The patch titled Subject: procfs: speed up /proc/pid/stat, statm has been added to the -mm tree. Its filename is procfs-speed-up-proc-pid-stat-statm.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** The -mm tree is included into linux-next and is updated there every 3-4 working days ------------------------------------------------------ From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> Subject: procfs: speed up /proc/pid/stat, statm Process accounting applications as top, ps visit some files under /proc/<pid>. With seq_put_decimal_ull(), we can optimize /proc/<pid>/stat and /proc/<pid>/statm files. This patch adds - seq_put_decimal_ll() for signed values. - allow delimiter == 0. - convert seq_printf() to seq_put_decimal_ull/ll in /proc/stat, statm. Test result on a system with 2000+ procs. Before patch: [kamezawa@bluextal test]$ top -b -n 1 | wc -l 2223 [kamezawa@bluextal test]$ time top -b -n 1 > /dev/null real 0m0.675s user 0m0.044s sys 0m0.121s [kamezawa@bluextal test]$ time ps -elf > /dev/null real 0m0.236s user 0m0.056s sys 0m0.176s After patch: kamezawa@bluextal ~]$ time top -b -n 1 > /dev/null real 0m0.657s user 0m0.052s sys 0m0.100s [kamezawa@bluextal ~]$ time ps -elf > /dev/null real 0m0.198s user 0m0.050s sys 0m0.145s Considering top, ps tend to scan /proc periodically, this will reduce cpu consumption by top/ps to some extent. Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> Cc: Alexey Dobriyan <adobriyan@xxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- fs/proc/array.c | 119 +++++++++++++++++++------------------ fs/seq_file.c | 21 ++++++ include/linux/seq_file.h | 2 3 files changed, 86 insertions(+), 56 deletions(-) diff -puN fs/proc/array.c~procfs-speed-up-proc-pid-stat-statm fs/proc/array.c --- a/fs/proc/array.c~procfs-speed-up-proc-pid-stat-statm +++ a/fs/proc/array.c @@ -462,59 +462,56 @@ static int do_task_stat(struct seq_file /* convert nsec -> ticks */ start_time = nsec_to_clock_t(start_time); - seq_printf(m, "%d (%s) %c %d %d %d %d %d %u %lu \ -%lu %lu %lu %lu %lu %ld %ld %ld %ld %d 0 %llu %lu %ld %lu %lu %lu %lu %lu \ -%lu %lu %lu %lu %lu %lu %lu %lu %d %d %u %u %llu %lu %ld %lu %lu %lu\n", - pid_nr_ns(pid, ns), - tcomm, - state, - ppid, - pgid, - sid, - tty_nr, - tty_pgrp, - task->flags, - min_flt, - cmin_flt, - maj_flt, - cmaj_flt, - cputime_to_clock_t(utime), - cputime_to_clock_t(stime), - cputime_to_clock_t(cutime), - cputime_to_clock_t(cstime), - priority, - nice, - num_threads, - start_time, - vsize, - mm ? get_mm_rss(mm) : 0, - rsslim, - mm ? (permitted ? mm->start_code : 1) : 0, - mm ? (permitted ? mm->end_code : 1) : 0, - (permitted && mm) ? mm->start_stack : 0, - esp, - eip, - /* The signal information here is obsolete. - * It must be decimal for Linux 2.0 compatibility. - * Use /proc/#/status for real-time signals. - */ - task->pending.signal.sig[0] & 0x7fffffffUL, - task->blocked.sig[0] & 0x7fffffffUL, - sigign .sig[0] & 0x7fffffffUL, - sigcatch .sig[0] & 0x7fffffffUL, - wchan, - 0UL, - 0UL, - task->exit_signal, - task_cpu(task), - task->rt_priority, - task->policy, - (unsigned long