The test is a bit different from the others, in that it does not run
if $membarrier is empty. For meson, the default can simply be disabled;
if one day we will toggle the default, no change is needed in meson.build.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
For consistency with other tests, --enable-avx2 and --enable-avx512f
fail to compile on x86 systems if cpuid.h is not available.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The method is now in 0.59, using it simplifies some conditionals.
There is a small change, which is to build virtfs-proxy-helper in a
tools-only build. This is done for consistency with other tools,
which are not culled by the absence of system emulator binaries.
.disable_auto_if() would also be useful to check for packages,
for example
-linux_io_uring = not_found
-if not get_option('linux_io_uring').auto() or have_block
- linux_io_uring = dependency('liburing', required: get_option('linux_io_uring'),
- method: 'pkg-config', kwargs: static_kwargs)
-endif
+linux_io_uring = dependency('liburing',
+ required: get_option('linux_io_uring').disable_auto_if(not have_block),
+ method: 'pkg-config', kwargs: static_kwargs)
This change however is much larger and I am not sure about the improved
readability, so I am not performing it right now.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The method is now in 0.59, using it simplifies some boolean conditions.
The other new methods .require() and .disable_auto_if() can be used too,
but introducing them is not just a matter of search-and-replace.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
QEMU can now easily crash with two continuous migration carried out:
(qemu) migrate -d exec:cat>out
(qemu) migrate_cancel
(qemu) migrate -d exec:cat>out
[crash] ../softmmu/memory.c:2782: memory_global_dirty_log_start: Assertion
`!(global_dirty_tracking & flags)' failed.
It's because memory API provides a way to postpone dirty log stop if the VM is
stopped, and that'll be re-done until the next VM start. It was added in 2017
with commit 1931076077 ("migration: optimize the downtime", 2017-08-01).
However the recent work on allowing dirty tracking to be bitmask broke it,
which is commit 63b41db4bc ("memory: make global_dirty_tracking a bitmask",
2021-11-01).
The fix proposed in this patch contains two things:
(1) Instead of passing over the flags to postpone stop dirty track, we add a
global variable (along with current vmstate_change variable) to record
what flags to stop dirty tracking.
(2) When start dirty tracking, instead if remove the vmstate hook directly,
we also execute the postponed stop process so that we make sure all the
starts and stops will be paired.
This procedure is overlooked in the bitmask-ify work in 2021.
Cc: Hyman Huang <huangy81@chinatelecom.cn>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2044818
Fixes: 63b41db4bc ("memory: make global_dirty_tracking a bitmask")
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220207123019.27223-1-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fourth RISC-V PR for QEMU 7.0
* Remove old Ibex PLIC header file
* Allow writing 8 bytes with generic loader
* Fixes for RV128
* Refactor RISC-V CPU configs
* Initial support for XVentanaCondOps custom extension
* Fix for vill field in vtype
* Fix trap cause for RV32 HS-mode CSR access from RV64 HS-mode
* Support for svnapot, svinval and svpbmt extensions
# gpg: Signature made Wed 16 Feb 2022 06:24:52 GMT
# gpg: using RSA key F6C4AC46D4934868D3B8CE8F21E10D29DF977054
# gpg: Good signature from "Alistair Francis <alistair@alistair23.me>" [full]
# Primary key fingerprint: F6C4 AC46 D493 4868 D3B8 CE8F 21E1 0D29 DF97 7054
* remotes/alistair/tags/pull-riscv-to-apply-20220216: (35 commits)
docs/system: riscv: Update description of CPU
target/riscv: add support for svpbmt extension
target/riscv: add support for svinval extension
target/riscv: add support for svnapot extension
target/riscv: add PTE_A/PTE_D/PTE_U bits check for inner PTE
target/riscv: Ignore reserved bits in PTE for RV64
hw/intc: Add RISC-V AIA APLIC device emulation
target/riscv: Allow users to force enable AIA CSRs in HART
hw/riscv: virt: Use AIA INTC compatible string when available
target/riscv: Implement AIA IMSIC interface CSRs
target/riscv: Implement AIA xiselect and xireg CSRs
target/riscv: Implement AIA mtopi, stopi, and vstopi CSRs
target/riscv: Implement AIA interrupt filtering CSRs
target/riscv: Implement AIA hvictl and hviprioX CSRs
target/riscv: Implement AIA CSRs for 64 local interrupts on RV32
target/riscv: Implement AIA local interrupt priorities
target/riscv: Allow AIA device emulation to set ireg rmw callback
target/riscv: Add defines for AIA CSRs
target/riscv: Add AIA cpu feature
target/riscv: Allow setting CPU feature from machine/device emulation
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The RISC-V AIA (Advanced Interrupt Architecture) defines a new
interrupt controller for wired interrupts called APLIC (Advanced
Platform Level Interrupt Controller). The APLIC is capabable of
forwarding wired interupts to RISC-V HARTs directly or as MSIs
(Message Signaled Interupts).
