From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
"Suggested-by: Clark Williams" <williams@redhat.com>,
Juri Lelli <juri.lelli@redhat.com>,
Bagas Sanjaya <bagasdotme@gmail.com>,
Daniel Bristot de Oliveira <bristot@kernel.org>,
Jonathan Corbet <corbet@lwn.net>
Subject: [for-next][PATCH 24/25] tracing/osnoise: Add preempt and/or irq disabled options
Date: Sat, 10 Dec 2022 08:58:14 -0500 [thread overview]
Message-ID: <20221210135826.943530120@goodmis.org> (raw)
In-Reply-To: 20221210135750.425719934@goodmis.org
From: Daniel Bristot de Oliveira <bristot@kernel.org>
The osnoise workload runs with preemption and IRQs enabled in such
a way as to allow all sorts of noise to disturb osnoise's execution.
hwlat tracer has a similar workload but works with irq disabled,
allowing only NMIs and the hardware to generate noise.
While thinking about adding an options file to hwlat tracer to
allow the system to panic, and other features I was thinking
to add, like having a tracepoint at each noise detection, it
came to my mind that is easier to make osnoise and also do
hardware latency detection than making hwlat "feature compatible"
with osnoise.
Other points are:
- osnoise already has an independent cpu file.
- osnoise has a more intuitive interface, e.g., runtime/period vs.
window/width (and people often need help remembering what it is).
- osnoise: tracepoints
- osnoise stop options
- osnoise options file itself
Moreover, the user-space side (in rtla) is simplified by reusing the
existing osnoise code.
Finally, people have been asking me about using osnoise for hw latency
detection, and I have to explain that it was sufficient but not
necessary. These options make it sufficient and necessary.
Adding a Suggested-by Clark, as he often asked me about this
possibility.
Link: https://lkml.kernel.org/r/d9c6c19135497054986900f94c8e47410b15316a.1670623111.git.bristot@kernel.org
Cc: Suggested-by: Clark Williams <williams@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: Daniel Bristot de Oliveira <bristot@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Daniel Bristot de Oliveira <bristot@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
---
kernel/trace/trace_osnoise.c | 48 ++++++++++++++++++++++++++++++++----
1 file changed, 43 insertions(+), 5 deletions(-)
diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c
index 5a7613942223..94c1b5eb1dc0 100644
--- a/kernel/trace/trace_osnoise.c
+++ b/kernel/trace/trace_osnoise.c
@@ -55,10 +55,17 @@ enum osnoise_options_index {
OSN_DEFAULTS = 0,
OSN_WORKLOAD,
OSN_PANIC_ON_STOP,
+ OSN_PREEMPT_DISABLE,
+ OSN_IRQ_DISABLE,
OSN_MAX
};
-static const char * const osnoise_options_str[OSN_MAX] = { "DEFAULTS", "OSNOISE_WORKLOAD", "PANIC_ON_STOP" };
+static const char * const osnoise_options_str[OSN_MAX] = {
+ "DEFAULTS",
+ "OSNOISE_WORKLOAD",
+ "PANIC_ON_STOP",
+ "OSNOISE_PREEMPT_DISABLE",
+ "OSNOISE_IRQ_DISABLE" };
#define OSN_DEFAULT_OPTIONS 0x2
static unsigned long osnoise_options = OSN_DEFAULT_OPTIONS;
@@ -1308,18 +1315,26 @@ static void notify_new_max_latency(u64 latency)
*/
static int run_osnoise(void)
{
+ bool disable_irq = test_bit(OSN_IRQ_DISABLE, &osnoise_options);
struct osnoise_variables *osn_var = this_cpu_osn_var();
u64 start, sample, last_sample;
u64 last_int_count, int_count;
s64 noise = 0, max_noise = 0;
s64 total, last_total = 0;
struct osnoise_sample s;
+ bool disable_preemption;
unsigned int threshold;
u64 runtime, stop_in;
u64 sum_noise = 0;
int hw_count = 0;
int ret = -1;
+ /*
+ * Disabling preemption is only required if IRQs are enabled,
+ * and the options is set on.
+ */
+ disable_preemption = !disable_irq && test_bit(OSN_PREEMPT_DISABLE, &osnoise_options);
+
/*
* Considers the current thread as the workload.
