* [PATCH v3 13/17] rv: Adapt the sco monitor to the new set_state
[not found] <20250715071434.22508-1-gmonaco@redhat.com>
@ 2025-07-15 7:14 ` Gabriele Monaco
2025-07-15 7:14 ` [PATCH v3 14/17] rv: Extend snroc model Gabriele Monaco
` (3 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-15 7:14 UTC (permalink / raw)
To: linux-kernel, Steven Rostedt, Jonathan Corbet, Masami Hiramatsu,
linux-trace-kernel, linux-doc
Cc: Gabriele Monaco, Ingo Molnar, Peter Zijlstra, Nam Cao,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
The sched_set_state tracepoint changed prototype adding a new argument,
this argument can differentiate between an explicit set_state called by
a task and a set state to runnable by the scheduler due to a pending
signal.
Expand the model to handle the new set_state flavour, the monitor was
making sure set state happens only outside of the scheduler, if the
event occurs with the new argument (from_signal) set to true, we instead
expect it to be inside the scheduler.
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
Documentation/trace/rv/monitor_sched.rst | 35 +++++++++++++-----------
kernel/trace/rv/monitors/sco/sco.c | 5 +++-
kernel/trace/rv/monitors/sco/sco.h | 4 +++
tools/verification/models/sched/sco.dot | 1 +
4 files changed, 28 insertions(+), 17 deletions(-)
diff --git a/Documentation/trace/rv/monitor_sched.rst b/Documentation/trace/rv/monitor_sched.rst
index 24b2c62a3bc26..6f76bba94d9fb 100644
--- a/Documentation/trace/rv/monitor_sched.rst
+++ b/Documentation/trace/rv/monitor_sched.rst
@@ -64,22 +64,25 @@ Monitor sco
~~~~~~~~~~~
The scheduling context operations (sco) monitor ensures changes in a task state
-happen only in thread context::
-
-
- |
- |
- v
- sched_set_state +------------------+
- +------------------ | |
- | | thread_context |
- +-----------------> | | <+
- +------------------+ |
- | |
- | schedule_entry | schedule_exit
- v |
- |
- scheduling_context -+
+happen only in thread context, the only exception is a special kind of set
+state that occurs if a task about to sleep has a pending signal. This set state
+is not called by the thread but by the scheduler itself::
+
+ |
+ |
+ v
+ sched_set_state +------------------+
+ +---------------------------------- | |
+ | | thread_context |
+ +---------------------------------> | | <+
+ +------------------+ |
+ | |
+ | schedule_entry | schedule_exit
+ v |
+ sched_set_state_runnable_signal |
+ +---------------------------------- |
+ | scheduling_context |
+ +---------------------------------> -+
Monitor snroc
~~~~~~~~~~~~~
diff --git a/kernel/trace/rv/monitors/sco/sco.c b/kernel/trace/rv/monitors/sco/sco.c
index c9206aa12c319..6457ff2469d08 100644
--- a/kernel/trace/rv/monitors/sco/sco.c
+++ b/kernel/trace/rv/monitors/sco/sco.c
@@ -22,7 +22,10 @@ DECLARE_DA_MON_PER_CPU(sco, unsigned char);
static void handle_sched_set_state(void *data, struct task_struct *tsk,
int state, bool from_signal)
{
- da_handle_start_event_sco(sched_set_state_sco);
+ if (from_signal)
+ da_handle_event_sco(sched_set_state_runnable_signal_sco);
+ else
+ da_handle_start_event_sco(sched_set_state_sco);
}
static void handle_schedule_entry(void *data, bool preempt, unsigned long ip)
diff --git a/kernel/trace/rv/monitors/sco/sco.h b/kernel/trace/rv/monitors/sco/sco.h
index 83ca9a03331af..2d05ab882b2a6 100644
--- a/kernel/trace/rv/monitors/sco/sco.h
+++ b/kernel/trace/rv/monitors/sco/sco.h
@@ -15,6 +15,7 @@ enum states_sco {
enum events_sco {
sched_set_state_sco = 0,
+ sched_set_state_runnable_signal_sco,
schedule_entry_sco,
schedule_exit_sco,
event_max_sco
@@ -35,17 +36,20 @@ static const struct automaton_sco automaton_sco = {
},
.event_names = {
"sched_set_state",
+ "sched_set_state_runnable_signal",
"schedule_entry",
"schedule_exit"
},
.function = {
{
thread_context_sco,
+ INVALID_STATE,
scheduling_context_sco,
INVALID_STATE
},
{
INVALID_STATE,
+ scheduling_context_sco,
INVALID_STATE,
thread_context_sco
},
diff --git a/tools/verification/models/sched/sco.dot b/tools/verification/models/sched/sco.dot
index 20b0e3b449a6b..4e44ed58c62a3 100644
--- a/tools/verification/models/sched/sco.dot
+++ b/tools/verification/models/sched/sco.dot
@@ -7,6 +7,7 @@ digraph state_automaton {
{node [shape = plaintext] "thread_context"};
"__init_thread_context" -> "thread_context";
"scheduling_context" [label = "scheduling_context"];
+ "scheduling_context" -> "scheduling_context" [ label = "sched_set_state_runnable_signal" ];
"scheduling_context" -> "thread_context" [ label = "schedule_exit" ];
"thread_context" [label = "thread_context", color = green3];
"thread_context" -> "scheduling_context" [ label = "schedule_entry" ];
--
2.50.1
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH v3 14/17] rv: Extend snroc model
[not found] <20250715071434.22508-1-gmonaco@redhat.com>
2025-07-15 7:14 ` [PATCH v3 13/17] rv: Adapt the sco monitor to the new set_state Gabriele Monaco
@ 2025-07-15 7:14 ` Gabriele Monaco
2025-07-15 7:14 ` [PATCH v3 15/17] rv: Replace tss monitor with more complete sts Gabriele Monaco
` (2 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-15 7:14 UTC (permalink / raw)
To: linux-kernel, Steven Rostedt, Jonathan Corbet, Masami Hiramatsu,
linux-trace-kernel, linux-doc
Cc: Gabriele Monaco, Ingo Molnar, Peter Zijlstra, Nam Cao,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
The snroc monitor ensures changes in a task state happens only in the
respective task's context. This is the only monitor enforcing a task
is to be switched in after being switched out, and vice-versa.
Although this is a trivial claim, it is useful to complete other claims
on when scheduling is possible while keeping models simple.
Add the sched_switch_vain event to the model, enforcing that a vain
switch doesn't change the context but can only occur while a task is
running (e.g. a call to schedule that re-selects the current task).
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
Documentation/trace/rv/monitor_sched.rst | 32 ++++++++++++-----------
kernel/trace/rv/monitors/snroc/snroc.c | 9 +++++++
kernel/trace/rv/monitors/snroc/snroc.h | 8 ++++--
tools/verification/models/sched/snroc.dot | 2 +-
4 files changed, 33 insertions(+), 18 deletions(-)
diff --git a/Documentation/trace/rv/monitor_sched.rst b/Documentation/trace/rv/monitor_sched.rst
index 6f76bba94d9fb..5cb58ac441d83 100644
--- a/Documentation/trace/rv/monitor_sched.rst
+++ b/Documentation/trace/rv/monitor_sched.rst
@@ -89,21 +89,23 @@ Monitor snroc
The set non runnable on its own context (snroc) monitor ensures changes in a
task state happens only in the respective task's context. This is a per-task
-monitor::
-
- |
- |
- v
- +------------------+
- | other_context | <+
- +------------------+ |
- | |
- | sched_switch_in | sched_switch_out
- v |
- sched_set_state |
- +------------------ |
- | own_context |
- +-----------------> -+
+monitor. A task is in its own context after switching in and leaves the context
+when switched out, a vain switch maintains the context::
+
+ |
+ |
+ v
+ +------------------+
+ | other_context | <+
+ +------------------+ |
+ | |
+ | sched_switch_in | sched_switch_out
+ v |
+ sched_set_state |
+ sched_switch_vain |
+ +-------------------- own_context |
+ | |
+ +-------------------> -+
Monitor scpd
~~~~~~~~~~~~
diff --git a/kernel/trace/rv/monitors/snroc/snroc.c b/kernel/trace/rv/monitors/snroc/snroc.c
index 2651f589d1554..11a56b58665e7 100644
--- a/kernel/trace/rv/monitors/snroc/snroc.c
+++ b/kernel/trace/rv/monitors/snroc/snroc.c
@@ -34,6 +34,13 @@ static void handle_sched_switch(void *data, bool preempt,
da_handle_event_snroc(next, sched_switch_in_snroc);
}
+static void handle_sched_switch_vain(void *data, bool preempt,
+ struct task_struct *tsk,
+ unsigned int tsk_state)
+{
+ da_handle_event_snroc(tsk, sched_switch_vain_snroc);
+}
+
static int enable_snroc(void)
{
int retval;
@@ -44,6 +51,7 @@ static int enable_snroc(void)
rv_attach_trace_probe("snroc", sched_set_state_tp, handle_sched_set_state);
rv_attach_trace_probe("snroc", sched_switch, handle_sched_switch);
+ rv_attach_trace_probe("snroc", sched_switch_vain_tp, handle_sched_switch_vain);
return 0;
}
@@ -54,6 +62,7 @@ static void disable_snroc(void)
rv_detach_trace_probe("snroc", sched_set_state_tp, handle_sched_set_state);
rv_detach_trace_probe("snroc", sched_switch, handle_sched_switch);
+ rv_detach_trace_probe("snroc", sched_switch_vain_tp, handle_sched_switch_vain);
da_monitor_destroy_snroc();
}
diff --git a/kernel/trace/rv/monitors/snroc/snroc.h b/kernel/trace/rv/monitors/snroc/snroc.h
index be46f7b9ebb87..064d98fb9796d 100644
--- a/kernel/trace/rv/monitors/snroc/snroc.h
+++ b/kernel/trace/rv/monitors/snroc/snroc.h
@@ -17,6 +17,7 @@ enum events_snroc {
sched_set_state_snroc = 0,
sched_switch_in_snroc,
sched_switch_out_snroc,
+ sched_switch_vain_snroc,
event_max_snroc
};
@@ -36,18 +37,21 @@ static const struct automaton_snroc automaton_snroc = {
.event_names = {
"sched_set_state",
"sched_switch_in",
- "sched_switch_out"
+ "sched_switch_out",
+ "sched_switch_vain"
},
.function = {
{
INVALID_STATE,
own_context_snroc,
+ INVALID_STATE,
INVALID_STATE
},
{
own_context_snroc,
INVALID_STATE,
- other_context_snroc
+ other_context_snroc,
+ own_context_snroc
},
},
.initial_state = other_context_snroc,
diff --git a/tools/verification/models/sched/snroc.dot b/tools/verification/models/sched/snroc.dot
index 8b71c32d4dca4..5b1a787d0350b 100644
--- a/tools/verification/models/sched/snroc.dot
+++ b/tools/verification/models/sched/snroc.dot
@@ -10,7 +10,7 @@ digraph state_automaton {
"other_context" -> "own_context" [ label = "sched_switch_in" ];
"own_context" [label = "own_context"];
"own_context" -> "other_context" [ label = "sched_switch_out" ];
- "own_context" -> "own_context" [ label = "sched_set_state" ];
+ "own_context" -> "own_context" [ label = "sched_set_state\nsched_switch_vain" ];
{ rank = min ;
"__init_other_context";
"other_context";
--
2.50.1
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH v3 15/17] rv: Replace tss monitor with more complete sts
[not found] <20250715071434.22508-1-gmonaco@redhat.com>
2025-07-15 7:14 ` [PATCH v3 13/17] rv: Adapt the sco monitor to the new set_state Gabriele Monaco
2025-07-15 7:14 ` [PATCH v3 14/17] rv: Extend snroc model Gabriele Monaco
@ 2025-07-15 7:14 ` Gabriele Monaco
2025-07-15 7:14 ` [PATCH v3 16/17] rv: Add nrp and sssw per-task monitors Gabriele Monaco
2025-07-15 7:14 ` [PATCH v3 17/17] rv: Add opid per-cpu monitor Gabriele Monaco
4 siblings, 0 replies; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-15 7:14 UTC (permalink / raw)
To: linux-kernel, Steven Rostedt, Jonathan Corbet, Masami Hiramatsu,
linux-trace-kernel, linux-doc
Cc: Gabriele Monaco, Ingo Molnar, Peter Zijlstra, Nam Cao,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
The tss monitor currently guarantees context switches can happen only
while scheduling, but it doesn't enforce that each scheduling call
implies a task switch. This can be implied only if we rely on the newly
introduced sched_switch_vain tracepoint, which represents a
scheduler call where the previously running task is the same that is
picked to run next, in fact no context is switched.
