This patch moves SIGNAL_STOP_DEQUEUED from signal_struct->flags to
task_struct->group_stop, and thus makes it per-thread.
Like SIGNAL_STOP_DEQUEUED, GROUP_STOP_DEQUEUED can be false-positive
after return from get_signal_to_deliver(), this is fine. The only
purpose of this bit is: we can drop ->siglock after __dequeue_signal()
returns the sig_kernel_stop() signal and before we call
do_signal_stop(), in this case we must not miss SIGCONT if it comes in
between.
But, unlike SIGNAL_STOP_DEQUEUED, GROUP_STOP_DEQUEUED can not be
false-positive in do_signal_stop() if multiple threads dequeue the
sig_kernel_stop() signal at the same time.
Consider two threads T1 and T2, SIGTTIN has a hanlder.
- T1 dequeues SIGTSTP and sets SIGNAL_STOP_DEQUEUED, then
it drops ->siglock
- SIGCONT comes and clears SIGNAL_STOP_DEQUEUED, SIGTSTP
should be cancelled.
- T2 dequeues SIGTTIN and sets SIGNAL_STOP_DEQUEUED again.
Since we have a handler we should not stop, T2 returns
to usermode to run the handler.
- T1 continues, calls do_signal_stop() and wrongly starts
the group stop because SIGNAL_STOP_DEQUEUED was restored
in between.
With or without this change:
- we need to do something with ptrace_signal() which can
return SIGSTOP, but this needs another discussion
- SIGSTOP can be lost if it races with the mt exec, will
be fixed later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
PF_EXITING or TASK_STOPPED has already called task_participate_group_stop()
and cleared its ->group_stop. No need to do task_clear_group_stop_pending()
when we start the new group stop.
Add a small comment to explain the !task_is_stopped() check. Note that this
check is not exactly right and it can lead to unnecessary stop later if the
thread is TASK_PTRACED. What we need is task_participated_in_group_stop(),
this will be solved later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
prepare_signal(SIGCONT) should never set TIF_SIGPENDING or wake up
the TASK_INTERRUPTIBLE threads. We are going to call complete_signal()
which should pick the right thread correctly. All we need is to wake
up the TASK_STOPPED threads.
If the task was stopped, it can't return to usermode without taking
->siglock. Otherwise we don't care, and the spurious TIF_SIGPENDING
can't be useful.
The comment says:
* If there is a handler for SIGCONT, we must make
* sure that no thread returns to user mode before
* we post the signal
It is not clear what this means. Probably, "when there's only a single
thread" and this continues to be true. Otherwise, even if this SIGCONT
is not private, with or without this change only one thread can dequeue
SIGCONT, other threads can happily return to user mode before before
that thread handles this signal.
Note also that wake_up_state(t, __TASK_STOPPED) can't race with the task
which changes its state, TASK_STOPPED state is protected by ->siglock as
well.
In short: when it comes to signal delivery, SIGCONT is the normal signal
and does not need any special support.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Commit da48524eb2 ("Prevent rt_sigqueueinfo and rt_tgsigqueueinfo
from spoofing the signal code") made the check on si_code too strict.
There are several legitimate places where glibc wants to queue a
negative si_code different from SI_QUEUE:
- This was first noticed with glibc's aio implementation, which wants
to queue a signal with si_code SI_ASYNCIO; the current kernel
causes glibc's tst-aio4 test to fail because rt_sigqueueinfo()
fails with EPERM.
- Further examination of the glibc source shows that getaddrinfo_a()
wants to use SI_ASYNCNL (which the kernel does not even define).
The timer_create() fallback code wants to queue signals with SI_TIMER.
As suggested by Oleg Nesterov <oleg@redhat.com>, loosen the check to
forbid only the problematic SI_TKILL case.
Reported-by: Klaus Dittrich <kladit@arcor.de>
Acked-by: Julien Tinnes <jln@google.com>
Cc: <stable@kernel.org>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Changelog:
Dec 8: Fixed bug in my check_kill_permission pointed out by
Eric Biederman.
Dec 13: Apply Eric's suggestion to pass target task into kill_ok_by_cred()
for clarity
Dec 31: address comment by Eric Biederman:
don't need cred/tcred in check_kill_permission.
Jan 1: use const cred struct.
Jan 11: Per Bastian Blank's advice, clean up kill_ok_by_cred().
Feb 16: kill_ok_by_cred: fix bad parentheses
Feb 23: per akpm, let compiler inline kill_ok_by_cred
Signed-off-by: Serge E. Hallyn <serge.hallyn@canonical.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Daniel Lezcano <daniel.lezcano@free.fr>
Acked-by: David Howells <dhowells@redhat.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Just as group_exit_code shouldn't be generated when a PTRACE_CONT'd
task re-enters job control stop, notifiction for the event should be
suppressed too. The logic is the same as the group_exit_code
generation suppression in do_signal_stop(), if SIGNAL_STOP_STOPPED is
already set, the task is re-entering job control stop without
intervening SIGCONT and the notifications should be suppressed.
Test case follows.
