This change is to cleanup some dead_code warnings that appear if certain
features aren't enabled.
This also updates the Cargo.lock when changed due to zeroize being added
to libchromeos-rs.
TEST=cargo check --all-features
BUG=None
Change-Id: I5347b584a7426dc37f3933b1e907b23a71145749
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2753128
Reviewed-by: Allen Webb <allenwebb@google.com>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Zach Reizner <zachr@chromium.org>
Commit-Queue: Zach Reizner <zachr@chromium.org>
This change is similar to http://crrev.com/c/2736520, which made the
path of the KVM device configurable. Similarly, most users will want
to keep the default paths of `/dev/vhost-vsock` and `/dev/vhost-net`.
In certain environments, namely Borg, those device nodes may be located
elsewhere.
BUG=None
TEST=./ci/builder --vm ./run_tests
Change-Id: I4bd7944d8f84fc0e7d255a3930c27f48a980e617
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2749235
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dylan Reid <dgreid@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Set the SECBIT_NO_SETUID_FIXUP securebit so that we don't lose
capabilities when changing the thread uid/gid. This allows us to
simplify the create and mkdir functions so that all the checks we
currently carry out are only done once by the host kernel.
To ensure that the setuid and setgid bits still get dropped when a file
is modified by a process that doesn't hold CAP_FSETID, check for
WRITE_KILL_PRIV in the write flags and temporarily drop CAP_FSETID when
it is set.
BUG=none
TEST=Check that default posix acls, setgid bits, and file/directory
creation via membership of a supplementary group all work as
expected.
Change-Id: I420484e357a970e997cb3e968a433278e82d8ad4
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2684067
Auto-Submit: Chirantan Ekbote <chirantan@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Chirantan Ekbote <chirantan@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
For historical reasons, crosvm switched to sending `IntoIobuf`, this
isn't implementable for a slice of slices as it would require an
allocation. The rust-vmm fork of sys_util still supports sending
`&[&[u8]]`. Adding that ability back makes it easier to use their
crates.
In addition crosvm allows receiving to slices by not iovecs, again the
opposite of the rust-vmm version. Add that ability as well.
Change-Id: I4566f232adb47a0802e39e3f6c09c03f97190eb3
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2740533
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Reviewed-by: Chirantan Ekbote <chirantan@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Keiichi Watanabe <keiichiw@chromium.org>
Commit-Queue: Keiichi Watanabe <keiichiw@chromium.org>
Use tempfile rather than SharedMemory for temporary files in the
composite disk unit tests. This fixes an API incompatibility that was
missed because these tests are only enabled with the composite-disk
feature.
This also enables the disk tests in test_all with the privileged option
so that they run in the VM; this allows the memfd_create syscall used in
the async_read/write tests to work.
BUG=b:181674168
TEST=cargo test -p disk --features=composite-disk
TEST=./test_all
Change-Id: I869b5daa29904b3558440e7b650713e8e5085d81
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2743401
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Make the forked child processess easier to distinguish.
Also tweak the debug_label for virtio-pci devices so that more of the
name can fit into a limited-length thread name.
BUG=None
TEST=pstree
Change-Id: I74a8c1f5ab869e814bed4f2bd71c3de5179f7855
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2740526
Reviewed-by: Zach Reizner <zachr@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Kokoro was failing since http://crrev.com/c/2744968 depends on recent
cras changes that were not included in the current Kokoro manifest.
BUG=None
TEST=./ci/kokoro/simulate_all passes
Change-Id: I470844af73b9f89ac8cbf9f5fcf8706d01924279
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2746014
Reviewed-by: Zach Reizner <zachr@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
This change updates the test VM to debian bullseye, primarily to enable
io_uring features that are not available in the 4.16 kernel shipping
with debian buster.
We are not updating the builder containers to bullseye, as newer
versions of qemu fail to start on kokoro. See b/181359683
Since it was needed for debugging this CL, it includes a feature to
debug kokoro builders via SSH. It can be enabled by setting
DEBUG_SSH_KEY from the fusion2 UI.