This patch adds device emulation for RISC-V AIA APLIC.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-id: 20220204174700.534953-19-anup@brainfault.org
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The AIA specificaiton adds interrupt filtering support for M-mode
and HS-mode. Using AIA interrupt filtering M-mode and H-mode can
take local interrupt 13 or above and selectively inject same local
interrupt to lower privilege modes.
At the moment, we don't have any local interrupts above 12 so we
add dummy implementation (i.e. read zero and ignore write) of AIA
interrupt filtering CSRs.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-id: 20220204174700.534953-13-anup@brainfault.org
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The AIA spec defines programmable 8-bit priority for each local interrupt
at M-level, S-level and VS-level so we extend local interrupt processing
to consider AIA interrupt priorities. The AIA CSRs which help software
configure local interrupt priorities will be added by subsequent patches.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20220204174700.534953-10-anup@brainfault.org
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The machine or device emulation should be able to force set certain
CPU features because:
1) We can have certain CPU features which are in-general optional
but implemented by RISC-V CPUs on the machine.
2) We can have devices which require a certain CPU feature. For example,
AIA IMSIC devices expect AIA CSRs implemented by RISC-V CPUs.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-id: 20220204174700.534953-6-anup@brainfault.org
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The guest external interrupts from an interrupt controller are
delivered only when the Guest/VM is running (i.e. V=1). This means
any guest external interrupt which is triggered while the Guest/VM
is not running (i.e. V=0) will be missed on QEMU resulting in Guest
with sluggish response to serial console input and other I/O events.
To solve this, we check and inject interrupt after setting V=1.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-id: 20220204174700.534953-5-anup@brainfault.org
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
To split up the decoder into multiple functions (both to support
vendor-specific opcodes in separate files and to simplify maintenance
of orthogonal extensions), this changes decode_op to iterate over a
table of decoders predicated on guard functions.
This commit only adds the new structure and the table, allowing for
the easy addition of additional decoders in the future.
Signed-off-by: Philipp Tomsich <philipp.tomsich@vrull.eu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20220202005249.3566542-6-philipp.tomsich@vrull.eu>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
As the number of extensions is growing, copying them individiually
into the DisasContext will scale less and less... instead we populate
a pointer to the RISCVCPUConfig structure in the DisasContext.
This adds an extra indirection when checking for the availability of
an extension (compared to copying the fields into DisasContext).
While not a performance problem today, we can always (shallow) copy
the entire structure into the DisasContext (instead of putting a
pointer to it) if this is ever deemed necessary.
Signed-off-by: Philipp Tomsich <philipp.tomsich@vrull.eu>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20220202005249.3566542-3-philipp.tomsich@vrull.eu>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
The documentation for the generic loader says that "the maximum size of
the data is 8 bytes". However, attempts to set data-len=8 trigger the
following assertion failure:
../hw/core/generic-loader.c:59: generic_loader_reset: Assertion `s->data_len < sizeof(s->data)' failed.
The type of s->data is uint64_t (i.e. 8 bytes long), so I believe this
assert should use <= instead of <.