#include <stdio.h>
#include <unistd.h>
#include <signal.h>
#include <time.h>
#include <sys/ptrace.h>
#include <sys/wait.h>
static const struct timespec ts100ms = { .tv_nsec = 100000000 };
static pid_t tracee, tracer;
static const char *pid_who(pid_t pid)
{
return pid == tracee ? "tracee" : (pid == tracer ? "tracer" : "mommy ");
}
static void sigchld_sigaction(int signo, siginfo_t *si, void *ucxt)
{
printf("%s: SIG status=%02d code=%02d (%s)\n",
pid_who(getpid()), si->si_status, si->si_code,
pid_who(si->si_pid));
}
int main(void)
{
const struct sigaction chld_sa = { .sa_sigaction = sigchld_sigaction,
.sa_flags = SA_SIGINFO|SA_RESTART };
siginfo_t si;
sigaction(SIGCHLD, &chld_sa, NULL);
tracee = fork();
if (!tracee) {
tracee = getpid();
while (1)
pause();
}
kill(tracee, SIGSTOP);
waitid(P_PID, tracee, &si, WSTOPPED);
tracer = fork();
if (!tracer) {
tracer = getpid();
ptrace(PTRACE_ATTACH, tracee, NULL, NULL);
waitid(P_PID, tracee, &si, WSTOPPED);
ptrace(PTRACE_CONT, tracee, NULL, (void *)(long)si.si_status);
waitid(P_PID, tracee, &si, WSTOPPED);
ptrace(PTRACE_CONT, tracee, NULL, (void *)(long)si.si_status);
waitid(P_PID, tracee, &si, WSTOPPED);
printf("tracer: detaching\n");
ptrace(PTRACE_DETACH, tracee, NULL, NULL);
return 0;
}
while (1)
pause();
return 0;
}
Before the patch, the parent gets the second notification for the
tracee after the tracer detaches. si_status is zero because
group_exit_code is not set by the group stop completion which
triggered this notification.
mommy : SIG status=19 code=05 (tracee)
tracer: SIG status=00 code=05 (tracee)
tracer: SIG status=19 code=04 (tracee)
tracer: SIG status=00 code=05 (tracee)
tracer: detaching
mommy : SIG status=00 code=05 (tracee)
mommy : SIG status=00 code=01 (tracer)
^C
After the patch, the duplicate notification is gone.
mommy : SIG status=19 code=05 (tracee)
tracer: SIG status=00 code=05 (tracee)
tracer: SIG status=19 code=04 (tracee)
tracer: SIG status=00 code=05 (tracee)
tracer: detaching
mommy : SIG status=00 code=01 (tracer)
^C
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
With recent changes, job control and ptrace stopped states are
properly separated and accessible to the real parent and the ptracer
respectively; however, notifications of job control stopped/continued
events to the real parent while ptraced are still missing.
A ptracee participates in group stop in ptrace_stop() but the
completion isn't notified. If participation results in completion of
group stop, notify the real parent of the event. The ptrace and group
stops are separate and can be handled as such.
However, when the real parent and the ptracer are in the same thread
group, only the ptrace stop event is visible through wait(2) and the
duplicate notifications are different from the current behavior and
are confusing. Suppress group stop notification in such cases.
The continued state is shared between the real parent and the ptracer
but is only meaningful to the real parent. Always notify the real
parent and notify the ptracer too for backward compatibility. Similar
to stop notification, if the real parent is the ptracer, suppress a
duplicate notification.
Test case follows.
#include <stdio.h>
#include <unistd.h>
#include <time.h>
#include <errno.h>
#include <sys/types.h>
#include <sys/ptrace.h>
#include <sys/wait.h>
int main(void)
{
const struct timespec ts100ms = { .tv_nsec = 100000000 };
pid_t tracee, tracer;
siginfo_t si;
int i;
tracee = fork();
if (tracee == 0) {
while (1) {
printf("tracee: SIGSTOP\n");
raise(SIGSTOP);
nanosleep(&ts100ms, NULL);
printf("tracee: SIGCONT\n");
raise(SIGCONT);
nanosleep(&ts100ms, NULL);
}
}
waitid(P_PID, tracee, &si, WSTOPPED | WNOHANG | WNOWAIT);
tracer = fork();
if (tracer == 0) {
nanosleep(&ts100ms, NULL);
ptrace(PTRACE_ATTACH, tracee, NULL, NULL);
for (i = 0; i < 11; i++) {
si.si_pid = 0;
waitid(P_PID, tracee, &si, WSTOPPED);
if (si.si_pid && si.si_code == CLD_TRAPPED)
ptrace(PTRACE_CONT, tracee, NULL,
(void *)(long)si.si_status);
}
printf("tracer: EXITING\n");
return 0;
}
while (1) {
si.si_pid = 0;
waitid(P_PID, tracee, &si, WSTOPPED | WCONTINUED | WEXITED);
if (si.si_pid)
printf("mommy : WAIT status=%02d code=%02d\n",
si.si_status, si.si_code);
}
return 0;
}
Before this patch, while ptraced, the real parent doesn't get
notifications for job control events, so although it can access those
events, the later waitid(2) call never wakes up.
tracee: SIGSTOP
mommy : WAIT status=19 code=05
tracee: SIGCONT
tracee: SIGSTOP
tracee: SIGCONT
tracee: SIGSTOP
tracee: SIGCONT
tracee: SIGSTOP
tracer: EXITING
mommy : WAIT status=19 code=05
^C
After this patch, it works as expected.
tracee: SIGSTOP
mommy : WAIT status=19 code=05
tracee: SIGCONT
mommy : WAIT status=18 code=06
tracee: SIGSTOP
mommy : WAIT status=19 code=05
tracee: SIGCONT
mommy : WAIT status=18 code=06
tracee: SIGSTOP
mommy : WAIT status=19 code=05
tracee: SIGCONT
mommy : WAIT status=18 code=06
tracee: SIGSTOP
tracer: EXITING
mommy : WAIT status=19 code=05
^C
-v2: Oleg pointed out that
* Group stop notification to the real parent should also happen
when ptracer detach races with ptrace_stop().
* real_parent_is_ptracer() should be testing thread group
equality not the task itself as wait(2) and stop/cont
notifications are normally thread-group wide.
Both issues are fixed accordingly.
-v3: real_parent_is_ptracer() updated to test child->real_parent
instead of child->group_leader->real_parent per Oleg's
suggestion.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
The stopped notifications in do_signal_stop() and exit_signals() are
always for the completion of job control. The one in do_signal_stop()
may be delivered to the ptracer if PTRACE_ATTACH races with
notification and the one in exit_signals() if task exits while
ptraced.