long)delayacct_blkio_ticks(task), - cputime_to_clock_t(gtime), - cputime_to_clock_t(cgtime), - (mm && permitted) ? mm->start_data : 0, - (mm && permitted) ? mm->end_data : 0, - (mm && permitted) ? mm->start_brk : 0); + seq_printf(m,"%d (%s) %c", pid_nr_ns(pid, ns), tcomm, state); + seq_put_decimal_ll(m, ' ', ppid); + seq_put_decimal_ll(m, ' ', pgid); + seq_put_decimal_ll(m, ' ', sid); + seq_put_decimal_ll(m, ' ', tty_nr); + seq_put_decimal_ll(m, ' ', tty_pgrp); + seq_put_decimal_ull(m, ' ', task->flags); + seq_put_decimal_ull(m, ' ', min_flt); + seq_put_decimal_ull(m, ' ', cmin_flt); + seq_put_decimal_ull(m, ' ', maj_flt); + seq_put_decimal_ull(m, ' ', cmaj_flt); + seq_put_decimal_ull(m, ' ', cputime_to_clock_t(utime)); + seq_put_decimal_ull(m, ' ', cputime_to_clock_t(stime)); + seq_put_decimal_ll(m, ' ', cputime_to_clock_t(cutime)); + seq_put_decimal_ll(m, ' ', cputime_to_clock_t(cstime)); + seq_put_decimal_ll(m, ' ', priority); + seq_put_decimal_ll(m, ' ', nice); + seq_put_decimal_ll(m, ' ', num_threads); + seq_put_decimal_ull(m, ' ', 0); + seq_put_decimal_ull(m, ' ', start_time); + seq_put_decimal_ull(m, ' ', vsize); + seq_put_decimal_ll(m, ' ', mm ? get_mm_rss(mm) : 0); + seq_put_decimal_ull(m, ' ', rsslim); + seq_put_decimal_ull(m, ' ', mm ? (permitted ? mm->start_code : 1) : 0); + seq_put_decimal_ull(m, ' ', mm ? (permitted ? mm->end_code : 1) : 0); + seq_put_decimal_ull(m, ' ', (permitted && mm) ? mm->start_stack : 0); + seq_put_decimal_ull(m, ' ', esp); + seq_put_decimal_ull(m, ' ', eip); + /* The signal information here is obsolete. + * It must be decimal for Linux 2.0 compatibility. + * Use /proc/#/status for real-time signals. + */ + seq_put_decimal_ull(m, ' ', task->pending.signal.sig[0] & 0x7fffffffUL); + seq_put_decimal_ull(m, ' ', task->blocked.sig[0] & 0x7fffffffUL); + seq_put_decimal_ull(m, ' ', sigign.sig[0] & 0x7fffffffUL); + seq_put_decimal_ull(m, ' ', sigcatch.sig[0] & 0x7fffffffUL); + seq_put_decimal_ull(m, ' ', wchan); + seq_put_decimal_ull(m, ' ', 0); + seq_put_decimal_ull(m, ' ', 0); + seq_put_decimal_ll(m, ' ', task->exit_signal); + seq_put_decimal_ll(m, ' ', task_cpu(task)); + seq_put_decimal_ull(m, ' ', task->rt_priority); + seq_put_decimal_ull(m, ' ', task->policy); + seq_put_decimal_ull(m, ' ', delayacct_blkio_ticks(task)); + seq_put_decimal_ull(m, ' ', cputime_to_clock_t(gtime)); + seq_put_decimal_ll(m, ' ', cputime_to_clock_t(cgtime)); + seq_put_decimal_ull(m, ' ', (mm && permitted) ? mm->start_data : 0); + seq_put_decimal_ull(m, ' ', (mm && permitted) ? mm->end_data : 0); + seq_put_decimal_ull(m, ' ', (mm && permitted) ? mm->start_brk : 0); + seq_putc(m, '\n'); if (mm) mmput(mm); return 0; @@ -542,8 +539,20 @@ int proc_pid_statm(struct seq_file *m, s size = task_statm(mm, &shared, &text, &data, &resident); mmput(mm); } - seq_printf(m, "%lu %lu %lu %lu 0 %lu 0\n", - size, resident, shared, text, data); + /* + * For quick read, open code by putting numbers directly + * expected format is + * seq_printf(m, "%lu %lu %lu %lu 0 %lu 0\n", + * size, resident, shared, text, data); + */ + seq_put_decimal_ull(m, 