Replace the monitor with a more comprehensive specification which
implies tss but also ensures that:
* each scheduler call switches context (or has a vain switch)
* each context switch happens with interrupts disabled
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
Documentation/trace/rv/monitor_sched.rst | 68 +++++++---
kernel/trace/rv/Kconfig | 2 +-
kernel/trace/rv/Makefile | 2 +-
kernel/trace/rv/monitors/sts/Kconfig | 18 +++
kernel/trace/rv/monitors/sts/sts.c | 117 ++++++++++++++++++
kernel/trace/rv/monitors/sts/sts.h | 97 +++++++++++++++
.../{tss/tss_trace.h => sts/sts_trace.h} | 8 +-
kernel/trace/rv/monitors/tss/Kconfig | 14 ---
kernel/trace/rv/monitors/tss/tss.c | 90 --------------
kernel/trace/rv/monitors/tss/tss.h | 47 -------
kernel/trace/rv/rv_trace.h | 2 +-
tools/verification/models/sched/sts.dot | 29 +++++
tools/verification/models/sched/tss.dot | 18 ---
13 files changed, 316 insertions(+), 196 deletions(-)
create mode 100644 kernel/trace/rv/monitors/sts/Kconfig
create mode 100644 kernel/trace/rv/monitors/sts/sts.c
create mode 100644 kernel/trace/rv/monitors/sts/sts.h
rename kernel/trace/rv/monitors/{tss/tss_trace.h => sts/sts_trace.h} (67%)
delete mode 100644 kernel/trace/rv/monitors/tss/Kconfig
delete mode 100644 kernel/trace/rv/monitors/tss/tss.c
delete mode 100644 kernel/trace/rv/monitors/tss/tss.h
create mode 100644 tools/verification/models/sched/sts.dot
delete mode 100644 tools/verification/models/sched/tss.dot
diff --git a/Documentation/trace/rv/monitor_sched.rst b/Documentation/trace/rv/monitor_sched.rst
index 5cb58ac441d83..e4171aadef1c2 100644
--- a/Documentation/trace/rv/monitor_sched.rst
+++ b/Documentation/trace/rv/monitor_sched.rst
@@ -40,26 +40,6 @@ defined in by Daniel Bristot in [1].
Currently we included the following:
-Monitor tss
-~~~~~~~~~~~
-
-The task switch while scheduling (tss) monitor ensures a task switch happens
-only in scheduling context, that is inside a call to `__schedule`::
-
- |
- |
- v
- +-----------------+
- | thread | <+
- +-----------------+ |
- | |
- | schedule_entry | schedule_exit
- v |
- sched_switch |
- +--------------- |
- | sched |
- +--------------> -+
-
Monitor sco
~~~~~~~~~~~
@@ -170,6 +150,54 @@ schedule is not called with interrupt disabled::
|
cant_sched -+
+Monitor sts
+~~~~~~~~~~~
+
+The schedule implies task switch (sts) monitor ensures a task switch happens in
+every scheduling context, that is inside a call to ``__schedule``, as well as no
+task switch can happen without scheduling and before interrupts are disabled.
+This require the special type of switch called vain, which occurs when the next
+task picked for execution is the same as the previously running one, in fact no
+real task switch occurs::
+
+ |
+ |
+ v
+ #====================# irq_disable
+ H H irq_enable
+ H thread H --------------+
+ H H |
+ +-------------> H H <-------------+
+ | #====================#
+ | |
+ | | schedule_entry
+ | v
+ | +--------------------+
+ | | scheduling | <+
+ | +--------------------+ |
+ | | |
+ | | irq_disable | irq_enable
+ | v |
+ | +--------------------+ |
+ | | disable_to_switch | -+
+ | schedule_exit +--------------------+
+ | |
+ | | sched_switch
+ | | sched_switch_vain
+ | v
+ | +--------------------+
+ | | switching |
+ | +--------------------+
+ | |
+ | | irq_enable
+ | v
+ | +--------------------+ irq_disable
+ | | | irq_enable
+ | | enable_to_exit | --------------+
+ | | | |
+ +-------------- | | <-------------+
+ +--------------------+
+
References
----------
diff --git a/kernel/trace/rv/Kconfig b/kernel/trace/rv/Kconfig
index 26017378f79b8..da62d97d329aa 100644
--- a/kernel/trace/rv/Kconfig
+++ b/kernel/trace/rv/Kconfig
@@ -45,12 +45,12 @@ source "kernel/trace/rv/monitors/wip/Kconfig"
source "kernel/trace/rv/monitors/wwnr/Kconfig"
source "kernel/trace/rv/monitors/sched/Kconfig"
-source "kernel/trace/rv/monitors/tss/Kconfig"
source "kernel/trace/rv/monitors/sco/Kconfig"
source "kernel/trace/rv/monitors/snroc/Kconfig"
source "kernel/trace/rv/monitors/scpd/Kconfig"
source "kernel/trace/rv/monitors/snep/Kconfig"
source "kernel/trace/rv/monitors/sncid/Kconfig"
+source "kernel/trace/rv/monitors/sts/Kconfig"
# Add new sched monitors here
source "kernel/trace/rv/monitors/rtapp/Kconfig"
diff --git a/kernel/trace/rv/Makefile b/kernel/trace/rv/Makefile
index 13ec2944c6650..db8653f7cedf3 100644
--- a/kernel/trace/rv/Makefile
+++ b/kernel/trace/rv/Makefile
@@ -6,7 +6,6 @@ obj-$(CONFIG_RV) += rv.o
obj-$(CONFIG_RV_MON_WIP) += monitors/wip/wip.o
obj-$(CONFIG_RV_MON_WWNR) += monitors/wwnr/wwnr.o
obj-$(CONFIG_RV_MON_SCHED) += monitors/sched/sched.o
-obj-$(CONFIG_RV_MON_TSS) += monitors/tss/tss.o
obj-$(CONFIG_RV_MON_SCO) += monitors/sco/sco.o
obj-$(CONFIG_RV_MON_SNROC) += monitors/snroc/snroc.o
obj-$(CONFIG_RV_MON_SCPD) += monitors/scpd/scpd.o
@@ -15,6 +14,7 @@ obj-$(CONFIG_RV_MON_SNCID) += monitors/sncid/sncid.o
obj-$(CONFIG_RV_MON_RTAPP) += monitors/rtapp/rtapp.o
obj-$(CONFIG_RV_MON_PAGEFAULT) += monitors/pagefault/pagefault.o
obj-$(CONFIG_RV_MON_SLEEP) += monitors/sleep/sleep.o
+obj-$(CONFIG_RV_MON_STS) += monitors/sts/sts.o
# Add new monitors here
obj-$(CONFIG_RV_REACTORS) += rv_reactors.o
obj-$(CONFIG_RV_REACT_PRINTK) += reactor_printk.o
diff --git a/kernel/trace/rv/monitors/sts/Kconfig b/kernel/trace/rv/monitors/sts/Kconfig
new file mode 100644
index 0000000000000..3fb1b5387566c
--- /dev/null
+++ b/kernel/trace/rv/monitors/sts/Kconfig
@@ -0,0 +1,18 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+config RV_MON_STS
+ depends on RV
+ depends on TRACE_IRQFLAGS
+ depends on RV_MON_SCHED
+ default y
+ select DA_MON_EVENTS_IMPLICIT
+ bool "sts monitor"
+ help
+ Monitor to ensure relationships between scheduler and switches
+ * each call to the scheduler implies a switch
+ * switches only happen inside the scheduler
+ * switches happen with interrupt disabled
+ This monitor is part of the sched monitors collection.
+
+ For further information, see:
+ Documentation/trace/rv/monitor_sched.rst
diff --git a/kernel/trace/rv/monitors/sts/sts.c b/kernel/trace/rv/monitors/sts/sts.c
new file mode 100644
index 0000000000000..deb18a9d48f3b
--- /dev/null
+++ b/kernel/trace/rv/monitors/sts/sts.c
@@ -0,0 +1,117 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/ftrace.h>
+#include <linux/tracepoint.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/rv.h>
+#include <rv/instrumentation.h>
+#include <rv/da_monitor.h>
+
+#define MODULE_NAME "sts"
+
+#include <trace/events/sched.h>
+#include <trace/events/preemptirq.h>
+#include <rv_trace.h>
+#include <monitors/sched/sched.h>
+
+#include "sts.h"
+
+static struct rv_monitor rv_sts;
+DECLARE_DA_MON_PER_CPU(sts, unsigned char);
+
+static void handle_irq_disable(void *data, unsigned long ip, unsigned long parent_ip)
+{
+ da_handle_event_sts(irq_disable_sts);
+}
+
+static void handle_irq_enable(void *data, unsigned long ip, unsigned long parent_ip)
+{
+ da_handle_event_sts(irq_enable_sts);
+}
+
+static void handle_sched_switch(void *data, bool preempt,
+ struct task_struct *prev,
+ struct task_struct *next,
+ unsigned int prev_state)
+{
+ da_handle_event_sts(sched_switch_sts);
+}
+
+static void handle_sched_switch_vain(void *data, bool preempt,
+ struct task_struct *tsk,
+ unsigned int tsk_state)
+{
+ da_handle_event_sts(sched_switch_vain_sts);
+}
+
+static void handle_schedule_entry(void *data, bool preempt, unsigned long ip)
+{
+ da_handle_event_sts(schedule_entry_sts);
+}
+
+static void handle_schedule_exit(void *data, bool is_switch, unsigned long ip)
+{
+ da_handle_start_event_sts(schedule_exit_sts);
+}
+
+static int enable_sts(void)
+{
+ int retval;
+
+ retval = da_monitor_init_sts();
+ if (retval)
+ return retval;
+
+ rv_attach_trace_probe("sts", irq_disable, handle_irq_disable);
+ rv_attach_trace_probe("sts", irq_enable, handle_irq_enable);
+ rv_attach_trace_probe("sts", sched_switch, handle_sched_switch);
+ rv_attach_trace_probe("sts", sched_switch_vain_tp, handle_sched_switch_vain);
+ rv_attach_trace_probe("sts", sched_entry_tp, handle_schedule_entry);
+ rv_attach_trace_probe("sts", sched_exit_tp, handle_schedule_exit);
+
+ return 0;
+}
+
+static void disable_sts(void)
+{
+ rv_sts.enabled = 0;
+
+ rv_detach_trace_probe("sts", irq_disable, handle_irq_disable);
+ rv_detach_trace_probe("sts", irq_enable, handle_irq_enable);
+ rv_detach_trace_probe("sts", sched_switch, handle_sched_switch);
+ rv_detach_trace_probe("sts", sched_switch_vain_tp, handle_sched_switch_vain);
+ rv_detach_trace_probe("sts", sched_entry_tp, handle_schedule_entry);
+ rv_detach_trace_probe("sts", sched_exit_tp, handle_schedule_exit);
+
+ da_monitor_destroy_sts();
+}
+
+/*
+ * This is the monitor register section.
