(). When XDP is enabled on the TUN/TAP, TUN/TAP can process XDP inside a loop without caring GUP thus it can do batch map flushing. When XDP is not enabled or not supported, the underlayer socket need to build skb and pass it to network core. The batched packet submission allows us to do batching like netif_receive_skb_list() in the future. This saves lots of indirect calls for better cache utilization. For the case that we can't so batching e.g when sndbuf is limited or packet size is too large, we will go for usual one packet per sendmsg() way. Doing testpmd on various setups gives us: Test /+pps% XDP_DROP on TAP /+44.8% XDP_REDIRECT on TAP /+29% macvtap (skb) /+26% Netperf tests shows obvious improvements for small packet transmission: size/session/+thu%/+normalize% 64/ 1/ +2%/ 0% 64/ 2/ +3%/ +1% 64/ 4/ +7%/ +5% 64/ 8/ +8%/ +6% 256/ 1/ +3%/ 0% 256/ 2/ +10%/ +7% 256/ 4/ +26%/ +22% 256/ 8/ +27%/ +23% 512/ 1/ +3%/ +2% 512/ 2/ +19%/ +14% 512/ 4/ +43%/ +40% 512/ 8/ +45%/ +41% 1024/ 1/ +4%/ 0% 1024/ 2/ +27%/ +21% 1024/ 4/ +38%/ +73% 1024/ 8/ +15%/ +24% 2048/ 1/ +10%/ +7% 2048/ 2/ +16%/ +12% 2048/ 4/ 0%/ +2% 2048/ 8/ 0%/ +2% 4096/ 1/ +36%/ +60% 4096/ 2/ -11%/ -26% 4096/ 4/ 0%/ +14% 4096/ 8/ 0%/ +4% 16384/ 1/ -1%/ +5% 16384/ 2/ 0%/ +2% 16384/ 4/ 0%/ -3% 16384/ 8/ 0%/ +4% 65535/ 1/ 0%/ +10% 65535/ 2/ 0%/ +8% 65535/ 4/ 0%/ +1% 65535/ 8/ 0%/ +3% Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 11:17:09 +08:00
kfree(n->vqs[VHOST_NET_VQ_TX].xdp);
kfree(n->dev.vqs);
if (n->page_frag.page)
__page_frag_cache_drain(n->page_frag.page, n->refcnt_bias);
kvfree(n);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
return 0;
}
static struct socket *get_raw_socket(int fd)
{
net: make getname() functions return length rather than use int* parameter Changes since v1: Added changes in these files: drivers/infiniband/hw/usnic/usnic_transport.c drivers/staging/lustre/lnet/lnet/lib-socket.c drivers/target/iscsi/iscsi_target_login.c drivers/vhost/net.c fs/dlm/lowcomms.c fs/ocfs2/cluster/tcp.c security/tomoyo/network.c Before: All these functions either return a negative error indicator, or store length of sockaddr into "int *socklen" parameter and return zero on success. "int *socklen" parameter is awkward. For example, if caller does not care, it still needs to provide on-stack storage for the value it does not need. None of the many FOO_getname() functions of various protocols ever used old value of *socklen. They always just overwrite it. This change drops this parameter, and makes all these functions, on success, return length of sockaddr. It's always >= 0 and can be differentiated from an error. Tests in callers are changed from "if (err)" to "if (err < 0)", where needed. rpc_sockname() lost "int buflen" parameter, since its only use was to be passed to kernel_getsockname() as &buflen and subsequently not used in any way. Userspace API is not changed. text data bss dec hex filename 30108430 2633624 873672 33615726 200ef6e vmlinux.before.o 30108109 2633612 873672 33615393 200ee21 vmlinux.o Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com> CC: David S. Miller <davem@davemloft.net> CC: linux-kernel@vger.kernel.org CC: netdev@vger.kernel.org CC: linux-bluetooth@vger.kernel.org CC: linux-decnet-user@lists.sourceforge.net CC: linux-wireless@vger.kernel.org CC: linux-rdma@vger.kernel.org CC: linux-sctp@vger.kernel.org CC: linux-nfs@vger.kernel.org CC: linux-x25@vger.kernel.org Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 03:00:20 +08:00
int r;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
