Zhao Liu [Mon, 23 Oct 2023 09:46:28 +0000 (17:46 +0800)]
tests: bios-tables-test: Extend smbios core count2 test to cover general topology
The commit 196ea60a734c3 ("hw/smbios: Fix core count in type4") fixed
the miscalculation of cores per socket.
The original core count2 test (with the topology configured by
"-smp 275") didn't recognize that topology-related but because it just
created a special topology with only one socket and one die by default,
ignoring the effect of more topology levels (between socket and core) on
the cores per socket calculation.
So for the topology in this case, there're the following considerations:
1. cores per socket should be more than 255 to ensure we could cover
the core count2 field.
2. The original bug was that cores per socket was miscalculated, so now
we should include as many topology levels as possible (multiple
sockets or dies, no module since x86 hasn't supported it) to cover
more general topology scenarios, to ensure that the cores per socket
encoded in the core count2 field is correct.
Based on these considerations, select the topology with multiple dies:
-smp 260,dies=2,cores=130,threads=1
Note, here we doesn't configure multiple sockets to avoid the error
("kvm_init_vcpu: kvm_get_vcpu failed (*): Too many open files") if user
uses the default ulimit seeting on his machine.
And the cores per socket calculation for multiple sockets has already
been covered by the core count test case, so that only multiple dies
configuration is enough.
The expected core count2 = cores per socket = cores (130) * dies (2) =
260.
Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-10-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zhao Liu [Mon, 23 Oct 2023 09:46:27 +0000 (17:46 +0800)]
tests: bios-tables-test: Prepare the ACPI table change for smbios type4 core count2 test
Following the guidelines in tests/qtest/bios-tables-test.c, this
is step 1 - 3.
List the ACPI tables that will be changed about the type 4 core count2
test case.
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-9-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-8-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zhao Liu [Mon, 23 Oct 2023 09:46:25 +0000 (17:46 +0800)]
tests: bios-tables-test: Add test for smbios type4 core count
This tests the commit 196ea60a734c3 ("hw/smbios: Fix core count in
type4").
In smbios_build_type_4_table() (hw/smbios/smbios.c), if the number of
cores in the socket is not more than 255, then smbios type4 table
encodes cores per socket into the core count field.
So for the topology in this case, there're the following considerations:
1. cores per socket should be not more than 255 to ensure we could cover
the core count field.
2. The original bug was that cores per socket was miscalculated, so now
we should include as many topology levels as possible (mutiple
sockets & dies, no module since x86 hasn't supported it) to cover
more general topology scenarios, to ensure that the cores per socket
encoded in the core count field is correct.
Based on these considerations, select the topology with multiple sockets
and dies:
-smp 54,sockets=2,dies=3,cores=3,threads=3
The expected core count = cores per socket = cores (3) * dies (3) = 9.
Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-7-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zhao Liu [Mon, 23 Oct 2023 09:46:24 +0000 (17:46 +0800)]
tests: bios-tables-test: Prepare the ACPI table change for smbios type4 core count test
Following the guidelines in tests/qtest/bios-tables-test.c, this
is step 1 - 3.
List the ACPI tables that will be added to test the type 4 core count
field.
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-6-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-5-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zhao Liu [Mon, 23 Oct 2023 09:46:22 +0000 (17:46 +0800)]
tests: bios-tables-test: Add test for smbios type4 count
This tests the commit d79a284a44bb7 ("hw/smbios: Fix smbios_smp_sockets
calculation").
In smbios_get_tables() (hw/smbios/smbios.c), smbios type4 table is built
for each socket, so the count of type4 tables should be equal to the
number of sockets.
Thus for the topology in this case, there're the following considerations:
1. The topology should include multiple sockets to ensure smbios could
create type4 tables for each socket.
2. In addition to sockets, for the more general topology, we should also
configure as many topology levels as possible (multiple dies, no
module since x86 hasn't supported it), to ensure that smbios is able
to exclude the effect of other topology levels to create the type4
tables only for sockets.
3. The original miscalculation bug also misused "smp.cpus", so it's
necessary to configure "cpus" (presented threads for machine) and
"maxcpus" (total threads for machine) as well to make sure that
configuring unpluged CPUs in smp (cpus < maxcpus) does not affect
the correctness of the count of type4 tables.
Based on these considerations, select the topology as the follow:
Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-4-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zhao Liu [Mon, 23 Oct 2023 09:46:21 +0000 (17:46 +0800)]
tests: bios-tables-test: Prepare the ACPI table change for smbios type4 count test
Following the guidelines in tests/qtest/bios-tables-test.c, this
is step 1 - 3.
List the ACPI tables that will be added to test the type 4 count.
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-3-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Test the commit a1d027be95bc3 ("machine: Add helpers to get cores/
threads per socket").
Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20231023094635.1588282-2-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hawkins Jiawei [Wed, 25 Oct 2023 01:08:06 +0000 (09:08 +0800)]
vdpa: Allow VIRTIO_NET_F_RSS in SVQ
Enable SVQ with VIRTIO_NET_F_RSS feature.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <626449eb303207de408126b3dc7c155cd72b028b.1698195059.git.yin31149@gmail.com> Acked-by: Eugenio Pérez <eperezma@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hawkins Jiawei [Wed, 25 Oct 2023 01:08:05 +0000 (09:08 +0800)]
vdpa: Restore receive-side scaling state
This patch reuses vhost_vdpa_net_load_rss() with some
refactorings to restore the receive-side scaling state
at device's startup.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <cf5b78a16ed0318982ceffb195f2227f6aad4ac1.1698195059.git.yin31149@gmail.com> Acked-by: Eugenio Pérez <eperezma@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hawkins Jiawei [Wed, 25 Oct 2023 01:08:04 +0000 (09:08 +0800)]
vdpa: Add SetSteeringEBPF method for NetClientState
At present, to enable the VIRTIO_NET_F_RSS feature, eBPF must
be loaded for the vhost backend.
Given that vhost-vdpa is one of the vhost backend, we need to
implement the SetSteeringEBPF method to support RSS for vhost-vdpa,
even if vhost-vdpa calculates the rss hash in the hardware device
instead of in the kernel by eBPF.
Although this requires QEMU to be compiled with `--enable-bpf`
configuration even if the vdpa device does not use eBPF to
calculate the rss hash, this can avoid adding the specific
conditional statements for vDPA case to enable the VIRTIO_NET_F_RSS
feature, which reduces code maintainbility.
Suggested-by: Eugenio Pérez <eperezma@redhat.com> Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <280e20ddce55b6de60f1552ba0865bffffe909b2.1698195059.git.yin31149@gmail.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hawkins Jiawei [Wed, 25 Oct 2023 01:02:25 +0000 (09:02 +0800)]
vdpa: Allow VIRTIO_NET_F_HASH_REPORT in SVQ
Enable SVQ with VIRTIO_NET_F_HASH_REPORT feature.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <d66b0aee501cdad7954231900c35a11cad1e13db.1698194366.git.yin31149@gmail.com> Acked-by: Eugenio Pérez <eperezma@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hawkins Jiawei [Wed, 25 Oct 2023 01:02:24 +0000 (09:02 +0800)]
vdpa: Restore hash calculation state
This patch introduces vhost_vdpa_net_load_rss() to restore
the hash calculation state at device's startup.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <dbf699acff8c226596136a55a6abe35ebfeac8b0.1698194366.git.yin31149@gmail.com> Acked-by: Eugenio Pérez <eperezma@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This commit adds basic documentation for using virtio-snd.
Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <e7fb941cf7636fdff40cbdcdcd660dec5f15ca3c.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
To perform audio capture we duplicate the TX logic of the previous
commit with the following difference: we receive data from the QEMU
audio backend and write it in the virt queue IO buffers the guest sends
to QEMU. When they are full (i.e. they have `period_bytes` amount of
data) or when recording stops in QEMU's audio backend, the buffer is
returned to the guest by notifying it.
Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <e56a17741a24ccadfbbea19d3c60c9406b795b23.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Handle output IO messages in the transmit (TX) virtqueue.
It allocates a VirtIOSoundPCMBuffer for each IO message and copies the
data buffer to it. When the IO buffer is written to the host's sound
card, the guest will be notified that it has been consumed.
The lifetime of an IO message is:
1. Guest sends IO message to TX virtqueue.
2. QEMU adds it to the appropriate stream's IO buffer queue.
3. Sometime later, the host audio backend calls the output callback,
virtio_snd_pcm_out_cb(), which is defined with an AUD_open_out()
call. The callback gets an available number of bytes the backend can
receive. Then it writes data from the IO buffer queue to the backend.
If at any time a buffer is exhausted, it is returned to the guest as
completed.
4. If the guest releases the stream, its buffer queue is flushed by
attempting to write any leftover data to the audio backend and
releasing all IO messages back to the guest. This is how according to
the spec the guest knows the release was successful.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <b7c6fc458c763d09a4abbcb620ae9b220afa5b8f.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Handle the PCM release control request, which is necessary for flushing
pending sound IO. No IO is handled yet so currently it only replies to
the request.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <ae0afa16461429df1a2f268313d5bfcca27479ec.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Handles the PCM prepare control request. It initializes a PCM stream
when the guests asks for it.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <c6a9c437ef48e45f083fc957dcf7fe18a028e657.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Handle the set parameters control request. It reconfigures a stream
based on a guest's preference if the values are valid and supported.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <d0d19928691f9375bfd83388806786cb7b161301.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Handle the start and stop control messages for a stream_id. This request
does nothing at the moment except for replying to it. Audio playback
or capture will be started/stopped here in follow-up commits.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <9657dbfe3cb4a48ceb033ceb5977dc08669dfefd.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Respond to the VIRTIO_SND_R_PCM_INFO control request with the parameters
of each requested PCM stream.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <5ecea6ba2fb0e3957d7d90bc4dbac521a3d1f678.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Receive guest requests in the control (CTRL) queue of the virtio sound
device and reply with a NOT SUPPORTED error to all control commands.
The receiving handler is virtio_snd_handle_ctrl(). It stores all control
messages in the queue in the device's command queue. Then it calls
virtio_snd_process_cmdq() to handle each message.
The handler is process_cmd() which replies with VIRTIO_SND_S_NOT_SUPP.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <3224aff87e7c4f2777bfe1bbbbca93b72525992c.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch adds a PCI wrapper device for the virtio-sound device.
It is necessary to instantiate a virtio-snd device in a guest.
