2025-10-02T14:56:23Z Nam Cao :
> Thomas Weißschuh writes:
>> I am wondering if it would make sense to add a new tracepoint that
>> fires in addition of the reactors. That would allow multiple
>> simultaneous consumers and also bespoke handlers in userspace.
>
> We do have tracepoints for each mon
sense.
Reviewed-by: Gabriele Monaco
Thanks,
Gabriele
>
> Fixes: 9162620eb604 ("rv: Add rtapp_pagefault monitor")
> Signed-off-by: Nam Cao
> Reported-by: kernel test robot
> Closes:
> https://lore.kernel.org/oe-kbuild-all/202509260455.6z9vkty4-...@intel.com/
>
as reported by Nathan.
>
> Convert everything to use struct list_head * as iterator. This also makes
> enabled_monitors consistent with available_monitors.
>
Looks good to me and passes my tests.
Reviewed-by: Gabriele Monaco
Thanks,
Gabriele
> Fixes: de090d1ccae
On Thu, 2025-10-02 at 11:14 +0200, Nam Cao wrote:
> Gabriele Monaco writes:
> > +/*
> > + * da_get_task - return the task associated to the monitor
> > + */
> > +static inline struct task_struct *da_get_task(struct da_monitor *da_mon)
> > +{
> > + return
On Thu, 2025-10-02 at 13:03 +0200, Nam Cao wrote:
> Gabriele Monaco writes:
>
> > Currently the automata parser assumes event strings don't have any
> > space, this stands true for event names, but can be a wrong assumption
> > if we want to store other informat
the
printk reactor works. Also validate disabling reacting_on or
monitoring_on prevents reactions.
These tests use the ftracetest suite.
Signed-off-by: Gabriele Monaco
---
MAINTAINERS | 1 +
.../testing/selftests/verification/.gitignore | 2 +
tools
names like initialize_ftrace to initialize_system.
Add the --rv argument to set up the test for rv, basically changing the
trace directory to $TRACING_DIR/rv and displaying an error if that
cannot be found.
This prepares for rv selftests inclusion.
Signed-off-by: Gabriele Monaco
---
tools/testing
in [1].
[1] - https://lore.kernel.org/lkml/87tt0t4u19@yellow.woof
To: Steven Rostedt
To: Nam Cao
Cc: Masami Hiramatsu
Cc: John Kacur
Cc: Waylon Cude
Cc: linux-trace-kernel@vger.kernel.org
Cc: linux-kselft...@vger.kernel.org
Gabriele Monaco (2):
selftest/ftrace: Generalise ftracetest to
On Tue, 2025-09-23 at 10:44 +0200, Nam Cao wrote:
> Gabriele Monaco writes:
> > Looks reasonable, can you work on the fix?
> > I see Steve is out for conferences so this won't be too urgent.
>
> The fix is in my previous email.
I confirm enabled_monitors works with you
On Tue, 2025-09-23 at 07:28 +0200, Nam Cao wrote:
> Hi Nathan,
>
Thanks for finding this!
> Nathan Chancellor writes:
> > I am seeing a crash when reading from
> > /sys/kernel/tracing/rv/enabled_monitors
> > on a couple of my arm64 boxes running Fedora after this change, which
> > landed in mai
l.org/lkml/20250814150809.140739-1-gmon...@redhat.com
To: Steven Rostedt
To: Nam Cao
Cc: Tomas Glozar
Cc: Juri Lelli
Cc: Clark Williams
Cc: John Kacur
Cc: linux-trace-kernel@vger.kernel.org
Gabriele Monaco (20):
rv: Refactor da_monitor to minimise macros
rv: Cleanup da_monitor after refactor
rv
The monitor storage (e.g. the rv monitor, pointer to the target, etc.)
is stored in a hash table indexed by id. Monitor storage objects are
automatically allocated unless specified otherwise (e.g. if the creation
context is unsafe for allocation).