struct socket *sock = sockfd_lookup(fd, &r);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (!sock)
return ERR_PTR(-ENOTSOCK);
/* Parameter checking */
if (sock->sk->sk_type != SOCK_RAW) {
r = -ESOCKTNOSUPPORT;
goto err;
}
if (sock->sk->sk_family != AF_PACKET) {
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
r = -EPFNOSUPPORT;
goto err;
}
return sock;
err:
sockfd_put(sock);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
return ERR_PTR(r);
}
static struct ptr_ring *get_tap_ptr_ring(int fd)
{
struct ptr_ring *ring;
struct file *file = fget(fd);
if (!file)
return NULL;
ring = tun_get_tx_ring(file);
if (!IS_ERR(ring))
goto out;
ring = tap_get_ptr_ring(file);
if (!IS_ERR(ring))
goto out;
ring = NULL;
out:
fput(file);
return ring;
}
static struct socket *get_tap_socket(int fd)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
struct file *file = fget(fd);
struct socket *sock;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (!file)
return ERR_PTR(-EBADF);
sock = tun_get_socket(file);
if (!IS_ERR(sock))
return sock;
sock = tap_get_socket(file);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (IS_ERR(sock))
fput(file);
return sock;
}
static struct socket *get_socket(int fd)
{
struct socket *sock;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
/* special case to disable backend */
if (fd == -1)
return NULL;
sock = get_raw_socket(fd);
if (!IS_ERR(sock))
return sock;
sock = get_tap_socket(fd);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (!IS_ERR(sock))
return sock;
return ERR_PTR(-ENOTSOCK);
}
static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
{
struct socket *sock, *oldsock;
struct vhost_virtqueue *vq;
struct vhost_net_virtqueue *nvq;
struct vhost_net_ubuf_ref *ubufs, *oldubufs = NULL;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
int r;
mutex_lock(&n->dev.mutex);
r = vhost_dev_check_owner(&n->dev);
if (r)
goto err;
if (index >= VHOST_NET_VQ_MAX) {
r = -ENOBUFS;
goto err;
}
vq = &n->vqs[index].vq;
nvq = &n->vqs[index];
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_lock(&vq->mutex);
/* Verify that ring has been setup correctly. */
if (!vhost_vq_access_ok(vq)) {
r = -EFAULT;
goto err_vq;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
sock = get_socket(fd);
if (IS_ERR(sock)) {
r = PTR_ERR(sock);
goto err_vq;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
/* start polling new socket */
oldsock = vhost_vq_get_backend(vq);
if (sock != oldsock) {
ubufs = vhost_net_ubuf_alloc(vq,
sock && vhost_sock_zcopy(sock));
if (IS_ERR(ubufs)) {
r = PTR_ERR(ubufs);
goto err_ubufs;
}
vhost_net_disable_vq(n, vq);
vhost_vq_set_backend(vq, sock);
vhost_net_buf_unproduce(nvq);
r = vhost_vq_init_access(vq);
if (r)
goto err_used;
r = vhost_net_enable_vq(n, vq);
if (r)
goto err_used;
if (index == VHOST_NET_VQ_RX)
nvq->rx_ring = get_tap_ptr_ring(fd);
oldubufs = nvq->ubufs;
nvq->ubufs = ubufs;
n->tx_packets = 0;
n->tx_zcopy_err = 0;
n->tx_flush = false;
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_unlock(&vq->mutex);
if (oldubufs) {
vhost_net_ubuf_put_wait_and_free(oldubufs);
mutex_lock(&vq->mutex);
vhost_zerocopy_signal_used(n, vq);
mutex_unlock(&vq->mutex);
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (oldsock) {
vhost_net_flush_vq(n, index);
sockfd_put(oldsock);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
mutex_unlock(&n->dev.mutex);
return 0;
err_used:
vhost_vq_set_backend(vq, oldsock);
vhost_net_enable_vq(n, vq);
if (ubufs)
vhost_net_ubuf_put_wait_and_free(ubufs);
err_ubufs:
if (sock)
sockfd_put(sock);
err_vq:
mutex_unlock(&vq->mutex);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
err:
mutex_unlock(&n->dev.mutex);
return r;
}
static long vhost_net_reset_owner(struct vhost_net *n)
{
struct socket *tx_sock = NULL;
struct socket *rx_sock = NULL;
long err;
struct vhost_iotlb *umem;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_lock(&n->dev.mutex);
err = vhost_dev_check_owner(&n->dev);
if (err)
goto done;
umem = vhost_dev_reset_owner_prepare();
if (!umem) {
err = -ENOMEM;
goto done;
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
vhost_net_stop(n, &tx_sock, &rx_sock);
vhost_net_flush(n);
vhost_dev_stop(&n->dev);
vhost_dev_reset_owner(&n->dev, umem);
vhost_net_vq_reset(n);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
done:
mutex_unlock(&n->dev.mutex);
if (tx_sock)
sockfd_put(tx_sock);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
if (rx_sock)
sockfd_put(rx_sock);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
return err;
}
static int vhost_net_set_features(struct vhost_net *n, u64 features)
{
size_t vhost_hlen, sock_hlen, hdr_len;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
int i;
hdr_len = (features & ((1ULL << VIRTIO_NET_F_MRG_RXBUF) |
(1ULL << VIRTIO_F_VERSION_1))) ?
sizeof(struct virtio_net_hdr_mrg_rxbuf) :
sizeof(struct virtio_net_hdr);
if (features & (1 << VHOST_NET_F_VIRTIO_NET_HDR)) {
/* vhost provides vnet_hdr */
vhost_hlen = hdr_len;
sock_hlen = 0;
} else {
/* socket provides vnet_hdr */
vhost_hlen = 0;
sock_hlen = hdr_len;
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_lock(&n->dev.mutex);
if ((features & (1 << VHOST_F_LOG_ALL)) &&
2016-06-23 14:04:32 +08:00
!vhost_log_access_ok(&n->dev))
goto out_unlock;
if ((features & (1ULL << VIRTIO_F_ACCESS_PLATFORM))) {
2016-06-23 14:04:32 +08:00
if (vhost_init_device_iotlb(&n->dev, true))
goto out_unlock;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
2016-06-23 14:04:32 +08:00
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
mutex_lock(&n->vqs[i].vq.mutex);
n->vqs[i].vq.acked_features = features;
n->vqs[i].vhost_hlen = vhost_hlen;
n->vqs[i].sock_hlen = sock_hlen;
mutex_unlock(&n->vqs[i].vq.mutex);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
mutex_unlock(&n->dev.mutex);
return 0;
2016-06-23 14:04:32 +08:00
out_unlock:
mutex_unlock(&n->dev.mutex);
return -EFAULT;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
static long vhost_net_set_owner(struct vhost_net *n)
{
int r;
mutex_lock(&n->dev.mutex);
if (vhost_dev_has_owner(&n->dev)) {
r = -EBUSY;
goto out;
}
r = vhost_net_set_ubuf_info(n);
if (r)
goto out;
r = vhost_dev_set_owner(&n->dev);
if (r)
vhost_net_clear_ubuf_info(n);
vhost_net_flush(n);
out:
mutex_unlock(&n->dev.mutex);
return r;
}
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
static long vhost_net_ioctl(struct file *f, unsigned int ioctl,
unsigned long arg)
{
struct vhost_net *n = f->private_data;
void __user *argp = (void __user *)arg;
u64 __user *featurep = argp;
struct vhost_vring_file backend;
u64 features;
int r;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
switch (ioctl) {
case VHOST_NET_SET_BACKEND:
if (copy_from_user(&backend, argp, sizeof backend))
return -EFAULT;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
return vhost_net_set_backend(n, backend.index, backend.fd);
case VHOST_GET_FEATURES:
features = VHOST_NET_FEATURES;
if (copy_to_user(featurep, &features, sizeof features))
return -EFAULT;
return 0;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
case VHOST_SET_FEATURES:
if (copy_from_user(&features, featurep, sizeof features))
return -EFAULT;
if (features & ~VHOST_NET_FEATURES)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
return -EOPNOTSUPP;
return vhost_net_set_features(n, features);
case VHOST_GET_BACKEND_FEATURES:
features = VHOST_NET_BACKEND_FEATURES;
if (copy_to_user(featurep, &features, sizeof(features)))
return -EFAULT;
return 0;
case VHOST_SET_BACKEND_FEATURES:
if (copy_from_user(&features, featurep, sizeof(features)))
return -EFAULT;
if (features & ~VHOST_NET_BACKEND_FEATURES)
return -EOPNOTSUPP;
vhost_set_backend_features(&n->dev, features);
return 0;
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
case VHOST_RESET_OWNER:
return vhost_net_reset_owner(n);
case VHOST_SET_OWNER:
return vhost_net_set_owner(n);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
default:
mutex_lock(&n->dev.mutex);
r = vhost_dev_ioctl(&n->dev, ioctl, argp);
if (r == -ENOIOCTLCMD)
r = vhost_vring_ioctl(&n->dev, ioctl, argp);
else
vhost_net_flush(n);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
mutex_unlock(&n->dev.mutex);
return r;
}
}
2016-06-23 14:04:32 +08:00
static ssize_t vhost_net_chr_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct file *file = iocb->ki_filp;
struct vhost_net *n = file->private_data;
struct vhost_dev *dev = &n->dev;
int noblock = file->f_flags & O_NONBLOCK;
return vhost_chr_read_iter(dev, to, noblock);
}
static ssize_t vhost_net_chr_write_iter(struct kiocb *iocb,
struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct vhost_net *n = file->private_data;
struct vhost_dev *dev = &n->dev;
return vhost_chr_write_iter(dev, from);
}
static __poll_t vhost_net_chr_poll(struct file *file, poll_table *wait)
2016-06-23 14:04:32 +08:00
{
struct vhost_net *n = file->private_data;
struct vhost_dev *dev = &n->dev;
return vhost_chr_poll(file, dev, wait);
}
static const struct file_operations vhost_net_fops = {
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
.owner = THIS_MODULE,
.release = vhost_net_release,
2016-06-23 14:04:32 +08:00
.read_iter = vhost_net_chr_read_iter,
.write_iter = vhost_net_chr_write_iter,
.poll = vhost_net_chr_poll,
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
.unlocked_ioctl = vhost_net_ioctl,
.compat_ioctl = compat_ptr_ioctl,
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
.open = vhost_net_open,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-16 00:52:59 +08:00
.llseek = noop_llseek,
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
};
static struct miscdevice vhost_net_misc = {
.minor = VHOST_NET_MINOR,
.name = "vhost-net",
.fops = &vhost_net_fops,
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
};
static int vhost_net_init(void)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
if (experimental_zcopytx)
vhost_net_enable_zcopy(VHOST_NET_VQ_TX);
return misc_register(&vhost_net_misc);
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
}
module_init(vhost_net_init);
static void vhost_net_exit(void)
vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox <alan@linux.intel.com>, David L Stevens <dlstevens@us.ibm.com>, Chris Wright <chrisw@redhat.com>) Acked-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-14 14:17:27 +08:00
{
misc_deregister(&vhost_net_misc);
}
module_exit(vhost_net_exit);
MODULE_VERSION("0.0.1");
MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Michael S. Tsirkin");
MODULE_DESCRIPTION("Host kernel accelerator for virtio net");
MODULE_ALIAS_MISCDEV(VHOST_NET_MINOR);
MODULE_ALIAS("devname:vhost-net");