linux/fs/proc/array.c

520 lines
12 KiB
C
Raw Normal View History

/*
* linux/fs/proc/array.c
*
* Copyright (C) 1992 by Linus Torvalds
* based on ideas by Darren Senn
*
* Fixes:
* Michael. K. Johnson: stat,statm extensions.
* <johnsonm@stolaf.edu>
*
* Pauline Middelink : Made cmdline,envline only break at '\0's, to
* make sure SET_PROCTITLE works. Also removed
* bad '!' which forced address recalculation for
* EVERY character on the current page.
* <middelin@polyware.iaf.nl>
*
* Danny ter Haar : added cpuinfo
* <dth@cistron.nl>
*
* Alessandro Rubini : profile extension.
* <rubini@ipvvis.unipv.it>
*
* Jeff Tranter : added BogoMips field to cpuinfo
* <Jeff_Tranter@Mitel.COM>
*
* Bruno Haible : remove 4K limit for the maps file
* <haible@ma2s2.mathematik.uni-karlsruhe.de>
*
* Yves Arrouye : remove removal of trailing spaces in get_array.
* <Yves.Arrouye@marin.fdn.fr>
*
* Jerome Forissier : added per-CPU time information to /proc/stat
* and /proc/<pid>/cpu extension
* <forissier@isia.cma.fr>
* - Incorporation and non-SMP safe operation
* of forissier patch in 2.1.78 by
* Hans Marcus <crowbar@concepts.nl>
*
* aeb@cwi.nl : /proc/partitions
*
*
* Alan Cox : security fixes.
* <alan@lxorguk.ukuu.org.uk>
*
* Al Viro : safe handling of mm_struct
*
* Gerhard Wichert : added BIGMEM support
* Siemens AG <Gerhard.Wichert@pdb.siemens.de>
*
* Al Viro & Jeff Garzik : moved most of the thing into base.c and
* : proc_misc.c. The rest may eventually go into
* : base.c too.
*/
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/time.h>
#include <linux/kernel.h>
#include <linux/kernel_stat.h>
#include <linux/tty.h>
#include <linux/string.h>
#include <linux/mman.h>
#include <linux/proc_fs.h>
#include <linux/ioport.h>
#include <linux/uaccess.h>
#include <linux/io.h>
#include <linux/mm.h>
#include <linux/hugetlb.h>
#include <linux/pagemap.h>
#include <linux/swap.h>
#include <linux/slab.h>
#include <linux/smp.h>
#include <linux/signal.h>
#include <linux/highmem.h>
#include <linux/file.h>
#include <linux/fdtable.h>
#include <linux/times.h>
#include <linux/cpuset.h>
#include <linux/rcupdate.h>
#include <linux/delayacct.h>
#include <linux/seq_file.h>
#include <linux/pid_namespace.h>
#include <linux/tracehook.h>
#include <asm/pgtable.h>
#include <asm/processor.h>
#include "internal.h"
static inline void task_name(struct seq_file *m, struct task_struct *p)
{
int i;
char *buf, *end;
char *name;
char tcomm[sizeof(p->comm)];
get_task_comm(tcomm, p);
seq_printf(m, "Name:\t");
end = m->buf + m->size;
buf = m->buf + m->count;
name = tcomm;
i = sizeof(tcomm);
while (i && (buf < end)) {
unsigned char c = *name;
name++;
i--;
*buf = c;
if (!c)
break;
if (c == '\\') {
buf++;
if (buf < end)
*buf++ = c;
continue;
}
if (c == '\n') {
*buf++ = '\\';
if (buf < end)
*buf++ = 'n';
continue;
}
buf++;
}
m->count = buf - m->buf;
seq_printf(m, "\n");
}
/*
* The task state array is a strange "bitmap" of
* reasons to sleep. Thus "running" is zero, and
* you can test for combinations of others with
* simple bit tests.