*/
@@ -1335,6 +1350,15 @@ static int run_osnoise(void)
*/
threshold = tracing_thresh ? : 5000;
+ /*
+ * Apply PREEMPT and IRQ disabled options.
+ */
+ if (disable_irq)
+ local_irq_disable();
+
+ if (disable_preemption)
+ preempt_disable();
+
/*
* Make sure NMIs see sampling first
*/
@@ -1422,16 +1446,21 @@ static int run_osnoise(void)
* cond_resched()
*/
if (IS_ENABLED(CONFIG_PREEMPT_RCU)) {
- local_irq_disable();
+ if (!disable_irq)
+ local_irq_disable();
+
rcu_momentary_dyntick_idle();
- local_irq_enable();
+
+ if (!disable_irq)
+ local_irq_enable();
}
/*
* For the non-preemptive kernel config: let threads runs, if
- * they so wish.
+ * they so wish, unless set not do to so.
*/
- cond_resched();
+ if (!disable_irq && !disable_preemption)
+ cond_resched();
last_sample = sample;
last_int_count = int_count;
@@ -1450,6 +1479,15 @@ static int run_osnoise(void)
*/
barrier();
+ /*
+ * Return to the preemptive state.
+ */
+ if (disable_preemption)
+ preempt_enable();
+
+ if (disable_irq)
+ local_irq_enable();
+
/*
* Save noise info.
*/
--
2.35.1
next prev parent reply other threads:[~2022-12-10 13:59 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-10 13:57 [for-next][PATCH 00/25] tracing: Updates for 6.2 Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 01/25] tracing/user_events: Fix call print_fmt leak Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 02/25] tracing: Update MAINTAINERS file for new patchwork and mailing list Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 03/25] ftrace/x86: Add back ftrace_expected for ftrace bug reports Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 04/25] tracing: Allow multiple hitcount values in histograms Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 05/25] tracing: Add .percent suffix option to histogram values Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 06/25] tracing: Add .graph suffix option to histogram value Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 07/25] tracing: Add nohitcount option for suppressing display of raw hitcount Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 08/25] tracing: docs: Update histogram doc for .percent/.graph and nohitcount Steven Rostedt
2022-12-10 13:57 ` [for-next][PATCH 09/25] trace/kprobe: remove duplicated calls of ring_buffer_event_data Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 10/25] tracing/probes: Handle system names with hyphens Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 11/25] tracing: Fix complicated dependency of CONFIG_TRACER_MAX_TRACE Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 12/25] x86/mm/kmmio: Switch to arch_spin_lock() Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 13/25] x86/mm/kmmio: Use rcu_read_lock_sched_notrace() Steven Rostedt
2022-12-10 17:47 ` Paul E. McKenney
2022-12-10 18:34 ` Steven Rostedt
2022-12-10 21:34 ` Paul E. McKenney
2022-12-10 22:32 ` Steven Rostedt
2022-12-11 5:52 ` Paul E. McKenney
2022-12-10 23:30 ` Thomas Gleixner
2022-12-10 23:55 ` Steven Rostedt
2022-12-12 10:51 ` Thomas Gleixner
2022-12-12 15:42 ` Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 14/25] tracing/hist: Fix wrong return value in parse_action_params() Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 15/25] tracing/hist: Fix out-of-bound write on action_data.var_ref_idx Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 16/25] tracing: Fix issue of missing one synthetic field Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 17/25] tracing/hist: Fix issue of losting command info in error_log Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 18/25] ring-buffer: Handle resize in early boot up Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 19/25] tracing: remove unnecessary trace_trigger ifdef Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 20/25] tracing/osnoise: Make osnoise_options static Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 21/25] tracing: Fix some checker warnings Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 22/25] Documentation/osnoise: Escape underscore of NO_ prefix Steven Rostedt
2022-12-10 13:58 ` [for-next][PATCH 23/25] tracing/osnoise: Add PANIC_ON_STOP option Steven Rostedt
2022-12-10 13:58 ` Steven Rostedt [this message]
2022-12-10 13:58 ` [for-next][PATCH 25/25] Documentation/osnoise: Add osnoise/options documentation Steven Rostedt
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20221210135826.943530120@goodmis.org \
--to=rostedt@goodmis.org \
--cc=akpm@linux-foundation.org \
--cc=bagasdotme@gmail.com \
--cc=bristot@kernel.org \
--cc=corbet@lwn.net \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=williams@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox