- Use vpython to call crovm tooling (This provides a predictable python
environment and allow us to use argh).
- Add build_context() to provide a simple API that readies source and
container.
- Nest preparation steps to reduce noise on builder page.
BUG=b:233913643
TEST=recipes.py test run
Change-Id: I6800e55f7311b32c6ef61918d13795af1fb3b588
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3674217
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Anton Romanov <romanton@google.com>
The builders do not have depot_tools installed globally, but
the recipes checkout contains them, so look up the proper
paths to the tools.
BUG=b:233913797
TEST=./recipes.py run build_chromeos
Change-Id: I5d50871fa0ccee769704b339a20d3406d23e0510
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3674216
Reviewed-by: Anton Romanov <romanton@google.com>
Tested-by: Dennis Kempin <denniskempin@google.com>
As a part of full guest suspension/resumption support extend crosvm
suspend command about generating sleepbtn press. This allows to emulate
PM1 and inject the vSCI to the guest. When the guest kernel is built
with ACPI button support enabled, it allows to suspend guest with use of
some guest daemon listening on ACPI events e.g. acpid
(acpid detects the ACPI sleepbtn event and triggers the guest
suspension).
During suspension process (VmRequest::Suspend), before VmResponse::Ok is
advertise, there is need to wait for actual guest suspension.
This is required, e.g. when crosvm suspend command is triggered
automatically by vm_concierge during host suspension process and
prevents suspending host before the guest is suspended.
Above is achieved by taking advantage of KVM_SYSTEM_EVENT_S2IDLE
request sent by hypervisor on behalf of a non-privileged VM. It is used
to wake up blocked thread, which is awaiting non-privileged guest
suspension to finish.
The enhanced suspend process introduced by this patch is triggered
only when the --s2idle flag is passed.
BUG=b:194391015
TEST=Suspend the guest by issuing "crosvm suspend
/run/vm/vm.<hash>/crosvm.sock" and make sure that the guest actually
triggers suspend process and finalizes it.
Change-Id: I38cf5a786275fc4afa4ba7642ef0be779f8f1548
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3602869
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Grzegorz Jaszczyk <jaszczyk@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
As a part of full guest suspension/resumption support extend crosvm
resume about generating powerbtn press during resume. This allows to
emulate PM1 and inject the vSCI to the guest. When the guest kernel
is built with ACPI button support enabled it allows to wakeup guest from
suspend (ACPI powerbtn fixed event is registered for the guest kernel
and since SCI is wakeup armed it allows to wakeup the guest).
Complementary patch for suspend: CL:3602869
The enhanced resume process introduced by this patch is triggered only
when the --s2idle flag is passed.
BUG=b:194391015
TEST=Suspend the guest by "echo mem > /sys/power/state" and wake it up
by issuing "crosvm resume /run/vm/vm.<hash>/crosvm.sock".
Change-Id: Ifa6626801b495cf96508531783c556804d5a51c6
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3602868
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Grzegorz Jaszczyk <jaszczyk@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
This just creates unnecessary noise. Just create a single dry run
merge commit each time.
BUG=b:227224259
TEST=None
Change-Id: I54385fd80f01771bb83e02f2b48e604ad2d68d1e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3673021
Reviewed-by: Anton Romanov <romanton@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The set_nonblocking() implementation of the VFIO Listener was marked as
unimplemented and caused a crash. However, given that accept() on a VFIO
listener never blocks, it is just as safe to mark the operation as
successful, since the behavior of accept() won't be impacted and it will
behave properly both in blocking and non-blocking modes.
Removing this panic will allow us to safely use the Listener's interface
set_nonblocking() method in generic code.
BUG=b:229554679
TEST=vhost-user console device works.
TEST=vvu console device works.
Change-Id: Ibbde7bcd048505fc3a84eb4881a74285bae6bf54
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3591111
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
vmm_vhost comes with its own set of listeners that implement the same
interface. Switch to use its socket listener as that will allow us to
further factorize code.
BUG=b:229554679
TEST=vhost-user console device works.
TEST=vvu console device works.
Change-Id: I19d0c9b95a0cca57eefd6c72b875e8a166d06b89
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3591110
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
The SlaveReqHandler should be the sole owner of a VhostUserBackend since
it calls its mutable methods, so we shouldn't need an Arc here. If
sharing is necessary users can use their own locking mechanism.
Single ownership will help to retrieve the backend after a client
disconnects, to reuse it for another connection if needed.
BUG=b:229554679
BUG=b:216407443
TEST=cargo test -p vmm_vhost
TEST=vhost-user console device works.
TEST=cargo test -p devices
Change-Id: I2358c807ac3ddb1ee4b29d97df0ade5a3e30a85a
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3591108
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Richard Zhang <rizhang@google.com>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Fix a few link-related (either dead links or typos) in the base crate.