All sound logic will be added to the virtio-snd device in the following
commits.
To add this device with a guest, you'll need a >=5.13 kernel compiled
with CONFIG_SND_VIRTIO=y, which at the time of writing most distros have
off by default.
Use with following flags in the invocation:
Pulseaudio:
-audio driver=pa,model=virtio
or
-audio driver=pa,model=virtio,server=/run/user/1000/pulse/native
sdl:
-audio driver=sdl,model=virtio
coreaudio (macos/darwin):
-audio driver=coreaudio,model=virtio
etc.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <b223598d59f56ead6a6d8d9bb6801e17489ddaa4.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Add a new VIRTIO device for the virtio sound device id. Functionality
will be added in the following commits.
Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin <Igor.Skalkin@opensynergy.com> Signed-off-by: Anton Yakovlev <Anton.Yakovlev@opensynergy.com> Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <f9678a41fe97b5886c1b04795f1be046509de866.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:43 +0000 (15:42 +0200)]
vhost-user-fs: Implement internal migration
A virtio-fs device's VM state consists of:
- the virtio device (vring) state (VMSTATE_VIRTIO_DEVICE)
- the back-end's (virtiofsd's) internal state
We get/set the latter via the new vhost operations to transfer migratory
state. It is its own dedicated subsection, so that for external
migration, it can be disabled.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-8-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:42 +0000 (15:42 +0200)]
vhost: Add high-level state save/load functions
vhost_save_backend_state() and vhost_load_backend_state() can be used by
vhost front-ends to easily save and load the back-end's state to/from
the migration stream.
Because we do not know the full state size ahead of time,
vhost_save_backend_state() simply reads the data in 1 MB chunks, and
writes each chunk consecutively into the migration stream, prefixed by
its length. EOF is indicated by a 0-length chunk.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-7-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:41 +0000 (15:42 +0200)]
vhost-user: Interface for migration state transfer
Add the interface for transferring the back-end's state during migration
as defined previously in vhost-user.rst.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-6-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:40 +0000 (15:42 +0200)]
vhost-user.rst: Migrating back-end-internal state
For vhost-user devices, qemu can migrate the virtio state, but not the
back-end's internal state. To do so, we need to be able to transfer
this internal state between front-end (qemu) and back-end.
At this point, this new feature is added for the purpose of virtio-fs
migration. Because virtiofsd's internal state will not be too large, we
believe it is best to transfer it as a single binary blob after the
streaming phase.
These are the additions to the protocol:
- New vhost-user protocol feature VHOST_USER_PROTOCOL_F_DEVICE_STATE
- SET_DEVICE_STATE_FD function: Front-end and back-end negotiate a file
descriptor over which to transfer the state.
- CHECK_DEVICE_STATE: After the state has been transferred through the
file descriptor, the front-end invokes this function to verify
success. There is no in-band way (through the file descriptor) to
indicate failure, so we need to check explicitly.
Once the transfer FD has been established via SET_DEVICE_STATE_FD
(which includes establishing the direction of transfer and migration
phase), the sending side writes its data into it, and the reading side
reads it until it sees an EOF. Then, the front-end will check for
success via CHECK_DEVICE_STATE, which on the destination side includes
checking for integrity (i.e. errors during deserialization).
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-5-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:39 +0000 (15:42 +0200)]
vhost-user.rst: Introduce suspended state
In vDPA, GET_VRING_BASE does not stop the queried vring, which is why
SUSPEND was introduced so that the returned index would be stable. In
vhost-user, it does stop the vring, so under the same reasoning, it can
get away without SUSPEND.
Still, we do want to clarify that if the device is completely stopped,
i.e. all vrings are stopped, the back-end should cease to modify any
state relating to the guest. Do this by calling it "suspended".
Suggested-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-4-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:38 +0000 (15:42 +0200)]
vhost-user.rst: Clarify enabling/disabling vrings
Currently, the vhost-user documentation says that rings are to be
initialized in a disabled state when VHOST_USER_F_PROTOCOL_FEATURES is
negotiated. However, by the time of feature negotiation, all rings have
already been initialized, so it is not entirely clear what this means.
At least the vhost-user-backend Rust crate's implementation interpreted
it to mean that whenever this feature is negotiated, all rings are to
put into a disabled state, which means that every SET_FEATURES call
would disable all rings, effectively halting the device. This is
problematic because the VHOST_F_LOG_ALL feature is also set or cleared
this way, which happens during migration. Doing so should not halt the
device.
Other implementations have interpreted this to mean that the device is
to be initialized with all rings disabled, and a subsequent SET_FEATURES
call that does not set VHOST_USER_F_PROTOCOL_FEATURES will enable all of
them. Here, SET_FEATURES will never disable any ring.
This interpretation does not suffer the problem of unintentionally
halting the device whenever features are set or cleared, so it seems
better and more reasonable.
We can clarify this in the documentation by making it explicit that the
enabled/disabled state is tracked even while the vring is stopped.
Every vring is initialized in a disabled state, and SET_FEATURES without
VHOST_USER_F_PROTOCOL_FEATURES simply becomes one way to enable all
vrings.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-3-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Hanna Czenczek [Mon, 16 Oct 2023 13:42:37 +0000 (15:42 +0200)]
vhost-user.rst: Improve [GS]ET_VRING_BASE doc
GET_VRING_BASE does not mention that it stops the respective ring. Fix
that.