+ */
+static struct rv_monitor rv_sts = {
+ .name = "sts",
+ .description = "schedule implies task switch.",
+ .enable = enable_sts,
+ .disable = disable_sts,
+ .reset = da_monitor_reset_all_sts,
+ .enabled = 0,
+};
+
+static int __init register_sts(void)
+{
+ return rv_register_monitor(&rv_sts, &rv_sched);
+}
+
+static void __exit unregister_sts(void)
+{
+ rv_unregister_monitor(&rv_sts);
+}
+
+module_init(register_sts);
+module_exit(unregister_sts);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>");
+MODULE_DESCRIPTION("sts: schedule implies task switch.");
diff --git a/kernel/trace/rv/monitors/sts/sts.h b/kernel/trace/rv/monitors/sts/sts.h
new file mode 100644
index 0000000000000..537ea1b2c023a
--- /dev/null
+++ b/kernel/trace/rv/monitors/sts/sts.h
@@ -0,0 +1,97 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Automatically generated C representation of sts automaton
+ * For further information about this format, see kernel documentation:
+ * Documentation/trace/rv/deterministic_automata.rst
+ */
+
+enum states_sts {
+ thread_sts = 0,
+ disable_to_switch_sts,
+ enable_to_exit_sts,
+ scheduling_sts,
+ switching_sts,
+ state_max_sts
+};
+
+#define INVALID_STATE state_max_sts
+
+enum events_sts {
+ irq_disable_sts = 0,
+ irq_enable_sts,
+ sched_switch_sts,
+ sched_switch_vain_sts,
+ schedule_entry_sts,
+ schedule_exit_sts,
+ event_max_sts
+};
+
+struct automaton_sts {
+ char *state_names[state_max_sts];
+ char *event_names[event_max_sts];
+ unsigned char function[state_max_sts][event_max_sts];
+ unsigned char initial_state;
+ bool final_states[state_max_sts];
+};
+
+static const struct automaton_sts automaton_sts = {
+ .state_names = {
+ "thread",
+ "disable_to_switch",
+ "enable_to_exit",
+ "scheduling",
+ "switching"
+ },
+ .event_names = {
+ "irq_disable",
+ "irq_enable",
+ "sched_switch",
+ "sched_switch_vain",
+ "schedule_entry",
+ "schedule_exit"
+ },
+ .function = {
+ {
+ thread_sts,
+ thread_sts,
+ INVALID_STATE,
+ INVALID_STATE,
+ scheduling_sts,
+ INVALID_STATE
+ },
+ {
+ INVALID_STATE,
+ scheduling_sts,
+ switching_sts,
+ switching_sts,
+ INVALID_STATE,
+ INVALID_STATE
+ },
+ {
+ enable_to_exit_sts,
+ enable_to_exit_sts,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ thread_sts
+ },
+ {
+ disable_to_switch_sts,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE
+ },
+ {
+ INVALID_STATE,
+ enable_to_exit_sts,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE
+ },
+ },
+ .initial_state = thread_sts,
+ .final_states = { 1, 0, 0, 0, 0 },
+};
diff --git a/kernel/trace/rv/monitors/tss/tss_trace.h b/kernel/trace/rv/monitors/sts/sts_trace.h
similarity index 67%
rename from kernel/trace/rv/monitors/tss/tss_trace.h
rename to kernel/trace/rv/monitors/sts/sts_trace.h
index 4619dbb50cc06..d78beb58d5b3d 100644
--- a/kernel/trace/rv/monitors/tss/tss_trace.h
+++ b/kernel/trace/rv/monitors/sts/sts_trace.h
@@ -4,12 +4,12 @@
* Snippet to be included in rv_trace.h
*/
-#ifdef CONFIG_RV_MON_TSS
-DEFINE_EVENT(event_da_monitor, event_tss,
+#ifdef CONFIG_RV_MON_STS
+DEFINE_EVENT(event_da_monitor, event_sts,
TP_PROTO(char *state, char *event, char *next_state, bool final_state),
TP_ARGS(state, event, next_state, final_state));
-DEFINE_EVENT(error_da_monitor, error_tss,
+DEFINE_EVENT(error_da_monitor, error_sts,
TP_PROTO(char *state, char *event),
TP_ARGS(state, event));
-#endif /* CONFIG_RV_MON_TSS */
+#endif /* CONFIG_RV_MON_STS */
diff --git a/kernel/trace/rv/monitors/tss/Kconfig b/kernel/trace/rv/monitors/tss/Kconfig
deleted file mode 100644
index 479f86f52e60d..0000000000000
--- a/kernel/trace/rv/monitors/tss/Kconfig
+++ /dev/null
@@ -1,14 +0,0 @@
-# SPDX-License-Identifier: GPL-2.0-only
-#
-config RV_MON_TSS
- depends on RV
- depends on RV_MON_SCHED
- default y
- select DA_MON_EVENTS_IMPLICIT
- bool "tss monitor"
- help
- Monitor to ensure sched_switch happens only in scheduling context.
- This monitor is part of the sched monitors collection.
-
- For further information, see:
- Documentation/trace/rv/monitor_sched.rst
diff --git a/kernel/trace/rv/monitors/tss/tss.c b/kernel/trace/rv/monitors/tss/tss.c
deleted file mode 100644
index 0452fcd9edcfe..0000000000000
--- a/kernel/trace/rv/monitors/tss/tss.c
+++ /dev/null
@@ -1,90 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-#include <linux/ftrace.h>
-#include <linux/tracepoint.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/rv.h>
-#include <rv/instrumentation.h>
-#include <rv/da_monitor.h>
-
-#define MODULE_NAME "tss"
-
-#include <trace/events/sched.h>
-#include <rv_trace.h>
-#include <monitors/sched/sched.h>
-
-#include "tss.h"
-
-static struct rv_monitor rv_tss;
-DECLARE_DA_MON_PER_CPU(tss, unsigned char);
-
-static void handle_sched_switch(void *data, bool preempt,
- struct task_struct *prev,
- struct task_struct *next,
- unsigned int prev_state)
-{
- da_handle_event_tss(sched_switch_tss);
-}
-
-static void handle_schedule_entry(void *data, bool preempt, unsigned long ip)
-{
- da_handle_event_tss(schedule_entry_tss);
-}
-
-static void handle_schedule_exit(void *data, bool is_switch, unsigned long ip)
-{
- da_handle_start_event_tss(schedule_exit_tss);
-}
-
-static int enable_tss(void)
-{
- int retval;
-
- retval = da_monitor_init_tss();
- if (retval)
- return retval;
-
- rv_attach_trace_probe("tss", sched_switch, handle_sched_switch);
- rv_attach_trace_probe("tss", sched_entry_tp, handle_schedule_entry);
- rv_attach_trace_probe("tss", sched_exit_tp, handle_schedule_exit);
-
- return 0;
-}
-
-static void disable_tss(void)
-{
- rv_tss.enabled = 0;
-
- rv_detach_trace_probe("tss", sched_switch, handle_sched_switch);
- rv_detach_trace_probe("tss", sched_entry_tp, handle_schedule_entry);
- rv_detach_trace_probe("tss", sched_exit_tp, handle_schedule_exit);
-
- da_monitor_destroy_tss();
-}
-
-static struct rv_monitor rv_tss = {
- .name = "tss",
- .description = "task switch while scheduling.",
- .enable = enable_tss,
- .disable = disable_tss,
- .reset = da_monitor_reset_all_tss,
- .enabled = 0,
-};
-
-static int __init register_tss(void)
-{
- return rv_register_monitor(&rv_tss, &rv_sched);
-}
-
-static void __exit unregister_tss(void)
-{
- rv_unregister_monitor(&rv_tss);
-}
-
-module_init(register_tss);
-module_exit(unregister_tss);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>");
-MODULE_DESCRIPTION("tss: task switch while scheduling.");
diff --git a/kernel/trace/rv/monitors/tss/tss.h b/kernel/trace/rv/monitors/tss/tss.h
deleted file mode 100644
index f0a36fda1b873..0000000000000
--- a/kernel/trace/rv/monitors/tss/tss.h
+++ /dev/null
@@ -1,47 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * Automatically generated C representation of tss automaton
- * For further information about this format, see kernel documentation:
- * Documentation/trace/rv/deterministic_automata.rst
- */
-
-enum states_tss {
- thread_tss = 0,
- sched_tss,
- state_max_tss
-};
-
-#define INVALID_STATE state_max_tss
-
-enum events_tss {
- sched_switch_tss = 0,
- schedule_entry_tss,
- schedule_exit_tss,
- event_max_tss
-};
-
-struct automaton_tss {
- char *state_names[state_max_tss];
- char *event_names[event_max_tss];
- unsigned char function[state_max_tss][event_max_tss];
- unsigned char initial_state;
- bool final_states[state_max_tss];
-};
-
-static const struct automaton_tss automaton_tss = {
- .state_names = {
- "thread",
- "sched"
- },
- .event_names = {
- "sched_switch",
- "schedule_entry",
- "schedule_exit"
- },
- .function = {
- { INVALID_STATE, sched_tss, INVALID_STATE },
- { sched_tss, INVALID_STATE, thread_tss },
- },
- .initial_state = thread_tss,
- .final_states = { 1, 0 },
-};
diff --git a/kernel/trace/rv/rv_trace.h b/kernel/trace/rv/rv_trace.h
index d38e0d3abdfde..dbd2f2ef513a7 100644
--- a/kernel/trace/rv/rv_trace.h
+++ b/kernel/trace/rv/rv_trace.h
@@ -58,11 +58,11 @@ DECLARE_EVENT_CLASS(error_da_monitor,
);
#include <monitors/wip/wip_trace.h>
-#include <monitors/tss/tss_trace.h>
#include <monitors/sco/sco_trace.h>
#include <monitors/scpd/scpd_trace.h>
#include <monitors/snep/snep_trace.h>
#include <monitors/sncid/sncid_trace.h>
+#include <monitors/sts/sts_trace.h>
// Add new monitors based on CONFIG_DA_MON_EVENTS_IMPLICIT here
#endif /* CONFIG_DA_MON_EVENTS_IMPLICIT */
diff --git a/tools/verification/models/sched/sts.dot b/tools/verification/models/sched/sts.dot
new file mode 100644
index 0000000000000..8152675b4f528
--- /dev/null
+++ b/tools/verification/models/sched/sts.dot
@@ -0,0 +1,29 @@
+digraph state_automaton {
+ center = true;
+ size = "7,11";
+ {node [shape = circle] "disable_to_switch"};
+ {node [shape = circle] "enable_to_exit"};
+ {node [shape = circle] "scheduling"};
+ {node [shape = circle] "switching"};
+ {node [shape = plaintext, style=invis, label=""] "__init_thread"};
+ {node [shape = doublecircle] "thread"};
+ {node [shape = circle] "thread"};
+ "__init_thread" -> "thread";
+ "disable_to_switch" [label = "disable_to_switch"];
+ "disable_to_switch" -> "scheduling" [ label = "irq_enable" ];
+ "disable_to_switch" -> "switching" [ label = "sched_switch\nsched_switch_vain" ];
+ "enable_to_exit" [label = "enable_to_exit"];
+ "enable_to_exit" -> "enable_to_exit" [ label = "irq_disable\nirq_enable" ];
+ "enable_to_exit" -> "thread" [ label = "schedule_exit" ];
+ "scheduling" [label = "scheduling"];
+ "scheduling" -> "disable_to_switch" [ label = "irq_disable" ];
+ "switching" [label = "switching"];
+ "switching" -> "enable_to_exit" [ label = "irq_enable" ];
+ "thread" [label = "thread", color = green3];
+ "thread" -> "scheduling" [ label = "schedule_entry" ];
+ "thread" -> "thread" [ label = "irq_disable\nirq_enable" ];
+ { rank = min ;
+ "__init_thread";
+ "thread";
+ }
+}
diff --git a/tools/verification/models/sched/tss.dot b/tools/verification/models/sched/tss.dot
deleted file mode 100644
index 7dfa1d9121bbd..0000000000000
--- a/tools/verification/models/sched/tss.dot
+++ /dev/null
@@ -1,18 +0,0 @@
-digraph state_automaton {
- center = true;
- size = "7,11";
- {node [shape = plaintext] "sched"};
- {node [shape = plaintext, style=invis, label=""] "__init_thread"};
- {node [shape = ellipse] "thread"};
- {node [shape = plaintext] "thread"};
- "__init_thread" -> "thread";
- "sched" [label = "sched"];
- "sched" -> "sched" [ label = "sched_switch" ];
- "sched" -> "thread" [ label = "schedule_exit" ];
- "thread" [label = "thread", color = green3];
- "thread" -> "sched" [ label = "schedule_entry" ];
- { rank = min ;
- "__init_thread";
- "thread";
- }
-}
--
2.50.1
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH v3 16/17] rv: Add nrp and sssw per-task monitors
[not found] <20250715071434.22508-1-gmonaco@redhat.com>
` (2 preceding siblings ...)