In both cases, the notifications are meaningless and confusing to the
ptracer as it never accesses the group stop state while the real
parent would miss notifications for the events it is watching.
Make sure these notifications always go to the real parent by calling
do_notify_parent_cld_stop() with %false @for_ptrace.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Currently, do_notify_parent_cldstop() determines whether the
notification is for the real parent or ptracer. Move the decision to
the caller by adding @for_ptrace parameter to
do_notify_parent_cldstop(). All the callers are updated to pass
task_ptrace(target_task), so this patch doesn't cause any behavior
difference.
While at it, add function comment to do_notify_parent_cldstop().
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
While ptraced, a task may be resumed while the containing process is
still job control stopped. If the task receives another stop signal
in this state, it will still initiate group stop, which generates
group_exit_code, which the real parent would be able to see once the
ptracer detaches.
In this scenario, the real parent may see two consecutive CLD_STOPPED
events from two stop signals without intervening SIGCONT, which
normally is impossible.
Test case follows.
#include <stdio.h>
#include <unistd.h>
#include <sys/ptrace.h>
#include <sys/wait.h>
int main(void)
{
pid_t tracee;
siginfo_t si;
tracee = fork();
if (!tracee)
while (1)
pause();
kill(tracee, SIGSTOP);
waitid(P_PID, tracee, &si, WSTOPPED);
if (!fork()) {
ptrace(PTRACE_ATTACH, tracee, NULL, NULL);
waitid(P_PID, tracee, &si, WSTOPPED);
ptrace(PTRACE_CONT, tracee, NULL, (void *)(long)si.si_status);
waitid(P_PID, tracee, &si, WSTOPPED);
ptrace(PTRACE_CONT, tracee, NULL, (void *)(long)si.si_status);
waitid(P_PID, tracee, &si, WSTOPPED);
ptrace(PTRACE_DETACH, tracee, NULL, NULL);
return 0;
}
while (1) {
si.si_pid = 0;
waitid(P_PID, tracee, &si, WSTOPPED | WNOHANG);
if (si.si_pid)
printf("st=%02d c=%02d\n", si.si_status, si.si_code);
}
return 0;
}
Before the patch, the latter waitid() in polling mode reports the
second stopped event generated by the implied SIGSTOP of
PTRACE_ATTACH.
st=19 c=05
^C
After the patch, the second event is not reported.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Currently, if the task is STOPPED on ptrace attach, it's left alone
and the state is silently changed to TRACED on the next ptrace call.
The behavior breaks the assumption that arch_ptrace_stop() is called
before any task is poked by ptrace and is ugly in that a task
manipulates the state of another task directly.
With GROUP_STOP_PENDING, the transitions between TASK_STOPPED and
TRACED can be made clean. The tracer can use the flag to tell the
tracee to retry stop on attach and detach. On retry, the tracee will
enter the desired state in the correct way. The lower 16bits of
task->group_stop is used to remember the signal number which caused
the last group stop. This is used while retrying for ptrace attach as
the original group_exit_code could have been consumed with wait(2) by
then.
As the real parent may wait(2) and consume the group_exit_code
anytime, the group_exit_code needs to be saved separately so that it
can be used when switching from regular sleep to ptrace_stop(). This
is recorded in the lower 16bits of task->group_stop.
If a task is already stopped and there's no intervening SIGCONT, a
ptrace request immediately following a successful PTRACE_ATTACH should
always succeed even if the tracer doesn't wait(2) for attach
completion; however, with this change, the tracee might still be
TASK_RUNNING trying to enter TASK_TRACED which would cause the
following request to fail with -ESRCH.
This intermediate state is hidden from the ptracer by setting
GROUP_STOP_TRAPPING on attach and making ptrace_check_attach() wait
for it to clear on its signal->wait_chldexit. Completing the
transition or getting killed clears TRAPPING and wakes up the tracer.
Note that the STOPPED -> RUNNING -> TRACED transition is still visible
to other threads which are in the same group as the ptracer and the
reverse transition is visible to all. Please read the comments for
details.
Oleg:
* Spotted a race condition where a task may retry group stop without
proper bookkeeping. Fixed by redoing bookkeeping on retry.
* Spotted that the transition is visible to userland in several
different ways. Most are fixed with GROUP_STOP_TRAPPING. Unhandled
corner case is documented.
* Pointed out not setting GROUP_STOP_SIGMASK on an already stopped
task would result in more consistent behavior.
* Pointed out that calling ptrace_stop() from do_signal_stop() in
TASK_STOPPED can race with group stop start logic and then confuse
the TRAPPING wait in ptrace_check_attach(). ptrace_stop() is now
called with TASK_RUNNING.
* Suggested using signal->wait_chldexit instead of bit wait.
* Spotted a race condition between TRACED transition and clearing of
TRAPPING.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Jan Kratochvil <jan.kratochvil@redhat.com>
A ptraced task would still stop at do_signal_stop() when it's stopping
for stop signals and do_signal_stop() behaves the same whether the
task is ptraced or not. However, in addition to stopping,
ptrace_stop() also does ptrace specific stuff like calling
architecture specific callbacks, so this behavior makes the code more
fragile and difficult to understand.
This patch makes do_signal_stop() test whether the task is ptraced and
use ptrace_stop() if so. This renders tracehook_notify_jctl() rather
pointless as the ptrace notification is now handled by ptrace_stop()
regardless of the return value from the tracehook. It probably is a
good idea to update it.
This doesn't solve the whole problem as tasks already in stopped state
would stay in the regular stop when ptrace attached. That part will
be handled by the next patch.
Oleg pointed out that this makes a userland-visible change. Before,
SIGCONT would be able to wake up a task in group stop even if the task
is ptraced if the tracer hasn't issued another ptrace command
afterwards (as the next ptrace commands transitions the state into
TASK_TRACED which ignores SIGCONT wakeups). With this and the next
patch, SIGCONT may race with the transition into TASK_TRACED and is
ignored if the tracee already entered TASK_TRACED.
Another userland visible change of this and the next patch is that the
ptracee's state would now be TASK_TRACED where it used to be
TASK_STOPPED, which is visible via fs/proc.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Jan Kratochvil <jan.kratochvil@redhat.com>
Currently, ptrace_stop() unconditionally participates in group stop
bookkeeping. This is unnecessary and inaccurate. Make it only
participate if the task is trapping for group stop - ie. if @why is
CLD_STOPPED. As ptrace_stop() currently is not used when trapping for
group stop, this equals to disabling group stop participation from
ptrace_stop().
A visible behavior change is increased likelihood of delayed group
stop completion if the thread group contains one or more ptraced
tasks.
This is to preapre for further cleanup of the interaction between
group stop and ptrace.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Currently task->signal->group_stop_count is used to decide whether to
stop for group stop. However, if there is a task in the group which
is taking a long time to stop, other tasks which are continued by
ptrace would repeatedly stop for the same group stop until the group
stop is complete.
Conversely, if a ptraced task is in TASK_TRACED state, the debugger
won't get notified of group stops which is inconsistent compared to
the ptraced task in any other state.
This patch introduces GROUP_STOP_PENDING which tracks whether a task
is yet to stop for the group stop in progress. The flag is set when a
group stop starts and cleared when the task stops the first time for
the group stop, and consulted whenever whether the task should
participate in a group stop needs to be determined. Note that now
tasks in TASK_TRACED also participate in group stop.
This results in the following behavior changes.
* For a single group stop, a ptracer would see at most one stop
reported.
* A ptracee in TASK_TRACED now also participates in group stop and the
tracer would get the notification. However, as a ptraced task could
be in TASK_STOPPED state or any ptrace trap could consume group
stop, the notification may still be missing. These will be
addressed with further patches.
* A ptracee may start a group stop while one is still in progress if
the tracer let it continue with stop signal delivery. Group stop
code handles this correctly.
Oleg:
* Spotted that a task might skip signal check even when its
GROUP_STOP_PENDING is set. Fixed by updating
recalc_sigpending_tsk() to check GROUP_STOP_PENDING instead of
group_stop_count.
* Pointed out that task->group_stop should be cleared whenever
task->signal->group_stop_count is cleared. Fixed accordingly.
* Pointed out the behavior inconsistency between TASK_TRACED and
RUNNING and the last behavior change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
task->signal->group_stop_count is used to track the progress of group
stop. It's initialized to the number of tasks which need to stop for
group stop to finish and each stopping or trapping task decrements.
However, each task doesn't keep track of whether it decremented the
counter or not and if woken up before the group stop is complete and
stops again, it can decrement the counter multiple times.
Please consider the following example code.
static void *worker(void *arg)
{
while (1) ;
return NULL;
}
int main(void)
{
pthread_t thread;
pid_t pid;
int i;
pid = fork();
if (!pid) {
for (i = 0; i < 5; i++)
pthread_create(&thread, NULL, worker, NULL);
while (1) ;
return 0;
}
ptrace(PTRACE_ATTACH, pid, NULL, NULL);
while (1) {
waitid(P_PID, pid, NULL, WSTOPPED);
ptrace(PTRACE_SINGLESTEP, pid, NULL, (void *)(long)SIGSTOP);
}
return 0;
}
The child creates five threads and the parent continuously traps the
first thread and whenever the child gets a signal, SIGSTOP is
delivered. If an external process sends SIGSTOP to the child, all
other threads in the process should reliably stop. However, due to
the above bug, the first thread will often end up consuming
group_stop_count multiple times and SIGSTOP often ends up stopping
none or part of the other four threads.
This patch adds a new field task->group_stop which is protected by
siglock and uses GROUP_STOP_CONSUME flag to track which task is still
to consume group_stop_count to fix this bug.
task_clear_group_stop_pending() and task_participate_group_stop() are
added to help manipulating group stop states. As ptrace_stop() now
also uses task_participate_group_stop(), it will set
SIGNAL_STOP_STOPPED if it completes a group stop.
There still are many issues regarding the interaction between group
stop and ptrace. Patches to address them will follow.
- Oleg spotted duplicate GROUP_STOP_CONSUME. Dropped.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
To prepare for cleanup of the interaction between group stop and
ptrace, add @why to ptrace_stop(). Existing users are updated such
that there is no behavior change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Roland McGrath <roland@redhat.com>
tracehook_notify_jctl() aids in determining whether and what to report
to the parent when a task is stopped or continued. The function also
adds an extra requirement that siglock may be released across it,
which is currently unused and quite difficult to satisfy in
well-defined manner.
As job control and the notifications are about to receive major
overhaul, remove the tracehook and open code it. If ever necessary,
let's factor it out after the overhaul.
* Oleg spotted incorrect CLD_CONTINUED/STOPPED selection when ptraced.
Fixed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
do_signal_stop() is used only by get_signal_to_deliver() and after a
successful signal stop, it always calls try_to_freeze(), so the
try_to_freeze() loop around schedule() in do_signal_stop() is
superflous and confusing. Remove it.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
After a task receives SIGCONT, its parent is notified via SIGCHLD with
its siginfo describing what the notified event is. If SIGCONT is
received while the child process is stopped, the code should be
CLD_CONTINUED. If SIGCONT is recieved while the child process is in
the process of being stopped, it should be CLD_STOPPED. Which code to
use is determined in prepare_signal() and recorded in signal->flags
using SIGNAL_CLD_CONTINUED|STOP flags.
get_signal_deliver() should test these flags and then notify
accoringly; however, it incorrectly tested SIGNAL_STOP_CONTINUED
instead of SIGNAL_CLD_CONTINUED, thus incorrectly notifying
CLD_CONTINUED if the signal is delivered before the task is wait(2)ed
and CLD_STOPPED if the state was fetched already.