Signed-off-by: Gabriele Monaco
---
include/linux
: Gabriele Monaco
---
tools/verification/rvgen/rvgen/dot2k.py | 6 +++--
.../rvgen/rvgen/templates/dot2k/main.c| 25 +++
2 files changed, 13 insertions(+), 18 deletions(-)
diff --git a/tools/verification/rvgen/rvgen/dot2k.py
b/tools/verification/rvgen/rvgen/dot2k.py
ng on
PREEMPT_RT only and allows the monitor to be built on kernels without
the preemptirqs tracepoints.
[1] - https://lore.kernel.org/lkml/20250625120823.60600-1-gmon...@redhat.com
Signed-off-by: Gabriele Monaco
---
Documentation/trace/rv/monitor_sched.rst | 62 +++-
kernel/trace/rv/monitors
2025-09-19T15:36:50Z Steven Rostedt :
> On Fri, 19 Sep 2025 16:09:50 +0200
> Gabriele Monaco wrote:
>
>> The tracepoints sched_entry, sched_exit and sched_set_need_resched
>> are not exported to tracefs as trace events, this allows only kernel
>> code to access them.
y) on state and events.
Signed-off-by: Gabriele Monaco
---
include/linux/rv.h | 38 +++
include/rv/da_monitor.h| 59 -
include/rv/ha_monitor.h| 469 +
kernel/trace/rv/Kconfig| 13 +
kernel/trace/rv/rv_trace.h | 63 +
5 files changed
ge in struct
da_monitor is fixed to unsigned int.
Remove empty and no longer required macros and substitute the type
parameter with the appropriate enum.
Additionally break long line and adjust the format overall.
Signed-off-by: Gabriele Monaco
---
include/rv/automata.h | 24 +++-
inclu
es or events are added.
Remove the assignment to 0 and add a comma to last elements, this
simplifies the logic for the code generator.
Signed-off-by: Gabriele Monaco
---
kernel/trace/rv/monitors/nrp/nrp.h | 20 +++---
kernel/trace/rv/monitors/opid/opid.h| 22 +++
kernel/
Add a sample monitor to showcase hybrid/timed automata.
The stall monitor identifies tasks stalled for longer than a threshold
and reacts when that happens.
Signed-off-by: Gabriele Monaco
---
Documentation/tools/rv/index.rst | 1 +
Documentation/tools/rv/rv-mon-stall.rst
monitors.
Signed-off-by: Gabriele Monaco
---
include/rv/da_monitor.h | 303 +++-
1 file changed, 142 insertions(+), 161 deletions(-)
diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h
index 7c3540d0f64e..05b216f90f93 100644
--- a/include/rv
deadiline
Signed-off-by: Gabriele Monaco
---
Documentation/trace/rv/index.rst | 1 +
Documentation/trace/rv/monitor_deadline.rst | 151 ++
kernel/trace/rv/Kconfig | 5 +
kernel/trace/rv/Makefile | 3 +
kernel/trace/rv
hard):
Called when a deadline server is stopped (hard) or put to idle
waiting for the next period (!hard)
Those tracepoints can be useful to validate the deadline scheduler with
RV and are not exported to tracefs.
Signed-off-by: Gabriele Monaco
---
include/trace/events/sched.h
.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/dot2k.py | 6 ++
tools/verification/rvgen/rvgen/generator.py | 2 +-
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/tools/verification/rvgen/rvgen/dot2k.py
b/tools/verification/rvgen/rvgen/dot2k.py
index
tracepoints being exported.
Export the 3 not exported tracepoints.
Note that sched_set_state is already exported as the macro is called
from modules.
[1] - https://github.com/qais-yousef/sched_tp.git
Signed-off-by: Gabriele Monaco
---
kernel/sched/core.c | 3 +++
1 file changed, 3 insertions(+)
diff
(separated by \n).
