| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Given this is only relevant for BPF tail call maps, it is adding up space
and penalizing other map types. We also need to extend this with further
objects to track / compare to. Therefore, lets move this out into a separate
structure and dynamically allocate it only for BPF tail call maps.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/20250730234733.530041-2-daniel@iogearbox.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Add a cookie to BPF maps to uniquely identify BPF maps for the timespan
when the node is up. This is different to comparing a pointer or BPF map
id which could get rolled over and reused.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/20250730234733.530041-1-daniel@iogearbox.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
In order to support Clang's stack depth tracking (for Linux's kstack_erase
feature), the coverage sanitizer needed to be disabled for __init (and
__head) section code. Doing this universally (i.e. for GCC too) created
a number of unexpected problems, ranging from changes to inlining logic
to failures to DCE code on earlier GCC versions.
Since this change is only needed for Clang, specialize it so that GCC
doesn't see the change as it isn't needed there (the GCC implementation
of kstack_erase uses a GCC plugin that removes stack depth tracking
instrumentation from __init sections during a late pass in the IR).
Successfully build and boot tested with GCC 12 and Clang 22.
Fixes: 381a38ea53d2 ("init.h: Disable sanitizer coverage for __init and __head")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202507270258.neWuiXLd-lkp@intel.com/
Reported-by: syzbot+5245cb609175fb6e8122@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/6888d004.a00a0220.26d0e1.0004.GAE@google.com/
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Marco Elver <elver@google.com>
Link: https://lore.kernel.org/r/20250729234055.it.233-kees@kernel.org
Signed-off-by: Kees Cook <kees@kernel.org>
|
|
When FORTIFY_SOURCE reports about a run-time buffer overread, the wrong
buffer size was being shown in the error message. (The bounds checking
was correct.)
Fixes: 3d965b33e40d ("fortify: Improve buffer overflow reporting")
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20250729231817.work.023-kees@kernel.org
Signed-off-by: Kees Cook <kees@kernel.org>
|
|
Certain targets disable kstack_erase by filtering out KSTACK_ERASE_CFLAGS
rather than adding DISABLE_KSTACK_ERASE. The renaming to kstack_erase
missed the CFLAGS export, which broke those build targets (e.g. x86
vdso32).
Fixes: 76261fc7d1be ("stackleak: Split KSTACK_ERASE_CFLAGS from GCC_PLUGINS_CFLAGS")
Signed-off-by: Kees Cook <kees@kernel.org>
|
|
Don't populate the read-only 'type' on the stack at run time,
instead make it static.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://lore.kernel.org/20250714160858.1234719-1-colin.i.king@gmail.com
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
The intent was to create a single column table, however the markup used
was actually for a header which led to docs build failures:
Sphinx parallel build error:
docutils.utils.SystemMessage: Documentation/virt/kvm/devices/arm-vgic-v3.rst:128: (SEVERE/4) Unexpected section title or transition.
Fix the issue by converting the attempted table to an unordered list.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Closes: https://lore.kernel.org/linux-next/20250729142217.0d4e64cd@canb.auug.org.au/
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Message-ID: <20250729152242.3232229-1-oliver.upton@linux.dev>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
In dbAllocCtl(), read_metapage() increases the reference count of the
metapage. However, when dp->tree.budmin < 0, the function returns -EIO
without calling release_metapage() to decrease the reference count,
leading to a memory leak.
Add release_metapage(mp) before the error return to properly manage
the metapage reference count and prevent the leak.
Fixes: a5f5e4698f8abbb25fe4959814093fb5bfa1aa9d ("jfs: fix shift-out-of-bounds in dbSplit")
Signed-off-by: Zheng Yu <zheng.yu@northwestern.edu>
Signed-off-by: Dave Kleikamp <dave.kleikamp@oracle.com>
|
|
When testing a series of commits that also includes changes to the Linux
tools directory, it is useless to test the changes in tools as they may
not affect the kernel itself. Doing tests on the kernel for changes that
do not affect the kernel is a waste of time.
Add a PATCHCHECK_SKIP that takes a series of shas that will be skipped
while doing the individual commit tests.