Fixes: e481a1f63c ("generic-loader: Add a generic loader")
Signed-off-by: Petr Tesarik <ptesarik@suse.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20220120092715.7805-1-ptesarik@suse.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
hw/nvme updates
- fix CVE-2021-3929
- add zone random write area support
- misc cleanups from Philippe
# gpg: Signature made Mon 14 Feb 2022 08:01:34 GMT
# gpg: using RSA key 522833AA75E2DCE6A24766C04DE1AF316D4F0DE9
# gpg: Good signature from "Klaus Jensen <its@irrelevant.dk>" [unknown]
# gpg: aka "Klaus Jensen <k.jensen@samsung.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: DDCA 4D9C 9EF9 31CC 3468 4272 63D5 6FC5 E55D A838
# Subkey fingerprint: 5228 33AA 75E2 DCE6 A247 66C0 4DE1 AF31 6D4F 0DE9
* remotes/nvme/tags/nvme-next-pull-request:
hw/nvme: add support for zoned random write area
hw/nvme: add ozcs enum
hw/nvme: add struct for zone management send
hw/nvme/ctrl: Pass buffers as 'void *' types
hw/nvme/ctrl: Have nvme_addr_write() take const buffer
hw/nvme: fix CVE-2021-3929
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
# gpg: Signature made Mon 14 Feb 2022 03:51:14 GMT
# gpg: using RSA key EF04965B398D6211
# gpg: Good signature from "Jason Wang (Jason Wang on RedHat) <jasowang@redhat.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 215D 46F4 8246 689E C77F 3562 EF04 965B 398D 6211
* remotes/jasowang/tags/net-pull-request:
net/eth: Don't consider ESP to be an IPv6 option header
hw/net: e1000e: Clear ICR on read when using non MSI-X interrupts
net/filter: Optimize filter_send to coroutine
net/colo-compare.c: Update the default value comments
net/colo-compare.c: Optimize compare order for performance
net: Fix uninitialized data usage
net/tap: Set return code on failure
hw/net/vmxnet3: Log guest-triggerable errors using LOG_GUEST_ERROR
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Block layer patches
- Fix crash in blockdev-reopen with iothreads
- fdc-isa: Respect QOM properties when building AML
# gpg: Signature made Fri 11 Feb 2022 17:44:52 GMT
# gpg: using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg: issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74 56FE 7F09 B272 C88F 2FD6
* remotes/kwolf-gitlab/tags/for-upstream:
hw/block/fdc-isa: Respect QOM properties when building AML
iotests: Test blockdev-reopen with iothreads and throttling
block: Lock AioContext for drain_end in blockdev-reopen
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Coroutine pool size was 64 from long ago, and the basis was organized in the commit message in 4d68e86b.
At that time, virtio-blk queue-size and num-queue were not configuable, and equivalent values were 128 and 1.
Coroutine pool size 64 was fine then.
Later queue-size and num-queue got configuable, and default values were increased.
Coroutine pool with size 64 exhausts frequently with random disk IO in new size, and slows down.
This commit adjusts coroutine pool size adaptively with new values.
This commit adds 64 by default, but now coroutine is not only for block devices,
and is not too much burdon comparing with new default.
pool size of 128 * vCPUs.
Signed-off-by: Hiroki Narukawa <hnarukaw@yahoo-corp.jp>
Message-id: 20220214115302.13294-2-hnarukaw@yahoo-corp.jp
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The virtiofsd currently crashes when used with glibc 2.35.
That is due to the rseq system call being added to every thread
creation [1][2].
[1]: https://www.efficios.com/blog/2019/02/08/linux-restartable-sequences/
[2]: https://sourceware.org/pipermail/libc-alpha/2022-February/136040.html
This happens not at daemon start, but when a guest connects
/usr/lib/qemu/virtiofsd -f --socket-path=/tmp/testvfsd -o sandbox=chroot \
-o source=/var/guests/j-virtiofs --socket-group=kvm
virtio_session_mount: Waiting for vhost-user socket connection...
# start ok, now guest will connect
virtio_session_mount: Received vhost-user socket connection
virtio_loop: Entry
fv_queue_set_started: qidx=0 started=1
fv_queue_set_started: qidx=1 started=1
Bad system call (core dumped)
We have to put rseq on the seccomp allowlist to avoid that the daemon
is crashing in this case.
Reported-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-id: 20220209111456.3328420-1-christian.ehrhardt@canonical.com
[Moved rseq to its alphabetically ordered position in the seccomp
allowlist.
--Stefan]
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