The tool can now generate the appropriate code to validate constraints
at runtime according to the dot specification.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/__main__.py | 8 +-
tools/verification/rvgen/rvgen/automata.py| 141 +-
tools
Describe theory and implementation of hybrid automata in the dedicated
page hybrid_automata.rst
Include a section on how to integrate a hybrid automaton in
monitor_synthesis.rst
Also remove a hanging $ in deterministic_automata.rst
Signed-off-by: Gabriele Monaco
---
.../trace/rv
rings.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/automata.py | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/tools/verification/rvgen/rvgen/automata.py
b/tools/verification/rvgen/rvgen/automata.py
index 3f06aef8d4fd..977ba859c34e 100644
--- a/
Functions in automata.py, dot2c.py and dot2k.py don't have type
annotations and it can get complicated to remember how to use them.
Add minimal type annotations.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/automata.py | 20 +--
tools/verification/rvgen/
str.join() can do what __buff_to_string() does. Therefore replace
__buff_to_string() to make the scripts more pythonic.
Also clean and remove some intermediate functions.
Reviewed-by: Nam Cao
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/dot2c.py | 35
The monitor container source files contained a declaration and a
definition for the rv_monitor variable. The former is superfluous and
can be removed.
Remove the variable declaration from the template as well as the
existing monitor containers.
Reviewed-by: Nam Cao
Signed-off-by: Gabriele
documentation to reflect the changes.
Signed-off-by: Gabriele Monaco
---
Documentation/trace/rv/monitor_synthesis.rst | 44 ++--
1 file changed, 21 insertions(+), 23 deletions(-)
diff --git a/Documentation/trace/rv/monitor_synthesis.rst
b/Documentation/trace/rv
mpty macro definitions to ease review of this change with
diff tools, but cleanup is required.
Also adapt existing monitors to keep the build working.
Signed-off-by: Gabriele Monaco
---
include/linux/rv.h | 4 +
include/rv/automata.h | 132 ++--
include/rv/da
On Fri, 2025-09-19 at 12:49 +0200, Thomas Weißschuh wrote:
> Reactions of the existing reactors are not easily detectable from programs
> and also not easily attributable to the triggering task.
> This makes it hard to test the monitors themselves programmatically.
>
> The signal reactors allows a
On Thu, 2025-09-18 at 11:48 +0200, Tomas Glozar wrote:
> čt 18. 9. 2025 v 10:36 odesílatel Gabriele Monaco napsal:
> >
> > Yeah totally, I have the feeling that with the kernel there's no such a
> > thing as a "theoretical bug", kinda like a good cons
On Wed, 2025-09-17 at 16:07 +0200, Nam Cao wrote:
> Gabriele Monaco writes:
> > On Wed, 2025-09-17 at 12:57 +, Xiu Jianfeng wrote:
> > > From: Xiu Jianfeng
> > >
> > > When booting kernel with lockdown=confidentiality parameter, the
> > > sys
On Wed, 2025-09-17 at 12:57 +, Xiu Jianfeng wrote:
> From: Xiu Jianfeng
>
> When booting kernel with lockdown=confidentiality parameter, the
> system
> will hang at rv_register_reactor() due to waiting for
> rv_interface_lock,
> as shown in the following log,
>
Thanks for finding this, the
: Frederic Weisbecker
Signed-off-by: Gabriele Monaco
---
include/trace/events/timer_migration.h | 4 ++--
kernel/time/timer_migration.c | 22 +++---
kernel/time/timer_migration.h | 2 +-
3 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/include
> However, ltl_monitor was not adapted, resulting in a faulty callback.
> This also resulted in an lkp build warning due to
> -Wincompatible-pointer-types.
>
Looks good to me, thanks!
Reviewed-by: Gabriele Monaco
Adding Nam (author of LTL) to the loop.