Fix it by testing SIGNAL_CLD_CONTINUED. While at it, uncompress the
?: test into if/else clause for better readability.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Userland should be able to trust the pid and uid of the sender of a
signal if the si_code is SI_TKILL.
Unfortunately, the kernel has historically allowed sigqueueinfo() to
send any si_code at all (as long as it was negative - to distinguish it
from kernel-generated signals like SIGILL etc), so it could spoof a
SI_TKILL with incorrect siginfo values.
Happily, it looks like glibc has always set si_code to the appropriate
SI_QUEUE, so there are probably no actual user code that ever uses
anything but the appropriate SI_QUEUE flag.
So just tighten the check for si_code (we used to allow any negative
value), and add a (one-time) warning in case there are binaries out
there that might depend on using other si_code values.
Signed-off-by: Julien Tinnes <jln@google.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
lock_task_sighand() grabs sighand->siglock in case of returning non-NULL
but unlock_task_sighand() releases it unconditionally. This leads sparse
to complain about the lock context imbalance. Rename and wrap
lock_task_sighand() using __cond_lock() macro to make sparse happy.
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The original hwpoison code added a new siginfo field si_addr_lsb to
pass the granuality of the fault address to user space. Unfortunately
this field was never copied to user space. Fix this here.
I added explicit checks for the MCEERR codes to avoid having
to patch all potential callers to initialize the field.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Commit 8f92054e7c ("CRED: Fix __task_cred()'s lockdep check and banner
comment") fixed the lockdep checks on __task_cred(). This has shown up
a place in the signalling code where a lock should be held - namely that
check_kill_permission() requires its callers to hold the RCU lock.
Fix group_send_sig_info() to get the RCU read lock around its call to
check_kill_permission().
Without this patch, the following warning can occur:
===================================================
[ INFO: suspicious rcu_dereference_check() usage. ]
---------------------------------------------------
kernel/signal.c:660 invoked rcu_dereference_check() without protection!
...
Reported-by: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Change zap_other_threads() to return the number of other sub-threads found
on ->thread_group list.
Other changes are cosmetic:
- change the code to use while_each_thread() helper
- remove the obsolete comment about SIGKILL/SIGSTOP
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andrew Tridgell reports that aio_read(SIGEV_SIGNAL) can fail if the
notification from the helper thread races with setresuid(), see
http://samba.org/~tridge/junkcode/aio_uid.c
This happens because check_kill_permission() doesn't permit sending a
signal to the task with the different cred->xids. But there is not any
security reason to check ->cred's when the task sends a signal (private or
group-wide) to its sub-thread. Whatever we do, any thread can bypass all
security checks and send SIGKILL to all threads, or it can block a signal
SIG and do kill(gettid(), SIG) to deliver this signal to another
sub-thread. Not to mention that CLONE_THREAD implies CLONE_VM.
Change check_kill_permission() to avoid the credentials check when the
sender and the target are from the same thread group.
Also, move "cred = current_cred()" down to avoid calling get_current()
twice.
Note: David Howells pointed out we could relax this even more, the
CLONE_SIGHAND (without CLONE_THREAD) case probably does not need
these checks too.
Roland said:
: The glibc (libpthread) that does set*id across threads has
: been in use for a while (2.3.4?), probably in distro's using kernels as old
: or older than any active -stable streams. In the race in question, this
: kernel bug is breaking valid POSIX application expectations.
Reported-by: Andrew Tridgell <tridge@samba.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Acked-by: David Howells <dhowells@redhat.com>
Cc: Eric Paris <eparis@parisplace.org>
Cc: Jakub Jelinek <jakub@redhat.com>
Cc: James Morris <jmorris@namei.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Cc: <stable@kernel.org> [all kernel versions]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch contains the hooks and instrumentation into kernel which
live outside the kernel/debug directory, which the kdb core
will call to run commands like lsmod, dmesg, bt etc...
CC: linux-arch@vger.kernel.org
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Martin Hicks <mort@sgi.com>
Make sure compiler won't do weird things with limits. E.g. fetching them
twice may return 2 different values after writable limits are implemented.
I.e. either use rlimit helpers added in commit 3e10e716ab ("resource:
add helpers for fetching rlimits") or ACCESS_ONCE if not applicable.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This makes sure that we pick the synchronous signals caused by a
processor fault over any pending regular asynchronous signals sent to
use by [t]kill().
This is not strictly required semantics, but it makes it _much_ easier
for programs like Wine that expect to find the fault information in the
signal stack.
Without this, if a non-synchronous signal gets picked first, the delayed
asynchronous signal will have its signal context pointing to the new
signal invocation, rather than the instruction that caused the SIGSEGV
or SIGBUS in the first place.
This is not all that pretty, and we're discussing making the synchronous
signals more explicit rather than have these kinds of implicit
preferences of SIGSEGV and friends. See for example
http://bugzilla.kernel.org/show_bug.cgi?id=15395
for some of the discussion. But in the meantime this is a simple and
fairly straightforward work-around, and the whole
if (x & Y)
x &= Y;
thing can be compiled into (and gcc does do it) just three instructions:
movq %rdx, %rax
andl $Y, %eax
cmovne %rax, %rdx
so it is at least a simple solution to a subtle issue.
Reported-and-tested-by: Pavel Vilim <wylda@volny.cz>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When print-fatal-signals is enabled it's possible to dump any memory
reachable by the kernel to the log by simply jumping to that address from
user space.
Or crash the system if there's some hardware with read side effects.
The fatal signals handler will dump 16 bytes at the execution address,
which is fully controlled by ring 3.