BUG=None
TEST=cargo doc
Change-Id: I520dd2fbfc16de72fcf19ecdd90943a7a963103c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3671057
Reviewed-by: Junichi Uekawa <uekawa@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Otherwise we may exit thinking we are trying to run an interactive
command in a non-interactive terminal.
BUG=None
TEST=./tools/dev_container --stop </dev/null
Change-Id: Ie34cf662b82090b100533306572552c5a550211e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3671132
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
This re-creates the Kokoro equivalent crosvm/ci/build-chromeos.sh.
It is a very inefficent builder that does a fresh checkout of
chromeos for every build.
We may want to consider using some of the chromiumos recipes
for caching the repo or sdk between builds.
BUG=b:233230344
TEST=./recipes.py test run
./recipes.py run build_chromeos
Change-Id: I5d545bae4817cae5b79c5d65901081d9b1391466
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664842
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Adds a builder for both pre- and post-submit, executes the newly
added health_check recipe.
BUG=b:233913455
TEST=lucicfg validate main.star
Change-Id: I69e9738a02bb298d3b2556058d2df1ee9979bda4
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3669815
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
glibc 2.33 uses the fstatat64 syscall to implement statx(), which is
called by the USB emulation code when a device is added. Allow it along
with fstat64 to fix a crash on USB device insertion.
aarch64 and x86_64 use the newfstatat syscall instead, so they do not
need to be adjusted.
BUG=chromium:1328120
TEST=Attach yubikey to Crostini on kevin
Change-Id: I6a592e25126a5baebdbc8839ba11b971950f4575
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3671085
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Dmitry Torokhov <dtor@chromium.org>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Submit after crrev.com/c/3664859 is merged.
BUG=b:215043150, b:183217901
TEST=HEVC playback works in ExoPlayer on guybrush w/ full set of changes
Change-Id: If775435040a77f4620a55d6838353836c40291c4
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664289
Commit-Queue: Jeffrey Kardatzke <jkardatzke@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
The ioctl allow list for the emulated xhci USB controller device was
copied from the 32-bit ARM version, but the structure sizes are
incorrect for 64-bit aarch64. Update the constants for ioctls that take
a structure with a different size:
- USBDEVFS_REAPURBNDELAY: struct usbdevfs_urb
- USBDEVFS_SUBMITURB: struct usbdevfs_urb
- USBDEVFS_CONTROL: struct usbdevfs_ctrltransfer
BUG=chromium:1328120
TEST=Share USB device with Crostini on trogdor64
Change-Id: I45d69021257af55cfeaf72dd33a500a358adc5ee
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3670055
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
This is a reland of commit 129840471d
Original change's description:
> devices: debugcon: add bochs-style debugcon device
>
> This allows for creating a simple write-only debug connection that uses
> a single IO port. This method is supported by both bochs and qemu[1].
>
> The main motivation here is to help in improving crosvm's ability to
> boot OVMF, which makes use of this simple debug device on port 0x402.
> For that one would use `--serial
> type=stdout,hardware=debugcon,debugcon_port=1026` or something similar.
>
> Note that serde_keyvalue doesn't support parsing hex values, hence the
> need to use `1026`, but I intend to address that as well to allow the
> more natural `debugcon_port=0x402` phrasing.
>
> [1]: https://github.com/qemu/qemu/blob/master/hw/char/debugcon.c
>
> BUG=b:233610263
>
> Change-Id: I73238e5c35925668d133bfaa542fcbcf1e28765f
> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3661255
> Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
> Commit-Queue: Mike Gerow <gerow@google.com>
> Tested-by: kokoro <noreply+kokoro@google.com>
Bug: b:233610263
Change-Id: I2f2125da95e8907f4f0c16930952e9b765d33e6c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3671087
Tested-by: kokoro <noreply+kokoro@google.com>
Auto-Submit: Mike Gerow <gerow@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
This reverts commit 129840471d.
Reason for revert: This conflicts with https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3600167 breaking the build
Original change's description:
> devices: debugcon: add bochs-style debugcon device
>
> This allows for creating a simple write-only debug connection that uses
> a single IO port. This method is supported by both bochs and qemu[1].
>
> The main motivation here is to help in improving crosvm's ability to
> boot OVMF, which makes use of this simple debug device on port 0x402.
> For that one would use `--serial
> type=stdout,hardware=debugcon,debugcon_port=1026` or something similar.
>
> Note that serde_keyvalue doesn't support parsing hex values, hence the
> need to use `1026`, but I intend to address that as well to allow the
> more natural `debugcon_port=0x402` phrasing.
