External/cloud-hypervisor
1
0
Fork 0
mirror of https://github.com/cloud-hypervisor/cloud-hypervisor.git synced 2024-11-05 03:21:13 +00:00
Code Issues Packages Projects Releases Wiki Activity
1,280 Commits 16 Branches 60 Tags 56 MiB
378dd81204
Commit Graph

218 Commits

Author SHA1 Message Date
Sergio Lopez
5c06b7f862 vhost_user_block: Implement optional static polling 
Actively polling the virtqueue significantly reduces the latency of
each I/O operation, at the expense of using more CPU time. This
features is specially useful when using low-latency devices (SSD,
NVMe) as the backend.

This change implements static polling. When a request arrives after
being idle, vhost_user_block will keep checking the virtqueue for new
requests, until POLL_QUEUE_US (50us) has passed without finding one.

POLL_QUEUE_US is defined to be 50us, based on the current latency of
enterprise SSDs (< 30us) and the overhead of the emulation.

This feature is enabled by default, and can be disabled by using the
"poll_queue" parameter of "block-backend".

This is a test using null_blk as a backend for the image, with the
following parameters:

 - null_blk gb=20 nr_devices=1 irqmode=2 completion_nsec=0 no_sched=1

With "poll_queue=false":

fio --ioengine=sync --bs=4k --rw randread --name randread --direct=1
--filename=/dev/vdb --time_based --runtime=10

randread: (g=0): rw=randread, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=sync, iodepth=1
fio-3.14
Starting 1 process
Jobs: 1 (f=1): [r(1)][100.0%][r=169MiB/s][r=43.2k IOPS][eta 00m:00s]
randread: (groupid=0, jobs=1): err= 0: pid=433: Tue Feb 18 11:12:59 2020
  read: IOPS=43.2k, BW=169MiB/s (177MB/s)(1688MiB/10001msec)
    clat (usec): min=17, max=836, avg=21.64, stdev= 3.81
     lat (usec): min=17, max=836, avg=21.77, stdev= 3.81
    clat percentiles (nsec):
     |  1.00th=[19328],  5.00th=[19840], 10.00th=[20352], 20.00th=[21120],
     | 30.00th=[21376], 40.00th=[21376], 50.00th=[21376], 60.00th=[21632],
     | 70.00th=[21632], 80.00th=[21888], 90.00th=[22144], 95.00th=[22912],
     | 99.00th=[28544], 99.50th=[30336], 99.90th=[39168], 99.95th=[42752],
     | 99.99th=[71168]
   bw (  KiB/s): min=168440, max=188496, per=100.00%, avg=172912.00, stdev=3975.63, samples=19
   iops        : min=42110, max=47124, avg=43228.00, stdev=993.91, samples=19
  lat (usec)   : 20=5.90%, 50=94.08%, 100=0.02%, 250=0.01%, 500=0.01%
  lat (usec)   : 750=0.01%, 1000=0.01%
  cpu          : usr=10.35%, sys=25.82%, ctx=432417, majf=0, minf=10
  IO depths    : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     issued rwts: total=432220,0,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=1

Run status group 0 (all jobs):
   READ: bw=169MiB/s (177MB/s), 169MiB/s-169MiB/s (177MB/s-177MB/s), io=1688MiB (1770MB), run=10001-10001msec

Disk stats (read/write):
  vdb: ios=427867/0, merge=0/0, ticks=7346/0, in_queue=0, util=99.04%

With "poll_queue=true" (default):

fio --ioengine=sync --bs=4k --rw randread --name randread --direct=1
--filename=/dev/vdb --time_based --runtime=10

randread: (g=0): rw=randread, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=sync, iodepth=1
fio-3.14
Starting 1 process
Jobs: 1 (f=1): [r(1)][100.0%][r=260MiB/s][r=66.7k IOPS][eta 00m:00s]
randread: (groupid=0, jobs=1): err= 0: pid=422: Tue Feb 18 11:14:47 2020
  read: IOPS=68.5k, BW=267MiB/s (280MB/s)(2674MiB/10001msec)
    clat (usec): min=10, max=966, avg=13.60, stdev= 3.49
     lat (usec): min=10, max=966, avg=13.70, stdev= 3.50
    clat percentiles (nsec):
     |  1.00th=[11200],  5.00th=[11968], 10.00th=[11968], 20.00th=[12224],
     | 30.00th=[12992], 40.00th=[13504], 50.00th=[13760], 60.00th=[13888],
     | 70.00th=[14016], 80.00th=[14144], 90.00th=[14272], 95.00th=[14656],
     | 99.00th=[20352], 99.50th=[23936], 99.90th=[35072], 99.95th=[36096],
     | 99.99th=[47872]
   bw (  KiB/s): min=265456, max=296456, per=100.00%, avg=274229.05, stdev=13048.14, samples=19
   iops        : min=66364, max=74114, avg=68557.26, stdev=3262.03, samples=19
  lat (usec)   : 20=98.84%, 50=1.15%, 100=0.01%, 250=0.01%, 500=0.01%
  lat (usec)   : 750=0.01%, 1000=0.01%
  cpu          : usr=8.24%, sys=21.15%, ctx=684669, majf=0, minf=10
  IO depths    : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     issued rwts: total=684611,0,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=1