For example, the runtime verification may have a series of commits like:
$ git log --abbrev-commit --pretty=oneline fac5493251a6~1..HEAD
3d3800b4f7f4 rv: Remove rv_reactor's reference counter
3d3c376118b5 rv: Merge struct rv_reactor_def into struct rv_reactor
24cbfe18d55a rv: Merge struct rv_monitor_def into struct rv_monitor
b0c08dd5348d rv: Remove unused field in struct rv_monitor_def
58d5f0d437a8 (debiantesting-x86-64/trace/rv/core) rv: Return init error when registering monitors
560473f2e2d7 verification/rvgen: Organise Kconfig entries for nested monitors
9efcf590827c tools/dot2c: Fix generated files going over 100 column limit
1160ccaf772f tools/rv: Stop gracefully also on SIGTERM
f60227f34489 tools/rv: Do not skip idle in trace
f3735df6281e verification/rvgen: Do not generate unused variables
6fb37c2a27eb verification/rvgen: Generate each variable definition only once
8cfcf9b0e92f verification/rvgen: Support the 'next' operator
fac5493251a6 rv: Allow to configure the number of per-task monitor
Where the first commit touches the kernel followed by a series of commits
that do not, and ends with commits that do. Instead of having to add
multiple patchcheck tests to handle the gaps, just include the commits
that should not be tested:
$ git log --abbrev-commit --pretty=oneline fac5493251a6~1..HEAD |
grep -e verification -e tools/ | cut -d' ' -f1 |
while read a ; do echo -n "$a "; done
560473f2e2d7 9efcf590827c 1160ccaf772f f60227f34489 f3735df6281e 6fb37c2a27eb 8cfcf9b0e92f
Then set PATCHCHECK_SKIP to that, and those commits will be skipped.
PATCHCHECK_SKIP = 560473f2e2d7 9efcf590827c 1160ccaf772f f60227f34489 f3735df6281e 6fb37c2a27eb 8cfcf9b0e92f
Cc: John 'Warthog9' Hawley <warthog9@kernel.org>
Cc: Dhaval Giani <dhaval.giani@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/20250725112153.1dd06b84@gandalf.local.home
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
|
Since preempt_count_add/del() are tracable functions, it is not allowed
to use preempt_disable/enable() in ftrace handlers. Without this fix,
probing on `preempt_count_add%return` will cause an infinite recursion
of fprobes.
To fix this problem, use preempt_disable/enable_notrace() in
fprobe_return().
Link: https://lore.kernel.org/all/175374642359.1471729.1054175011228386560.stgit@mhiramat.tok.corp.google.com/
Fixes: 4346ba160409 ("fprobe: Rewrite fprobe on function-graph tracer")
Cc: stable@vger.kernel.org
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
Delete fexit_noreturns.c files and migrate the cases into
tracing_failure.c files.
The result:
$ tools/testing/selftests/bpf/test_progs -t tracing_failure/fexit_noreturns
#467/4 tracing_failure/fexit_noreturns:OK
#467 tracing_failure:OK
Summary: 1/1 PASSED, 0 SKIPPED, 0 FAILED
Signed-off-by: KaFai Wan <kafai.wan@linux.dev>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20250724151454.499040-5-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
The result:
$ tools/testing/selftests/bpf/test_progs -t tracing_failure/tracing_deny
#468/3 tracing_failure/tracing_deny:OK
#468 tracing_failure:OK
Summary: 1/1 PASSED, 0 SKIPPED, 0 FAILED
Signed-off-by: KaFai Wan <kafai.wan@linux.dev>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20250724151454.499040-4-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Show the rejected function name when attaching tracing programs to
functions in deny list.
With this change, we know why tracing programs can't attach to functions
like __rcu_read_lock() from log.
$ ./fentry
libbpf: prog '__rcu_read_lock': BPF program load failed: -EINVAL
libbpf: prog '__rcu_read_lock': -- BEGIN PROG LOAD LOG --
Attaching tracing programs to function '__rcu_read_lock' is rejected.
Suggested-by: Leon Hwang <leon.hwang@linux.dev>
Signed-off-by: KaFai Wan <kafai.wan@linux.dev>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20250724151454.499040-3-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
With this change, we know the precise rejected function name when
attaching fexit/fmod_ret to __noreturn functions from log.
$ ./fexit
libbpf: prog 'fexit': BPF program load failed: -EINVAL
libbpf: prog 'fexit': -- BEGIN PROG LOAD LOG --
Attaching fexit/fmod_ret to __noreturn function 'do_exit' is rejected.