2025-07-15 7:14 ` [PATCH v3 15/17] rv: Replace tss monitor with more complete sts Gabriele Monaco
@ 2025-07-15 7:14 ` Gabriele Monaco
2025-07-15 7:14 ` [PATCH v3 17/17] rv: Add opid per-cpu monitor Gabriele Monaco
4 siblings, 0 replies; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-15 7:14 UTC (permalink / raw)
To: linux-kernel, Steven Rostedt, Jonathan Corbet, Masami Hiramatsu,
linux-trace-kernel, linux-doc
Cc: Gabriele Monaco, Ingo Molnar, Peter Zijlstra, Nam Cao,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
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: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
Documentation/trace/rv/monitor_sched.rst | 168 +++++++++++++++++++++
kernel/trace/rv/Kconfig | 2 +
kernel/trace/rv/Makefile | 2 +
kernel/trace/rv/monitors/nrp/Kconfig | 14 ++
kernel/trace/rv/monitors/nrp/nrp.c | 146 ++++++++++++++++++
kernel/trace/rv/monitors/nrp/nrp.h | 87 +++++++++++
kernel/trace/rv/monitors/nrp/nrp_trace.h | 15 ++
kernel/trace/rv/monitors/sched/Kconfig | 1 +
kernel/trace/rv/monitors/sssw/Kconfig | 15 ++
kernel/trace/rv/monitors/sssw/sssw.c | 115 ++++++++++++++
kernel/trace/rv/monitors/sssw/sssw.h | 97 ++++++++++++
kernel/trace/rv/monitors/sssw/sssw_trace.h | 15 ++
kernel/trace/rv/rv_trace.h | 2 +
tools/verification/models/sched/nrp.dot | 29 ++++
tools/verification/models/sched/sssw.dot | 24 +++
15 files changed, 732 insertions(+)
create mode 100644 kernel/trace/rv/monitors/nrp/Kconfig
create mode 100644 kernel/trace/rv/monitors/nrp/nrp.c
create mode 100644 kernel/trace/rv/monitors/nrp/nrp.h
create mode 100644 kernel/trace/rv/monitors/nrp/nrp_trace.h
create mode 100644 kernel/trace/rv/monitors/sssw/Kconfig
create mode 100644 kernel/trace/rv/monitors/sssw/sssw.c
create mode 100644 kernel/trace/rv/monitors/sssw/sssw.h
create mode 100644 kernel/trace/rv/monitors/sssw/sssw_trace.h
create mode 100644 tools/verification/models/sched/nrp.dot
create mode 100644 tools/verification/models/sched/sssw.dot
diff --git a/Documentation/trace/rv/monitor_sched.rst b/Documentation/trace/rv/monitor_sched.rst
index e4171aadef1c2..4cb6590284786 100644
--- a/Documentation/trace/rv/monitor_sched.rst
+++ b/Documentation/trace/rv/monitor_sched.rst
@@ -198,6 +198,174 @@ real task switch occurs::
+-------------- | | <-------------+
+--------------------+
+Monitor nrp
+-----------
+
+The need resched preempts (nrp) monitor ensures preemption requires need
+resched. Only kernel preemptions are considered, since preemptions while
+returning to userspace, for this monitor, are indistinguishable from
+``sched_switch_yield`` (described in the sssw monitor).
+A preemption is any of the following types of ``sched_switch``:
+
+* ``sched_switch_preempt``:
+ a task is ``preempted``, this can happen after the need for ``rescheduling``
+ has been set. This is not valid for the *lazy* variant of the flag, which
+ causes only userspace preemptions.
+* ``sched_switch_vain_preempt``:
+ a task goes through the scheduler from a preemption context but it is picked
+ as the next task to run, hence no real task switch occurs. Since the
+ scheduler runs, this clears the need to reschedule.
+
+In theory, a preemption can only occur after the ``need_resched`` flag is set. In
+practice, however, it is possible to see a preemption where the flag is not
+set. This can happen in one specific condition::
+
+ need_resched
+ preempt_schedule()
+ preempt_schedule_irq()
+ __schedule()
+ !need_resched
+ __schedule()
+
+In the situation above, standard preemption starts (e.g. from preempt_enable
+when the flag is set), an interrupt occurs before scheduling and, on its exit
+path, it schedules, which clears the ``need_resched`` flag.
+When the preempted task runs again, the standard preemption started earlier
+resumes, although the flag is no longer set. The monitor considers this a
+``nested_preemption``, this allows another preemption without re-setting the
+flag. This condition relaxes the monitor constraints and may catch false
+negatives (i.e. no real ``nested_preemptions``) but makes the monitor more
+robust and able to validate other scenarios.
+For simplicity, the monitor starts in ``preempt_irq``, although no interrupt
+occurred, as the situation above is tricky to pinpoint.
+
+This monitor ignores when the task is switched in, as this complicates things
+when different types of ``sched_switch`` occur (e.g. sleeping or yielding, here
+marked as ``sched_switch_other`` or ``sched_switch_vain``). The snroc monitor
+ensures a task is switched in before it can be switched out again.
+For this reason, the ``any_thread_running`` state does not imply that the
+monitored task is not running, simply it is not set for rescheduling::
+
+ sched_switch_other
+ sched_switch_vain
+ irq_entry #===========================================#
+ +-------------------------- H H
+ | H H
+ +-------------------------> H any_thread_running H
+ H H
+ +-------------------------> H H
+ | #===========================================#
+ | | ^
+ | sched_switch_preempt | | sched_switch_preempt
+ | sched_switch_vain_preempt | sched_switch_vain_preempt
+ | sched_switch_other | sched_need_resched | sched_switch_other
+ | sched_switch_vain | | sched_switch_vain
+ | v |
+ | +----------------------+ |
+ | +--- | | |
+ | sched_need_resched | | rescheduling | -+
+ | +--> | |
+ | +----------------------+
+ | |
+ | | irq_exit
+ | v
+ | +----------------------+
+ | | | ---+
+ | ---> | | | sched_need_resched
+ | | preempt_irq | | irq_exit
+ | | | <--+
+ | | | <--+
+ | +----------------------+ |
+ | | |
+ | | sched_switch_preempt |
+ | sched_switch_vain_preempt |
+ | | sched_switch_other | sched_need_resched
+ | | sched_switch_vain |
+ | v |
+ | +-----------------------+ |
+ +-------------------------- | nested_preempt | --+
+ +-----------------------+
+ ^ irq_entry |
+ +-------------------+
+
+Due to how the ``need_resched`` flag on the preemption count works on arm64,
+this monitor is unstable on that architecture, as it often records preemptions
+when the flag is not set, even in presence of the workaround above.
+For the time being, the monitor is disabled by default on arm64.
+
+Monitor sssw
+------------
+
+The set state sleep and wakeup (sssw) monitor ensures ``sched_set_state`` to
+sleepable leads to sleeping and sleeping tasks require wakeup.
+It includes the following types of ``sched_switch``:
+
+* ``switch_suspend``:
+ a task puts itself to sleep, this can happen only after explicitly setting
+ the task to ``sleepable``. After a task is suspended, it needs to be woken up
+ (``waking`` state) before being switched in again.
+ Setting the task's state to ``sleepable`` can be reverted before switching if it
+ is woken up or set to ``runnable``.
+* ``switch_blocked``:
+ a special case of a ``switch_suspend`` where the task is waiting on a
+ sleeping RT lock (``PREEMPT_RT`` only), it is common to see wakeup and set
+ state events racing with each other and this leads the model to perceive this
+ type of switch when the task is not set to sleepable. This is a limitation of
+ the model in SMP system and workarounds may slow down the system.
+* ``switch_yield``:
+ a task explicitly calls the scheduler, this looks like a preemption as the
+ task is still runnable but the ``need_resched`` flag is not set. It can
+ happen after a ``yield`` system call or from the idle task. By definition,
+ a task cannot yield while ``sleepable`` as that would be a suspension.
+* ``switch_vain``:
+ a task explicitly calls the scheduler but it is picked as the next task to run,
+ hence no real task switch occurs. This can occur as a yield, which is not
+ valid when the task is sleepable. A special case of a yield is when a task in
+ ``TASK_INTERRUPTIBLE`` calls the scheduler while a signal is pending. The
+ task doesn't go through the usual blocking/waking and is set back to
+ runnable, the resulting switch looks like a yield.
+
+As for the nrp monitor, this monitor doesn't include a running state,
+``sleepable`` and ``runnable`` are only referring to the task's desired
+state, which could be scheduled out (e.g. due to preemption). However, it does
+include the event ``sched_switch_in`` to represent when a task is allowed to
+become running. This can be triggered also by preemption, but cannot occur
+after the task got to ``sleeping`` until a ``wakeup``::
+
+ sched_set_state_runnable
+ sched_wakeup
+ sched_switch_vain_preempt |
+ sched_switch_preempt |
+ sched_switch_yield v
+ sched_switch_vain #=============================================#
+ +-----------------> H H
+ | H H
+ +------------------ H runnable H
+ H H
+ +-----------------> H H
+ | #=============================================#
+ sched_set_state_runnable | | ^
+ sched_wakeup | sched_switch_blocking |
+ | sched_set_state_sleepable | |
+ | v | |
+ | +------------------------+ | sched_wakeup
+ +------------------ | | | |
+ | | | |
+ +-----------------> | sleepable | | |
+ | | | | |
+ +------------------ | | | |
+ sched_switch_in +------------------------+ | |
+ sched_switch_preempt | | |
+ sched_switch_vain_preempt sched_switch_suspend | |
+ sched_set_state_sleepable sched_switch_blocking | |
+ | | |
+ v | |
+ +------------------------+ | |
+ | sleeping | <-+ |
+ +------------------------+ |
+ | |
+ +----------------------------+
+
References
----------
diff --git a/kernel/trace/rv/Kconfig b/kernel/trace/rv/Kconfig
index da62d97d329aa..2cce1f22892a0 100644
--- a/kernel/trace/rv/Kconfig
+++ b/kernel/trace/rv/Kconfig
@@ -51,6 +51,8 @@ source "kernel/trace/rv/monitors/scpd/Kconfig"
source "kernel/trace/rv/monitors/snep/Kconfig"
source "kernel/trace/rv/monitors/sncid/Kconfig"
source "kernel/trace/rv/monitors/sts/Kconfig"
+source "kernel/trace/rv/monitors/nrp/Kconfig"
+source "kernel/trace/rv/monitors/sssw/Kconfig"
# Add new sched monitors here
source "kernel/trace/rv/monitors/rtapp/Kconfig"
diff --git a/kernel/trace/rv/Makefile b/kernel/trace/rv/Makefile
index db8653f7cedf3..3f517fe639c5a 100644
--- a/kernel/trace/rv/Makefile
+++ b/kernel/trace/rv/Makefile
@@ -15,6 +15,8 @@ obj-$(CONFIG_RV_MON_RTAPP) += monitors/rtapp/rtapp.o
obj-$(CONFIG_RV_MON_PAGEFAULT) += monitors/pagefault/pagefault.o
obj-$(CONFIG_RV_MON_SLEEP) += monitors/sleep/sleep.o
obj-$(CONFIG_RV_MON_STS) += monitors/sts/sts.o
+obj-$(CONFIG_RV_MON_NRP) += monitors/nrp/nrp.o
+obj-$(CONFIG_RV_MON_SSSW) += monitors/sssw/sssw.o
# Add new monitors here
obj-$(CONFIG_RV_REACTORS) += rv_reactors.o
obj-$(CONFIG_RV_REACT_PRINTK) += reactor_printk.o
diff --git a/kernel/trace/rv/monitors/nrp/Kconfig b/kernel/trace/rv/monitors/nrp/Kconfig
new file mode 100644
index 0000000000000..f37ff70e8d204
--- /dev/null
+++ b/kernel/trace/rv/monitors/nrp/Kconfig
@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+config RV_MON_NRP
+ depends on RV
+ depends on RV_MON_SCHED
+ default y if !ARCH_ARM64
+ select DA_MON_EVENTS_ID
+ bool "nrp monitor"
+ help
+ Monitor to ensure preemption requires need resched.
+ This monitor is part of the sched monitors collection.