*/
static const char *task_state_array[] = {
"R (running)", /* 0 */
"S (sleeping)", /* 1 */
"D (disk sleep)", /* 2 */
"T (stopped)", /* 4 */
"T (tracing stop)", /* 8 */
"Z (zombie)", /* 16 */
"X (dead)" /* 32 */
};
static inline const char *get_task_state(struct task_struct *tsk)
{
unsigned int state = (tsk->state & TASK_REPORT) | tsk->exit_state;
const char **p = &task_state_array[0];
while (state) {
p++;
state >>= 1;
}
return *p;
}
static inline void task_state(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *p)
{
struct group_info *group_info;
int g;
struct fdtable *fdt = NULL;
pid_t ppid, tpid;
rcu_read_lock();
ppid = pid_alive(p) ?
task_tgid_nr_ns(rcu_dereference(p->real_parent), ns) : 0;
tpid = 0;
if (pid_alive(p)) {
struct task_struct *tracer = tracehook_tracer_task(p);
if (tracer)
tpid = task_pid_nr_ns(tracer, ns);
}
seq_printf(m,
"State:\t%s\n"
"Tgid:\t%d\n"
"Pid:\t%d\n"
"PPid:\t%d\n"
"TracerPid:\t%d\n"
"Uid:\t%d\t%d\t%d\t%d\n"
"Gid:\t%d\t%d\t%d\t%d\n",
get_task_state(p),
task_tgid_nr_ns(p, ns),
pid_nr_ns(pid, ns),
ppid, tpid,
p->uid, p->euid, p->suid, p->fsuid,
p->gid, p->egid, p->sgid, p->fsgid);
task_lock(p);
if (p->files)
fdt = files_fdtable(p->files);
seq_printf(m,
"FDSize:\t%d\n"
"Groups:\t",
fdt ? fdt->max_fds : 0);
rcu_read_unlock();
group_info = p->group_info;
get_group_info(group_info);
task_unlock(p);
for (g = 0; g < min(group_info->ngroups, NGROUPS_SMALL); g++)
seq_printf(m, "%d ", GROUP_AT(group_info, g));
put_group_info(group_info);
seq_printf(m, "\n");
}
static void render_sigset_t(struct seq_file *m, const char *header,
sigset_t *set)
{
int i;
seq_printf(m, "%s", header);
i = _NSIG;
do {
int x = 0;
i -= 4;
if (sigismember(set, i+1)) x |= 1;
if (sigismember(set, i+2)) x |= 2;
if (sigismember(set, i+3)) x |= 4;
if (sigismember(set, i+4)) x |= 8;
seq_printf(m, "%x", x);
} while (i >= 4);
seq_printf(m, "\n");
}
static void collect_sigign_sigcatch(struct task_struct *p, sigset_t *ign,
sigset_t *catch)
{
struct k_sigaction *k;
int i;
k = p->sighand->action;
for (i = 1; i <= _NSIG; ++i, ++k) {
if (k->sa.sa_handler == SIG_IGN)
sigaddset(ign, i);
else if (k->sa.sa_handler != SIG_DFL)
sigaddset(catch, i);
}
}
static inline void task_sig(struct seq_file *m, struct task_struct *p)
{
unsigned long flags;
sigset_t pending, shpending, blocked, ignored, caught;
int num_threads = 0;
unsigned long qsize = 0;
unsigned long qlim = 0;
sigemptyset(&pending);
sigemptyset(&shpending);
sigemptyset(&blocked);
sigemptyset(&ignored);
sigemptyset(&caught);
if (lock_task_sighand(p, &flags)) {
pending = p->pending.signal;
shpending = p->signal->shared_pending.signal;
blocked = p->blocked;
collect_sigign_sigcatch(p, &ignored, &caught);
num_threads = atomic_read(&p->signal->count);
qsize = atomic_read(&p->user->sigpending);
qlim = p->signal->rlim[RLIMIT_SIGPENDING].rlim_cur;
unlock_task_sighand(p, &flags);
}
seq_printf(m, "Threads:\t%d\n", num_threads);
seq_printf(m, "SigQ:\t%lu/%lu\n", qsize, qlim);
/* render them all */
render_sigset_t(m, "SigPnd:\t", &pending);
render_sigset_t(m, "ShdPnd:\t", &shpending);
render_sigset_t(m, "SigBlk:\t", &blocked);
render_sigset_t(m, "SigIgn:\t", &ignored);
render_sigset_t(m, "SigCgt:\t", &caught);
}
static void render_cap_t(struct seq_file *m, const char *header,
kernel_cap_t *a)
{
unsigned __capi;
seq_printf(m, "%s", header);
CAP_FOR_EACH_U32(__capi) {
seq_printf(m, "%08x",
capabilities: remain source compatible with 32-bit raw legacy capability support. Source code out there hard-codes a notion of what the _LINUX_CAPABILITY_VERSION #define means in terms of the semantics of the raw capability system calls capget() and capset(). Its unfortunate, but true. Since the confusing header file has been in a released kernel, there is software that is erroneously using 64-bit capabilities with the semantics of 32-bit compatibilities. These recently compiled programs may suffer corruption of their memory when sys_getcap() overwrites more memory than they are coded to expect, and the raising of added capabilities when using sys_capset(). As such, this patch does a number of things to clean up the situation for all. It 1. forces the _LINUX_CAPABILITY_VERSION define to always retain its legacy value. 2. adopts a new #define strategy for the kernel's internal implementation of the preferred magic. 3. deprecates v2 capability magic in favor of a new (v3) magic number. The functionality of v3 is entirely equivalent to v2, the only difference being that the v2 magic causes the kernel to log a "deprecated" warning so the admin can find applications that may be using v2 inappropriately. [User space code continues to be encouraged to use the libcap API which protects the application from details like this. libcap-2.10 is the first to support v3 capabilities.] Fixes issue reported in https://bugzilla.redhat.com/show_bug.cgi?id=447518. Thanks to Bojan Smojver for the report. [akpm@linux-foundation.org: s/depreciate/deprecate/g] [akpm@linux-foundation.org: be robust about put_user size] [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Andrew G. Morgan <morgan@kernel.org> Cc: Serge E. Hallyn <serue@us.ibm.com> Cc: Bojan Smojver <bojan@rexursive.com> Cc: stable@kernel.org Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2008-05-28 05:05:17 +00:00
a->cap[(_KERNEL_CAPABILITY_U32S-1) - __capi]);
}
seq_printf(m, "\n");
}
static inline void task_cap(struct seq_file *m, struct task_struct *p)
{
render_cap_t(m, "CapInh:\t", &p->cap_inheritable);
render_cap_t(m, "CapPrm:\t", &p->cap_permitted);
render_cap_t(m, "CapEff:\t", &p->cap_effective);
render_cap_t(m, "CapBnd:\t", &p->cap_bset);
}
static inline void task_context_switch_counts(struct seq_file *m,
struct task_struct *p)
{
seq_printf(m, "voluntary_ctxt_switches:\t%lu\n"
"nonvoluntary_ctxt_switches:\t%lu\n",
p->nvcsw,
p->nivcsw);
}
int proc_pid_status(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
struct mm_struct *mm = get_task_mm(task);
task_name(m, task);
task_state(m, ns, pid, task);
if (mm) {
task_mem(m, mm);
mmput(mm);
}
task_sig(m, task);
task_cap(m, task);
cpuset_task_status_allowed(m, task);
#if defined(CONFIG_S390)
task_show_regs(m, task);
#endif
task_context_switch_counts(m, task);
return 0;
}
static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task, int whole)
{
unsigned long vsize, eip, esp, wchan = ~0UL;
long priority, nice;
int tty_pgrp = -1, tty_nr = 0;
sigset_t sigign, sigcatch;
char state;
pid_t ppid = 0, pgid = -1, sid = -1;
int num_threads = 0;
struct mm_struct *mm;
unsigned long long start_time;
unsigned long cmin_flt = 0, cmaj_flt = 0;
unsigned long min_flt = 0, maj_flt = 0;
cputime_t cutime, cstime, utime, stime;
cputime_t cgtime, gtime;
unsigned long rsslim = 0;
char tcomm[sizeof(task->comm)];
unsigned long flags;
state = *get_task_state(task);
vsize = eip = esp = 0;
mm = get_task_mm(task);
if (mm) {
vsize = task_vsize(mm);
eip = KSTK_EIP(task);
esp = KSTK_ESP(task);
}
get_task_comm(tcomm, task);
sigemptyset(&sigign);
sigemptyset(&sigcatch);
cutime = cstime = utime = stime = cputime_zero;
cgtime = gtime = cputime_zero;
if (lock_task_sighand(task, &flags)) {
struct signal_struct *sig = task->signal;
if (sig->tty) {
struct pid *pgrp = tty_get_pgrp(sig->tty);
tty_pgrp = pid_nr_ns(pgrp, ns);
put_pid(pgrp);
tty_nr = new_encode_dev(tty_devnum(sig->tty));
}
num_threads = atomic_read(&sig->count);
collect_sigign_sigcatch(task, &sigign, &sigcatch);
cmin_flt = sig->cmin_flt;
cmaj_flt = sig->cmaj_flt;
cutime = sig->cutime;
cstime = sig->cstime;
cgtime = sig->cgtime;
rsslim = sig->rlim[RLIMIT_RSS].rlim_cur;
/* add up live thread stats at the group level */
if (whole) {