Suggested-by: Leon Hwang <leon.hwang@linux.dev>
Signed-off-by: KaFai Wan <kafai.wan@linux.dev>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20250724151454.499040-2-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
audit_policy() does not support any other algorithm, so the crypto_shash
abstraction provides no value. Just use the SHA-256 library API
instead, which is much simpler and easier to use.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Fan Wu <wufan@kernel.org>
|
|
DTS example in the bindings should be indented with 2- or 4-spaces and
aligned with opening '- |', so correct any differences like 3-spaces or
mixtures 2- and 4-spaces in one binding.
No functional changes here, but saves some comments during reviews of
new patches built on existing code.
Acked-by: Ulf Hansson <ulf.hansson@linaro.org> # For MMC
Acked-by: Lee Jones <lee@kernel.org>
Acked-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> # renesas
Link: https://lore.kernel.org/r/20250107131456.247610-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20250725100241.120106-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
|
|
When gmin_get_config_var() calls efi.get_variable() and the EFI variable
is larger than the expected buffer size, two behaviors combine to create
a stack buffer overflow:
1. gmin_get_config_var() does not return the proper error code when
efi.get_variable() fails. It returns the stale 'ret' value from
earlier operations instead of indicating the EFI failure.
2. When efi.get_variable() returns EFI_BUFFER_TOO_SMALL, it updates
*out_len to the required buffer size but writes no data to the output
buffer. However, due to bug #1, gmin_get_var_int() believes the call
succeeded.
The caller gmin_get_var_int() then performs:
- Allocates val[CFG_VAR_NAME_MAX + 1] (65 bytes) on stack
- Calls gmin_get_config_var(dev, is_gmin, var, val, &len) with len=64
- If EFI variable is >64 bytes, efi.get_variable() sets len=required_size
- Due to bug #1, thinks call succeeded with len=required_size
- Executes val[len] = 0, writing past end of 65-byte stack buffer
This creates a stack buffer overflow when EFI variables are larger than
64 bytes. Since EFI variables can be controlled by firmware or system
configuration, this could potentially be exploited for code execution.
Fix the bug by returning proper error codes from gmin_get_config_var()
based on EFI status instead of stale 'ret' value.
The gmin_get_var_int() function is called during device initialization
for camera sensor configuration on Intel Bay Trail and Cherry Trail
platforms using the atomisp camera stack.
Reported-by: zepta <z3ptaa@gmail.com>
Closes: https://lore.kernel.org/all/CAPBS6KoQyM7FMdPwOuXteXsOe44X4H3F8Fw+y_qWq6E+OdmxQA@mail.gmail.com
Fixes: 38d4f74bc148 ("media: atomisp_gmin_platform: stop abusing efivar API")
Reviewed-by: Hans de Goede <hansg@kernel.org>
Link: https://lore.kernel.org/r/20250724080756.work.741-kees@kernel.org
Signed-off-by: Kees Cook <kees@kernel.org>
|
|
Fix a typo that uses ',' instead of ';' for line delimiter.
Link: https://lore.kernel.org/linux-trace-kernel/175366879192.487099.5714468217360139639.stgit@mhiramat.tok.corp.google.com/
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
When introduced in commit 9eb22f7fedfc ("fs: add ioctl to query metadata
and protection info capabilities") the stub of blk_get_meta_cap() for
!BLK_DEV_INTEGRITY always returns -EOPNOTSUPP. The motivation was that
while the command was unsupported in that configuration it was still
recognized.
A later change instead assumed -ENOIOCTLCMD as is required for unknown
ioctl commands per Documentation/driver-api/ioctl.rst. The result being
that on !BLK_DEV_INTEGRITY configs, any ioctl which reaches
blkdev_common_ioctl() will return -EOPNOTSUPP.
Change the stub to return -ENOIOCTLCMD, fixing the issue and better
matching with expectations.
[ The blkdev_common_ioctl() confusion has been fixed, but -ENOIOCTLCMD
is the right thing to return for unrecognized ioctls, so the patch
remains the right thing to do. - Linus ]
Link: https://lore.kernel.org/lkml/CACzX3AsRd__fXb9=CJPTTJC494SDnYAtYrN2=+bZgMCvM6UQDg@mail.gmail.com
Fixes: 42b0ef01e6b5 ("block: fix FS_IOC_GETLBMD_CAP parsing in blkdev_common_ioctl()")
Signed-off-by: Klara Modin <klarasmodin@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Remove incorrect page alignment check for the writeback len arg in
fuse_iomap_writeback_range(). len will always be block-aligned as
passed in by iomap.
On regular fuse filesystems, i_blkbits is set to PAGE_SHIFT so this is
not a problem but for fuseblk filesystems, the block size is set to a
default of 512 bytes or a block size passed in at mount time.