+
+ For further information, see:
+ Documentation/trace/rv/monitor_sched.rst
diff --git a/kernel/trace/rv/monitors/nrp/nrp.c b/kernel/trace/rv/monitors/nrp/nrp.c
new file mode 100644
index 0000000000000..90600497dc487
--- /dev/null
+++ b/kernel/trace/rv/monitors/nrp/nrp.c
@@ -0,0 +1,146 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/ftrace.h>
+#include <linux/tracepoint.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/rv.h>
+#include <rv/instrumentation.h>
+#include <rv/da_monitor.h>
+
+#define MODULE_NAME "nrp"
+
+#include <trace/events/irq.h>
+#include <trace/events/sched.h>
+#include <rv_trace.h>
+#include <monitors/sched/sched.h>
+
+#include "nrp.h"
+
+static struct rv_monitor rv_nrp;
+DECLARE_DA_MON_PER_TASK(nrp, unsigned char);
+#ifdef CONFIG_X86_LOCAL_APIC
+#include <asm/trace/irq_vectors.h>
+
+static void handle_vector_irq_entry(void *data, int vector)
+{
+ da_handle_event_nrp(current, irq_entry_nrp);
+}
+
+static void attach_vector_irq(void)
+{
+ rv_attach_trace_probe("nrp", local_timer_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_IRQ_WORK))
+ rv_attach_trace_probe("nrp", irq_work_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_SMP)) {
+ rv_attach_trace_probe("nrp", reschedule_entry, handle_vector_irq_entry);
+ rv_attach_trace_probe("nrp", call_function_entry, handle_vector_irq_entry);
+ rv_attach_trace_probe("nrp", call_function_single_entry, handle_vector_irq_entry);
+ }
+}
+
+static void detach_vector_irq(void)
+{
+ rv_detach_trace_probe("nrp", local_timer_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_IRQ_WORK))
+ rv_detach_trace_probe("nrp", irq_work_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_SMP)) {
+ rv_detach_trace_probe("nrp", reschedule_entry, handle_vector_irq_entry);
+ rv_detach_trace_probe("nrp", call_function_entry, handle_vector_irq_entry);
+ rv_detach_trace_probe("nrp", call_function_single_entry, handle_vector_irq_entry);
+ }
+}
+
+#else
+/* We assume irq_entry tracepoints are sufficient on other architectures */
+static void attach_vector_irq(void) { }
+static void detach_vector_irq(void) { }
+#endif
+static void handle_irq_entry(void *data, int irq, struct irqaction *action)
+{
+ da_handle_event_nrp(current, irq_entry_nrp);
+}
+
+static void handle_sched_need_resched(void *data, struct task_struct *tsk,
+ int cpu, int tif)
+{
+ if (tif == TIF_NEED_RESCHED)
+ da_handle_start_event_nrp(tsk, sched_need_resched_nrp);
+}
+
+static void handle_sched_switch(void *data, bool preempt,
+ struct task_struct *prev,
+ struct task_struct *next,
+ unsigned int prev_state)
+{
+ if (preempt)
+ da_handle_event_nrp(prev, sched_switch_preempt_nrp);
+ else
+ da_handle_event_nrp(prev, sched_switch_other_nrp);
+}
+
+static void handle_sched_switch_vain(void *data, bool preempt,
+ struct task_struct *tsk,
+ unsigned int tsk_state)
+{
+ if (preempt)
+ da_handle_event_nrp(tsk, sched_switch_vain_preempt_nrp);
+ else
+ da_handle_event_nrp(tsk, sched_switch_vain_nrp);
+}
+
+static int enable_nrp(void)
+{
+ int retval;
+
+ retval = da_monitor_init_nrp();
+ if (retval)
+ return retval;
+
+ rv_attach_trace_probe("nrp", irq_handler_entry, handle_irq_entry);
+ rv_attach_trace_probe("nrp", sched_set_need_resched_tp, handle_sched_need_resched);
+ rv_attach_trace_probe("nrp", sched_switch, handle_sched_switch);
+ rv_attach_trace_probe("nrp", sched_switch_vain_tp, handle_sched_switch_vain);
+ attach_vector_irq();
+
+ return 0;
+}
+
+static void disable_nrp(void)
+{
+ rv_nrp.enabled = 0;
+
+ rv_detach_trace_probe("nrp", irq_handler_entry, handle_irq_entry);
+ rv_detach_trace_probe("nrp", sched_set_need_resched_tp, handle_sched_need_resched);
+ rv_detach_trace_probe("nrp", sched_switch, handle_sched_switch);
+ rv_detach_trace_probe("nrp", sched_switch_vain_tp, handle_sched_switch_vain);
+ detach_vector_irq();
+
+ da_monitor_destroy_nrp();
+}
+
+static struct rv_monitor rv_nrp = {
+ .name = "nrp",
+ .description = "need resched preempts.",
+ .enable = enable_nrp,
+ .disable = disable_nrp,
+ .reset = da_monitor_reset_all_nrp,
+ .enabled = 0,
+};
+
+static int __init register_nrp(void)
+{
+ return rv_register_monitor(&rv_nrp, &rv_sched);
+}
+
+static void __exit unregister_nrp(void)
+{
+ rv_unregister_monitor(&rv_nrp);
+}
+
+module_init(register_nrp);
+module_exit(unregister_nrp);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>");
+MODULE_DESCRIPTION("nrp: need resched preempts.");
diff --git a/kernel/trace/rv/monitors/nrp/nrp.h b/kernel/trace/rv/monitors/nrp/nrp.h
new file mode 100644
index 0000000000000..376c5a03dd2e9
--- /dev/null
+++ b/kernel/trace/rv/monitors/nrp/nrp.h
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Automatically generated C representation of nrp automaton
+ * For further information about this format, see kernel documentation:
+ * Documentation/trace/rv/deterministic_automata.rst
+ */
+
+enum states_nrp {
+ preempt_irq_nrp = 0,
+ any_thread_running_nrp,
+ nested_preempt_nrp,
+ rescheduling_nrp,
+ state_max_nrp
+};
+
+#define INVALID_STATE state_max_nrp
+
+enum events_nrp {
+ irq_entry_nrp = 0,
+ sched_need_resched_nrp,
+ sched_switch_other_nrp,
+ sched_switch_preempt_nrp,
+ sched_switch_vain_nrp,
+ sched_switch_vain_preempt_nrp,
+ event_max_nrp
+};
+
+struct automaton_nrp {
+ char *state_names[state_max_nrp];
+ char *event_names[event_max_nrp];
+ unsigned char function[state_max_nrp][event_max_nrp];
+ unsigned char initial_state;
+ bool final_states[state_max_nrp];
+};
+
+static const struct automaton_nrp automaton_nrp = {
+ .state_names = {
+ "preempt_irq",
+ "any_thread_running",
+ "nested_preempt",
+ "rescheduling"
+ },
+ .event_names = {
+ "irq_entry",
+ "sched_need_resched",
+ "sched_switch_other",
+ "sched_switch_preempt",
+ "sched_switch_vain",
+ "sched_switch_vain_preempt"
+ },
+ .function = {
+ {
+ preempt_irq_nrp,
+ preempt_irq_nrp,
+ nested_preempt_nrp,
+ nested_preempt_nrp,
+ nested_preempt_nrp,
+ nested_preempt_nrp
+ },
+ {
+ any_thread_running_nrp,
+ rescheduling_nrp,
+ any_thread_running_nrp,
+ INVALID_STATE,
+ any_thread_running_nrp,
+ INVALID_STATE
+ },
+ {
+ nested_preempt_nrp,
+ preempt_irq_nrp,
+ any_thread_running_nrp,
+ any_thread_running_nrp,
+ any_thread_running_nrp,
+ any_thread_running_nrp
+ },
+ {
+ preempt_irq_nrp,
+ rescheduling_nrp,
+ any_thread_running_nrp,
+ any_thread_running_nrp,
+ any_thread_running_nrp,
+ any_thread_running_nrp
+ },
+ },
+ .initial_state = preempt_irq_nrp,
+ .final_states = { 0, 1, 0, 0 },
+};
diff --git a/kernel/trace/rv/monitors/nrp/nrp_trace.h b/kernel/trace/rv/monitors/nrp/nrp_trace.h
new file mode 100644
index 0000000000000..2e13497de3b6f
--- /dev/null
+++ b/kernel/trace/rv/monitors/nrp/nrp_trace.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+/*
+ * Snippet to be included in rv_trace.h
+ */
+
+#ifdef CONFIG_RV_MON_NRP
+DEFINE_EVENT(event_da_monitor_id, event_nrp,
+ TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state),
+ TP_ARGS(id, state, event, next_state, final_state));
+
+DEFINE_EVENT(error_da_monitor_id, error_nrp,
+ TP_PROTO(int id, char *state, char *event),
+ TP_ARGS(id, state, event));
+#endif /* CONFIG_RV_MON_NRP */
diff --git a/kernel/trace/rv/monitors/sched/Kconfig b/kernel/trace/rv/monitors/sched/Kconfig
index ae3eb410abd78..aa16456da8647 100644
--- a/kernel/trace/rv/monitors/sched/Kconfig
+++ b/kernel/trace/rv/monitors/sched/Kconfig
@@ -2,6 +2,7 @@
#
config RV_MON_SCHED
depends on RV
+ depends on RV_PER_TASK_MONITORS >= 3
bool "sched monitor"
help
Collection of monitors to check the scheduler behaves according to specifications.
diff --git a/kernel/trace/rv/monitors/sssw/Kconfig b/kernel/trace/rv/monitors/sssw/Kconfig
new file mode 100644
index 0000000000000..23b7eeb38bbfc
--- /dev/null
+++ b/kernel/trace/rv/monitors/sssw/Kconfig
@@ -0,0 +1,15 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+config RV_MON_SSSW
+ depends on RV
+ depends on RV_MON_SCHED
+ default y
+ select DA_MON_EVENTS_ID
+ bool "sssw monitor"
+ help
+ Monitor to ensure sched_set_state to sleepable leads to sleeping and
+ sleeping tasks require wakeup.
+ This monitor is part of the sched monitors collection.