Gabriele
> Fixes: edd3cb
On Thu, 2025-09-04 at 10:20 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:08:08PM +0200, Gabriele Monaco wrote:
> > + def fill_per_obj_definitions(self) -> list:
> > + if self.monitor_type == "per_obj":
> > + return [""&qu
On Thu, 2025-09-04 at 09:19 +0200, Nam Cao wrote:
> Am I hallucinating, or the fix tag is already there?
I should stop doing 3 things at the same time, my bad :')
>
> > (Adding Nam to the loop, as author of that patch)
>
> Thanks for CC!
>
> Reviewed-by: Nam Cao
>
> I have been debating whet
or path jump to out_unlock, ensuring that the
> mutex is always released before returning.
>
> Fixes: 24cbfe18d55a ("rv: Merge struct rv_monitor_def into struct
> rv_monitor")
> Signed-off-by: Zhen Ni
Good catch, thank you!
Reviewed-by: Gabriele Monaco
I would also add a:
On Mon, 2025-08-25 at 11:55 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:08:02PM +0200, Gabriele Monaco wrote:
> > for i in event.split("\\n"):
> > + if ";" in i:
> > + # if the event c
On Mon, 2025-08-25 at 12:06 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:08:02PM +0200, Gabriele Monaco wrote:
> > + def fill_constr_func(self) -> list[str]:
> > + buff = []
> > + if self.constraints:
> > + buff.append(
> > +&q
On Mon, 2025-08-25 at 10:13 +0200, Nam Cao wrote:
> On Mon, Aug 25, 2025 at 09:48:23AM +0200, Gabriele Monaco wrote:
> > On Thu, 2025-08-21 at 14:18 +0200, Nam Cao wrote:
> > > On Thu, Aug 14, 2025 at 05:08:00PM +0200, Gabriele Monaco wrote:
> > > > Deterministi
On Thu, 2025-08-21 at 10:49 +0200, Gabriele Monaco wrote:
> On Thu, 2025-08-21 at 10:14 +0200, Nam Cao wrote:
>
> > > +#define RV_AUTOMATON_NAME CONCATENATE(automaton_, MONITOR_NAME)
> > > +#define EVENT_MAX CONCATENATE(event_max_, MONITOR_NAME)
> > > +#define
On Thu, 2025-08-21 at 14:18 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:08:00PM +0200, Gabriele Monaco wrote:
> > Deterministic automata define which events are allowed in every
> > state,
> > but cannot define more sophisticated constraint taking into account
2025-08-21T15:15:36Z Nam Cao :
> We could also rewrite the parser using ply with a well-defined grammar and
> tokenizer, like how the LTL parser is implemented. Doing it this way would
> be easier to validate as well, because the grammar would be mostly
> copy-pasted from the specification.
Good i
On Thu, 2025-08-21 at 14:22 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:08:01PM +0200, Gabriele Monaco wrote:
> > Currently the automata parser assumes event strings don't have any
> > space, this stands true for event names, but can be a wrong
> > assumption
>
On Thu, 2025-08-21 at 14:38 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:08:02PM +0200, Gabriele Monaco wrote:
> > +import re
>
> Oh no..
I was tempted to suggest rewriting the dot parser with regex but I
guess I'll hold that back!
>
> > +from typing imp
On Thu, 2025-08-21 at 10:29 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:07:57PM +0200, Gabriele Monaco wrote:
> > Functions in automata.py, dot2c.py and dot2k.py don't have type
> > annotations and it can get complicated to remember how to use them.
> >
> &
On Thu, 2025-08-21 at 10:14 +0200, Nam Cao wrote:
> On Thu, Aug 14, 2025 at 05:07:53PM +0200, Gabriele Monaco wrote:
> > The da_monitor helper functions are generated from macros of the
> > type:
> >
> > DECLARE_DA_FUNCTION(name, type) \
> > static void da_func_x
On Tue, 2025-08-19 at 16:02 +0200, Juri Lelli wrote:
> On 19/08/25 12:34, Gabriele Monaco wrote:
> >
> >
> > On Tue, 2025-08-19 at 12:12 +0200, Peter Zijlstra wrote:
> > > On Tue, Aug 19, 2025 at 11:56:57AM +0200, Juri Lelli wrote:
> > > > Hi!