Furthermore, it is not fully clear what the "base offset" these
commands' documentation refers to is; an offset could be many things.
Be more precise and verbose about it, especially given that these
commands use different payload structures depending on whether the vring
is split or packed.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20231016134243.68248-2-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
* tag 'pull-sp-20231105' of https://gitlab.com/rth7680/qemu: (21 commits)
target/sparc: Check for invalid cond in gen_compare_reg
target/sparc: Implement UDIV inline
target/sparc: Implement UDIVX and SDIVX inline
target/sparc: Discard cpu_cond at the end of each insn
target/sparc: Record entire jump condition in DisasContext
target/sparc: Merge gen_op_next_insn into only caller
target/sparc: Pass displacement to advance_jump_cond
target/sparc: Merge advance_jump_uncond_{never,always} into advance_jump_cond
target/sparc: Merge gen_branch2 into advance_pc
target/sparc: Do flush_cond in advance_jump_cond
target/sparc: Always copy conditions into a new temporary
target/sparc: Change DisasCompare.c2 to int
target/sparc: Remove DisasCompare.is_bool
target/sparc: Remove CC_OP leftovers
target/sparc: Remove CC_OP_TADDTV, CC_OP_TSUBTV
target/sparc: Remove CC_OP_SUB, CC_OP_SUBX, CC_OP_TSUB
target/sparc: Remove CC_OP_ADD, CC_OP_ADDX, CC_OP_TADD
target/sparc: Remove CC_OP_DIV
target/sparc: Remove CC_OP_LOGIC
target/sparc: Split psr and xcc into components
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
* tag 'migration-20231103-pull-request' of https://gitlab.com/juan.quintela/qemu:
migration: Unlock mutex in error case
docs/migration: Add the dirty limit section
tests/migration: Introduce dirty-limit into guestperf
tests/migration: Introduce dirty-ring-size option into guestperf
tests: Add migration dirty-limit capability test
system/dirtylimit: Drop the reduplicative check
system/dirtylimit: Fix a race situation
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Stefan Hajnoczi [Mon, 6 Nov 2023 00:36:47 +0000 (08:36 +0800)]
Merge tag 'dump-pull-request' of https://gitlab.com/marcandre.lureau/qemu into staging
dump queue
Hi
The "dump" queue, with:
- [PATCH v3 qemu 0/3] Allow dump-guest-memory to output standard kdump format
- [PATCH v2 0/5] dump: Minor fixes & improvements
* tag 'dump-pull-request' of https://gitlab.com/marcandre.lureau/qemu:
dump: Drop redundant check for empty dump
dump: Improve some dump-guest-memory error messages
dump: Recognize "fd:" protocols on Windows hosts
dump: Fix g_array_unref(NULL) in dump-guest-memory
dump: Rename qmp_dump_guest_memory() parameter to match QAPI schema
dump: Add command interface for kdump-raw formats
dump: Allow directly outputting raw kdump format
dump: Pass DumpState to write_ functions
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
# -----BEGIN PGP SIGNATURE-----
#
# iLMEAAEKAB0WIQS4/x2g0v3LLaCcbCxAov/yOSY+3wUCZUSQIgAKCRBAov/yOSY+
# 31aIBADj5FzdUxyFB813SouAiEiyMdI4bN98AunomAk3Kt8PF1XPoP8kPzcjxcMI
# kCW4eoHb12MVs9OclkqFY3VyaxtSD3YSG/h8W9YxaDyU+L/q89RS+J4r6CAZ8ylg
# J4uxs3Lv8nwPEvRb4zITAt8JQllLey1100j/uu4fU0Rx7vUcMA==
# =9RMx
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 03 Nov 2023 14:16:02 HKT
# gpg: using RSA key B8FF1DA0D2FDCB2DA09C6C2C40A2FFF239263EDF
# gpg: Good signature from "Song Gao <m17746591750@163.com>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: B8FF 1DA0 D2FD CB2D A09C 6C2C 40A2 FFF2 3926 3EDF
* tag 'pull-loongarch-20231103' of https://gitlab.com/gaosong/qemu:
linux-user/loongarch64: Add LASX sigcontext save/restore
linux-user/loongarch64: Add LSX sigcontext save/restore
linux-user/loongarch64: Use abi_{ulong,uint} types
linux-user/loongarch64: setup_sigframe() set 'end' context size 0
linux-user/loongarch64: Fix setup_extcontext alloc wrong fpu_context size
linux-user/loongarch64: Use traps to track LSX/LASX usage
target/loongarch: Support 4K page size
target/loongarch: Implement query-cpu-model-expansion
target/loongarch: Allow user enable/disable LSX/LASX features
target/loongarch: Add cpu model 'max'
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
target/sparc: Check for invalid cond in gen_compare_reg
Consolidate the test here; drop the "inverted logic".
Fix MOVr and FMOVR, which were missing the invalid test.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
target/sparc: Discard cpu_cond at the end of each insn
If the insn raises no exceptions, there will be no path in which
cpu_cond is used, and so the computation may be optimized away.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
target/sparc: Record entire jump condition in DisasContext
Use the original condition instead of consuming cpu_cond,
which will now only be live along exception paths.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
target/sparc: Merge gen_op_next_insn into only caller
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
target/sparc: Pass displacement to advance_jump_cond
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
target/sparc: Merge advance_jump_uncond_{never,always} into advance_jump_cond
Handle these via TCG_COND_{ALWAYS,NEVER}.
