| Message ID | 1593614103-23574-1-git-send-email-pizhenwei@bytedance.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | fs/proc: add short desc for /proc/softirqs | expand |
On Wed, Jul 01, 2020 at 10:35:03PM +0800, zhenwei pi wrote: > Only softirq name is not friendly to end-users, typically 'HI' is > difficult to understand. During developing irqtop/lsirq utilities > for util-linux, Karel Zak considered that we should give more > information to end-users. Discuss about this: > https://github.com/karelzak/util-linux/pull/1079 > > Add short desc for /proc/softirqs in this patch, then /proc/softirqs > gets more human-readable. > --- a/fs/proc/softirqs.c > +++ b/fs/proc/softirqs.c > @@ -20,7 +20,7 @@ static int show_softirqs(struct seq_file *p, void *v) > seq_printf(p, "%12s:", softirq_to_name[i]); > for_each_possible_cpu(j) > seq_printf(p, " %10u", kstat_softirqs_cpu(i, j)); > - seq_putc(p, '\n'); > + seq_printf(p, " %s\n", softirq_to_desc[i]); This could break parsers. I'd rather leave it as is and update proc(5). Of course this file doesn't need more characters in it. :-\
diff --git a/fs/proc/softirqs.c b/fs/proc/softirqs.c index 12901dc..fcd21f3 100644 --- a/fs/proc/softirqs.c +++ b/fs/proc/softirqs.c @@ -20,7 +20,7 @@ static int show_softirqs(struct seq_file *p, void *v) seq_printf(p, "%12s:", softirq_to_name[i]); for_each_possible_cpu(j) seq_printf(p, " %10u", kstat_softirqs_cpu(i, j)); - seq_putc(p, '\n'); + seq_printf(p, " %s\n", softirq_to_desc[i]); } return 0; } diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 5db970b..1d51397 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -543,10 +543,11 @@ enum #define SOFTIRQ_STOP_IDLE_MASK (~(1 << RCU_SOFTIRQ)) -/* map softirq index to softirq name. update 'softirq_to_name' in - * kernel/softirq.c when adding a new softirq. +/* map softirq index to softirq name. update 'softirq_to_name' & + * 'softirq_to_desc' in kernel/softirq.c when adding a new softirq. */ extern const char * const softirq_to_name[NR_SOFTIRQS]; +extern const char * const softirq_to_desc[NR_SOFTIRQS]; /* softirq mask and active fields moved to irq_cpustat_t in * asm/hardirq.h to get better cache usage. KAO diff --git a/kernel/softirq.c b/kernel/softirq.c index c4201b7f..74eca3b 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -61,6 +61,18 @@ const char * const softirq_to_name[NR_SOFTIRQS] = { "TASKLET", "SCHED", "HRTIMER", "RCU" }; +const char * const softirq_to_desc[NR_SOFTIRQS] = { + "high priority tasklet softirq", + "timer softirq", + "network transmit softirq", + "network receive softirq", + "block device softirq", + "IO poll softirq", + "normal priority tasklet softirq", + "schedule softirq", + "high resolution timer softirq", + "RCU softirq" +}; /* * we cannot loop indefinitely here to avoid userspace starvation, * but we also don't want to introduce a worst case 1/HZ latency
Only softirq name is not friendly to end-users, typically 'HI' is difficult to understand. During developing irqtop/lsirq utilities for util-linux, Karel Zak considered that we should give more information to end-users. Discuss about this: https://github.com/karelzak/util-linux/pull/1079 Add short desc for /proc/softirqs in this patch, then /proc/softirqs gets more human-readable. Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- fs/proc/softirqs.c | 2 +- include/linux/interrupt.h | 5 +++-- kernel/softirq.c | 12 ++++++++++++ 3 files changed, 16 insertions(+), 3 deletions(-)