BUG=b:178228512
TEST=./test_all passes
Change-Id: I22f1678577183a0aace80245a87c31907221cc09
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2721062
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Zach Reizner <zachr@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Crosvm pre-allocate mmio for device, but it doesn't enable memory/io
space in pci command config register, then OVMF doesn't use the pre-allocated
mmio and reallocate device mmio.
BUG=b:179053182
TEST='crosvm run -bios OVMF.fd' and check device info in efi shell
Change-Id: I7176e7f9716d829efff1ea023666eb705b525e5c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2741920
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
All virtio devices have virtio version 1.0 as base feature, but this revision
id isn't in pci configuration register, then OVMF won't start virtio 1.0
driver, and virtio device couldn't be used in OVMF.
BUG=b:179053182
TEST='crosvm run -bios OVMF.fd' and check virito blk device in EFI shell.
Change-Id: I8cbcd71b9b6ccef07b56853b7450b74e4dcbae1b
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2741919
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Ovmf goes through all entries in smbios table, without end entry,
assert() failure occurs.
BUG=b:179053182
TEST='crosvm run -bios OVMF.fd' and boot OVMF into efi shell
Change-Id: I0068133c75583dc4a3477d01a58f0ee8775b509e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2741918
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
This essentially duplicates the same implicit fd-passing support in
create_block_device. By accepting the FD directly instead of opening the
file under `/proc/self/fd/...` we can inherit the same permissions that
a caller to concierge, for example, gets. This simultaneously allows
crosvm to access files it might not otherwise be able to, and prevents
it from misusing its privileges to potentially elevate a read-only FD to
wa read/write FD.
BUG=b:181347894
TEST=Start pmem VM passed from concierge via FD. Ensured the FD itself
was used instead of crosvm re-opening the file under `/proc/self/fd...`.
Change-Id: Ic4d4e6155a1978b45e82141609fdadff45ca987b
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2724473
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Chirantan Ekbote <chirantan@chromium.org>
Commit-Queue: Mike Gerow <gerow@google.com>
To track arc in VM in UMA and to separate it
from other linux VMs.
Changes:
- Add client_type options to Ac97Parameters.
- Add client_type option for the ac97 devices with CRAS backend.
BUG=b:177393225
TEST=Apply full patch set and start audio in ARCVM with
`cras_test_client --dump_a`
Cq-Depend: chromium:2744525
Change-Id: I27201aa65baed0ee59cf689dd7f22b5b91f00946
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2744968
Reviewed-by: Chirantan Ekbote <chirantan@chromium.org>
Tested-by: Chih-Yang Hsia <paulhsia@chromium.org>
Commit-Queue: Chih-Yang Hsia <paulhsia@chromium.org>
Most users will want to keep the default `/dev/kvm` path. However, in
certain environments, namely Borg, the KVM device node may be located
elsewhere.
This is the first of a set of related changes that will make hard-coded
device paths configurable.
BUG=None
TEST=./ci/builder --vm ./run_tests
Change-Id: I6087879c535be3779e20eff1f8fb5080f80cf020
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2736520
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Dylan Reid <dgreid@chromium.org>
Starting with 2.32 glibc nanosleep() was refactored to use the
clock_nanosleep syscall so various software will fail unless
the new syscall is allowed. We can't just drop the old nanosleep
syscall because it will break glibc 2.27 which is still used.
See glibc commits:
807edded25 nptl: Refactor thrd_sleep in terms of clock_nanosleep
3537ecb49c Refactor nanosleep in terms of clock_nanosleep
79a547b162 nptl: Move nanosleep implementation to libc
This is a bulk edit done with the following command:
git grep -rl 'nanosleep: 1' | xargs sed -i \
'/^nanosleep: 1/a clock_nanosleep: 1'
BUG=chromium:1171287
TEST=Local builds and booting on kevin/64/eve/minnie.
Change-Id: I975535078d88200f52319c7eea3a4c7ebf299933
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2735575
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Manoj Gupta <manojgupta@chromium.org>
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Reviewed-by: Stephen Barber <smbarber@chromium.org>
While a host virtio device provides |num_queues| virtqueues, a guest virtio driver doesn't necessarily use all of them. For example, the virtio-blk driver uses only |nr_cpu_ids| virtqueues at most [1].
To avoid checking whether each queue is ready in each device implementation, we can filter them before starting device activation.
[1]:
https://patchwork.kernel.org/project/linux-block/cover/1553682995-5682-1-git-send-email-dongli.zhang@oracle.com/
BUG=b:179671351, b:181753022
TEST=CQ
Change-Id: I29d21d8d9db2d99aa9591ca55c18d06d2368797e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2732735
Tested-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Keiichi Watanabe <keiichiw@chromium.org>
Add #[repr(packed)] to struct virtio_blk_config to make its size same with the
the origianl C struct. The packed annotation will remove 4-byte padding at the
end of the struct and make the size of the struct smaller. (64 bytes -> 60 bytes)
Since it won't affect offsets of any fields, it shouldn't change any behavior
when the guest reads a config field. But, it can matter when the entire config
struct is passed via vhost-user protocol.
BUG=none
TEST=run a VM
Change-Id: I4dca9f1bdd93166192eca4d05d542ce851852aa7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2726059
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Keiichi Watanabe <keiichiw@chromium.org>
After the userspace streamoff the input queue, the crosvm should not
return the previous frames. However, VDA might still return frames
before notifying reset is done. This CL drops the decoded frames after
calling VDA::Reset() until reset is completed.
BUG=b:181087034
TEST=android.media.cts.AdaptivePlaybackTest
Change-Id: Ieaa40ef27f1b37a262c80f9f30698c03ef16bdb9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2728584
Tested-by: Chih-Yu Huang <akahuang@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Commit-Queue: Chih-Yu Huang <akahuang@chromium.org>
When the video is flushed, V4L2DecodeComponent streamoff V4L2 output
and input queue. Then crosvm releases all output buffers and calls
VDA::Reset(). However, VaapiVDA implementation doesn't release output
buffer at Reset(). If Vaapi decodes the following frame before
V4L2DecodeComponent QBUF any output buffer, then crosvm will drop the
decoded frame.
This CL makes crosvm postpone sending the decoded frame if this
situation happens. Crosvm would sends the decoded frame when receiving
the buffer again.
BUG=b:181087034
TEST=emerge-hatch-arc-r crosvm
TEST=android.media.cts.AdaptivePlaybackTest
TEST=seek video many times and check no error occurs at the end of video
Change-Id: I0c8e59e2a206d6b2cd2009fd70380e7d5a366953
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2719245
Tested-by: Chih-Yu Huang <akahuang@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Commit-Queue: Chih-Yu Huang <akahuang@chromium.org>
Originally, process_cmd function can only return the response of the
procesesed cmd. However, we need to return the response of events for
some commands. This CL makes the process_cmd function could return
the responses of both command and event.
BUG=b:181087034
TEST=emerge-hatch-arc-r crosvm
Change-Id: Ie781795f8cee1c66e8462c602f876043b0dea9bc
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2719244
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Chih-Yu Huang <akahuang@chromium.org>
Originally process_cmd() return VideoResult<VideoCmdResponseType>.
However, VideoCmdResponseType could return the error result by
Sync(CmdResponse::Error). VideoResult is not needed.
This CL change the returned type of process_cmd() to
VideoCmdResponseType to reduce code complexity.
BUG=b:181087034
TEST=emerge-hatch-arc-r crosvm
Change-Id: I1795a3eb09fe36076f5ad43fdd8d1eb9e21ffcd9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2731607
Tested-by: Chih-Yu Huang <akahuang@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Commit-Queue: Chih-Yu Huang <akahuang@chromium.org>
By using these traits in crosvm, we also add serde as a dependency.