Please note that non-page-aligned lengths are fine for the logic in
fuse_iomap_writeback_range(). The check was originally added as a
safeguard to detect conspicuously wrong ranges.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Fixes: ef7e7cbb323f ("fuse: use iomap for writeback")
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Link: https://lore.kernel.org/linux-fsdevel/CA+G9fYs5AdVM-T2Tf3LciNCwLZEHetcnSkHsjZajVwwpM2HmJw@mail.gmail.com/
Reported-by: Sasha Levin <sashal@kernel.org>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Add a per-cpu monitor as part of the sched model:
* opid: operations with preemption and irq disabled
Monitor to ensure wakeup and need_resched occur with irq and
preemption disabled or in irq handlers.
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Link: https://lore.kernel.org/20250728135022.255578-10-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Acked-by: Nam Cao <namcao@linutronix.de>
Tested-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Add 2 per-task monitors as part of the sched model:
* nrp: need-resched preempts
Monitor to ensure preemption requires need resched.
* sssw: set state sleep and wakeup
Monitor to ensure sched_set_state to sleepable leads to sleeping and
sleeping tasks require wakeup.
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/20250728135022.255578-9-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Acked-by: Nam Cao <namcao@linutronix.de>
Tested-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
The tss monitor currently guarantees task switches can happen only while
scheduling, whereas the sncid monitor enforces scheduling occurs with
interrupt disabled.
Replace the monitors with a more comprehensive specification which
implies both but also ensures that:
* each scheduler call disable interrupts to switch
* each task switch happens with interrupts disabled
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Nam Cao <namcao@linutronix.de>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/20250728135022.255578-8-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Add the following tracepoint:
* sched_set_need_resched(tsk, cpu, tif)
Called when a task is set the need resched [lazy] flag
Remove the unused ip parameter from sched_entry and sched_exit and alter
sched_entry to have a value of preempt consistent with the one used in
sched_switch.
Also adapt all monitors using sched_{entry,exit} to avoid breaking build.
These tracepoints are useful to describe the Linux task model and are
adapted from the patches by Daniel Bristot de Oliveira
(https://bristot.me/linux-task-model/).
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Nam Cao <namcao@linutronix.de>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Link: https://lore.kernel.org/20250728135022.255578-7-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
DA monitor can be accessed from multiple cores simultaneously, this is
likely, for instance when dealing with per-task monitors reacting on
events that do not always occur on the CPU where the task is running.
This can cause race conditions where two events change the next state
and we see inconsistent values. E.g.:
[62] event_srs: 27: sleepable x sched_wakeup -> running (final)
[63] event_srs: 27: sleepable x sched_set_state_sleepable -> sleepable
[63] error_srs: 27: event sched_switch_suspend not expected in the state running
In this case the monitor fails because the event on CPU 62 wins against
the one on CPU 63, although the correct state should have been
sleepable, since the task get suspended.
Detect if the current state was modified by using try_cmpxchg while
storing the next value. If it was, try again reading the current state.
After a maximum number of failed retries, react by calling a special
tracepoint, print on the console and reset the monitor.
Remove the functions da_monitor_curr_state() and da_monitor_set_state()
as they only hide the underlying implementation in this case.
Monitors where this type of condition can occur must be able to account
for racing events in any possible order, as we cannot know the winner.
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/20250728135022.255578-6-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Reviewed-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
RV monitors relying on the preemptirqs tracepoints are set as dependent
on PREEMPT_TRACER and IRQSOFF_TRACER. In fact, those configurations do
enable the tracepoints but are not the minimal configurations enabling
them, which are TRACE_PREEMPT_TOGGLE and TRACE_IRQFLAGS (not selectable
manually).
Set TRACE_PREEMPT_TOGGLE and TRACE_IRQFLAGS as dependencies for
monitors.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Link: https://lore.kernel.org/20250728135022.255578-5-gmonaco@redhat.com
Fixes: fbe6c09b7eb4 ("rv: Add scpd, snep and sncid per-cpu monitors")
Acked-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Using DA monitors tracepoints with KASAN enabled triggers the following
warning:
BUG: KASAN: global-out-of-bounds in do_trace_event_raw_event_event_da_monitor+0xd6/0x1a0
Read of size 32 at addr ffffffffaada8980 by task ...