Allow dc->npc to be variable, using gen_mov_pc_npc.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The function had only one caller. Canonicalize the cpu_cond
test to TCG_COND_NE, the "natural" sense of its value.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
target/sparc: Always copy conditions into a new temporary
This will allow the condition to live across changes to
the global cc variables.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We don't require c2 to be variable, so emphasize that.
We don't currently require c2 to be non-zero, but that will change.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Since we're going to feed cpu_cond to another comparison, we don't
reqire a boolean value -- anything non-zero is sufficient.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
All instructions have been converted to generate
full condition codes explicitly.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
These are all related and implementable with common code.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
These are all related and implementable with common code.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Return both result and overflow from helper_[us]div.
Compute all flags explicitly in gen_op_[us]divcc.
Marginally improve the INT64_MIN special case in helper_sdiv.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Step in removing CC_OP: change the representation of CC_OP_FLAGS.
The 8 bits are distributed between 6 variables, which should make
it easy to keep up to date.
The code within cc_helper.c is quite ugly but is only temporary.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Isolate linux-user from changes to icc representation.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Hyman Huang [Wed, 1 Nov 2023 14:04:08 +0000 (22:04 +0800)]
tests/migration: Introduce dirty-limit into guestperf
Currently, guestperf does not cover the dirty-limit
migration, support this feature.
Note that dirty-limit requires 'dirty-ring-size' set.
To enable dirty-limit, setting x-vcpu-dirty-limit-period
as 500ms and x-vcpu-dirty-limit as 10MB/s:
$ ./tests/migration/guestperf.py \
--dirty-ring-size 4096 \
--dirty-limit --x-vcpu-dirty-limit-period 500 \
--vcpu-dirty-limit 10 --output output.json \
To run the entire standardized set of dirty-limit-enabled
comparisons, with unix migration:
$ ./tests/migration/guestperf-batch.py \
--dirty-ring-size 4096 \
--dst-host localhost --transport unix \
--filter compr-dirty-limit* --output outputdir
Hyman Huang [Wed, 1 Nov 2023 14:04:06 +0000 (22:04 +0800)]
tests: Add migration dirty-limit capability test
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The test case does the following things:
1. start src, dst vm and enable dirty-limit capability
2. start migrate and set cancel it to check if dirty limit
stop working.
3. restart dst vm
4. start migrate and enable dirty-limit capability
5. check if migration satisfy the convergence condition
during pre-switchover phase.
Note that this test case involves many passes, so it runs
in slow mode only.
Signed-off-by: Hyman Huang <yong.huang@smartx.com> Acked-by: Peter Xu <peterx@redhat.com> Reviewed-by: Fabiano Rosas <farosas@suse.de> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <e55a302df9da7dbc00ad825f47f57c1a756d303e.1698847223.git.yong.huang@smartx.com>
Hyman Huang [Wed, 1 Nov 2023 14:04:05 +0000 (22:04 +0800)]
system/dirtylimit: Drop the reduplicative check
Checking if dirty limit is in service is done by the
dirtylimit_query_all function, drop the reduplicative
check in the qmp_query_vcpu_dirty_limit function.
Signed-off-by: Hyman Huang <yong.huang@smartx.com> Reviewed-by: Fabiano Rosas <farosas@suse.de> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <df9c3514933ff6750ef88068af18d3054bedf746.1698847223.git.yong.huang@smartx.com>
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231101030816.2353416-7-gaosong@loongson.cn>
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231101030816.2353416-6-gaosong@loongson.cn>
Song Gao [Wed, 1 Nov 2023 03:08:14 +0000 (11:08 +0800)]
linux-user/loongarch64: Use abi_{ulong,uint} types
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231101030816.2353416-5-gaosong@loongson.cn>
The kernel setup_sigcontext() set end context size 0.
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231101030816.2353416-4-gaosong@loongson.cn>
The alloc size is sizeof(struct target_fpu_context).
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231101030816.2353416-3-gaosong@loongson.cn>
linux-user/loongarch64: Use traps to track LSX/LASX usage
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231101030816.2353416-2-gaosong@loongson.cn> Signed-off-by: Song Gao <gaosong@loongson.cn>
3. start with '-cpu la132,lasx=off'
qemu-system-loongarch64: can't apply global la132-loongarch-cpu.lasx=off: Property 'la132-loongarch-cpu.lasx' not found
4. start with '-cpu max,lasx=off' or start with '-cpu la464,lasx=off' query cpu model la132
(QEMU) query-cpu-model-expansion type=static model={"name":"la132"}
{"return": {"model": {"name": "la132"}}}
Acked-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231020084925.3457084-4-gaosong@loongson.cn>
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231020084925.3457084-3-gaosong@loongson.cn>
Song Gao [Fri, 20 Oct 2023 08:49:23 +0000 (16:49 +0800)]
target/loongarch: Add cpu model 'max'
We use cpu la464 for the 'max' cpu.
Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231020084925.3457084-2-gaosong@loongson.cn>
Stefan Hajnoczi [Fri, 3 Nov 2023 02:04:12 +0000 (10:04 +0800)]
Merge tag 'pull-target-arm-20231102' of https://git.linaro.org/people/pmaydell/qemu-arm into staging
target-arm queue:
* linux-user/elfload: Add missing arm64 hwcap values
* stellaris-gamepad: Convert to qdev
* docs/specs: Convert various txt docs to rST
* MAINTAINERS: Make sure that gicv3_internal.h is covered, too
* hw/arm/pxa2xx_gpio: Pass CPU using QOM link property
* hw/watchdog/wdt_imx2: Trace MMIO access and timer activity
* hw/misc/imx7_snvs: Trace MMIO access
* hw/misc/imx6_ccm: Convert DPRINTF to trace events
* hw/i2c/pm_smbus: Convert DPRINTF to trace events
* target/arm: Enable FEAT_MOPS insns in user-mode emulation
* linux-user: Report AArch64 hwcap2 fields above bit 31
* target/arm: Make FEAT_MOPS SET* insns handle Xs == XZR correctly
* target/arm: Fix SVE STR increment
* hw/char/stm32f2xx_usart: implement TX interrupts
* target/arm: Correctly propagate stage 1 BTI guarded bit in a two-stage walk
* xlnx-versal-virt: Add AMD/Xilinx TRNG device
* tag 'pull-target-arm-20231102' of https://git.linaro.org/people/pmaydell/qemu-arm: (33 commits)
tests/qtest: Introduce tests for AMD/Xilinx Versal TRNG device
hw/arm: xlnx-versal-virt: Add AMD/Xilinx TRNG device
hw/misc: Introduce AMD/Xilix Versal TRNG device
target/arm: Correctly propagate stage 1 BTI guarded bit in a two-stage walk
hw/char/stm32f2xx_usart: Add more definitions for CR1 register
hw/char/stm32f2xx_usart: Update IRQ when DR is written
hw/char/stm32f2xx_usart: Extract common IRQ update code to update_irq()
target/arm: Fix SVE STR increment
target/arm: Make FEAT_MOPS SET* insns handle Xs == XZR correctly
linux-user: Report AArch64 hwcap2 fields above bit 31
target/arm: Enable FEAT_MOPS insns in user-mode emulation
hw/i2c/pm_smbus: Convert DPRINTF to trace events
hw/misc/imx6_ccm: Convert DPRINTF to trace events
hw/misc/imx7_snvs: Trace MMIO access
hw/watchdog/wdt_imx2: Trace timer activity
hw/watchdog/wdt_imx2: Trace MMIO access
hw/arm/pxa2xx_gpio: Pass CPU using QOM link property
MAINTAINERS: Make sure that gicv3_internal.h is covered, too
docs/specs/vmgenid: Convert to rST
docs/specs/vmcoreinfo: Convert to rST
...
Conflicts:
hw/input/stellaris_input.c
The qdev conversion in this pull request ("stellaris-gamepad: Convert
to qdev") eliminates the vmstate_register() call that was converted to
vmstate_register_any() in the conflicting migration pull request.
vmstate_register_any() is no longer necessary now that this device has
been converted to qdev, so take this pull request's version of
stellaris_gamepad.c over the previous pull request's
stellaris_input.c (the file was renamed).
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
- migration reboot mode (steve)
* I disabled the test because our CI don't like programs using so
much shared memory. Searching for a fix.
- test for postcopy recover (fabiano)
- MigrateAddress QAPI (het)
- better return path error handling (peter)
- traces for downtime (peter)
- vmstate_register() check for duplicates (juan)
thomas find better solutions for s390x and ipmi.
now also works on s390x
* tag 'migration-20231102-pull-request' of https://gitlab.com/juan.quintela/qemu: (40 commits)
migration: modify test_multifd_tcp_none() to use new QAPI syntax.
migration: Implement MigrateChannelList to hmp migration flow.
migration: Implement MigrateChannelList to qmp migration flow.
migration: modify migration_channels_and_uri_compatible() for new QAPI syntax
migration: New migrate and migrate-incoming argument 'channels'
migration: Convert the file backend to the new QAPI syntax
migration: convert exec backend to accept MigrateAddress.
migration: convert rdma backend to accept MigrateAddress
migration: convert socket backend to accept MigrateAddress
migration: convert migration 'uri' into 'MigrateAddress'
migration: New QAPI type 'MigrateAddress'
migration: Change ram_dirty_bitmap_reload() retval to bool
tests/migration-test: Add a test for postcopy hangs during RECOVER
migration: Allow network to fail even during recovery
migration: Refactor error handling in source return path
tests/qtest: migration: add reboot mode test
cpr: reboot mode
cpr: relax vhost migration blockers
cpr: relax blockdev migration blockers
migration: per-mode blockers
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Tong Ho [Tue, 31 Oct 2023 18:46:11 +0000 (11:46 -0700)]
tests/qtest: Introduce tests for AMD/Xilinx Versal TRNG device
Signed-off-by: Tong Ho <tong.ho@amd.com>
Message-id: 20231031184611.3029156-4-tong.ho@amd.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Connect the support for Versal True Random Number Generator
(TRNG) device.