BUG=b:176847362
TEST=run_tests
Cq-Depend: chromium:2729637
Change-Id: Icd16a6e163d9e1dedbe3924cb94f29d777eb2216
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2613690
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Zach Reizner <zachr@chromium.org>
Commit-Queue: Zach Reizner <zachr@chromium.org>
This is replaced by test_all and the new builders for kokoro.
See ci/README.md for details.
BUG=b:178233937
TEST=None
Change-Id: Id8bf721e026e71d9f284f920ef8287beb414c621
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2705292
Tested-by: Dennis Kempin <denniskempin@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Zach Reizner <zachr@chromium.org>
The args may be interpreted as pointers for some capabilities, so the
caller must ensure that any such pointers are allocated appropriately.
BUG=b:181564686
TEST=cargo test
Change-Id: I244f4d9417e588a6be5681f4718bb9ad7b262c3e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2726709
Auto-Submit: Andrew Walbran <qwandor@google.com>
Commit-Queue: Dylan Reid <dgreid@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
This removes some unsafe code, improving the interface so that it cannot
be misused (e.g. previously, different Vec<u8> instances could be passed
to fdt functions that did not validate the contents).
The new implementation also adds some extra error checking to catch
invalid string values in all API entry points that accept strings, as
well as out-of-order node nesting that would result in DTB data that did
not conform to the spec.
BUG=b:175729255
TEST=cargo test -p arch
TEST=Boot Crostini on kevin (arm)
TEST=diff old and new dts from `dtc -I fs -O dts /proc/device-tree`
Change-Id: I567d562ed1b447aa0d282b68c71758edea13ffc0
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2713569
Reviewed-by: Zach Reizner <zachr@chromium.org>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
- Add an address space region for the protected KVM firmware.
- Query firmware size, mmap something that size and create a memslot.
BUG=b:163789172
TEST=cargo test
Change-Id: I054cf5d763c980d073c17bce70e85a781816b64d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2623942
Auto-Submit: Andrew Walbran <qwandor@google.com>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Commit-Queue: Andrew Walbran <qwandor@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This CL replaces BorrowedIoVec with VolatileSlice, since VolatileSlice
is a superset of the BorrowedIoVec interface. Also uring_mem -> mem
since that interface will not be exclusively used by uring.
BUG=none
TEST=builds
Change-Id: I33e23483e7afc263c76d71f88736cf38fd5e520e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2724863
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Noah Gold <nkgold@google.com>
This is a simple python script to uprev ebuild files where Cargo.lock
has a newer version, and vice versa.
The script is run to fix up some libraries that got out of sync.
The protobuf dependencies should be upreved, but newer versions
have deprecated some of the methods we use, causing clippy warnings.
BUG=b:175120466
TEST=Tests in Kokoro and CQ pass
Cq-Depend: chromium:2704533
Change-Id: I1485fbffba61e72502f8398320094dfe2c7ffeea
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2705681
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: kokoro <noreply+kokoro@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
In a long-running system, there is no reason to expect that a
significant number of freed pages are consecutive. However, batching is
relatively simple and can result in significant gains in a newly booted
system, so it's worth attempting.
BUG=None
TEST=arc.Boot.vm
Change-Id: Ia7dff4ab095d640a2a23ac4976bc277b09d9ea79
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2666412
Commit-Queue: David Stevens <stevensd@chromium.org>
Tested-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Chirantan Ekbote <chirantan@chromium.org>
The max values for the multitouch slot ID and tracking/finger IDs were
set to zero previously, making it impossible to track multiple
fingers. This CL updates the max values to allow for 10 MT contact
points.
BUG=None
TEST=applied known working code from another branch.