Call Trace:
<TASK>
[...]
do_trace_event_raw_event_event_da_monitor+0xd6/0x1a0
? __pfx_do_trace_event_raw_event_event_da_monitor+0x10/0x10
? trace_event_sncid+0x83/0x200
trace_event_sncid+0x163/0x200
[...]
The buggy address belongs to the variable:
automaton_snep+0x4e0/0x5e0
This is caused by the tracepoints reading 32 bytes __array instead of
__string from the automata definition. Such strings are literals and
reading 32 bytes ends up in out of bound memory accesses (e.g. the next
automaton's data in this case).
The error is harmless as, while printing the string, we stop at the null
terminator, but it should still be fixed.
Use the __string facilities while defining the tracepoints to avoid
reading out of bound memory.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Link: https://lore.kernel.org/20250728135022.255578-4-gmonaco@redhat.com
Fixes: 792575348ff7 ("rv/include: Add deterministic automata monitor definition via C macros")
Reviewed-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
RV event tracepoints print a line with the format:
"event_xyz: S0 x event -> S1 "
"event_xyz: S1 x event -> S0 (final)"
While printing an event leading to a non-final state, the line
has a trailing white space (visible above before the closing ").
Adapt the format string not to print the trailing whitespace if we are
not printing "(final)".
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Link: https://lore.kernel.org/20250728135022.255578-3-gmonaco@redhat.com
Reviewed-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
The RV da_monitor API allows to start monitors in two ways:
da_handle_start_event_NAME and da_handle_start_run_event_NAME.
The former is used when the event is followed by the initial state of
the module, so we ignore the event but we know the monitor is in the
initial state and can start monitoring, the latter can be used if the
event can only occur in the initial state, so we do handle the event as
if the monitor was in the initial state.
This latter API is defined for implicit monitors but not per-task ones.
Define da_handle_start_run_event_NAME macro also for per-task monitors.
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Link: https://lore.kernel.org/20250728135022.255578-2-gmonaco@redhat.com
Reviewed-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
This patch fixes several minor typos in comments within the BPF verifier.
No changes in functionality.
Signed-off-by: Suchit Karunakaran <suchitkarunakaran@gmail.com>
Link: https://lore.kernel.org/r/20250727081754.15986-1-suchitkarunakaran@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Commit d7f008738171 ("bpf: try harder to deduce register bounds from
different numeric domains") added a second call to __reg_deduce_bounds
in reg_bounds_sync because a single call wasn't enough to converge to a
fixed point in terms of register bounds.
With patch "bpf: Improve bounds when s64 crosses sign boundary" from
this series, Eduard noticed that calling __reg_deduce_bounds twice isn't
enough anymore to converge. The first selftest added in "selftests/bpf:
Test cross-sign 64bits range refinement" highlights the need for a third
call to __reg_deduce_bounds. After instruction 7, reg_bounds_sync
performs the following bounds deduction:
reg_bounds_sync entry: scalar(smin=-655,smax=0xeffffeee,smin32=-783,smax32=-146)
__update_reg_bounds: scalar(smin=-655,smax=0xeffffeee,smin32=-783,smax32=-146)
__reg_deduce_bounds:
__reg32_deduce_bounds: scalar(smin=-655,smax=0xeffffeee,smin32=-783,smax32=-146,umin32=0xfffffcf1,umax32=0xffffff6e)
__reg64_deduce_bounds: scalar(smin=-655,smax=0xeffffeee,smin32=-783,smax32=-146,umin32=0xfffffcf1,umax32=0xffffff6e)
__reg_deduce_mixed_bounds: scalar(smin=-655,smax=0xeffffeee,umin=umin32=0xfffffcf1,umax=0xffffffffffffff6e,smin32=-783,smax32=-146,umax32=0xffffff6e)
__reg_deduce_bounds:
__reg32_deduce_bounds: scalar(smin=-655,smax=0xeffffeee,umin=umin32=0xfffffcf1,umax=0xffffffffffffff6e,smin32=-783,smax32=-146,umax32=0xffffff6e)
__reg64_deduce_bounds: scalar(smin=-655,smax=smax32=-146,umin=0xfffffffffffffd71,umax=0xffffffffffffff6e,smin32=-783,umin32=0xfffffcf1,umax32=0xffffff6e)
__reg_deduce_mixed_bounds: scalar(smin=-655,smax=smax32=-146,umin=0xfffffffffffffd71,umax=0xffffffffffffff6e,smin32=-783,umin32=0xfffffcf1,umax32=0xffffff6e)
__reg_bound_offset: scalar(smin=-655,smax=smax32=-146,umin=0xfffffffffffffd71,umax=0xffffffffffffff6e,smin32=-783,umin32=0xfffffcf1,umax32=0xffffff6e,var_off=(0xfffffffffffffc00; 0x3ff))
__update_reg_bounds: scalar(smin=-655,smax=smax32=-146,umin=0xfffffffffffffd71,umax=0xffffffffffffff6e,smin32=-783,umin32=0xfffffcf1,umax32=0xffffff6e,var_off=(0xfffffffffffffc00; 0x3ff))
In particular, notice how:
1. In the first call to __reg_deduce_bounds, __reg32_deduce_bounds
learns new u32 bounds.