timers: fix itimer/many thread hang Overview This patch reworks the handling of POSIX CPU timers, including the ITIMER_PROF, ITIMER_VIRT timers and rlimit handling. It was put together with the help of Roland McGrath, the owner and original writer of this code. The problem we ran into, and the reason for this rework, has to do with using a profiling timer in a process with a large number of threads. It appears that the performance of the old implementation of run_posix_cpu_timers() was at least O(n*3) (where "n" is the number of threads in a process) or worse. Everything is fine with an increasing number of threads until the time taken for that routine to run becomes the same as or greater than the tick time, at which point things degrade rather quickly. This patch fixes bug 9906, "Weird hang with NPTL and SIGPROF." Code Changes This rework corrects the implementation of run_posix_cpu_timers() to make it run in constant time for a particular machine. (Performance may vary between one machine and another depending upon whether the kernel is built as single- or multiprocessor and, in the latter case, depending upon the number of running processors.) To do this, at each tick we now update fields in signal_struct as well as task_struct. The run_posix_cpu_timers() function uses those fields to make its decisions. We define a new structure, "task_cputime," to contain user, system and scheduler times and use these in appropriate places: struct task_cputime { cputime_t utime; cputime_t stime; unsigned long long sum_exec_runtime; }; This is included in the structure "thread_group_cputime," which is a new substructure of signal_struct and which varies for uniprocessor versus multiprocessor kernels. For uniprocessor kernels, it uses "task_cputime" as a simple substructure, while for multiprocessor kernels it is a pointer: struct thread_group_cputime { struct task_cputime totals; }; struct thread_group_cputime { struct task_cputime *totals; }; We also add a new task_cputime substructure directly to signal_struct, to cache the earliest expiration of process-wide timers, and task_cputime also replaces the it_*_expires fields of task_struct (used for earliest expiration of thread timers). The "thread_group_cputime" structure contains process-wide timers that are updated via account_user_time() and friends. In the non-SMP case the structure is a simple aggregator; unfortunately in the SMP case that simplicity was not achievable due to cache-line contention between CPUs (in one measured case performance was actually _worse_ on a 16-cpu system than the same test on a 4-cpu system, due to this contention). For SMP, the thread_group_cputime counters are maintained as a per-cpu structure allocated using alloc_percpu(). The timer functions update only the timer field in the structure corresponding to the running CPU, obtained using per_cpu_ptr(). We define a set of inline functions in sched.h that we use to maintain the thread_group_cputime structure and hide the differences between UP and SMP implementations from the rest of the kernel. The thread_group_cputime_init() function initializes the thread_group_cputime structure for the given task. The thread_group_cputime_alloc() is a no-op for UP; for SMP it calls the out-of-line function thread_group_cputime_alloc_smp() to allocate and fill in the per-cpu structures and fields. The thread_group_cputime_free() function, also a no-op for UP, in SMP frees the per-cpu structures. The thread_group_cputime_clone_thread() function (also a UP no-op) for SMP calls thread_group_cputime_alloc() if the per-cpu structures haven't yet been allocated. The thread_group_cputime() function fills the task_cputime structure it is passed with the contents of the thread_group_cputime fields; in UP it's that simple but in SMP it must also safely check that tsk->signal is non-NULL (if it is it just uses the appropriate fields of task_struct) and, if so, sums the per-cpu values for each online CPU. Finally, the three functions account_group_user_time(), account_group_system_time() and account_group_exec_runtime() are used by timer functions to update the respective fields of the thread_group_cputime structure. Non-SMP operation is trivial and will not be mentioned further. The per-cpu structure is always allocated when a task creates its first new thread, via a call to thread_group_cputime_clone_thread() from copy_signal(). It is freed at process exit via a call to thread_group_cputime_free() from cleanup_signal(). All functions that formerly summed utime/stime/sum_sched_runtime values from from all threads in the thread group now use thread_group_cputime() to snapshot the values in the thread_group_cputime structure or the values in the task structure itself if the per-cpu structure hasn't been allocated. Finally, the code in kernel/posix-cpu-timers.c has changed quite a bit. The run_posix_cpu_timers() function has been split into a fast path and a slow path; the former safely checks whether there are any expired thread timers and, if not, just returns, while the slow path does the heavy lifting. With the dedicated thread group fields, timers are no longer "rebalanced" and the process_timer_rebalance() function and related code has gone away. All summing loops are gone and all code that used them now uses the thread_group_cputime() inline. When process-wide timers are set, the new task_cputime structure in signal_struct is used to cache the earliest expiration; this is checked in the fast path. Performance The fix appears not to add significant overhead to existing operations. It generally performs the same as the current code except in two cases, one in which it performs slightly worse (Case 5 below) and one in which it performs very significantly better (Case 2 below). Overall it's a wash except in those two cases. I've since done somewhat more involved testing on a dual-core Opteron system. Case 1: With no itimer running, for a test with 100,000 threads, the fixed kernel took 1428.5 seconds, 513 seconds more than the unfixed system, all of which was spent in the system. There were twice as many voluntary context switches with the fix as without it. Case 2: With an itimer running at .01 second ticks and 4000 threads (the most an unmodified kernel can handle), the fixed kernel ran the test in eight percent of the time (5.8 seconds as opposed to 70 seconds) and had better tick accuracy (.012 seconds per tick as opposed to .023 seconds per tick). Case 3: A 4000-thread test with an initial timer tick of .01 second and an interval of 10,000 seconds (i.e. a timer that ticks only once) had very nearly the same performance in both cases: 6.3 seconds elapsed for the fixed kernel versus 5.5 seconds for the unfixed kernel. With fewer threads (eight in these tests), the Case 1 test ran in essentially the same time on both the modified and unmodified kernels (5.2 seconds versus 5.8 seconds). The Case 2 test ran in about the same time as well, 5.9 seconds versus 5.4 seconds but again with much better tick accuracy, .013 seconds per tick versus .025 seconds per tick for the unmodified kernel. Since the fix affected the rlimit code, I also tested soft and hard CPU limits. Case 4: With a hard CPU limit of 20 seconds and eight threads (and an itimer running), the modified kernel was very slightly favored in that while it killed the process in 19.997 seconds of CPU time (5.002 seconds of wall time), only .003 seconds of that was system time, the rest was user time. The unmodified kernel killed the process in 20.001 seconds of CPU (5.014 seconds of wall time) of which .016 seconds was system time. Really, though, the results were too close to call. The results were essentially the same with no itimer running. Case 5: With a soft limit of 20 seconds and a hard limit of 2000 seconds (where the hard limit would never be reached) and an itimer running, the modified kernel exhibited worse tick accuracy than the unmodified kernel: .050 seconds/tick versus .028 seconds/tick. Otherwise, performance was almost indistinguishable. With no itimer running this test exhibited virtually identical behavior and times in both cases. In times past I did some limited performance testing. those results are below. On a four-cpu Opteron system without this fix, a sixteen-thread test executed in 3569.991 seconds, of which user was 3568.435s and system was 1.556s. On the same system with the fix, user and elapsed time were about the same, but system time dropped to 0.007 seconds. Performance with eight, four and one thread were comparable. Interestingly, the timer ticks with the fix seemed more accurate: The sixteen-thread test with the fix received 149543 ticks for 0.024 seconds per tick, while the same test