Warning: unlike the TRNG component in a real device from the
Versal device familiy, the connected TRNG model is not of
cryptographic grade and is not intended for use cases when
cryptograpically strong TRNG is needed.
Signed-off-by: Tong Ho <tong.ho@amd.com> Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20231031184611.3029156-3-tong.ho@amd.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Tong Ho [Tue, 31 Oct 2023 18:46:09 +0000 (11:46 -0700)]
hw/misc: Introduce AMD/Xilix Versal TRNG device
This adds a non-cryptographic grade implementation of the
model for the True Random Number Generator (TRNG) component
in AMD/Xilinx Versal device family.
This implements all 3 modes defined by the actual hardware
specs, all of which selectable by guest software at will
at anytime:
1) PRNG mode, in which the generated sequence is required to
be reproducible after reseeded by the same 384-bit value
as supplied by guest software.
2) Test mode, in which the generated sequence is required to
be reproducible ater reseeded by the same 128-bit test
seed supplied by guest software.
3) TRNG mode, in which non-reproducible sequence is generated
based on periodic reseed by a suitable entropy source.
This model is only intended for non-real world testing of
guest software, where cryptographically strong PRNG or TRNG
is not needed.
This model supports versions 1 & 2 of the device, with
default to be version 2; the 'hw-version' uint32 property
can be set to 0x0100 to override the default.
Other implemented properties:
- 'forced-prng', uint64
When set to non-zero, mode 3's entropy source is implemented
as a deterministic sequence based on the given value and other
deterministic parameters.
This option allows the emulation to test guest software using
mode 3 and to reproduce data-dependent defects.
- 'fips-fault-events', uint32, bit-mask
bit 3: Triggers the SP800-90B entropy health test fault irq
bit 1: Triggers the FIPS 140-2 continuous test fault irq
Signed-off-by: Tong Ho <tong.ho@amd.com>
Message-id: 20231031184611.3029156-2-tong.ho@amd.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
dump_init() first computes the size of the dump, taking the filter
area into account, and fails if its zero. It then looks for memory in
the filter area, and fails if there is none.
This is redundant: if the size of the dump is zero, there is no
memory, and vice versa. Delete this check.
Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20231031104531.3169721-6-armbru@redhat.com>
dump: Improve some dump-guest-memory error messages
Zero @length is rejected with "Invalid parameter 'length'". Improve
to "parameter 'length' expects a non-zero length".
qemu_open_old() is a wrapper around qemu_open_internal() that throws
away error information. Switch to the wrapper that doesn't:
qemu_create(). Example improvement:
(qemu) dump-guest-memory /dev/fdset/x 0 1
Error: Could not open '/dev/fdset/x': Invalid argument
becomes
Error: Could not parse fdset /dev/fdset/x
@protocol values not starting with "fd:" or "file:" are rejected with
"Invalid parameter 'protocol'". Improve to "parameter 'protocol' must
start with 'file:' or 'fd:'".
While there, make the conditional checking @protocol a little more
obvious.
Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20231031104531.3169721-5-armbru@redhat.com>
A few QMP command can work with named file descriptors.
The only way to create a named file descriptor used to be QMP command
getfd, which only works on POSIX hosts. Thus, named file descriptors
were actually usable only there.
They became usable on Windows hosts when we added QMP command
get-win32-socket (commit 4cda177c601 "qmp: add 'get-win32-socket'").
Except in dump-guest-memory, because qmp_dump_guest_memory() compiles
its named file descriptor code only #if !defined(WIN32).
Compile it unconditionally, like we do for the other commands
supporting them.
Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20231031104531.3169721-4-armbru@redhat.com>
dump: Fix g_array_unref(NULL) in dump-guest-memory
When dump_init()'s check for non-zero @length fails, dump_cleanup()
passes null s->string_table_buf to g_array_unref(), which spews "GLib:
g_array_unref: assertion 'array' failed" to stderr.
Guard the g_array_unref().
Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20231031104531.3169721-3-armbru@redhat.com>
dump: Rename qmp_dump_guest_memory() parameter to match QAPI schema
The name of the second parameter differs between QAPI schema and C
implementation: it's @protocol in the former and @file in the latter.
Potentially confusing. Change the C implementation to match the QAPI
schema.
Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20231031104531.3169721-2-armbru@redhat.com>
Stephen Brennan [Mon, 18 Sep 2023 23:32:33 +0000 (16:32 -0700)]
dump: Add command interface for kdump-raw formats
The QMP dump API represents the dump format as an enumeration. Add three
new enumerators, one for each supported kdump compression, each named
"kdump-raw-*".
For the HMP command line, rather than adding a new flag corresponding to
each format, it seems more human-friendly to add a single flag "-R" to
switch the kdump formats to "raw" mode. The choice of "-R" also
correlates nicely to the "makedumpfile -R" option, which would serve to
reassemble a flattened vmcore.
Signed-off-by: Stephen Brennan <stephen.s.brennan@oracle.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
[ Marc-André: replace loff_t with off_t, indent fixes ] Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230918233233.1431858-4-stephen.s.brennan@oracle.com>
Stephen Brennan [Mon, 18 Sep 2023 23:32:32 +0000 (16:32 -0700)]
dump: Allow directly outputting raw kdump format
The flattened format (currently output by QEMU) is used by makedumpfile
only when it is outputting a vmcore to a file which is not seekable. The
flattened format functions essentially as a set of instructions of the
form "seek to the given offset, then write the given bytes out".