> >
On Tue, 2025-08-19 at 12:08 +0200, Juri Lelli wrote:
> On 19/08/25 11:48, Gabriele Monaco wrote:
> > That's a good point, I need to check the actual overhead..
> >
> > One thing to note is that this timer is used only on state
> > constraints,
> > one cou
On Tue, 2025-08-19 at 11:14 +0200, Juri Lelli wrote:
> On 19/08/25 10:53, Juri Lelli wrote:
> > Hi!
> >
> > On 14/08/25 17:08, Gabriele Monaco wrote:
>
> ...
>
> > > + static bool verify_constraint(enum states curr_state, enum
> > > events eve
On Tue, 2025-08-19 at 10:53 +0200, Juri Lelli wrote:
> Hi!
>
> On 14/08/25 17:08, Gabriele Monaco wrote:
> >
> > +
> > +Examples
> > +
>
> Maybe add subsection titles to better mark separation between
> different examples?
Sure, makes sense
On Tue, 2025-08-19 at 12:12 +0200, Peter Zijlstra wrote:
> On Tue, Aug 19, 2025 at 11:56:57AM +0200, Juri Lelli wrote:
> > Hi!
> >
> > On 14/08/25 17:08, Gabriele Monaco wrote:
> > > Add the following tracepoints:
> > >
> > > * sched_dl_throttle
On Tue, 2025-08-19 at 11:18 +0200, Juri Lelli wrote:
> Hi!
>
> On 14/08/25 17:08, Gabriele Monaco wrote:
>
> ...
>
> > +/*
> > + * ha_monitor_init_env - setup timer and reset all environment
> > + *
> > + * Called from a hook in the DA s
On Sun, 2025-08-10 at 16:42 +0530, Gopi Krishna Menon wrote:
> Specifically, fix spelling of "practice"
>
> Signed-off-by: Gopi Krishna Menon
Thanks for finding this!
Reviewed-by: Gabriele Monaco
> ---
> Documentation/trace/rv/monitor_synthesis.rst | 2 +-
>
.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/dot2k.py | 20
tools/verification/rvgen/rvgen/generator.py | 2 +-
2 files changed, 21 insertions(+), 1 deletion(-)
diff --git a/tools/verification/rvgen/rvgen/dot2k.py
b/tools/verification/rvgen/rvgen
rings.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/automata.py | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/tools/verification/rvgen/rvgen/automata.py
b/tools/verification/rvgen/rvgen/automata.py
index 76d7a3cfaec6..c637cf4ee749 100644
--- a/
hard):
Called when a deadline server is stopped (hard) or put to idle
waiting for the next period (!hard)
Those tracepoints can be useful to validate the deadline scheduler with
RV and are not exported to tracefs.
Signed-off-by: Gabriele Monaco
---
include/trace/events/sched.h
deadiline
Signed-off-by: Gabriele Monaco
---
Documentation/trace/rv/monitor_deadline.rst | 111
kernel/trace/rv/Kconfig | 5 +
kernel/trace/rv/Makefile | 3 +
kernel/trace/rv/monitors/deadline/Kconfig | 5 +
kernel/trace/rv/monitors
in a hash table indexed by id. Monitor storage objects are
automatically allocated unless specified otherwise (e.g. if the creation
context is unsafe for allocation).
Signed-off-by: Gabriele Monaco
---
include/linux/rv.h | 2 +-
include/rv/da_common.h | 21 +++
include/rv/da_monitor.h
Add a sample monitor to showcase hybrid/timed automata.
The stall monitor identifies tasks stalled for longer than a threshold
and reacts when that happens.
Signed-off-by: Gabriele Monaco
---
Documentation/tools/rv/rv-mon-stall.rst | 44 +++
Documentation/trace/rv/monitor_stall.rst
The monitor container source files contained a declaration and a
definition for the rv_monitor variable. The former is superfluous and
can be removed.