Change-Id: Ic2e9919c2b83368eb1bc2085122c672fdafbdc84
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2708669
Reviewed-by: Tristan Muntsinger <muntsinger@google.com>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Reviewed-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Noah Gold <nkgold@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Auto-Submit: Noah Gold <nkgold@google.com>
A BIOS can be larger or smaller than 1 MB; as long as the image is
placed so the reset vector lines up in the correct place at the end of
the 32-bit address space, everything should work.
BUG=b:179053182
TEST=`crosvm run --bios OVMF.fd` and observe debug messages
Change-Id: Ibafd9bb5ee4fd4b0ff2c28c38f022ff3b36dd95d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2651288
Reviewed-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Zach Reizner <zachr@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
When the file system implements zero message open support, the file
handle is meaningless and it needs to know the inode of the
file/directory on which the ioctl was called.
BUG=b:180565632
TEST=lsattr, chattr, both work when zero message open is enabled.
Android's FileBasedEncryptionPolicyTest[0] gets ENOTTY as an error
instead of EBADF
[0]: bfbc00c20d/tests/tests/security/native/encryption/FileBasedEncryptionPolicyTest.cpp
Change-Id: Ic55ee95df928d645874dd8a9c7dc579b708927fa
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2706370
Tested-by: Chirantan Ekbote <chirantan@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Auto-Submit: Chirantan Ekbote <chirantan@chromium.org>
Commit-Queue: Stephen Barber <smbarber@chromium.org>
Reviewed-by: Stephen Barber <smbarber@chromium.org>
Reviewed-by: kokoro <noreply+kokoro@google.com>
Having these function be generic over F: AsRawDescriptor with a &F
parameter means that we cannot use trait objects with them because the
compiler complains that `dyn AsRawDescriptor` does not have a
compile-time size.
Instead change the generic parameter to be a trait object. We're
already doing this in a few other places and we're about to make a
syscall so the cost of dynamic dispatch is probably overshadowed by the
cost of syscall itself.
BUG=b:180565632
TEST=unit tests
Change-Id: I38f696b621411e7cf3e13af71e426865b6509f6f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2706369
Tested-by: Chirantan Ekbote <chirantan@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Auto-Submit: Chirantan Ekbote <chirantan@chromium.org>
Commit-Queue: Stephen Barber <smbarber@chromium.org>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Stephen Barber <smbarber@chromium.org>
Reviewed-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
The arm version of libc uses the fcntl64 syscall (0xdd); update the
seccomp policy to match the other arm policies.
BUG=chromium:901139
TEST=Boot Crostini on kevin (arm)
Change-Id: I82eee1ec4918a56495d6c330913c34d4a4a20e4b
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2705290
Reviewed-by: Zach Reizner <zachr@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Reviewed-by: kokoro <noreply+kokoro@google.com>
Tested-by: Dylan Reid <dgreid@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Handle data offset for input bitstream buffers, as it is sometimes used
to skip headers at the start of buffers.
BUG=b:174531173
TEST=android.media.cts.MediaDrmClearkeyTest#testClearKeyPlaybackMpeg2ts
Change-Id: I6beee5cde24803ba90638c1dc130b75466f4847d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2692676
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: David Stevens <stevensd@chromium.org>
Fixes a clippy issue and disables the panic_safe test.
This test occasionally deadlocks (Maybe a 5% chance) when running
inside a test VM. I filed a bug to re-enable it.
BUG=b:179924844
TEST=kokoro passes
Change-Id: I6e9b5fc38e7bbab3fd4b2b8ba8cb5532dd14af7e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2697860
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Auto-Submit: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
This is the original CL with one minor exception: we don't bind
mount the camera socket with the GPU device. That was the prior
behavior, and for some reason it really doesn't work with Mali +
SECCOMP[1]. It's not really important for the Wayland prototype,
so we'll let the camera team figure it out if and when they are
so inclined.
