* [PATCH] watchdog: Don't need a definition of the perf fallback-attr
@ 2025-07-27 3:25 Qing Wong
2025-08-09 4:07 ` Qing Wang
0 siblings, 1 reply; 2+ messages in thread
From: Qing Wong @ 2025-07-27 3:25 UTC (permalink / raw)
To: peterz, linux-kernel; +Cc: mingo, akpm, song, Qing Wang
From: Qing Wang <wangqing7171@gmail.com>
The 'hardlockup_config_perf_event()' only overwrites the type and
config of the perf attr. It's wasteful that redefining a big static
variable to fallback. It’s just a matter of simply fallback these
two variables.
Signed-off-by: Qing Wang <wangqing7171@gmail.com>
---
kernel/watchdog_perf.c | 17 +++++++----------
1 file changed, 7 insertions(+), 10 deletions(-)
diff --git a/kernel/watchdog_perf.c b/kernel/watchdog_perf.c
index 75af12ff774e..08b9acbd2aa4 100644
--- a/kernel/watchdog_perf.c
+++ b/kernel/watchdog_perf.c
@@ -92,14 +92,6 @@ static struct perf_event_attr wd_hw_attr = {
.disabled = 1,
};
-static struct perf_event_attr fallback_wd_hw_attr = {
- .type = PERF_TYPE_HARDWARE,
- .config = PERF_COUNT_HW_CPU_CYCLES,
- .size = sizeof(struct perf_event_attr),
- .pinned = 1,
- .disabled = 1,
-};
-
/* Callback function for perf event subsystem */
static void watchdog_overflow_callback(struct perf_event *event,
struct perf_sample_data *data,
@@ -114,6 +106,12 @@ static void watchdog_overflow_callback(struct perf_event *event,
watchdog_hardlockup_check(smp_processor_id(), regs);
}
+static void fallback_wd_hw_attr(void)
+{
+ wd_hw_attr.type = PERF_TYPE_HARDWARE;
+ wd_hw_attr.config = PERF_COUNT_HW_CPU_CYCLES;
+}
+
static int hardlockup_detector_event_create(void)
{
unsigned int cpu;
@@ -133,8 +131,7 @@ static int hardlockup_detector_event_create(void)
evt = perf_event_create_kernel_counter(wd_attr, cpu, NULL,
watchdog_overflow_callback, NULL);
if (IS_ERR(evt)) {
- wd_attr = &fallback_wd_hw_attr;
- wd_attr->sample_period = hw_nmi_get_sample_period(watchdog_thresh);
+ fallback_wd_hw_attr();
evt = perf_event_create_kernel_counter(wd_attr, cpu, NULL,
watchdog_overflow_callback, NULL);
}
--
2.43.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] watchdog: Don't need a definition of the perf fallback-attr
2025-07-27 3:25 [PATCH] watchdog: Don't need a definition of the perf fallback-attr Qing Wong
@ 2025-08-09 4:07 ` Qing Wang
0 siblings, 0 replies; 2+ messages in thread
From: Qing Wang @ 2025-08-09 4:07 UTC (permalink / raw)
To: peterz, linux-kernel; +Cc: mingo, akpm, song
Hello,
PING.
Thanks,
Qing.
On 7/27/2025 11:25 AM, Qing Wong wrote:
> From: Qing Wang <wangqing7171@gmail.com>
>
> The 'hardlockup_config_perf_event()' only overwrites the type and
> config of the perf attr. It's wasteful that redefining a big static
> variable to fallback. It’s just a matter of simply fallback these
> two variables.
>
> Signed-off-by: Qing Wang <wangqing7171@gmail.com>
> ---
> kernel/watchdog_perf.c | 17 +++++++----------
> 1 file changed, 7 insertions(+), 10 deletions(-)
>
> diff --git a/kernel/watchdog_perf.c b/kernel/watchdog_perf.c
> index 75af12ff774e..08b9acbd2aa4 100644
> --- a/kernel/watchdog_perf.c
> +++ b/kernel/watchdog_perf.c
> @@ -92,14 +92,6 @@ static struct perf_event_attr wd_hw_attr = {
> .disabled = 1,
> };
>
> -static struct perf_event_attr fallback_wd_hw_attr = {
> - .type = PERF_TYPE_HARDWARE,
> - .config = PERF_COUNT_HW_CPU_CYCLES,
> - .size = sizeof(struct perf_event_attr),
> - .pinned = 1,
> - .disabled = 1,
> -};
> -
> /* Callback function for perf event subsystem */
> static void watchdog_overflow_callback(struct perf_event *event,
> struct perf_sample_data *data,
> @@ -114,6 +106,12 @@ static void watchdog_overflow_callback(struct perf_event *event,
> watchdog_hardlockup_check(smp_processor_id(), regs);
> }
>
> +static void fallback_wd_hw_attr(void)
> +{
> + wd_hw_attr.type = PERF_TYPE_HARDWARE;
> + wd_hw_attr.config = PERF_COUNT_HW_CPU_CYCLES;
> +}
> +
> static int hardlockup_detector_event_create(void)
> {
> unsigned int cpu;
> @@ -133,8 +131,7 @@ static int hardlockup_detector_event_create(void)
> evt = perf_event_create_kernel_counter(wd_attr, cpu, NULL,
> watchdog_overflow_callback, NULL);
> if (IS_ERR(evt)) {
> - wd_attr = &fallback_wd_hw_attr;
> - wd_attr->sample_period = hw_nmi_get_sample_period(watchdog_thresh);
> + fallback_wd_hw_attr();
> evt = perf_event_create_kernel_counter(wd_attr, cpu, NULL,
> watchdog_overflow_callback, NULL);
> }
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-08-09 4:07 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-07-27 3:25 [PATCH] watchdog: Don't need a definition of the perf fallback-attr Qing Wong
2025-08-09 4:07 ` Qing Wang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).