Remove the variable declaration from the template as well as the
existing monitor containers.
Signed-off-by: Gabriele Monaco
---
kernel/trace
Functions in automata.py, dot2c.py and dot2k.py don't have type
annotations and it can get complicated to remember how to use them.
Add minimal type annotations.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/automata.py | 18 ++---
tools/verification/rvgen/
str.join() can do what __buff_to_string() does. Therefore replace
__buff_to_string() to make the scripts more pythonic.
Also clean and remove some intermediate functions.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/dot2c.py | 35 +
1 file changed
y) on state and events.
Signed-off-by: Gabriele Monaco
---
include/linux/rv.h | 26 +++
include/rv/da_monitor.h| 45 -
include/rv/ha_monitor.h| 384 +
kernel/trace/rv/Kconfig| 13 ++
kernel/trace/rv/rv_trace.h | 63 ++
5 files ch
Signed-off-by: Gabriele Monaco
---
Documentation/trace/rv/monitor_synthesis.rst | 44 ++--
1 file changed, 21 insertions(+), 23 deletions(-)
diff --git a/Documentation/trace/rv/monitor_synthesis.rst
b/Documentation/trace/rv/monitor_synthesis.rst
index ac808a7554f5..ce0c1a5104d4
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/rvgen/dot2k.py | 6 --
.../rvgen/rvgen/templates/dot2k/main.c | 17 +
2 files changed, 9 insertions(+), 14 deletions(-)
diff --git a/tools/verification/rvgen/rvgen/dot2k.py
b/tools/verification
ge in struct
da_monitor is fixed to unsigned int.
Remove empty and no longer required macros and substitute the type
parameter with the appropriate enum.
Additionally break long line and adjust the format overall.
Signed-off-by: Gabriele Monaco
---
include/rv/automata.h | 24 +++--
inclu
ng on
PREEMPT_RT only and allows the monitor to be built on kernels without
the preemptirqs tracepoints.
[1] - https://lore.kernel.org/lkml/20250625120823.60600-1-gmon...@redhat.com
Signed-off-by: Gabriele Monaco
---
Documentation/trace/rv/monitor_sched.rst | 62 +++-
kernel/trace/rv/monitors
Describe theory and implementation of hybrid automata in the dedicated
page hybrid_automata.rst
Include a section on how to integrate a hybrid automaton in
monitor_synthesis.rst
Also remove a hanging $ in deterministic_automata.rst
Signed-off-by: Gabriele Monaco
---
.../trace/rv
(separated by \n).
The tool can now generate the appropriate code to validate constraints
at runtime according to the dot specification.
Signed-off-by: Gabriele Monaco
---
tools/verification/rvgen/__main__.py | 8 +-
tools/verification/rvgen/rvgen/automata.py| 128 +++-
tools
mpty macro definitions to ease review of this change with
diff tools, but cleanup is required.
Also adapt existing monitors to keep the build working.
Signed-off-by: Gabriele Monaco
---
include/linux/rv.h | 4 +
include/rv/automata.h | 132 ++--
include/rv/da
am Cao
Cc: Tomas Glozar
Cc: Juri Lelli
Cc: Clark Williams
Cc: John Kacur
Cc: linux-trace-kernel@vger.kernel.org
Gabriele Monaco (17):
rv: Refactor da_monitor to minimise macros
rv: Cleanup da_monitor after refactor
Documentation/rv: Adapt documentation after da_monitor refactoring
ve
On Mon, 2025-08-11 at 10:40 +0200, Nam Cao wrote:
> Add trace points into enqueue_task() and dequeue_task(). They are
> useful to implement RV monitor which validates RT scheduling.