+
+ For further information, see:
+ Documentation/trace/rv/monitor_sched.rst
diff --git a/kernel/trace/rv/monitors/sssw/sssw.c b/kernel/trace/rv/monitors/sssw/sssw.c
new file mode 100644
index 0000000000000..efa695dead36a
--- /dev/null
+++ b/kernel/trace/rv/monitors/sssw/sssw.c
@@ -0,0 +1,115 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/ftrace.h>
+#include <linux/tracepoint.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/rv.h>
+#include <rv/instrumentation.h>
+#include <rv/da_monitor.h>
+
+#define MODULE_NAME "sssw"
+
+#include <trace/events/sched.h>
+#include <rv_trace.h>
+#include <monitors/sched/sched.h>
+
+#include "sssw.h"
+
+static struct rv_monitor rv_sssw;
+DECLARE_DA_MON_PER_TASK(sssw, unsigned char);
+
+static void handle_sched_set_state(void *data, struct task_struct *tsk,
+ int state, bool from_signal)
+{
+ if (state == TASK_RUNNING)
+ da_handle_start_event_sssw(tsk, sched_set_state_runnable_sssw);
+ else
+ da_handle_event_sssw(tsk, sched_set_state_sleepable_sssw);
+}
+
+static void handle_sched_switch(void *data, bool preempt,
+ struct task_struct *prev,
+ struct task_struct *next,
+ unsigned int prev_state)
+{
+ if (preempt)
+ da_handle_event_sssw(prev, sched_switch_preempt_sssw);
+ else if (prev_state == TASK_RUNNING)
+ da_handle_start_event_sssw(prev, sched_switch_yield_sssw);
+ else if (prev_state == TASK_RTLOCK_WAIT)
+ /* special case of sleeping task with racy conditions */
+ da_handle_event_sssw(prev, sched_switch_blocking_sssw);
+ else
+ da_handle_event_sssw(prev, sched_switch_suspend_sssw);
+ da_handle_event_sssw(next, sched_switch_in_sssw);
+}
+
+static void handle_sched_switch_vain(void *data, bool preempt,
+ struct task_struct *tsk,
+ unsigned int tsk_state)
+{
+ if (preempt)
+ da_handle_event_sssw(tsk, sched_switch_vain_preempt_sssw);
+ else
+ da_handle_start_event_sssw(tsk, sched_switch_vain_sssw);
+}
+
+static void handle_sched_wakeup(void *data, struct task_struct *p)
+{
+ da_handle_start_event_sssw(p, sched_wakeup_sssw);
+}
+
+static int enable_sssw(void)
+{
+ int retval;
+
+ retval = da_monitor_init_sssw();
+ if (retval)
+ return retval;
+
+ rv_attach_trace_probe("sssw", sched_set_state_tp, handle_sched_set_state);
+ rv_attach_trace_probe("sssw", sched_switch, handle_sched_switch);
+ rv_attach_trace_probe("sssw", sched_switch_vain_tp, handle_sched_switch_vain);
+ rv_attach_trace_probe("sssw", sched_wakeup, handle_sched_wakeup);
+
+ return 0;
+}
+
+static void disable_sssw(void)
+{
+ rv_sssw.enabled = 0;
+
+ rv_detach_trace_probe("sssw", sched_set_state_tp, handle_sched_set_state);
+ rv_detach_trace_probe("sssw", sched_switch, handle_sched_switch);
+ rv_detach_trace_probe("sssw", sched_switch_vain_tp, handle_sched_switch_vain);
+ rv_detach_trace_probe("sssw", sched_wakeup, handle_sched_wakeup);
+
+ da_monitor_destroy_sssw();
+}
+
+static struct rv_monitor rv_sssw = {
+ .name = "sssw",
+ .description = "set state sleep and wakeup.",
+ .enable = enable_sssw,
+ .disable = disable_sssw,
+ .reset = da_monitor_reset_all_sssw,
+ .enabled = 0,
+};
+
+static int __init register_sssw(void)
+{
+ return rv_register_monitor(&rv_sssw, &rv_sched);
+}
+
+static void __exit unregister_sssw(void)
+{
+ rv_unregister_monitor(&rv_sssw);
+}
+
+module_init(register_sssw);
+module_exit(unregister_sssw);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>");
+MODULE_DESCRIPTION("sssw: set state sleep and wakeup.");
diff --git a/kernel/trace/rv/monitors/sssw/sssw.h b/kernel/trace/rv/monitors/sssw/sssw.h
new file mode 100644
index 0000000000000..2b29230f29cff
--- /dev/null
+++ b/kernel/trace/rv/monitors/sssw/sssw.h
@@ -0,0 +1,97 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Automatically generated C representation of sssw automaton
+ * For further information about this format, see kernel documentation:
+ * Documentation/trace/rv/deterministic_automata.rst
+ */
+
+enum states_sssw {
+ runnanble_sssw = 0,
+ sleepable_sssw,
+ sleeping_sssw,
+ state_max_sssw
+};
+
+#define INVALID_STATE state_max_sssw
+
+enum events_sssw {
+ sched_set_state_runnable_sssw = 0,
+ sched_set_state_sleepable_sssw,
+ sched_switch_blocking_sssw,
+ sched_switch_in_sssw,
+ sched_switch_preempt_sssw,
+ sched_switch_suspend_sssw,
+ sched_switch_vain_sssw,
+ sched_switch_vain_preempt_sssw,
+ sched_switch_yield_sssw,
+ sched_wakeup_sssw,
+ event_max_sssw
+};
+
+struct automaton_sssw {
+ char *state_names[state_max_sssw];
+ char *event_names[event_max_sssw];
+ unsigned char function[state_max_sssw][event_max_sssw];
+ unsigned char initial_state;
+ bool final_states[state_max_sssw];
+};
+
+static const struct automaton_sssw automaton_sssw = {
+ .state_names = {
+ "runnanble",
+ "sleepable",
+ "sleeping"
+ },
+ .event_names = {
+ "sched_set_state_runnable",
+ "sched_set_state_sleepable",
+ "sched_switch_blocking",
+ "sched_switch_in",
+ "sched_switch_preempt",
+ "sched_switch_suspend",
+ "sched_switch_vain",
+ "sched_switch_vain_preempt",
+ "sched_switch_yield",
+ "sched_wakeup"
+ },
+ .function = {
+ {
+ runnanble_sssw,
+ sleepable_sssw,
+ sleeping_sssw,
+ runnanble_sssw,
+ runnanble_sssw,
+ INVALID_STATE,
+ runnanble_sssw,
+ runnanble_sssw,
+ runnanble_sssw,
+ runnanble_sssw
+ },
+ {
+ runnanble_sssw,
+ sleepable_sssw,
+ sleeping_sssw,
+ sleepable_sssw,
+ sleepable_sssw,
+ sleeping_sssw,
+ INVALID_STATE,
+ sleepable_sssw,
+ INVALID_STATE,
+ runnanble_sssw
+ },
+ {
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ runnanble_sssw
+ },
+ },
+ .initial_state = runnanble_sssw,
+ .final_states = { 1, 0, 0 },
+};
diff --git a/kernel/trace/rv/monitors/sssw/sssw_trace.h b/kernel/trace/rv/monitors/sssw/sssw_trace.h
new file mode 100644
index 0000000000000..6c03cfc6960bf
--- /dev/null
+++ b/kernel/trace/rv/monitors/sssw/sssw_trace.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+/*
+ * Snippet to be included in rv_trace.h
+ */
+
+#ifdef CONFIG_RV_MON_SSSW
+DEFINE_EVENT(event_da_monitor_id, event_sssw,
+ TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state),
+ TP_ARGS(id, state, event, next_state, final_state));
+
+DEFINE_EVENT(error_da_monitor_id, error_sssw,
+ TP_PROTO(int id, char *state, char *event),
+ TP_ARGS(id, state, event));
+#endif /* CONFIG_RV_MON_SSSW */
diff --git a/kernel/trace/rv/rv_trace.h b/kernel/trace/rv/rv_trace.h
index dbd2f2ef513a7..b5cb1d1807fd2 100644
--- a/kernel/trace/rv/rv_trace.h
+++ b/kernel/trace/rv/rv_trace.h
@@ -124,6 +124,8 @@ DECLARE_EVENT_CLASS(error_da_monitor_id,
#include <monitors/wwnr/wwnr_trace.h>
#include <monitors/snroc/snroc_trace.h>
+#include <monitors/nrp/nrp_trace.h>
+#include <monitors/sssw/sssw_trace.h>
// Add new monitors based on CONFIG_DA_MON_EVENTS_ID here
#endif /* CONFIG_DA_MON_EVENTS_ID */
diff --git a/tools/verification/models/sched/nrp.dot b/tools/verification/models/sched/nrp.dot
new file mode 100644
index 0000000000000..337ea310f7f8a
--- /dev/null
+++ b/tools/verification/models/sched/nrp.dot
@@ -0,0 +1,29 @@
+digraph state_automaton {
+ center = true;
+ size = "7,11";
+ {node [shape = doublecircle] "any_thread_running"};
+ {node [shape = circle] "any_thread_running"};
+ {node [shape = circle] "nested_preempt"};
+ {node [shape = plaintext, style=invis, label=""] "__init_preempt_irq"};
+ {node [shape = circle] "preempt_irq"};
+ {node [shape = circle] "rescheduling"};
+ "__init_preempt_irq" -> "preempt_irq";
+ "any_thread_running" [label = "any_thread_running", color = green3];
+ "any_thread_running" -> "any_thread_running" [ label = "sched_switch_other\nsched_switch_vain\nirq_entry" ];
+ "any_thread_running" -> "rescheduling" [ label = "sched_need_resched" ];
+ "nested_preempt" [label = "nested_preempt"];
+ "nested_preempt" -> "any_thread_running" [ label = "sched_switch_preempt\nsched_switch_vain_preempt\nsched_switch_other\nsched_switch_vain" ];
+ "nested_preempt" -> "nested_preempt" [ label = "irq_entry" ];
+ "nested_preempt" -> "preempt_irq" [ label = "sched_need_resched" ];
+ "preempt_irq" [label = "preempt_irq"];
+ "preempt_irq" -> "nested_preempt" [ label = "sched_switch_preempt\nsched_switch_vain_preempt\nsched_switch_other\nsched_switch_vain" ];
+ "preempt_irq" -> "preempt_irq" [ label = "irq_entry\nsched_need_resched" ];
+ "rescheduling" [label = "rescheduling"];
+ "rescheduling" -> "any_thread_running" [ label = "sched_switch_preempt\nsched_switch_vain_preempt\nsched_switch_other\nsched_switch_vain" ];
+ "rescheduling" -> "preempt_irq" [ label = "irq_entry" ];
+ "rescheduling" -> "rescheduling" [ label = "sched_need_resched" ];
+ { rank = min ;
+ "__init_preempt_irq";
+ "preempt_irq";
+ }
+}
diff --git a/tools/verification/models/sched/sssw.dot b/tools/verification/models/sched/sssw.dot
new file mode 100644
index 0000000000000..362e783f2bd58
--- /dev/null
+++ b/tools/verification/models/sched/sssw.dot
@@ -0,0 +1,24 @@
+digraph state_automaton {
+ center = true;
+ size = "7,11";
+ {node [shape = plaintext, style=invis, label=""] "__init_runnanble"};
+ {node [shape = doublecircle] "runnanble"};
+ {node [shape = circle] "runnanble"};
+ {node [shape = circle] "sleepable"};
+ {node [shape = circle] "sleeping"};
+ "__init_runnanble" -> "runnanble";
+ "runnanble" [label = "runnanble", color = green3];
+ "runnanble" -> "runnanble" [ label = "sched_set_state_runnable\nsched_wakeup\nsched_switch_in\nsched_switch_vain\nsched_switch_yield\nsched_switch_preempt\nsched_switch_vain_preempt" ];
+ "runnanble" -> "sleepable" [ label = "sched_set_state_sleepable" ];
+ "runnanble" -> "sleeping" [ label = "sched_switch_blocking" ];
+ "sleepable" [label = "sleepable"];
+ "sleepable" -> "runnanble" [ label = "sched_set_state_runnable\nsched_wakeup" ];
+ "sleepable" -> "sleepable" [ label = "sched_set_state_sleepable\nsched_switch_in\nsched_switch_preempt\nsched_switch_vain_preempt" ];
+ "sleepable" -> "sleeping" [ label = "sched_switch_suspend\nsched_switch_blocking" ];
+ "sleeping" [label = "sleeping"];
+ "sleeping" -> "runnanble" [ label = "sched_wakeup" ];
+ { rank = min ;
+ "__init_runnanble";
+ "runnanble";
+ }
+}
--
2.50.1
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH v3 17/17] rv: Add opid per-cpu monitor
[not found] <20250715071434.22508-1-gmonaco@redhat.com>
` (3 preceding siblings ...)
2025-07-15 7:14 ` [PATCH v3 16/17] rv: Add nrp and sssw per-task monitors Gabriele Monaco
@ 2025-07-15 7:14 ` Gabriele Monaco
2025-07-16 9:38 ` Nam Cao
4 siblings, 1 reply; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-15 7:14 UTC (permalink / raw)
To: linux-kernel, Steven Rostedt, Jonathan Corbet, Masami Hiramatsu,
linux-trace-kernel, linux-doc
Cc: Gabriele Monaco, Ingo Molnar, Peter Zijlstra, Nam Cao,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
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.
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
Documentation/trace/rv/monitor_sched.rst | 55 +++++++
kernel/trace/rv/Kconfig | 1 +
kernel/trace/rv/Makefile | 1 +
kernel/trace/rv/monitors/nrp/Kconfig | 2 +-
kernel/trace/rv/monitors/opid/Kconfig | 17 +++
kernel/trace/rv/monitors/opid/opid.c | 169 +++++++++++++++++++++
kernel/trace/rv/monitors/opid/opid.h | 104 +++++++++++++
kernel/trace/rv/monitors/opid/opid_trace.h | 15 ++
kernel/trace/rv/rv_trace.h | 1 +
tools/verification/models/sched/opid.dot | 35 +++++
10 files changed, 399 insertions(+), 1 deletion(-)
create mode 100644 kernel/trace/rv/monitors/opid/Kconfig
create mode 100644 kernel/trace/rv/monitors/opid/opid.c
create mode 100644 kernel/trace/rv/monitors/opid/opid.h
create mode 100644 kernel/trace/rv/monitors/opid/opid_trace.h
create mode 100644 tools/verification/models/sched/opid.dot
diff --git a/Documentation/trace/rv/monitor_sched.rst b/Documentation/trace/rv/monitor_sched.rst
index 4cb6590284786..6f796e22dbab6 100644
--- a/Documentation/trace/rv/monitor_sched.rst
+++ b/Documentation/trace/rv/monitor_sched.rst
@@ -366,6 +366,61 @@ after the task got to ``sleeping`` until a ``wakeup``::
| |
+----------------------------+
+Monitor opid
+------------
+
+The operations with preemption and irq disabled (opid) monitor ensures
+operations like ``wakeup`` and ``need_resched`` occur with interrupts and
+preemption disabled or during interrupt context, in such case preemption may
+not be disabled explicitly.
+``need_resched`` can be set by some RCU internals functions, in which case it
+doesn't match a task wakeup and might occur with only interrupts disabled::
+
+ | sched_need_resched
+ | sched_waking
+ | irq_entry
+ | +--------------------+
+ v v |
+ +------------------------------------------------------+
+ +----------- | disabled | <+
+ | +------------------------------------------------------+ |
+ | | ^ |
+ | | preempt_disable sched_need_resched |
+ | preempt_enable | +--------------------+ |
+ | v | v | |
+ | +------------------------------------------------------+ |
+ | | irq_disabled | |
+ | +------------------------------------------------------+ |
+ | | | ^ |
+ | irq_entry | | |
+ | sched_need_resched v | irq_disable |
+ | sched_waking +--------------+ | | |
+ | +----- | | irq_enable | |
+ | | | in_irq | | | |
+ | +----> | | | | |
+ | +--------------+ | | irq_disable
+ | | | | |
+ | irq_enable | irq_enable | | |
+ | v v | |
+ | #======================================================# |
+ | H enabled H |
+ | #======================================================# |
+ | | ^ ^ preempt_enable | |
+ | preempt_disable preempt_enable +--------------------+ |
+ | v | |
+ | +------------------+ | |
+ +----------> | preempt_disabled | -+ |
+ +------------------+ |
+ | |
+ +-------------------------------------------------------+
+
+This monitor is designed to work on ``PREEMPT_RT`` kernels, the special case of
+events occurring in interrupt context is a shortcut to identify valid scenarios
+where the preemption tracepoints might not be visible, during interrupts
+preemption is always disabled. On non- ``PREEMPT_RT`` kernels, the interrupts
+might invoke a softirq to set ``need_resched`` and wake up a task. This is
+another special case that is currently not supported by the monitor.
+
References
----------
diff --git a/kernel/trace/rv/Kconfig b/kernel/trace/rv/Kconfig
index 2cce1f22892a0..86d4c59ca4966 100644
--- a/kernel/trace/rv/Kconfig
+++ b/kernel/trace/rv/Kconfig
@@ -53,6 +53,7 @@ source "kernel/trace/rv/monitors/sncid/Kconfig"
source "kernel/trace/rv/monitors/sts/Kconfig"
source "kernel/trace/rv/monitors/nrp/Kconfig"
source "kernel/trace/rv/monitors/sssw/Kconfig"
+source "kernel/trace/rv/monitors/opid/Kconfig"
# Add new sched monitors here
source "kernel/trace/rv/monitors/rtapp/Kconfig"
diff --git a/kernel/trace/rv/Makefile b/kernel/trace/rv/Makefile
index 3f517fe639c5a..955d0947122ce 100644
--- a/kernel/trace/rv/Makefile
+++ b/kernel/trace/rv/Makefile
@@ -17,6 +17,7 @@ obj-$(CONFIG_RV_MON_SLEEP) += monitors/sleep/sleep.o
obj-$(CONFIG_RV_MON_STS) += monitors/sts/sts.o
obj-$(CONFIG_RV_MON_NRP) += monitors/nrp/nrp.o
obj-$(CONFIG_RV_MON_SSSW) += monitors/sssw/sssw.o
+obj-$(CONFIG_RV_MON_OPID) += monitors/opid/opid.o
# Add new monitors here
obj-$(CONFIG_RV_REACTORS) += rv_reactors.o
obj-$(CONFIG_RV_REACT_PRINTK) += reactor_printk.o
diff --git a/kernel/trace/rv/monitors/nrp/Kconfig b/kernel/trace/rv/monitors/nrp/Kconfig
index f37ff70e8d204..a175c430d351f 100644
--- a/kernel/trace/rv/monitors/nrp/Kconfig
+++ b/kernel/trace/rv/monitors/nrp/Kconfig
@@ -3,7 +3,7 @@
config RV_MON_NRP
depends on RV
depends on RV_MON_SCHED
- default y if !ARCH_ARM64
+ default y if !ARM64
select DA_MON_EVENTS_ID
bool "nrp monitor"
help
diff --git a/kernel/trace/rv/monitors/opid/Kconfig b/kernel/trace/rv/monitors/opid/Kconfig
new file mode 100644
index 0000000000000..23b43d2704153
--- /dev/null
+++ b/kernel/trace/rv/monitors/opid/Kconfig
@@ -0,0 +1,17 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+config RV_MON_OPID
+ depends on RV
+ depends on TRACE_IRQFLAGS
+ depends on TRACE_PREEMPT_TOGGLE
+ depends on RV_MON_SCHED
+ default y if PREEMPT_RT
+ select DA_MON_EVENTS_IMPLICIT
+ bool "opid monitor"
+ help
+ Monitor to ensure operations like wakeup and need resched occur with
+ interrupts and preemption disabled or during IRQs, where preemption
+ may not be disabled explicitly.
+
+ For further information, see:
+ Documentation/trace/rv/monitor_sched.rst
diff --git a/kernel/trace/rv/monitors/opid/opid.c b/kernel/trace/rv/monitors/opid/opid.c
new file mode 100644
index 0000000000000..4f1902f24805c
--- /dev/null
+++ b/kernel/trace/rv/monitors/opid/opid.c
@@ -0,0 +1,169 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/ftrace.h>
+#include <linux/tracepoint.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/rv.h>
+#include <rv/instrumentation.h>
+#include <rv/da_monitor.h>
+
+#define MODULE_NAME "opid"
+
+#include <trace/events/sched.h>
+#include <trace/events/irq.h>
+#include <trace/events/preemptirq.h>
+#include <rv_trace.h>
+#include <monitors/sched/sched.h>
+
+#include "opid.h"
+
+static struct rv_monitor rv_opid;
+DECLARE_DA_MON_PER_CPU(opid, unsigned char);
+
+#ifdef CONFIG_X86_LOCAL_APIC
+#include <asm/trace/irq_vectors.h>
+
+static void handle_vector_irq_entry(void *data, int vector)
+{
+ da_handle_event_opid(irq_entry_opid);
+}
+
+static void attach_vector_irq(void)
+{
+ rv_attach_trace_probe("opid", local_timer_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_IRQ_WORK))
+ rv_attach_trace_probe("opid", irq_work_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_SMP)) {
+ rv_attach_trace_probe("opid", reschedule_entry, handle_vector_irq_entry);
+ rv_attach_trace_probe("opid", call_function_entry, handle_vector_irq_entry);
+ rv_attach_trace_probe("opid", call_function_single_entry, handle_vector_irq_entry);
+ }
+}
+
+static void detach_vector_irq(void)
+{
+ rv_detach_trace_probe("opid", local_timer_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_IRQ_WORK))
+ rv_detach_trace_probe("opid", irq_work_entry, handle_vector_irq_entry);
+ if (IS_ENABLED(CONFIG_SMP)) {
+ rv_detach_trace_probe("opid", reschedule_entry, handle_vector_irq_entry);
+ rv_detach_trace_probe("opid", call_function_entry, handle_vector_irq_entry);
+ rv_detach_trace_probe("opid", call_function_single_entry, handle_vector_irq_entry);
+ }
+}
+
+#else
+/* We assume irq_entry tracepoints are sufficient on other architectures */
+static void attach_vector_irq(void) { }
+static void detach_vector_irq(void) { }
+#endif
+
+static void handle_irq_disable(void *data, unsigned long ip, unsigned long parent_ip)
+{
+ da_handle_event_opid(irq_disable_opid);
+}
+
+static void handle_irq_enable(void *data, unsigned long ip, unsigned long parent_ip)
+{
+ da_handle_event_opid(irq_enable_opid);
+}
+
+static void handle_irq_entry(void *data, int irq, struct irqaction *action)
+{
+ da_handle_event_opid(irq_entry_opid);
+}
+
+static void handle_preempt_disable(void *data, unsigned long ip, unsigned long parent_ip)
+{
+ da_handle_event_opid(preempt_disable_opid);
+}
+
+static void handle_preempt_enable(void *data, unsigned long ip, unsigned long parent_ip)
+{
+ da_handle_event_opid(preempt_enable_opid);
+}
+
+static void handle_sched_need_resched(void *data, struct task_struct *tsk, int cpu, int tif)
+{
+ /* The monitor's intitial state is not in_irq */
+ if (this_cpu_read(hardirq_context))
+ da_handle_event_opid(sched_need_resched_opid);
+ else
+ da_handle_start_event_opid(sched_need_resched_opid);
+}
+
+static void handle_sched_waking(void *data, struct task_struct *p)
+{
+ /* The monitor's intitial state is not in_irq */
+ if (this_cpu_read(hardirq_context))
+ da_handle_event_opid(sched_waking_opid);
+ else
+ da_handle_start_event_opid(sched_waking_opid);
+}
+
+static int enable_opid(void)
+{
+ int retval;
+
+ retval = da_monitor_init_opid();
+ if (retval)
+ return retval;
+
+ rv_attach_trace_probe("opid", irq_disable, handle_irq_disable);
+ rv_attach_trace_probe("opid", irq_enable, handle_irq_enable);
+ rv_attach_trace_probe("opid", irq_handler_entry, handle_irq_entry);
+ rv_attach_trace_probe("opid", preempt_disable, handle_preempt_disable);
+ rv_attach_trace_probe("opid", preempt_enable, handle_preempt_enable);
+ rv_attach_trace_probe("opid", sched_set_need_resched_tp, handle_sched_need_resched);
+ rv_attach_trace_probe("opid", sched_waking, handle_sched_waking);
+ attach_vector_irq();
+
+ return 0;
+}
+
+static void disable_opid(void)
+{
+ rv_opid.enabled = 0;
+
+ rv_detach_trace_probe("opid", irq_disable, handle_irq_disable);
+ rv_detach_trace_probe("opid", irq_enable, handle_irq_enable);
+ rv_detach_trace_probe("opid", irq_handler_entry, handle_irq_entry);
+ rv_detach_trace_probe("opid", preempt_disable, handle_preempt_disable);
+ rv_detach_trace_probe("opid", preempt_enable, handle_preempt_enable);
+ rv_detach_trace_probe("opid", sched_set_need_resched_tp, handle_sched_need_resched);
+ rv_detach_trace_probe("opid", sched_waking, handle_sched_waking);
+ detach_vector_irq();
+
+ da_monitor_destroy_opid();
+}
+
+/*
+ * This is the monitor register section.