without the fix received 58720 for 0.061 seconds per tick. Both cases were configured for an interval of 0.01 seconds. Again, the other tests were comparable. Each thread in this test computed the primes up to 25,000,000. I also did a test with a large number of threads, 100,000 threads, which is impossible without the fix. In this case each thread computed the primes only up to 10,000 (to make the runtime manageable). System time dominated, at 1546.968 seconds out of a total 2176.906 seconds (giving a user time of 629.938s). It received 147651 ticks for 0.015 seconds per tick, still quite accurate. There is obviously no comparable test without the fix. Signed-off-by: Frank Mayhar <fmayhar@google.com> Cc: Roland McGrath <roland@redhat.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-12 16:54:39 +00:00
struct task_cputime cputime;
struct task_struct *t = task;
do {
min_flt += t->min_flt;
maj_flt += t->maj_flt;
gtime = cputime_add(gtime, task_gtime(t));
t = next_thread(t);
} while (t != task);
min_flt += sig->min_flt;
maj_flt += sig->maj_flt;
timers: fix itimer/many thread hang Overview This patch reworks the handling of POSIX CPU timers, including the ITIMER_PROF, ITIMER_VIRT timers and rlimit handling. It was put together with the help of Roland McGrath, the owner and original writer of this code. The problem we ran into, and the reason for this rework, has to do with using a profiling timer in a process with a large number of threads. It appears that the performance of the old implementation of run_posix_cpu_timers() was at least O(n*3) (where "n" is the number of threads in a process) or worse. Everything is fine with an increasing number of threads until the time taken for that routine to run becomes the same as or greater than the tick time, at which point things degrade rather quickly. This patch fixes bug 9906, "Weird hang with NPTL and SIGPROF." Code Changes This rework corrects the implementation of run_posix_cpu_timers() to make it run in constant time for a particular machine. (Performance may vary between one machine and another depending upon whether the kernel is built as single- or multiprocessor and, in the latter case, depending upon the number of running processors.) To do this, at each tick we now update fields in signal_struct as well as task_struct. The run_posix_cpu_timers() function uses those fields to make its decisions. We define a new structure, "task_cputime," to contain user, system and scheduler times and use these in appropriate places: struct task_cputime { cputime_t utime; cputime_t stime; unsigned long long sum_exec_runtime; }; This is included in the structure "thread_group_cputime," which is a new substructure of signal_struct and which varies for uniprocessor versus multiprocessor kernels. For uniprocessor kernels, it uses "task_cputime" as a simple substructure, while for multiprocessor kernels it is a pointer: struct thread_group_cputime { struct task_cputime totals; }; struct thread_group_cputime { struct task_cputime *totals; }; We also add a new task_cputime substructure directly to signal_struct, to cache the earliest expiration of process-wide timers, and task_cputime also replaces the it_*_expires fields of task_struct (used for earliest expiration of thread timers). The "thread_group_cputime" structure contains process-wide timers that are updated via account_user_time() and friends. In the non-SMP case the structure is a simple aggregator; unfortunately in the SMP case that simplicity was not achievable due to cache-line contention between CPUs (in one measured case performance was actually _worse_ on a 16-cpu system than the same test on a 4-cpu system, due to this contention). For SMP, the thread_group_cputime counters are maintained as a per-cpu structure allocated using alloc_percpu(). The timer functions update only the timer field in the structure corresponding to the running CPU, obtained using per_cpu_ptr(). We define a set of inline functions in sched.h that we use to maintain the thread_group_cputime structure and hide the differences between UP