In addition when something jumps to a unmapped address there will be up to
16 additional useless page faults, which might be potentially slow (and at
least is not very efficient)
Fortunately this option is off by default and only there on i386.
But fix it by checking for kernel addresses and also stopping when there's
a page fault.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
sys: Fix missing rcu protection for __task_cred() access
signals: Fix more rcu assumptions
signal: Fix racy access to __task_cred in kill_pid_info_as_uid()
Move the call to do_signal_stop() down, after tracehook call. This makes
->group_stop_count condition visible to tracers before do_signal_stop()
will participate in this group-stop.
Currently the patch has no effect, tracehook_get_signal() always returns 0.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Trivial, s/0/SI_USER/ in collect_signal() for grep.
This is a bit confusing, we don't know the source of this signal.
But we don't care, and "info->si_code = 0" is imho worse.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Change send_signal() to use si_fromuser(). From now SEND_SIG_NOINFO
triggers the "from_ancestor_ns" check.
This fixes reparent_thread()->group_send_sig_info(pdeath_signal)
behaviour, before this patch send_signal() does not detect the
cross-namespace case when the child of the dying parent belongs to the
sub-namespace.
This patch can affect the behaviour of send_sig(), kill_pgrp() and
kill_pid() when the caller sends the signal to the sub-namespace with
"priv == 0" but surprisingly all callers seem to use them correctly,
including disassociate_ctty(on_exit).
Except: drivers/staging/comedi/drivers/addi-data/*.c incorrectly use
send_sig(priv => 0). But his is minor and should be fixed anyway.
Reported-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Reviewed-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
No changes in compiled code. The patch adds the new helper, si_fromuser()
and changes check_kill_permission() to use this helper.
The real effect of this patch is that from now we "officially" consider
SEND_SIG_NOINFO signal as "from user-space" signals. This is already true
if we look at the code which uses SEND_SIG_NOINFO, except __send_signal()
has another opinion - see the next patch.
The naming of these special SEND_SIG_XXX siginfo's is really bad
imho. From __send_signal()'s pov they mean
SEND_SIG_NOINFO from user
SEND_SIG_PRIV from kernel
SEND_SIG_FORCED no info
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Reviewed-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1) Remove the misleading comment in __sigqueue_alloc() which claims
that holding a spinlock is equivalent to rcu_read_lock().
2) Add a rcu_read_lock/unlock around the __task_cred() access
in __sigqueue_alloc()
This needs to be revisited to remove the remaining users of
read_lock(&tasklist_lock) but that's outside the scope of this patch.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <20091210004703.269843657@linutronix.de>
kill_pid_info_as_uid() accesses __task_cred() without being in a RCU
read side critical section. tasklist_lock is not protecting that when
CONFIG_TREE_PREEMPT_RCU=y.
Convert the whole tasklist_lock section to rcu and use
lock_task_sighand to prevent the exit race.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <20091210004703.232302055@linutronix.de>
Acked-by: Oleg Nesterov <oleg@redhat.com>
When the system has too many timers or too many aggregate
queued signals, the EAGAIN error is returned to application
from kernel, including timer_create() [POSIX.1b].
It means that the app exceeded the limit of pending signals,
but in general application writers do not expect this
outcome and the current silent failure can cause rare app
failures under very high load.
This patch adds a new message when we reach the limit
and if print_fatal_signals is enabled:
task/1234: reached RLIMIT_SIGPENDING, dropping signal
If you see this message and your system behaved unexpectedly,
you can run following command to lift the limit:
# ulimit -i unlimited
With help from Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>.
Signed-off-by: Naohiro Ooiwa <nooiwa@miraclelinux.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: oleg@redhat.com
LKML-Reference: <4AF6E7E2.9080406@miraclelinux.com>
[ Modified a few small details, gave surrounding code some love. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
__fatal_signal_pending inlines to one instruction on x86, probably two
instructions on other machines. It takes two longer x86 instructions just
to call it and test its return value, not to mention the function itself.
On my random x86_64 config, this saved 70 bytes of text (59 of those being
__fatal_signal_pending itself).
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce do_send_sig_info() and convert group_send_sig_info(),
send_sig_info(), do_send_specific() to use this helper.
Hopefully it will have more users soon, it allows to specify
specific/group behaviour via "bool group" argument.
Shaves 80 bytes from .text.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This changes tracehook_notify_jctl() so it's called with the siglock held,
and changes its argument and return value definition. These clean-ups
make it a better fit for what new tracing hooks need to check.
Tracing needs the siglock here, held from the time TASK_STOPPED was set,
to avoid potential SIGCONT races if it wants to allow any blocking in its
tracing hooks.
This also folds the finish_stop() function into its caller
do_signal_stop(). The function is short, called only once and only
unconditionally. It aids readability to fold it in.
[oleg@redhat.com: do not call tracehook_notify_jctl() in TASK_STOPPED state]
[oleg@redhat.com: introduce tracehook_finish_jctl() helper]
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The bug is old, it wasn't cause by recent changes.
Test case:
static void *tfunc(void *arg)
{
int pid = (long)arg;
assert(ptrace(PTRACE_ATTACH, pid, NULL, NULL) == 0);
kill(pid, SIGKILL);
sleep(1);
return NULL;
}
int main(void)
{
pthread_t th;
long pid = fork();
if (!pid)
pause();
signal(SIGCHLD, SIG_IGN);
assert(pthread_create(&th, NULL, tfunc, (void*)pid) == 0);
int r = waitpid(-1, NULL, __WNOTHREAD);
printf("waitpid: %d %m\n", r);
return 0;
}
Before the patch this program hangs, after this patch waitpid() correctly
fails with errno == -ECHILD.