+ */
+static struct rv_monitor rv_opid = {
+ .name = "opid",
+ .description = "operations with preemption and irq disabled.",
+ .enable = enable_opid,
+ .disable = disable_opid,
+ .reset = da_monitor_reset_all_opid,
+ .enabled = 0,
+};
+
+static int __init register_opid(void)
+{
+ rv_register_monitor(&rv_opid, &rv_sched);
+ return 0;
+}
+
+static void __exit unregister_opid(void)
+{
+ rv_unregister_monitor(&rv_opid);
+}
+
+module_init(register_opid);
+module_exit(unregister_opid);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>");
+MODULE_DESCRIPTION("opid: operations with preemption and irq disabled.");
diff --git a/kernel/trace/rv/monitors/opid/opid.h b/kernel/trace/rv/monitors/opid/opid.h
new file mode 100644
index 0000000000000..efdbcda8fff30
--- /dev/null
+++ b/kernel/trace/rv/monitors/opid/opid.h
@@ -0,0 +1,104 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Automatically generated C representation of opid automaton
+ * For further information about this format, see kernel documentation:
+ * Documentation/trace/rv/deterministic_automata.rst
+ */
+
+enum states_opid {
+ disabled_opid = 0,
+ enabled_opid,
+ in_irq_opid,
+ irq_disabled_opid,
+ preempt_disabled_opid,
+ state_max_opid
+};
+
+#define INVALID_STATE state_max_opid
+
+enum events_opid {
+ irq_disable_opid = 0,
+ irq_enable_opid,
+ irq_entry_opid,
+ preempt_disable_opid,
+ preempt_enable_opid,
+ sched_need_resched_opid,
+ sched_waking_opid,
+ event_max_opid
+};
+
+struct automaton_opid {
+ char *state_names[state_max_opid];
+ char *event_names[event_max_opid];
+ unsigned char function[state_max_opid][event_max_opid];
+ unsigned char initial_state;
+ bool final_states[state_max_opid];
+};
+
+static const struct automaton_opid automaton_opid = {
+ .state_names = {
+ "disabled",
+ "enabled",
+ "in_irq",
+ "irq_disabled",
+ "preempt_disabled"
+ },
+ .event_names = {
+ "irq_disable",
+ "irq_enable",
+ "irq_entry",
+ "preempt_disable",
+ "preempt_enable",
+ "sched_need_resched",
+ "sched_waking"
+ },
+ .function = {
+ {
+ INVALID_STATE,
+ preempt_disabled_opid,
+ disabled_opid,
+ INVALID_STATE,
+ irq_disabled_opid,
+ disabled_opid,
+ disabled_opid
+ },
+ {
+ irq_disabled_opid,
+ INVALID_STATE,
+ INVALID_STATE,
+ preempt_disabled_opid,
+ enabled_opid,
+ INVALID_STATE,
+ INVALID_STATE
+ },
+ {
+ INVALID_STATE,
+ enabled_opid,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ in_irq_opid,
+ in_irq_opid
+ },
+ {
+ INVALID_STATE,
+ enabled_opid,
+ in_irq_opid,
+ disabled_opid,
+ INVALID_STATE,
+ irq_disabled_opid,
+ INVALID_STATE
+ },
+ {
+ disabled_opid,
+ INVALID_STATE,
+ INVALID_STATE,
+ INVALID_STATE,
+ enabled_opid,
+ INVALID_STATE,
+ INVALID_STATE
+ },
+ },
+ .initial_state = disabled_opid,
+ .final_states = { 0, 1, 0, 0, 0 },
+};
diff --git a/kernel/trace/rv/monitors/opid/opid_trace.h b/kernel/trace/rv/monitors/opid/opid_trace.h
new file mode 100644
index 0000000000000..3df6ff955c300
--- /dev/null
+++ b/kernel/trace/rv/monitors/opid/opid_trace.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+/*
+ * Snippet to be included in rv_trace.h
+ */
+
+#ifdef CONFIG_RV_MON_OPID
+DEFINE_EVENT(event_da_monitor, event_opid,
+ TP_PROTO(char *state, char *event, char *next_state, bool final_state),
+ TP_ARGS(state, event, next_state, final_state));
+
+DEFINE_EVENT(error_da_monitor, error_opid,
+ TP_PROTO(char *state, char *event),
+ TP_ARGS(state, event));
+#endif /* CONFIG_RV_MON_OPID */
diff --git a/kernel/trace/rv/rv_trace.h b/kernel/trace/rv/rv_trace.h
index b5cb1d1807fd2..c2e841a080df6 100644
--- a/kernel/trace/rv/rv_trace.h
+++ b/kernel/trace/rv/rv_trace.h
@@ -63,6 +63,7 @@ DECLARE_EVENT_CLASS(error_da_monitor,
#include <monitors/snep/snep_trace.h>
#include <monitors/sncid/sncid_trace.h>
#include <monitors/sts/sts_trace.h>
+#include <monitors/opid/opid_trace.h>
// Add new monitors based on CONFIG_DA_MON_EVENTS_IMPLICIT here
#endif /* CONFIG_DA_MON_EVENTS_IMPLICIT */
diff --git a/tools/verification/models/sched/opid.dot b/tools/verification/models/sched/opid.dot
new file mode 100644
index 0000000000000..2d5e1df3405f2
--- /dev/null
+++ b/tools/verification/models/sched/opid.dot
@@ -0,0 +1,35 @@
+digraph state_automaton {
+ center = true;
+ size = "7,11";
+ {node [shape = plaintext, style=invis, label=""] "__init_disabled"};
+ {node [shape = circle] "disabled"};
+ {node [shape = doublecircle] "enabled"};
+ {node [shape = circle] "enabled"};
+ {node [shape = circle] "in_irq"};
+ {node [shape = circle] "irq_disabled"};
+ {node [shape = circle] "preempt_disabled"};
+ "__init_disabled" -> "disabled";
+ "disabled" [label = "disabled"];
+ "disabled" -> "disabled" [ label = "sched_need_resched\nsched_waking\nirq_entry" ];
+ "disabled" -> "irq_disabled" [ label = "preempt_enable" ];
+ "disabled" -> "preempt_disabled" [ label = "irq_enable" ];
+ "enabled" [label = "enabled", color = green3];
+ "enabled" -> "enabled" [ label = "preempt_enable" ];
+ "enabled" -> "irq_disabled" [ label = "irq_disable" ];
+ "enabled" -> "preempt_disabled" [ label = "preempt_disable" ];
+ "in_irq" [label = "in_irq"];
+ "in_irq" -> "enabled" [ label = "irq_enable" ];
+ "in_irq" -> "in_irq" [ label = "sched_need_resched\nsched_waking" ];
+ "irq_disabled" [label = "irq_disabled"];
+ "irq_disabled" -> "disabled" [ label = "preempt_disable" ];
+ "irq_disabled" -> "enabled" [ label = "irq_enable" ];
+ "irq_disabled" -> "in_irq" [ label = "irq_entry" ];
+ "irq_disabled" -> "irq_disabled" [ label = "sched_need_resched" ];
+ "preempt_disabled" [label = "preempt_disabled"];
+ "preempt_disabled" -> "disabled" [ label = "irq_disable" ];
+ "preempt_disabled" -> "enabled" [ label = "preempt_enable" ];
+ { rank = min ;
+ "__init_disabled";
+ "disabled";
+ }
+}
--
2.50.1
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: [PATCH v3 17/17] rv: Add opid per-cpu monitor
2025-07-15 7:14 ` [PATCH v3 17/17] rv: Add opid per-cpu monitor Gabriele Monaco
@ 2025-07-16 9:38 ` Nam Cao
2025-07-16 10:00 ` Gabriele Monaco
2025-07-18 10:26 ` Gabriele Monaco
0 siblings, 2 replies; 8+ messages in thread
From: Nam Cao @ 2025-07-16 9:38 UTC (permalink / raw)
To: Gabriele Monaco
Cc: linux-kernel, Steven Rostedt, Jonathan Corbet, Masami Hiramatsu,
linux-trace-kernel, linux-doc, Ingo Molnar, Peter Zijlstra,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
On Tue, Jul 15, 2025 at 09:14:34AM +0200, Gabriele Monaco wrote:
> diff --git a/kernel/trace/rv/monitors/nrp/Kconfig b/kernel/trace/rv/monitors/nrp/Kconfig
> index f37ff70e8d204..a175c430d351f 100644
> --- a/kernel/trace/rv/monitors/nrp/Kconfig
> +++ b/kernel/trace/rv/monitors/nrp/Kconfig
> @@ -3,7 +3,7 @@
> config RV_MON_NRP
> depends on RV
> depends on RV_MON_SCHED
> - default y if !ARCH_ARM64
> + default y if !ARM64
I think this is not supposed to be in this patch? It has nothing to do with
the opid monitor.
> select DA_MON_EVENTS_ID
> bool "nrp monitor"
> help
> diff --git a/kernel/trace/rv/monitors/opid/Kconfig b/kernel/trace/rv/monitors/opid/Kconfig
> new file mode 100644
> index 0000000000000..23b43d2704153
> --- /dev/null
> +++ b/kernel/trace/rv/monitors/opid/Kconfig
> @@ -0,0 +1,17 @@
> +# SPDX-License-Identifier: GPL-2.0-only
> +#
> +config RV_MON_OPID
> + depends on RV
> + depends on TRACE_IRQFLAGS
> + depends on TRACE_PREEMPT_TOGGLE
> + depends on RV_MON_SCHED
> + default y if PREEMPT_RT
> + select DA_MON_EVENTS_IMPLICIT
Shouldn't we add "depends on PREEMPT_RT"? I tried this monitor on
non-RT x86 kernel, and got some errors. That could confuse people.
And the monitor reports some errors on riscv64 with PREEMPT_RT=y:
root@riscv:~/rv-tests# uname -a
Linux riscv 6.16.0-rc6-00054-g7590637d9ca2 #87 SMP PREEMPT_RT Wed Jul 16 11:26:00 CEST 2025 riscv64 GNU/Linux
root@riscv:~/rv-tests# stress-ng --cpu-sched -1
stress-ng: info: [452] defaulting to a 1 day run per stressor
stress-ng: info: [452] dispatching hogs: 4 cpu-sched
[ 614.390462] rv: monitor opid does not allow event irq_entry on state in_irq
Best regards,
Nam
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v3 17/17] rv: Add opid per-cpu monitor
2025-07-16 9:38 ` Nam Cao
@ 2025-07-16 10:00 ` Gabriele Monaco
2025-07-18 10:26 ` Gabriele Monaco
1 sibling, 0 replies; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-16 10:00 UTC (permalink / raw)
To: Nam Cao
Cc: linux-kernel, Steven Rostedt, linux-trace-kernel, linux-doc,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
On Wed, 2025-07-16 at 11:38 +0200, Nam Cao wrote:
> On Tue, Jul 15, 2025 at 09:14:34AM +0200, Gabriele Monaco wrote:
> > diff --git a/kernel/trace/rv/monitors/nrp/Kconfig
> > b/kernel/trace/rv/monitors/nrp/Kconfig
> > index f37ff70e8d204..a175c430d351f 100644
> > --- a/kernel/trace/rv/monitors/nrp/Kconfig
> > +++ b/kernel/trace/rv/monitors/nrp/Kconfig
> > @@ -3,7 +3,7 @@
> > config RV_MON_NRP
> > depends on RV
> > depends on RV_MON_SCHED
> > - default y if !ARCH_ARM64
> > + default y if !ARM64
>
> I think this is not supposed to be in this patch? It has nothing to
> do with
> the opid monitor.
Damn, fixed up the wrong patch, will move it to the other one..
>
> > select DA_MON_EVENTS_ID
> > bool "nrp monitor"
> > help
> > diff --git a/kernel/trace/rv/monitors/opid/Kconfig
> > b/kernel/trace/rv/monitors/opid/Kconfig
> > new file mode 100644
> > index 0000000000000..23b43d2704153
> > --- /dev/null
> > +++ b/kernel/trace/rv/monitors/opid/Kconfig
> > @@ -0,0 +1,17 @@
> > +# SPDX-License-Identifier: GPL-2.0-only
> > +#
> > +config RV_MON_OPID
> > + depends on RV
> > + depends on TRACE_IRQFLAGS
> > + depends on TRACE_PREEMPT_TOGGLE
> > + depends on RV_MON_SCHED
> > + default y if PREEMPT_RT
> > + select DA_MON_EVENTS_IMPLICIT
>
> Shouldn't we add "depends on PREEMPT_RT"? I tried this monitor on
> non-RT x86 kernel, and got some errors. That could confuse people.
Mmh, my rationale was that it reports errors on non PREEMPT_RT, but it
does build. If someone wants to try it out there, they are free to do
it. We are just not supporting it officially.
The monitor might start working in the future also on non RT kernels,
or at least if someone wants to try whether it's the case, they can do
it easily.
Same idea for the ARM64 thing above.
But I should definitely mention this explicitly in the Kconfig entry
not to confuse people..
>
> And the monitor reports some errors on riscv64 with PREEMPT_RT=y:
>
> root@riscv:~/rv-tests# uname -a
> Linux riscv 6.16.0-rc6-00054-g7590637d9ca2 #87 SMP PREEMPT_RT Wed Jul
> 16 11:26:00 CEST 2025 riscv64 GNU/Linux
> root@riscv:~/rv-tests# stress-ng --cpu-sched -1
> stress-ng: info: [452] defaulting to a 1 day run per stressor
> stress-ng: info: [452] dispatching hogs: 4 cpu-sched
> [ 614.390462] rv: monitor opid does not allow event irq_entry on
> state in_irq
>
Mmh riscv.. I haven't tested it there, guess I need to start keeping a
VM somewhere.
Thanks,
Gabriele
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v3 17/17] rv: Add opid per-cpu monitor
2025-07-16 9:38 ` Nam Cao
2025-07-16 10:00 ` Gabriele Monaco
@ 2025-07-18 10:26 ` Gabriele Monaco
1 sibling, 0 replies; 8+ messages in thread
From: Gabriele Monaco @ 2025-07-18 10:26 UTC (permalink / raw)
To: Nam Cao
Cc: linux-kernel, Steven Rostedt, linux-trace-kernel, linux-doc,
Tomas Glozar, Juri Lelli, Clark Williams, John Kacur
On Wed, 2025-07-16 at 11:38 +0200, Nam Cao wrote:
> And the monitor reports some errors on riscv64 with PREEMPT_RT=y:
>
> root@riscv:~/rv-tests# uname -a
> Linux riscv 6.16.0-rc6-00054-g7590637d9ca2 #87 SMP PREEMPT_RT Wed Jul
> 16 11:26:00 CEST 2025 riscv64 GNU/Linux
> root@riscv:~/rv-tests# stress-ng --cpu-sched -1
> stress-ng: info: [452] defaulting to a 1 day run per stressor
> stress-ng: info: [452] dispatching hogs: 4 cpu-sched
> [ 614.390462] rv: monitor opid does not allow event irq_entry on
> state in_irq
Finally managed to bootstrap a riscv VM, that is an error I thought I'd
fixed and I'm genuinely surprised I didn't notice on other
architectures.
Thanks again for catching it though!
Gabriele
^ permalink raw reply [flat|nested] 8+ messages in thread