and SMP implementations from the rest of the kernel. The thread_group_cputime_init() function initializes the thread_group_cputime structure for the given task. The thread_group_cputime_alloc() is a no-op for UP; for SMP it calls the out-of-line function thread_group_cputime_alloc_smp() to allocate and fill in the per-cpu structures and fields. The thread_group_cputime_free() function, also a no-op for UP, in SMP frees the per-cpu structures. The thread_group_cputime_clone_thread() function (also a UP no-op) for SMP calls thread_group_cputime_alloc() if the per-cpu structures haven't yet been allocated. The thread_group_cputime() function fills the task_cputime structure it is passed with the contents of the thread_group_cputime fields; in UP it's that simple but in SMP it must also safely check that tsk->signal is non-NULL (if it is it just uses the appropriate fields of task_struct) and, if so, sums the per-cpu values for each online CPU. Finally, the three functions account_group_user_time(), account_group_system_time() and account_group_exec_runtime() are used by timer functions to update the respective fields of the thread_group_cputime structure. Non-SMP operation is trivial and will not be mentioned further. The per-cpu structure is always allocated when a task creates its first new thread, via a call to thread_group_cputime_clone_thread() from copy_signal(). It is freed at process exit via a call to thread_group_cputime_free() from cleanup_signal(). All functions that formerly summed utime/stime/sum_sched_runtime values from from all threads in the thread group now use thread_group_cputime() to snapshot the values in the thread_group_cputime structure or the values in the task structure itself if the per-cpu structure hasn't been allocated. Finally, the code in kernel/posix-cpu-timers.c has changed quite a bit. The run_posix_cpu_timers() function has been split into a fast path and a slow path; the former safely checks whether there are any expired thread timers and, if not, just returns, while the slow path does the heavy lifting. With the dedicated thread group fields, timers are no longer "rebalanced" and the process_timer_rebalance() function and related code has gone away. All summing loops are gone and all code that used them now uses the thread_group_cputime() inline. When process-wide timers are set, the new task_cputime structure in signal_struct is used to cache the earliest expiration; this is checked in the fast path. Performance The fix appears not to add significant overhead to existing operations. It generally performs the same as the current code except in two cases, one in which it performs slightly worse (Case 5 below) and one in which it performs very significantly better (Case 2 below). Overall it's a wash except in those two cases. I've since done somewhat more involved testing on a dual-core Opteron system. Case 1: With no itimer running, for a test with 100,000 threads, the fixed kernel took 1428.5 seconds, 513 seconds more than the unfixed system, all of which was spent in the system. There were twice as many voluntary context switches with the fix as without it. Case 2: With an itimer running at .01 second ticks and 4000 threads (the most an unmodified kernel can handle), the fixed kernel ran the test in eight percent of the time (5.8 seconds as opposed to 70 seconds) and had better tick accuracy (.012 seconds per tick as opposed to .023 seconds per tick). Case 3: A 4000-thread test with an initial timer tick of .01 second and an interval of 10,000 seconds (i.e. a timer that ticks only once) had very nearly the same performance in both cases: 6.3 seconds elapsed for the fixed kernel versus 5.5 seconds for the unfixed kernel. With fewer threads (eight in these tests), the Case 1 test ran in essentially the same time on both the modified and unmodified kernels (5.2 seconds versus 5.8 seconds). The Case 2 test ran in about the same time as well, 5.9 seconds versus 5.4 seconds but again with much better tick accuracy, .013 seconds per tick versus .025 seconds per tick for the unmodified