Run status group 0 (all jobs):
   READ: bw=267MiB/s (280MB/s), 267MiB/s-267MiB/s (280MB/s-280MB/s), io=2674MiB (2804MB), run=10001-10001msec

Disk stats (read/write):
  vdb: ios=677855/0, merge=0/0, ticks=7026/0, in_queue=0, util=99.04%

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-02-19 17:13:47 +00:00
Sergio Lopez
1ef6996207 vhost_user_backend: Add helpers for EVENT_IDX 
Add helpers to Vring and VhostUserSlaveReqHandler for EVENT_IDX, so
consumers of this crate can make use of this feature.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-02-19 17:13:47 +00:00
Sebastien Boeuf
ddf6caf955 ci: Improve test_memory_mergeable_on stability 
The integration test test_memory_mergeable_on has been fairly unstable
for quite some time now. Because it can take some time for the VM to be
spawned and to be able to perform a correct measure of the PSS, this
commit simply increases the time before such measure is done.
This should return more accurate PSS results, which should help
stabilize the test.

Fixes #781

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
 2020-02-19 12:36:28 +00:00
Liu Bo
4970e2f703 vhost-user-fs: add dax tests for vhost_user_fs rust daemon 
Now that vhost_user_fs rust daemon supports virtiofs's dax mode, this adds
the two dax tests accordingly.

Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
 2020-02-19 07:52:50 +00:00
Liu Bo
59958f0a61 vhost_user_fs: add the ability to set slave req fd 
This adds the missing part of supporting virtiofs dax on the slave end,
that is, receiving a socket pair fd from the master end to set up a
communication channel for sending setupmapping & removemapping messages.

Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
 2020-02-19 07:52:50 +00:00
Rob Bradford
f7378bc092 tests: Add self spawning vhost-user-block test 
Also rename the net self spawning test to differentiate it from the the
block one.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-18 08:43:47 +00:00
Rob Bradford
7485a0c1f7 Revert "build: Don't fail build on test_vfio failure" 
This reverts commit 014844d0da.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-17 08:23:02 +00:00
Rob Bradford
f03602a4c9 tests: Add self spawning vhost-user-net test 
Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-14 17:32:49 +00:00
Bo Chen
ebd83699dc main: Display git commit hash with the '--version' option 
Add a build-script to propagate the git commit hash to other crates at
compile time through environment variables, and display the hash along
with the '--version' option.

Fixes #729

Signed-off-by: Bo Chen <chen.bo@intel.com>
 2020-02-14 10:00:14 +01:00
Rob Bradford
e8e4f43d52 tests: Use hugepages for test_vfio 
Using hugepages with VFIO and virtiofs can improve the performance of
the test_vfio test.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-13 17:55:21 +01:00
Rob Bradford
014844d0da build: Don't fail build on test_vfio failure 
test_vfio has been failing consistently on the CI so mark it with
a "#[ignore]" and then forceably build it again but ignore the build
result.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-12 15:37:47 +00:00
Rob Bradford
da7f31d4bc bin: vhost_user_fs: Port to new exit event strategy 
Rather than handling the KILL_EVENT in the event handler itself use the
newly added support in VhostUserBackend for providing a kill event
framework.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-11 15:21:07 +01:00
Rob Bradford
7f032c8bb3 bin: vhost_user_fs: Shutdown worker thread on exit 
This will ensure a clean shutdown of the backend

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-10 09:16:49 +01:00
Rob Bradford
99cb8dc0a4 bin: vhost_user_fs use error! macro logging for consistency 
Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-10 09:16:49 +01:00
Rob Bradford
6e6b2b84fe scripts: Check the Rust formatting is valid 
Check the rust formatting rather than just reformatting code on the CI
agent.

Also fix a formatting error that slipped in whilst the cargo fmt check
was not working correctly.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-04 16:47:05 +01:00
Rob Bradford
bc053f1b13 main: Ignore error on log writing 
A previous version of this change attempted to avoid panicking by not
using .expect() when handling an error when attempting to write to the
log file. Unfortunately the macro eprintln!() that was used to replace
the .expect() also has the behaviour of panicking if stderr cannot be
used. Instead swallow the error completely as if writing to the log has
failed at logging time it is almost certainly the case that any message
about the log would also not be seen.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-04 12:22:23 +00:00
Yang Zhong
91739be120 main: Add help info for block-backend 
Since the vhost-user-blk binary will be removed and the newer
release will integrate this block backend into cloud-hypervisor
binary. The block backend code has been added num_queues cmdline
support, we need update multiple queues help info for this
block-backend in the cloud-hypervisor.