2. __reg64_deduce_bounds is unable to improve bounds at this point.
3. __reg_deduce_mixed_bounds derives new u64 bounds from the u32 bounds.
4. In the second call to __reg_deduce_bounds, __reg64_deduce_bounds
improves the smax and umin bounds thanks to patch "bpf: Improve
bounds when s64 crosses sign boundary" from this series.
5. Subsequent functions are unable to improve the ranges further (only
tnums). Yet, a better smin32 bound could be learned from the smin
bound.
__reg32_deduce_bounds is able to improve smin32 from smin, but for that
we need a third call to __reg_deduce_bounds.
As discussed in [1], there may be a better way to organize the deduction
rules to learn the same information with less calls to the same
functions. Such an optimization requires further analysis and is
orthogonal to the present patchset.
Link: https://lore.kernel.org/bpf/aIKtSK9LjQXB8FLY@mail.gmail.com/ [1]
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Co-developed-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Link: https://lore.kernel.org/r/79619d3b42e5525e0e174ed534b75879a5ba15de.1753695655.git.paul.chaignon@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
The improvement of the u64/s64 range refinement fixed the invariant
violation that was happening on this test for BPF_JSLT when crossing the
sign boundary.
After this patch, we have one test remaining with a known invariant
violation. It's the same test as fixed here but for 32 bits ranges.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Link: https://lore.kernel.org/r/ad046fb0016428f1a33c3b81617aabf31b51183f.1753695655.git.paul.chaignon@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
This patch adds coverage for the new cross-sign 64bits range refinement
logic. The three tests cover the cases when the u64 and s64 ranges
overlap (1) in the negative portion of s64, (2) in the positive portion
of s64, and (3) in both portions.
The first test is a simplified version of a BPF program generated by
syzkaller that caused an invariant violation [1]. It looks like
syzkaller could not extract the reproducer itself (and therefore didn't
report it to the mailing list), but I was able to extract it from the
console logs of a crash.
The principle is similar to the invariant violation described in
commit 6279846b9b25 ("bpf: Forget ranges when refining tnum after
JSET"): the verifier walks a dead branch, uses the condition to refine
ranges, and ends up with inconsistent ranges. In this case, the dead
branch is when we fallthrough on both jumps. The new refinement logic
improves the bounds such that the second jump is properly detected as
always-taken and the verifier doesn't end up walking a dead branch.
The second and third tests are inspired by the first, but rely on
condition jumps to prepare the bounds instead of ALU instructions. An
R10 write is used to trigger a verifier error when the bounds can't be
refined.
Link: https://syzkaller.appspot.com/bug?extid=c711ce17dd78e5d4fdcf [1]
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Link: https://lore.kernel.org/r/a0e17b00dab8dabcfa6f8384e7e151186efedfdd.1753695655.git.paul.chaignon@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
This patch updates the range refinement logic in the reg_bound test to
match the new logic from the previous commit. Without this change, tests
would fail because we end with more precise ranges than the tests
expect.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Link: https://lore.kernel.org/r/b7f6b1fbe03373cca4e1bb6a113035a6cd2b3ff7.1753695655.git.paul.chaignon@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
__reg64_deduce_bounds currently improves the s64 range using the u64
range and vice versa, but only if it doesn't cross the sign boundary.
This patch improves __reg64_deduce_bounds to cover the case where the
s64 range crosses the sign boundary but overlaps with the u64 range on
only one end. In that case, we can improve both ranges. Consider the
following example, with the s64 range crossing the sign boundary:
0 U64_MAX
| [xxxxxxxxxxxxxx u64 range xxxxxxxxxxxxxx] |
|----------------------------|----------------------------|
|xxxxx s64 range xxxxxxxxx] [xxxxxxx|
0 S64_MAX S64_MIN -1
The u64 range overlaps only with positive portion of the s64 range. We
can thus derive the following new s64 and u64 ranges.