kernel. Since the fix affected the rlimit code, I also tested soft and hard CPU limits. Case 4: With a hard CPU limit of 20 seconds and eight threads (and an itimer running), the modified kernel was very slightly favored in that while it killed the process in 19.997 seconds of CPU time (5.002 seconds of wall time), only .003 seconds of that was system time, the rest was user time. The unmodified kernel killed the process in 20.001 seconds of CPU (5.014 seconds of wall time) of which .016 seconds was system time. Really, though, the results were too close to call. The results were essentially the same with no itimer running. Case 5: With a soft limit of 20 seconds and a hard limit of 2000 seconds (where the hard limit would never be reached) and an itimer running, the modified kernel exhibited worse tick accuracy than the unmodified kernel: .050 seconds/tick versus .028 seconds/tick. Otherwise, performance was almost indistinguishable. With no itimer running this test exhibited virtually identical behavior and times in both cases. In times past I did some limited performance testing. those results are below. On a four-cpu Opteron system without this fix, a sixteen-thread test executed in 3569.991 seconds, of which user was 3568.435s and system was 1.556s. On the same system with the fix, user and elapsed time were about the same, but system time dropped to 0.007 seconds. Performance with eight, four and one thread were comparable. Interestingly, the timer ticks with the fix seemed more accurate: The sixteen-thread test with the fix received 149543 ticks for 0.024 seconds per tick, while the same test without the fix received 58720 for 0.061 seconds per tick. Both cases were configured for an interval of 0.01 seconds. Again, the other tests were comparable. Each thread in this test computed the primes up to 25,000,000. I also did a test with a large number of threads, 100,000 threads, which is impossible without the fix. In this case each thread computed the primes only up to 10,000 (to make the runtime manageable). System time dominated, at 1546.968 seconds out of a total 2176.906 seconds (giving a user time of 629.938s). It received 147651 ticks for 0.015 seconds per tick, still quite accurate. There is obviously no comparable test without the fix. Signed-off-by: Frank Mayhar <fmayhar@google.com> Cc: Roland McGrath <roland@redhat.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-12 16:54:39 +00:00
thread_group_cputime(task, &cputime);
utime = cputime.utime;
stime = cputime.stime;
gtime = cputime_add(gtime, sig->gtime);
}
sid = task_session_nr_ns(task, ns);
ppid = task_tgid_nr_ns(task->real_parent, ns);
pgid = task_pgrp_nr_ns(task, ns);
unlock_task_sighand(task, &flags);
}
if (!whole || num_threads < 2)
wchan = get_wchan(task);
if (!whole) {
min_flt = task->min_flt;
maj_flt = task->maj_flt;
utime = task_utime(task);
stime = task_stime(task);
gtime = task_gtime(task);
}
/* scale priority and nice values from timeslices to -20..20 */
/* to make it look like a "normal" Unix priority/nice value */
priority = task_prio(task);
nice = task_nice(task);
/* Temporary variable needed for gcc-2.96 */
/* convert timespec -> nsec*/
start_time =
(unsigned long long)task->real_start_time.tv_sec * NSEC_PER_SEC
+ task->real_start_time.tv_nsec;
/* 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\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 ? mm->start_code : 0,
mm ? mm->end_code : 0,
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));
if (mm)
mmput(mm);
return 0;
}
int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
return do_task_stat(m, ns, pid, task, 0);
}
int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
return do_task_stat(m, ns, pid, task, 1);
}
int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
int size = 0, resident = 0, shared = 0, text = 0, lib = 0, data = 0;
struct mm_struct *mm = get_task_mm(task);
if (mm) {
size = task_statm(mm, &shared, &text, &data, &resident);
mmput(mm);
}
seq_printf(m, "%d %d %d %d %d %d %d\n",
size, resident, shared, text, lib, data, 0);
return 0;
}