>
> [1]: https://github.com/qemu/qemu/blob/master/hw/char/debugcon.c
>
> BUG=b:233610263
>
> Change-Id: I73238e5c35925668d133bfaa542fcbcf1e28765f
> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3661255
> Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
> Commit-Queue: Mike Gerow <gerow@google.com>
> Tested-by: kokoro <noreply+kokoro@google.com>
Bug: b:233610263
Change-Id: Ia70540c3041611f59968cb7929b238eecfe2530d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3671082
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Auto-Submit: Mike Gerow <gerow@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Dennis Kempin <denniskempin@google.com>
This allows us to run multiple steps with the same intermediate
build data.
To make sure we start with a clean slate, any existing containers are
deleted at the beginning of each recipe.
BUG=b:233913455
TEST=./recipes.py test run
Change-Id: Ieda684d40bf08937183cba34242360c025724af8
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3668813
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
This allows for creating a simple write-only debug connection that uses
a single IO port. This method is supported by both bochs and qemu[1].
The main motivation here is to help in improving crosvm's ability to
boot OVMF, which makes use of this simple debug device on port 0x402.
For that one would use `--serial
type=stdout,hardware=debugcon,debugcon_port=1026` or something similar.
Note that serde_keyvalue doesn't support parsing hex values, hence the
need to use `1026`, but I intend to address that as well to allow the
more natural `debugcon_port=0x402` phrasing.
[1]: https://github.com/qemu/qemu/blob/master/hw/char/debugcon.c
BUG=b:233610263
Change-Id: I73238e5c35925668d133bfaa542fcbcf1e28765f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3661255
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Mike Gerow <gerow@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Using the new `SerialInput` trait, we can create an async input source
and get rid of the polling thread.
BUG=b:228912920
TEST=vhost-user console device is working (with input) on Linux.
Change-Id: I1be9ae499f9c782b4aacba44e1eca524f1b701c1
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3600168
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Introduce a new dedicated trait for types that can be used as serial
input. This trait adds the requirement that the type must implement
`AsRawDescriptor`, which will allow us to use the input as a poll source
instead of having to spin a dedicated thread to call `read()` on it.
BUG=b:228912920
TEST=cargo build
Change-Id: Ib07f7b4e63545fa0f27940553ad6124796f5b3cc
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3600167
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
The default is still to run all of them, but this will allow us to run
checks in separate luci steps to provide an easier to read result.
BUG=b:233913455
TEST=./tools/health-check python fmt
Change-Id: Iddc803f8f423db36ece53a13acfe564560b789a6
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3668812
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
We run a neverending command in the container to keep it alive in the
background.
By default, this would be bash, which does the job when --tty is
specified. But we no longer do that all the time.
Instead we can use sleep, which does not rely on such tricks.
BUG=b:233913455
TEST=./tools/dev_container true
Change-Id: I2d04aabbef6b670daf6b94fdc2c65c7bd759c61b
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3669682
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Without this cfg check, execution can reach an unreachable!() statement
in the arg handling code on non-gdb builds.
BUG=None
TEST=cargo build # without --features=gdb
Change-Id: I8bec7abbb39e658d5c8f45946fd4fc28f0ae97a2
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3668741
Reviewed-by: Anton Romanov <romanton@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Everything that takes a PCI BAR index should use the appropriate
PciBarIndex type rather than u8 (or usize) for better type safety and
descriptiveness.
BUG=b:232838930
TEST=tools/presubmit
Change-Id: I0a918a6da6327c963fb297ba765563774a7e5e49
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664282
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Anton Romanov <romanton@google.com>
The read_bar() and write_bar() functions had an open-coded equivalent of
BAR lookup to find out if the access was intended for the settings BAR
or one of the device-specific extra BARs. Instead, we can reuse the same
BAR lookup code to find the index of the relevant BAR and compare it
against the settings_bar field.
BUG=b:232838930
TEST=tools/presubmit
Change-Id: I2e5f3ccff47d31718e835ccb745db75dbda2e4ec
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664281
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Anton Romanov <romanton@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Define constants for the end of the ranges that exist within the
settings BAR so we can use the range syntax in the match instead of
manually checking against the start and end.
This uses inclusive range (..=) syntax, since exclusive range pattern
syntax (..) is experimental. Therefore, each range has a `_LAST`
constant that is equal to the address of the last valid byte in that
range (first + size - 1).
Rewriting the matches in this way allows us to clean up the two clippy
ignores and makes the code a bit easier to read.
BUG=b:232838930
TEST=tools/clippy
TEST=Boot x86-64 Crostini
Change-Id: Ibb8b5ea055e598c19504b7b1b6753706ad458875
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664280
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Anton Romanov <romanton@google.com>
We need to open vvu device before we go to jail since we don't have
access to sysfs in jail.
BUG=b:233690640
TEST=manual - fs is usable from a sibling guest when using VVU.