0, size); + seq_put_decimal_ull(m, ' ', resident); + seq_put_decimal_ull(m, ' ', shared); + seq_put_decimal_ull(m, ' ', text); + seq_put_decimal_ull(m, ' ', 0); + seq_put_decimal_ull(m, ' ', text); + seq_put_decimal_ull(m, ' ', 0); + seq_putc(m, '\n'); return 0; } diff -puN fs/seq_file.c~procfs-speed-up-proc-pid-stat-statm fs/seq_file.c --- a/fs/seq_file.c~procfs-speed-up-proc-pid-stat-statm +++ a/fs/seq_file.c @@ -659,7 +659,8 @@ int seq_put_decimal_ull(struct seq_file if (m->count + 2 >= m->size) /* we'll write 2 bytes at least */ goto overflow; - m->buf[m->count++] = delimiter; + if (delimiter) + m->buf[m->count++] = delimiter; if (num < 10) { m->buf[m->count++] = num + '0'; @@ -677,6 +678,24 @@ overflow: } EXPORT_SYMBOL(seq_put_decimal_ull); +int seq_put_decimal_ll(struct seq_file *m, char delimiter, + long long num) +{ + if (num < 0) { + if (m->count + 3 >= m->size) { + m->count = m->size; + return -1; + } + if (delimiter) + m->buf[m->count++] = delimiter; + num = -num; + delimiter = '-'; + } + return seq_put_decimal_ull(m, delimiter, num); + +} +EXPORT_SYMBOL(seq_put_decimal_ll); + /** * seq_write - write arbitrary data to buffer * @seq: seq_file identifying the buffer to which data should be written diff -puN include/linux/seq_file.h~procfs-speed-up-proc-pid-stat-statm include/linux/seq_file.h --- a/include/linux/seq_file.h~procfs-speed-up-proc-pid-stat-statm +++ a/include/linux/seq_file.h @@ -123,6 +123,8 @@ int seq_open_private(struct file *, cons int seq_release_private(struct inode *, struct file *); int seq_put_decimal_ull(struct seq_file *m, char delimiter, unsigned long long num); +int seq_put_decimal_ll(struct seq_file *m, char delimiter, + long long num); #define SEQ_START_TOKEN ((void *)1) /* _ Subject: Subject: procfs: speed up /proc/pid/stat, statm Patches currently in -mm which might be from kamezawa.hiroyu@xxxxxxxxxxxxxx are origin.patch mm-postpone-migrated-page-mapping-reset.patch linux-next.patch mm-memcontrolc-fix-warning-with-config_numa=n.patch mm-oom-avoid-looping-when-chosen-thread-detaches-its-mm.patch mm-oom-fold-oom_kill_task-into-oom_kill_process.patch mm-oom-do-not-emit-oom-killer-warning-if-chosen-thread-is-already-exiting.patch mm-add-rss-counters-consistency-check.patch mm-vmscanc-cleanup-with-s-reclaim_mode-isolate_mode.patch mm-make-get_mm_counter-static-inline.patch mm-vmscan-fix-misused-nr_reclaimed-in-shrink_mem_cgroup_zone.patch memcg-replace-mem_cont-by-mem_res_ctlr.patch memcg-replace-mem-and-mem_cont-stragglers.patch memcg-lru_size-instead-of-mem_cgroup_zstat.patch memcg-enum-lru_list-lru.patch memcg-remove-redundant-returns.patch memcg-remove-unnecessary-thp-check-in-page-stat-accounting.patch idr-make-idr_get_next-good-for-rcu_read_lock.patch cgroup-revert-ss_id_lock-to-spinlock.patch memcg-let-css_get_next-rely-upon-rcu_read_lock.patch memcg-remove-pcg_cache-page_cgroup-flag.patch memcg-remove-pcg_cache-page_cgroup-flag-checkpatch-fixes.patch proc-speedup-proc-stat-handling.patch procfs-add-num_to_str-to-speed-up-proc-stat.patch procfs-add-num_to_str-to-speed-up-proc-stat-fix.patch procfs-speed-up-proc-pid-stat-statm.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html