Bug: b:146066070
Bug: b:173630595
Bug: b:150239451
Bug: b:180126126
TEST=arc.Boot.vm
[1] audit(1613339319.226:43): auid=4294967295 uid=603 gid=603
ses=4294967295 subj=u:r:cros_camera_algo:s0 pid=17107
comm="cros_camera_alg" exe="/usr/bin/cros_camera_algo" sig=31
arch=40000028 syscall=54 compat=1 ip=0xe86a70b8 code=0x0
This reverts commit 51e1c4ad3e3a71a263501d2566d3b1ea59ba2070.
Change-Id: I74f49ece55656d7a9096900e3f19a528234b4224
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2695550
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Robert Tarasov <tutankhamen@chromium.org>
Enable the ZERO_MESSAGE_{OPEN,OPENDIR} features when the cache policy is
"always". This feature allows the kernel to skip the open message after
a successful lookup, reducing the amount of work that the server does.
This is implemented by changing the file descriptors stored in the
InodeData from O_PATH fds to O_RDONLY fds for files and directories.
Other types of directory entries (symlinks, special files, etc) still
use O_PATH fds.
If the kernel sends a write request for an fd opened in read-only mode
or a read request for an fd opened in write-only mode (can happen when
creating a new file), then we open a new fd in read-write mode before
performing the read/write. This only needs to happen the first time we
get a request that doesn't match the open flags.
This change should improve performance of opening and reading many small
files. It improves the blogbench read score by ~40% but reduces the
write score by ~25%. It also reduces the work done by the virtio-fs
server when loading roblox. The first load time is reduced by
~17% (3.04 seconds -> 2.52 seconds) and non-initial load times are
reduced by 50% (0.3 seconds -> 0.15 seconds).
BUG=none
TEST=vm.Virtiofs, vm.Blogbench.virtiofs, arc.PlayStore.vm, load roblox
inside arcvm
Change-Id: I042246a2fd9f7a0feeffc0f629073c594626392a
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2684066
Reviewed-by: Stephen Barber <smbarber@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Chirantan Ekbote <chirantan@chromium.org>
This reverts commit a9c4b3a749.
Reason for revert: This made ARCVM fail to boot on kukui-arc-r. See http://b/180126126.
Original change's description:
> rutabaga_gfx: cross-domain: a new year's miracle in February
>
> The cross-domain context is specialized for cross domain
> allocation/resource sharing. It takes direct inspiration from
> the pioneering virtio_wl device and tries to incorporate
> similiar functionality into virtio_gpu.
>
> The goal here is just to introduce the building blocks so we
> can continue experimenting. In particular, this change:
>
> * hooked up the RutabagaChannels. This is typically a socket to
> Wayland or Mojo for the camera use case.
>
> * added CROSS_DOMAIN_CMD_INIT and CROSS_DOMAIN_CMD_GET_IMAGE_REQS
> to the cross-domain protocol. Further commands (such as
> CROSS_DOMAIN_SEND) will be needed, but that requires more
> Sommelier refactorings.
>
> * added a path to RutabagaGralloc to allocate via minigbm or shared
> memory.
>
> * Recieves responses via a shared ring buffer of type BLOB_MEM_GUEST.
> The synchronization protocol looks positively primitive compared to
> the revolutionary Address Space Graphics (ASG) algorithm [1], but
> it may be sufficient for the Wayland use case.
>
> [1] https://goto.google.com/address-space-graphics
>
> BUG=b:146066070, b:173630595, b:150239451
> TEST=launch virtual machine with 2D mode
> TEST=launch virtual machine with 3D mode
> TEST=run sommelier with "wl-dmabuf" and "wl-shm"
>
> Change-Id: I46784f17040494ce3a646bdbde516800aa64bd5d
> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2626488
> Tested-by: kokoro <noreply+kokoro@google.com>
> Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
> Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
> Reviewed-by: Zach Reizner <zachr@chromium.org>
Bug: b:146066070
Bug: b:173630595
Bug: b:150239451
Bug: b:180126126
Change-Id: Ie33442fdcedcf43b6a24d25198fa2d88b5b96919
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2695056
Reviewed-by: Ryo Hashimoto <hashimoto@chromium.org>
Reviewed-by: Hiroki Sato <hirokisato@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Hiroki Sato <hirokisato@chromium.org>
Commit-Queue: Hiroki Sato <hirokisato@chromium.org>
Commit "1fcc8a8f io_uring: Make UringContext Sync" uncovered an issue
where wakeups were lost if a uring operation was added from a different
thread while the executor thread was blocked inside an io_uring_enter
call. To fix this, PendingOperation calls UringContext::submit
whenever the pending IO is not yet completed and has not been submitted.