0 U64_MAX
| [xxxxxx u64 range xxxxx] |
|----------------------------|----------------------------|
| [xxxxxx s64 range xxxxx] |
0 S64_MAX S64_MIN -1
The same logic can probably apply to the s32/u32 ranges, but this patch
doesn't implement that change.
In addition to the selftests, the __reg64_deduce_bounds change was
also tested with Agni, the formal verification tool for the range
analysis [1].
Link: https://github.com/bpfverif/agni [1]
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Link: https://lore.kernel.org/r/933bd9ce1f36ded5559f92fdc09e5dbc823fa245.1753695655.git.paul.chaignon@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
The caller has already passed in the memslot, and there are
two instances `{kvm_faultin_pfn/mark_page_dirty}` of retrieving
the memslot again in `kvm_riscv_gstage_map`, we can replace them
with `{__kvm_faultin_pfn/mark_page_dirty_in_slot}`.
Signed-off-by: Quan Zhou <zhouquan@iscas.ac.cn>
Reviewed-by: Anup Patel <anup@brainfault.org>
Link: https://lore.kernel.org/r/50989f0a02790f9d7dc804c2ade6387c4e7fbdbc.1749634392.git.zhouquan@iscas.ac.cn
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
There is already a helper function find_vma_intersection() in KVM
for searching intersecting VMAs, use it directly.
Signed-off-by: Quan Zhou <zhouquan@iscas.ac.cn>
Reviewed-by: Anup Patel <anup@brainfault.org>
Link: https://lore.kernel.org/r/230d6c8c8b8dd83081fcfd8d83a4d17c8245fa2f.1731552790.git.zhouquan@iscas.ac.cn
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
For `perf kvm stat` on the RISC-V, in order to avoid the
occurrence of `UNKNOWN` event names, interrupts should be
reported in addition to exceptions.
testing without patch:
Event name Samples Sample% Time(ns)
--------------------------- -------- -------- ------------
STORE_GUEST_PAGE_FAULT 1496461 53.00% 889612544
UNKNOWN 887514 31.00% 272857968
LOAD_GUEST_PAGE_FAULT 305164 10.00% 189186331
VIRTUAL_INST_FAULT 70625 2.00% 134114260
SUPERVISOR_SYSCALL 32014 1.00% 58577110
INST_GUEST_PAGE_FAULT 1 0.00% 2545
testing with patch:
Event name Samples Sample% Time(ns)
--------------------------- -------- -------- ------------
IRQ_S_TIMER 211271 58.00% 738298680600
EXC_STORE_GUEST_PAGE_FAULT 111279 30.00% 130725914800
EXC_LOAD_GUEST_PAGE_FAULT 22039 6.00% 25441480600
EXC_VIRTUAL_INST_FAULT 8913 2.00% 21015381600
IRQ_VS_EXT 4748 1.00% 10155464300
IRQ_S_EXT 2802 0.00% 13288775800
IRQ_S_SOFT 1998 0.00% 4254129300
Signed-off-by: Quan Zhou <zhouquan@iscas.ac.cn>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Link: https://lore.kernel.org/r/9693132df4d0f857b8be3a75750c36b40213fcc0.1726211632.git.zhouquan@iscas.ac.cn
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Enable ring-based dirty memory tracking on riscv:
- Enable CONFIG_HAVE_KVM_DIRTY_RING_ACQ_REL as riscv is weakly
ordered.
- Set KVM_DIRTY_LOG_PAGE_OFFSET for the ring buffer's physical page
offset.
- Add a check to kvm_vcpu_kvm_riscv_check_vcpu_requests for checking
whether the dirty ring is soft full.
To handle vCPU requests that cause exits to userspace, modified the
`kvm_riscv_check_vcpu_requests` to return a value (currently only
returns 0 or 1).
Signed-off-by: Quan Zhou <zhouquan@iscas.ac.cn>
Reviewed-by: Anup Patel <anup@brainfault.org>
Link: https://lore.kernel.org/r/20e116efb1f7aff211dd8e3cf8990c5521ed5f34.1749810735.git.zhouquan@iscas.ac.cn
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The Smnpm extension requires special handling because the guest ISA
extension maps to a different extension (Ssnpm) on the host side.
commit 1851e7836212 ("RISC-V: KVM: Allow Smnpm and Ssnpm extensions for
guests") missed that the vcpu->arch.isa bit is based only on the host
extension, so currently both KVM_RISCV_ISA_EXT_{SMNPM,SSNPM} map to
vcpu->arch.isa[RISCV_ISA_EXT_SSNPM]. This does not cause any problems
for the guest, because both extensions are force-enabled anyway when the
host supports Ssnpm, but prevents checking for (guest) Smnpm in the SBI
FWFT logic.