>
> Signed-off-by: Nam Cao
> ---
Peter, Ingo, this patch adds new tracepoints in the scheduler do agree
with the c
ry. Also, it shouldn't happen, but sometimes
the order of inclusion matters. Anyway this isn't the case here (builds
fine locally) and when it is, the test robot should notice.
Looks good to me overall, thanks!
Reviewed-by: Gabriele Monaco
>
> Fixes: 24cbfe18d55a ("rv: Merge
2025-08-09T06:36:58Z Akhilesh Patil :
> Remove redundant include to clean up the code.
> Fix this redundancy introduced by commit [1].
>
> Fixes: 24cbfe18d55a ("rv: Merge struct rv_monitor_def into struct
> rv_monitor") [1]
> Reported-by: kernel test robot
> Closes: https://lore.kernel.org/r/20
On Fri, 2025-08-08 at 07:29 +0200, Nam Cao wrote:
> Gabriele Monaco writes:
> > The monitor shows a violation also in case of priority inversion
> > boosting, e.g.:
> >
> > stress-ng --prio-inv 2
>
> This looks like something that would trigger the fair deadli
On Fri, 2025-08-08 at 07:12 +0200, Nam Cao wrote:
> Gabriele Monaco writes:
> > I get you're trying to be more type-agnostic, but I believe this
> > /online/ is a bit imprecise, unless you register a hotplug handler
> > and just initialise the online CPUs (much of an o
On Wed, 2025-08-06 at 10:01 +0200, Nam Cao wrote:
> Add "real-time scheduling" monitor, which validates that SCHED_RR and
> SCHED_FIFO tasks are scheduled before tasks with normal and
> extensible scheduling policies
>
> Signed-off-by: Nam Cao
> ---
The monitor shows a violation also in case of
On Wed, 2025-08-06 at 10:01 +0200, Nam Cao wrote:
> Add "real-time scheduling" monitor, which validates that SCHED_RR and
> SCHED_FIFO tasks are scheduled before tasks with normal and
> extensible scheduling policies
>
> Signed-off-by: Nam Cao
> ---
> v2:
> - use the new tracepoints
> - move
cpu"
Looks good to me.
Reviewed-by: Gabriele Monaco
Thanks,
Gabriele
> ---
> include/rv/ltl_monitor.h | 32 ++
> kernel/trace/rv/Kconfig | 4
> kernel/trace/rv/rv_trace.h | 46
> ++
> 3 files changed,
unless you register a hotplug handler and
just initialise the online CPUs (much of an overkill I'd say).
What about something like "this is false if the monitor exists already
before the monitor is enabled"
Other than that it looks good to me.
Reviewed-by: Gabriele Monaco
Thanks,
Gabrie
list_head, and this
> wrong type cast became a functional problem.
>
> Properly use container_of() instead.
Good catch, thanks!
Reviewed-by: Gabriele Monaco
>
> Fixes: 24cbfe18d55a ("rv: Merge struct rv_monitor_def into struct
> rv_monitor")
> Signed-off-by: Nam
On Wed, 2025-08-06 at 10:46 +0200, Nam Cao wrote:
> On Wed, Aug 06, 2025 at 10:15:48AM +0200, Gabriele Monaco wrote:
> > I didn't make it on time before your V2, I assume you solved
> > already so
> > you might ignore this.
> >
> > You kinda have some
On Tue, 2025-08-05 at 17:45 +0200, Nam Cao wrote:
> On Tue, Aug 05, 2025 at 02:22:17PM +0200, Nam Cao wrote:
> > On Tue, Aug 05, 2025 at 10:40:30AM +0200, Gabriele Monaco wrote:
> > > Hello Nam,
> > >
> > > I just built and booted up the monitor in a VM (vir
On Wed, 2025-07-30 at 14:45 +0200, Nam Cao wrote:
> Add "real-time scheduling" monitor, which validates that SCHED_RR and
> SCHED_FIFO tasks are scheduled before tasks with normal and
> extensible
> scheduling policies
>
> Signed-off-by: Nam Cao
Hello Nam,
I just built and booted up the monitor
On Mon, 2025-08-04 at 12:45 -0700, Palmer Dabbelt wrote:
> From: Palmer Dabbelt
>
> Some systems (like 32-bit RISC-V) only have the 64-bit time_t
> versions of syscalls. So handle the 32-bit time_t version of those
> being undefined.