Change-Id: I0340d105b667cce2ba88d7557e64771fa2bedf9e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3665907
Auto-Submit: Lepton Wu <lepton@chromium.org>
Commit-Queue: Lepton Wu <lepton@chromium.org>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
This was fixed in crrev.com/c/3659824, but a subsequent commit somehow
reverted that line.
BUG=None
TEST=`cargo build` does not display any warning.
Change-Id: Id50ee4f615117ada8c27552e073f7fb04d8d009e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664166
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Junichi Uekawa <uekawa@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
We want to eventually reuse these functions in the regular virtio device
code, which uses a different kind of interrupt, so make that parameter
generic.
BUG=b:228912920
TEST=cargo build
Change-Id: I1139ce6626ac7b952b7769db6e66d677607d3307
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3600166
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
The error path reproduces the code outside of the match arm, so it can
be removed.
BUG=b:228912920
TEST=cargo build
Change-Id: Ibd485a52441be03501041bc4136bab9293c7066b
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3600165
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Used tx buffers are returned to the driver with 0 bytes written by the
device. The current code makes it look like we return the number of
received bytes, even though this is not the case since
process_transmit_request always returns 0 for some reason.
Make this more clear by explicitly setting 0 bytes used, and making
process_transmit_request return nothing.
BUG=b:228912920
TEST=cargo build
Change-Id: I9094f7a43847d7e29d5390f274d6650ca5ec7c4d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3600164
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
This reverts commit 55a5c8d8f9.
In order to accommodate the removal of the
UnixSeqPacket+CloseNotifier change, I updated the controller.rs
to reference platform-specific internal implementations.
On Windows, CloseNotifier is used to detect closed Tubes.
On Linux, we rely on PollContext returning because the socket fd
is hung up.
Some minor adjustmets to the code were made just to allow as
litte duplication as possible.
In the end, very little logic has changed from the original CL,
it's just moved around.
TL;DR:
This fixes the downstream regression by removing its dependency
on the breaking changes to base.
BUG=b:232316549
FIXED=b:232316549
TEST=Crosvm tests
Change-Id: I946d5096f7a312538c3c694950697fab1be7f0ca
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3661257
Commit-Queue: Michael Hoyle <mikehoyle@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
On Unix, instead of getting random data from `/dev/urandom`, it will get
it from the `rand` platform agnostic crate instead.
OsRng.fill_bytes on unix will make a syscall to getrandom(2) if
available, otherwise it will read from `dev/urandom` after a succesful
poll to `dev/random`. Regardless of which way a random data is
retrieved, if the entropy pool is not intialized, `fill_bytes` will
block until it is intialized. This shouldn't be a problem because it is
a one time cost.
This CL will also upstream the Windows implementation of the rng device.
BUG=b:213149162
TEST=built and presubmits
Change-Id: Ic017f11795f8006e0bf2a04eb0478b3a3d336507
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3657812
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Richard Zhang <rizhang@google.com>
They have been dropped with the recent refactoring to not use argh.
BUG=None
TEST=./tools/dev_container -v
Change-Id: Ifa65b14d316e27dbb5cef9714fbd3adf110b79e6
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3664051
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
The builders use the existing dev_container to run tests
for the 3 currently supported architectures on linux:
x86_64, aarch64 and armhf.
A new recipe 'build_linux' has been added runs the
dev_container to build and execute tests.
To share code with other upcoming recipes, some of the
recipe code has been extracted into a shared crosvm
module.
Basic tests for the recipes have been added.
BUG=b:233230344
TEST=./recipes.py test run
./recipes.py run build_linux
Change-Id: I17ecb25c0a0eabdce56537831454ac22d4dc7021
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3654196
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
minigbm virtgpu_cross_domain backend checks for this capset. Eventually
once this check is removed in minigbm (once everyone is using the
context-types arg to control whether cross-domain is exposed) we can
drop the dummy capset. But we need it for now.
BUG=b:230100768
TEST=gfxbench, android games in arcvm
Change-Id: I7c955c365f439806c797d1d38daef2e1c9b3079f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3608703
Auto-Submit: Rob Clark <robdclark@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Add a way to control exposed capsets from cmdline. This will let us
enable certain context types only in certain VMs, with a single build
of crosvm.
BUG=b:230100768
TEST=gfxbench, android games in arcvm
Change-Id: I8d32a22c4c198566e984b06d6aa545796752c128
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3636797
Auto-Submit: Rob Clark <robdclark@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Expose support for the new virtgpu drm native contexts.
BUG=b:230100768
TEST=gfxbench, android games in arcvm
Change-Id: I068e9b14ac47782c9251d9b67156755f2888d17f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3469494
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Auto-Submit: Rob Clark <robdclark@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