Signed-off-by: Yang Zhong <yang.zhong@intel.com>
 2020-02-04 11:06:17 +00:00
Samuel Ortiz
ae6cf4c922 tests: integration: Add memory overhead test 
We measure the memory overhead that the VMM process adds to the guest VM
and compare it with a maximum acceptable limit. The test is run against
a simple VM, running 1 vCPU and 512MB of RAM. Although this is not by
any mean a comprehensive VMM overhead measurement, it will allow us to
detect when and if any PR makes our code cross an arbitrary memory
overhead threshold.

Fixes: #64

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
 2020-02-04 09:23:56 +00:00
Rob Bradford
7cb61d3960 main: Don't panic (by calling .expect()) if writing to the log fails 
As this can happen during the running of the VMM we should be very
careful not to panic() as that can lead to a thread being used by the VM
disappearing.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-02-03 19:08:08 +01:00
Yang Zhong
789a39a2d5 ci: Add MQ support in the test cases 
The CI need to add MQ support, and the vcpu number should
be bigger than or equal to queue number.

Signed-off-by: Yang Zhong <yang.zhong@intel.com>
 2020-02-03 09:49:27 +01:00
Yang Zhong
99da1dff90 vhost-user-blk: Add MQ support in backend 
Adding the num_queues parameter for vhost-user-blk backend, which
can enable MQ support in the backend.

This patch has enabled the MQ support from handle_event, and the
vhost-user-backend crate will enable multiple threads to call this
handle_event to handle read/write operations.

Signed-off-by: Yang Zhong <yang.zhong@intel.com>
 2020-02-03 09:49:27 +01:00
Rob Bradford
b4d04bdff6 tests: Add CLI <-> API validation test for --disk changes 
Check that the CLI generates the JSON data as expected.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-29 08:06:37 +00:00
Rob Bradford
12f4cd951a tests: Use "--disks" for vhost-user-block testing 
Use newly unified command line options for testing vhost-user based
block support.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-29 08:06:37 +00:00
Rob Bradford
aeeae661fc vmm: Support vhost-user-block via "--disks" 
Add a socket and vhost_user parameter to this option so that the same
configuration option can be used for both virtio-block and
vhost-user-block.  For now it is necessary to specify both vhost_user
and socket parameters as auto activation is not yet implemented. The wce
parameter for supporting "Write Cache Enabling" is also added to the
disk configuration.

The original command line parameter is still supported for now and will
be removed in a future release.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-29 08:06:37 +00:00
Rob Bradford
4888dd2e01 tests: Add CLI <-> API validation test for --net changes 
Check that the CLI generates the JSON data as expected.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-28 12:39:26 +00:00
Rob Bradford
8f096b8be5 tests: Use "--net" parameter for vhost-user-net setup 
Use the newly unified parameter for setting up vhost-user-net based
networking.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-28 12:39:26 +00:00
Rob Bradford
a831aa214c vmm: Support vhost-user-net via "--net" 
Add a socket and vhost_user parameter to this option so that the same
configuration option can be used for both virtio-net and vhost-user-net.
For now it is necessary to specify both vhost_user and socket parameters
as auto activation is not yet implemented. The original command line
parameter is still supported for now.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-28 12:39:26 +00:00
Sebastien Boeuf
9efaff73fa ci: Consolidate virtio-blk integration tests 
In order to reduce the amount of times VMs are being started through
integration tests, this commit consolidates very similar tests related
to virtio-blk into a single one.

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
 2020-01-28 09:26:53 +01:00
Sebastien Boeuf
a33e8342ee ci: Add integration test for virtio-blk multiqueue support 
Just add a new integration test to verify that multiqueue support is
correctly supported and that we can find the right amount of queues in
the guest.

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
 2020-01-28 09:26:53 +01:00
Sebastien Boeuf
08e47ebd4b vmm: Add num_queues and queue_size parameters to virtio-blk 
The number of queues and the size of each queue were not configurable.
In anticipation for adding multiqueue support, this commit introduces
some new parameters to let the user decide about the number of queues
and the queue size.

Note that the default values for each of these parameters are identical
to the default values used for vhost-user-blk, that is 1 for the number
of queues and 128 for the queue size.

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
 2020-01-28 09:26:53 +01:00
Sebastien Boeuf
aa5ae7616e src: Fix map_err losing the inner error 
Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
 2020-01-24 12:42:09 +01:00
Rob Bradford
cafab9e981 main: Add some command line argument conflicts 
Add explicit exclusions with --net-backend from --block-backend (and
vice-versa.) And also with "--kernel" as this is the option for "VM boot" that is never optional.