Redefine kvm_isa_ext_arr to look up the guest extension, since only the
guest -> host mapping is unambiguous. Factor out the logic for checking
for host support of an extension, so this special case only needs to be
handled in one place, and be explicit about which variables hold a host
vs a guest ISA extension.
Fixes: 1851e7836212 ("RISC-V: KVM: Allow Smnpm and Ssnpm extensions for guests")
Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Link: https://lore.kernel.org/r/20250111004702.2813013-2-samuel.holland@sifive.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Delegate illegal instruction fault to VS mode by default to avoid such
exceptions being trapped to HS and redirected back to VS.
The delegation of illegal instruction fault is particularly important
to guest applications that use vector instructions frequently. In such
cases, an illegal instruction fault will be raised when guest user thread
uses vector instruction the first time and then guest kernel will enable
user thread to execute following vector instructions.
The fw pmu event counter remains undeleted so that guest can still query
illegal instruction events via sbi call. Guest will only see zero count
on illegal instruction faults and know 'firmware' has delegated it.
Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Xu Lu <luxu.kernel@bytedance.com>
Link: https://lore.kernel.org/r/20250714094554.89151-1-luxu.kernel@bytedance.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Currently, all kvm_riscv_hfence_xyz() APIs assume VMID to be the
host VMID of the Guest/VM which resticts use of these APIs only
for host TLB maintenance. Let's allow passing VMID as a parameter
to all kvm_riscv_hfence_xyz() APIs so that they can be re-used
for nested virtualization related TLB maintenance.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-13-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The upcoming nested virtualization can share g-stage page table
management with the current host g-stage implementation hence
factor-out g-stage page table management as separate sources
and also use "kvm_riscv_mmu_" prefix for host g-stage functions.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-12-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Currently, the struct kvm_riscv_hfence does not have vmid field
and various hfence processing functions always pick vmid assigned
to the guest/VM. This prevents us from doing hfence operation on
arbitrary vmid hence add vmid field to struct kvm_riscv_hfence
and use it wherever applicable.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-11-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Introduce struct kvm_gstage_mapping which represents a g-stage
mapping at a particular g-stage page table level. Also, update
the kvm_riscv_gstage_map() to return the g-stage mapping upon
success.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-10-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The MMU, TLB, and VMID management for KVM RISC-V already exists as
seprate sources so create separate headers along these lines. This
further simplifies asm/kvm_host.h header.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-9-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The H-extension CSRs accessed by kvm_riscv_vcpu_trap_redirect() will
trap when KVM RISC-V is running as Guest/VM hence remove these traps
by using ncsr_xyz() instead of csr_xyz().
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-8-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The kvm_arch_flush_remote_tlbs_range() expected by KVM core can be
easily implemented for RISC-V using kvm_riscv_hfence_gvma_vmid_gpa()
hence provide it.
Also with kvm_arch_flush_remote_tlbs_range() available for RISC-V, the
mmu_wp_memory_region() can happily use kvm_flush_remote_tlbs_memslot()
instead of kvm_flush_remote_tlbs().
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-7-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The gstage_set_pte() and gstage_op_pte() should flush TLB only when
a leaf PTE changes so that unnecessary TLB flushes can be avoided.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-6-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The KVM_REQ_HFENCE_GVMA_VMID_ALL is same as KVM_REQ_TLB_FLUSH so
to avoid confusion let's replace KVM_REQ_HFENCE_GVMA_VMID_ALL with
KVM_REQ_TLB_FLUSH. Also, rename kvm_riscv_hfence_gvma_vmid_all_process()
to kvm_riscv_tlb_flush_process().
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-5-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>
|
Daniel Borkmann
Kees Cook
Colin Ian King
Oliver Upton
Zheng Yu
Steven Rostedt
Masami Hiramatsu (Google)
KaFai Wan
Eric Biggers
Krzysztof Kozlowski
Klara Modin
Joanne Koong
Gabriele Monaco
Suchit Karunakaran
Paul Chaignon
Quan Zhou
Samuel Holland
Xu Lu
Anup Patel