>
> Fixes: f74f8bb246cf ("rv: Add rtapp_sleep monitor")
> Sig
On Fri, 2025-08-01 at 15:26 +0530, K Prateek Nayak wrote:
> There are a few more cases with delayed dequeue:
>
> 1. A delayed task can be reenqueued before it is completely dequeued
> and
> will lead to a enqueue -> enqueue transition if we don't trace the
> first dequeue.
>
> 2. There are
On Fri, 2025-08-01 at 09:58 +0200, Nam Cao wrote:
> On Thu, Jul 31, 2025 at 09:47:10AM +0200, Gabriele Monaco wrote:
> > On Wed, 2025-07-30 at 14:45 +0200, Nam Cao wrote:
> > > Add "real-time scheduling" monitor, which validates that SCHED_RR
> > > and SCHED_
On Thu, 2025-07-31 at 11:28 +0200, Nam Cao wrote:
> On Thu, Jul 31, 2025 at 11:04:44AM +0200, Gabriele Monaco wrote:
> > I stole your solution to get rid of macros for the DA as well
> > (might
> > post it after this merge window or with the next changes) and I'm
>
patches going to coordinate, but I think it
may make sense to share those values for all monitor types.
What do you think?
For the rest this patch looks good to me, nice use of the typedef.
Reviewed-by: Gabriele Monaco
Thanks,
Gabriele
On Thu, 2025-07-31 at 09:35 +0200, Nam Cao wrote:
> On Wed, Jul 30, 2025 at 06:18:45PM +0200, Gabriele Monaco wrote:
> > Well, thinking about it again, these tracepoints might simplify
> > things
> > considerably when tasks change policy..
> >
> > Syscalls may fai
On Wed, 2025-07-30 at 14:45 +0200, Nam Cao wrote:
> Add support for per-cpu run-time verification linear temporal logic
> monitors. This is analogous to deterministic automaton per-cpu
> monitors.
>
> Signed-off-by: Nam Cao
> ---
>
> diff --git a/kernel/trace/rv/rv_trace.h b/kernel/trace/rv/rv_t
On Wed, 2025-07-30 at 14:45 +0200, Nam Cao wrote:
> Add "real-time scheduling" monitor, which validates that SCHED_RR and
> SCHED_FIFO tasks are scheduled before tasks with normal and
> extensible
> scheduling policies
>
Looks a very interesting monitor!
A few questions:
I assume this works with
On Wed, 2025-07-30 at 17:18 +0200, Nam Cao wrote:
> On Wed, Jul 30, 2025 at 03:53:14PM +0200, Gabriele Monaco wrote:
> > On Wed, 2025-07-30 at 14:45 +0200, Nam Cao wrote:
> > > Add trace points into enqueue_task_rt() and dequeue_task_rt().
> > > They
> > > are
On Wed, 2025-07-30 at 14:45 +0200, Nam Cao wrote:
> Add trace points into enqueue_task_rt() and dequeue_task_rt(). They
> are useful to implement RV monitor which validates RT scheduling.
>
I get it's much simpler this way, but is it that different to follow
the task's existing tracepoints?
* ta
On Tue, 2025-07-29 at 11:37 +0200, Nam Cao wrote:
> On Tue, Jul 29, 2025 at 11:25:12AM +0200, Nam Cao wrote:
> > Kernel:
> > - base: ftrace/for-next
I assume you mean rv/for-next ? The one that includes all changes as of
yesterday.
> > - config: defconfig + mod2noconfig + PREEMPT_RT + monitor
1 - 100 of 361 matches
Mail list logo