The flattened format can be reconstructed using makedumpfile -R, or
makedumpfile-R.pl, but it is a slow process because it requires copying
the entire vmcore. The flattened format can also be directly read by
crash, but still, it requires a lengthy reassembly phase.
To sum up, the flattened format is not an ideal one: it should only be
used on files which are actually not seekable. This is the exact
strategy which makedumpfile uses, as seen in the implementation of
"write_buffer()" in makedumpfile [1]. However, QEMU has always used the
flattened format. For compatibility it is best not to change the default
output format without warning. So, add a flag to DumpState which changes
the output to use the normal (i.e. raw) format. This flag will be added
to the QMP and HMP commands in the next change.
Stephen Brennan [Mon, 18 Sep 2023 23:32:31 +0000 (16:32 -0700)]
dump: Pass DumpState to write_ functions
For the next patch, we need a reference to DumpState when writing data.
Signed-off-by: Stephen Brennan <stephen.s.brennan@oracle.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230918233233.1431858-2-stephen.s.brennan@oracle.com>
Peter Maydell [Tue, 31 Oct 2023 17:37:23 +0000 (17:37 +0000)]
target/arm: Correctly propagate stage 1 BTI guarded bit in a two-stage walk
In a two-stage translation, the result of the BTI guarded bit should
be the guarded bit from the first stage of translation, as there is
no BTI guard information in stage two. Our code tried to do this,
but got it wrong, because we currently have two fields where the GP
bit information might live (ARMCacheAttrs::guarded and
CPUTLBEntryFull::extra::arm::guarded), and we were storing the GP bit
in the latter during the stage 1 walk but trying to copy the former
in combine_cacheattrs().
Remove the duplicated storage, and always use the field in
CPUTLBEntryFull; correctly propagate the stage 1 value to the output
in get_phys_addr_twostage().
Note for stable backports: in v8.0 and earlier the field is named
result->f.guarded, not result->f.extra.arm.guarded.
Cc: qemu-stable@nongnu.org
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1950 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20231031173723.26582-1-peter.maydell@linaro.org
Peter Maydell [Mon, 30 Oct 2023 17:40:00 +0000 (17:40 +0000)]
target/arm: Make FEAT_MOPS SET* insns handle Xs == XZR correctly
Most of the registers used by the FEAT_MOPS instructions cannot use
31 as a register field value; this is CONSTRAINED UNPREDICTABLE to
NOP or UNDEF (we UNDEF). However, it is permitted for the "source
value" register for the memset insns SET* to be 31, which (as usual
for most data-processing insns) means it should be the zero register
XZR. We forgot to handle this case, with the effect that trying to
set memory to zero with a "SET* Xd, Xn, XZR" sets the memory to
the value that happens to be in the low byte of SP.
Handle XZR when getting the SET* data value from the register file.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20231030174000.3792225-4-peter.maydell@linaro.org
Peter Maydell [Mon, 30 Oct 2023 17:39:59 +0000 (17:39 +0000)]
linux-user: Report AArch64 hwcap2 fields above bit 31
The AArch64 ELF hwcap2 field is 64 bits, but our get_elf_hwcap2()
works with uint32_t, so it accidentally fails to report any hwcaps
over bit 31. Use uint64_t here.
The Arm hwcap2 is only 32 bits (because the ELF format makes these
fields be the size of "long" in the ABI), but since it shares the
prototype declaration for get_elf_hwcap2() it is easier to also
expand it to 64 bits.
The only hwcap fields we implement already that are affected by this
are the HBC and MOPS ones, neither of which were implemented in a
previous release, so this doesn't need backporting to older stable
branches.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20231030174000.3792225-3-peter.maydell@linaro.org
Peter Maydell [Mon, 30 Oct 2023 17:39:58 +0000 (17:39 +0000)]
target/arm: Enable FEAT_MOPS insns in user-mode emulation
In user-mode emulation, we need to set the SCTLR_EL1.MSCEn
bit to avoid all the FEAT_MOPS insns UNDEFing.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20231030174000.3792225-2-peter.maydell@linaro.org
Bernhard Beschow [Sat, 28 Oct 2023 12:24:13 +0000 (14:24 +0200)]
hw/misc/imx6_ccm: Convert DPRINTF to trace events
Signed-off-by: Bernhard Beschow <shentey@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20231028122415.14869-5-shentey@gmail.com
[PMM: Add "Hz" unit indicator to frequency traces] Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Bernhard Beschow [Sat, 28 Oct 2023 12:24:12 +0000 (14:24 +0200)]
hw/misc/imx7_snvs: Trace MMIO access
Signed-off-by: Bernhard Beschow <shentey@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20231028122415.14869-4-shentey@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Bernhard Beschow [Sat, 28 Oct 2023 12:24:11 +0000 (14:24 +0200)]
hw/watchdog/wdt_imx2: Trace timer activity
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Message-id: 20231028122415.14869-3-shentey@gmail.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Bernhard Beschow [Sat, 28 Oct 2023 12:24:10 +0000 (14:24 +0200)]
hw/watchdog/wdt_imx2: Trace MMIO access
Signed-off-by: Bernhard Beschow <shentey@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20231028122415.14869-2-shentey@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>