The problem is, __ptrace_detach() reaps the EXIT_ZOMBIE tracee if its
->real_parent is our sub-thread and we ignore SIGCHLD. But in this case
we should wake up other threads which can sleep in do_wait().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Vitaly Mayatskikh <vmayatsk@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The previous commit ("do_sigaltstack: avoid copying 'stack_t' as a
structure to user space") fixed a real bug. This one just cleans up the
copy from user space to that gcc can generate better code for it (and so
that it looks the same as the later copy back to user space).
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ulrich Drepper correctly points out that there is generally padding in
the structure on 64-bit hosts, and that copying the structure from
kernel to user space can leak information from the kernel stack in those
padding bytes.
Avoid the whole issue by just copying the three members one by one
instead, which also means that the function also can avoid the need for
a stack frame. This also happens to match how we copy the new structure
from user space, so it all even makes sense.
[ The obvious solution of adding a memset() generates horrid code, gcc
does really stupid things. ]
Reported-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If the non-traced sub-thread calls do_notify_parent_cldstop(), we send the
notification to group_leader->real_parent and we report group_leader's
pid.
But, if group_leader is traced we use the wrong ->parent->nsproxy->pid_ns,
the tracer and parent can live in different namespaces. Change the code
to use "parent" instead of tsk->parent.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Acked-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
No functional changes.
- Nobody except ptrace.c & co should use ptrace flags directly, we have
task_ptrace() for that.
- No need to specially check PT_PTRACED, we must not have other PT_ bits
set without PT_PTRACED. And no need to know this flag exists.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This false positive is due to field padding in struct sigqueue. When
this dynamically allocated structure is copied to the stack (in arch-
specific delivery code), kmemcheck sees a read from the padding, which
is, naturally, uninitialized.
Hide the false positive using the __GFP_NOTRACK_FALSE_POSITIVE flag.
Also made the rlimit override code a bit clearer by introducing a new
variable.
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
* 'tracing-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (244 commits)
Revert "x86, bts: reenable ptrace branch trace support"
tracing: do not translate event helper macros in print format
ftrace/documentation: fix typo in function grapher name
tracing/events: convert block trace points to TRACE_EVENT(), fix !CONFIG_BLOCK
tracing: add protection around module events unload
tracing: add trace_seq_vprint interface
tracing: fix the block trace points print size
tracing/events: convert block trace points to TRACE_EVENT()
ring-buffer: fix ret in rb_add_time_stamp
ring-buffer: pass in lockdep class key for reader_lock
tracing: add annotation to what type of stack trace is recorded
tracing: fix multiple use of __print_flags and __print_symbolic
tracing/events: fix output format of user stack
tracing/events: fix output format of kernel stack
tracing/trace_stack: fix the number of entries in the header
ring-buffer: discard timestamps that are at the start of the buffer
ring-buffer: try to discard unneeded timestamps
ring-buffer: fix bug in ring_buffer_discard_commit
ftrace: do not profile functions when disabled
tracing: make trace pipe recognize latency format flag
...
sys_kill has the per thread counterpart sys_tgkill. sigqueueinfo is
missing a thread directed counterpart. Such an interface is important
for migrating applications from other OSes which have the per thread
delivery implemented.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Acked-by: Ulrich Drepper <drepper@redhat.com>
Split out the code from do_tkill to make it reusable by the follow up
patch which implements sys_rt_tgsigqueueinfo
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Don't flush inherited SIGKILL during execve() in SELinux's post cred commit
hook. This isn't really a security problem: if the SIGKILL came before the
credentials were changed, then we were right to receive it at the time, and
should honour it; if it came after the creds were changed, then we definitely
should honour it; and in any case, all that will happen is that the process
will be scrapped before it ever returns to userspace.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
Impact: clean up
Create a sub directory in include/trace called events to keep the
trace point headers in their own separate directory. Only headers that
declare trace points should be defined in this directory.
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Zhao Lei <zhaolei@cn.fujitsu.com>
Cc: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This patch lowers the number of places a developer must modify to add
new tracepoints. The current method to add a new tracepoint
into an existing system is to write the trace point macro in the
trace header with one of the macros TRACE_EVENT, TRACE_FORMAT or
DECLARE_TRACE, then they must add the same named item into the C file
with the macro DEFINE_TRACE(name) and then add the trace point.
This change cuts out the needing to add the DEFINE_TRACE(name).
Every file that uses the tracepoint must still include the trace/<type>.h
file, but the one C file must also add a define before the including
of that file.
#define CREATE_TRACE_POINTS
#include <trace/mytrace.h>
This will cause the trace/mytrace.h file to also produce the C code
necessary to implement the trace point.
Note, if more than one trace/<type>.h is used to create the C code
it is best to list them all together.
#define CREATE_TRACE_POINTS
#include <trace/foo.h>
#include <trace/bar.h>
#include <trace/fido.h>
Thanks to Mathieu Desnoyers and Christoph Hellwig for coming up with
the cleaner solution of the define above the includes over my first
design to have the C code include a "special" header.
This patch converts sched, irq and lockdep and skb to use this new
method.
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Zhao Lei <zhaolei@cn.fujitsu.com>
Cc: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When sending a signal to a descendant namespace, set ->si_pid to 0 since
the sender does not have a pid in the receiver's namespace.
Note:
- If rt_sigqueueinfo() sets si_code to SI_USER when sending a
signal across a pid namespace boundary, the value in ->si_pid
will be cleared to 0.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Normally SIG_DFL signals to global and container-init are dropped early.
But if a signal is blocked when it is posted, we cannot drop the signal
since the receiver may install a handler before unblocking the signal.
Once this signal is queued however, the receiver container-init has no way
of knowing if the signal was sent from an ancestor or descendant
namespace. This patch ensures that contianer-init drops all SIG_DFL
signals in get_signal_to_deliver() except SIGKILL/SIGSTOP.
If SIGSTOP/SIGKILL originate from a descendant of container-init they are
never queued (i.e dropped in sig_ignored() in an earler patch).