Ideally we would conflcit the backend arguments against the "vm-group"
however this does not work as it includes some arguments that have a
default value set and thus clap thinks those arguments are always
provided. Conflicting with "--kernel" is thus a reasonable compromise.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
611d13ed1b main: Move logging arguments to their own group 
Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
8aabb9012d main: Add "--vhost-user-blk" to the "vm-config" group 
Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
bdb0ad98c4 tests: Use cloud-hypervisor binary for "--{net,block}-backend" 
Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
45fb9187c5 main: Allow main binary to be used as net or block backend 
Use newly factorised out function to start the block or net
vhost-user-backend.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
7f73eebbdb vhost_user_block: Split launching backend into its own function 
Split the basic launching functionality into its own function in the
newly added vhost_user_block crate.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
9fd187c9fe vhost_user_net: Split launching backend into its own function 
Split the basic launching functionality into its own function in the
newly added vhost_user_net crate.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
b851887738 main: Split current VMM running code into it's own function 
This prepaees the main binary to start running vhost-user backends.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
1dd2451895 vhost_user_block: Refactor vhost_user_block backend code into a new crate 
Extract the majority of the code that provides the vhost-user-block
backend into its own crate and port the binary to use it.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
fb763c4043 vhost_user_net: Refactor vhost_user_net backend code into a new crate 
Extract the majority of the code that provides the vhost-user-net
backend into its own crate and port the binary to use it.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
dd82064aba main: Delay converting command line to VM params 
Generate the VM params until after the logging has been enabled. This
will make it possible to reuse the logging for vhost backends.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
e12179a3e6 bin: vhost_user_net: Rename "--backend" to "--net-backend" 
This will prevent it from conflicting when it is aggregated into the
cloud-hypervisor binary.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Rob Bradford
3ede2dc53a bin: vhost_user_blk: Rename "--backend" to "--block-backend" 
This will prevent it from conflicting when it is aggregated into the
cloud-hypervisor binary.

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
 2020-01-23 10:30:06 +00:00
Sergio Lopez
605fa8307b ci: Add integration test for virtio-blk with 'direct=on' 
Previous commit added 'direct' support for virtio-blk to be able to
open files backing disk images with O_DIRECT, so add an integration
test for this new feature.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-01-21 13:39:45 +00:00
Sergio Lopez
2bd90d9263 ci: Add integration test for virtio-blk with 'readonly=on' 
Previous commit added 'readonly' support for virtio-blk to be able to
configure read-only devices, so add an integration test for this new
feature.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-01-21 13:39:45 +00:00
Sergio Lopez
ed5459f268 ci: Add integration test for vhost_user_blk with 'direct' 
This test is a variant of test_boot_vhost_user_blk(), named
test_boot_vhost_user_blk_direct(), that instances the vhost-user-blk
daemon with 'direct=true', to test this recently introduced feature
for opening files with O_DIRECT.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-01-17 17:28:44 +00:00
Sergio Lopez
e0a8da2f46 vhost_user_blk: Add missing WCE property support 
Add missing WCE (write-cache enable) property support. This not only
an enhancement, but also a fix for a bug.

Right now, when vhost_user_blk uses a qcow2 image, it doesn't write
the QCOW2 metadata until the guest explicitly requests a flush. In
practice, this is equivalent to the write back semantic.

Without WCE, the guest assumes write through for the virtio_blk
device, and doesn't send those flush requests. Adding support for WCE,
and enabling it by default, we ensure the guest does send said
requests.

Supporting "WCE = false" would require updating our qcow2
implementation to ensure that, when required, it honors the write
through semantics by not deferring the updates to QCOW2 metadata.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-01-17 17:28:44 +00:00
Sergio Lopez
c7e9056c1e vhost_user_blk: implement support for direct (O_DIRECT) mode 
Add support for opening the disk images with O_DIRECT. This allows
bypassing the host's file system cache, which is useful to avoid
polluting its cache and for better data integrity.

This mode of operation can be enabled by adding the "direct=<bool>"
parameter to the "backend" argument:

./target/debug/vhost_user_blk --backend image=test.raw,sock=/tmp/vhostblk,direct=true

The "direct" parameter defaults to "false", to preserve the original
behavior.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-01-17 17:28:44 +00:00
Sergio Lopez
a14aee9213 qcow: Use RawFile as backend instead of File 
Use RawFile as backend instead of File. This allows us to abstract
the access to the actual image with a specialized layer, so we have a
place where we can deal with the low-level peculiarities.

Signed-off-by: Sergio Lopez <slp@redhat.com>
 2020-01-17 17:28:44 +00:00