Unfortunately, since we tend to write code like
`r.read_to_vec(..).await`, this meant that we would call
UringContext::submit every time after adding a new operation to the
submit queue, kind of defeating the purpose of batching multiple IO ops
in a single syscall. Instead only call UringContext::submit when the
current thread is not the same as the executor thread. The executor
will submit all pending operations anyway the next time it calls
UringContext::wait so there's no need to do it from PendingOperation
when it is polled on the executor thread.
BUG=none
TEST=unit tests
Change-Id: Ia95f3844790d3392e074e3ab55a9c6ef59f29db2
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2684063
Commit-Queue: Chirantan Ekbote <chirantan@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
Destroying a stream while there are outstanding async commands results
in the responses for those commands having no corresponding tracked
descriptor. Instead of trying to handle this case specifically, make
untracked async responses non-fatal errors, instead of shutting down the
decoder device completely.
BUG=b:177697115
TEST=android.security.cts.StagefrightTest
Change-Id: I142ec9814fd69ddb79ef16140b7b06cd0c9f0123
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2690728
Reviewed-by: Chih-Yu Huang <akahuang@chromium.org>
Tested-by: David Stevens <stevensd@chromium.org>
Commit-Queue: David Stevens <stevensd@chromium.org>
The newly added async primitives allow for increasing the separation of
the various tasks performed by balloon. Breaking each task in to an
asynchronous function.
BUG=chromium:901139
TEST=Boot crosvm, run 'crosvm balloon' to set the balloon size, check
'vmstat' inside the VM to verify the free memory is affected by the
balloon growing and shrinking.
run crosvm balloon_stats command and ensure that stats are reported
correctly.
Change-Id: I0ae2be5eb8e4be65b2eb74de90888357af6ecfd4
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/1993163
Tested-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Keiichi Watanabe <keiichiw@chromium.org>
Reviewed-by: Chirantan Ekbote <chirantan@chromium.org>
Reviewed-by: Dylan Reid <dgreid@chromium.org>
The cross-domain context is specialized for cross domain
allocation/resource sharing. It takes direct inspiration from
the pioneering virtio_wl device and tries to incorporate
similiar functionality into virtio_gpu.
The goal here is just to introduce the building blocks so we
can continue experimenting. In particular, this change:
* hooked up the RutabagaChannels. This is typically a socket to
Wayland or Mojo for the camera use case.
* added CROSS_DOMAIN_CMD_INIT and CROSS_DOMAIN_CMD_GET_IMAGE_REQS
to the cross-domain protocol. Further commands (such as
CROSS_DOMAIN_SEND) will be needed, but that requires more
Sommelier refactorings.
* added a path to RutabagaGralloc to allocate via minigbm or shared
memory.
* Recieves responses via a shared ring buffer of type BLOB_MEM_GUEST.
The synchronization protocol looks positively primitive compared to
the revolutionary Address Space Graphics (ASG) algorithm [1], but
it may be sufficient for the Wayland use case.
[1] https://goto.google.com/address-space-graphics
BUG=b:146066070, b:173630595, b:150239451
TEST=launch virtual machine with 2D mode
TEST=launch virtual machine with 3D mode
TEST=run sommelier with "wl-dmabuf" and "wl-shm"
Change-Id: I46784f17040494ce3a646bdbde516800aa64bd5d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/2626488
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Zach Reizner <zachr@chromium.org>