If SIGSTOP/SIGKILL originate from parent namespace, the signal is queued
and container-init processes the signal.
IOW, if get_signal_to_deliver() sees a sig_kernel_only() signal for global
or container-init, the signal must have been generated internally or must
have come from an ancestor ns and we process the signal.
Further, the signal_group_exit() check was needed to cover the case of a
multi-threaded init sending SIGKILL to other threads when doing an exit()
or exec(). But since the new sig_kernel_only() check covers the SIGKILL,
the signal_group_exit() check is no longer needed and can be removed.
Finally, now that we have all pieces in place, set SIGNAL_UNKILLABLE for
container-inits.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Drop early any SIG_DFL or SIG_IGN signals to container-init from within
the same container. But queue SIGSTOP and SIGKILL to the container-init
if they are from an ancestor container.
Blocked, fatal signals (i.e when SIG_DFL is to terminate) from within the
container can still terminate the container-init. That will be addressed
in the next patch.
Note: To be bisect-safe, SIGNAL_UNKILLABLE will be set for container-inits
in a follow-on patch. Until then, this patch is just a preparatory
step.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
send_signal() (or its helper) needs to determine the pid namespace of the
sender. But a signal sent via kill_pid_info_as_uid() comes from within
the kernel and send_signal() does not need to determine the pid namespace
of the sender. So define a helper for send_signal() which takes an
additional parameter, 'from_ancestor_ns' and have kill_pid_info_as_uid()
use that helper directly.
The 'from_ancestor_ns' parameter will be used in a follow-on patch.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(This is a modified version of the patch submitted by Oleg Nesterov
http://lkml.org/lkml/2008/11/18/249 and tries to address comments that
came up in that discussion)
init ignores the SIG_DFL signals but we queue them anyway, including
SIGKILL. This is mostly OK, the signal will be dropped silently when
dequeued, but the pending SIGKILL has 2 bad implications:
- it implies fatal_signal_pending(), so we confuse things
like wait_for_completion_killable/lock_page_killable.
- for the sub-namespace inits, the pending SIGKILL can
mask (legacy_queue) the subsequent SIGKILL from the
parent namespace which must kill cinit reliably.
(preparation, cinits don't have SIGNAL_UNKILLABLE yet)
The patch can't help when init is ptraced, but ptracing of init is not
"safe" anyway.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Container-init must behave like global-init to processes within the
container and hence it must be immune to unhandled fatal signals from
within the container (i.e SIG_DFL signals that terminate the process).
But the same container-init must behave like a normal process to processes
in ancestor namespaces and so if it receives the same fatal signal from a
process in ancestor namespace, the signal must be processed.
Implementing these semantics requires that send_signal() determine pid
namespace of the sender but since signals can originate from workqueues/
interrupt-handlers, determining pid namespace of sender may not always be
possible or safe.
This patchset implements the design/simplified semantics suggested by
Oleg Nesterov. The simplified semantics for container-init are:
- container-init must never be terminated by a signal from a
descendant process.
- container-init must never be immune to SIGKILL from an ancestor
namespace (so a process in parent namespace must always be able
to terminate a descendant container).
- container-init may be immune to unhandled fatal signals (like
SIGUSR1) even if they are from ancestor namespace. SIGKILL/SIGSTOP
are the only reliable signals to a container-init from ancestor
namespace.
This patch:
Based on an earlier patch submitted by Oleg Nesterov and comments from
Roland McGrath (http://lkml.org/lkml/2008/11/19/258).
The handler parameter is currently unused in the tracehook functions.
Besides, the tracehook functions are called with siglock held, so the
functions can check the handler if they later need to.
Removing the parameter simiplifies changes to sig_ignored() in a follow-on
patch.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch fixes bug #12208:
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=12208
Subject : uml is very slow on 2.6.28 host
This turned out to be not a scheduler regression, but an already
existing problem in ptrace being triggered by subtle scheduler
changes.
The problem is this:
- task A is ptracing task B
- task B stops on a trace event
- task A is woken up and preempts task B
- task A calls ptrace on task B, which does ptrace_check_attach()
- this calls wait_task_inactive(), which sees that task B is still on the runq
- task A goes to sleep for a jiffy
- ...
Since UML does lots of the above sequences, those jiffies quickly add
up to make it slow as hell.
This patch solves this by not rescheduling in read_unlock() after
ptrace_stop() has woken up the tracer.
Thanks to Oleg Nesterov and Ingo Molnar for the feedback.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
CC: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We're going to split the process wide cpu accounting into two parts:
- clocks; which can take all the time they want since they run
from user context.
- timers; which need constant time tracing but can affort the overhead
because they're default off -- and rare.
The clock readout will go back to a full sum of the thread group, for this
we need to re-add the exit stats that were removed in the initial itimer
rework (f06febc9: timers: fix itimer/many thread hang).
Furthermore, since that full sum can be rather slow for large thread groups
and we have the complete dead task stats, revert the do_notify_parent time
computation.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With print-fatal-signals=1 on a kernel with CONFIG_PREEMPT=y, sending an
unexpected signal to a process causes a BUG: using smp_processor_id() in
preemptible code.
get_signal_to_deliver() releases the siglock before calling
print_fatal_signal(), which calls show_regs(), which calls
smp_processor_id(), which is not supposed to be called from a
preemptible thread.
Make sure show_regs() runs with preemption disabled.
Signed-off-by: Ed Swierk <eswierk@aristanetworks.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Convert all system calls to return a long. This should be a NOP since all
converted types should have the same size anyway.
With the exception of sys_exit_group which returned void. But that doesn't
matter since the system call doesn't return.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
POSIX requires the si_pid to be the process id of the sender, so ->si_pid
should really be set to 'tgid'. This change does have following changes
in behavior:
- When sending pdeath_signal on re-parent to a sub-thread, ->si_pid
cannot be used to identify the thread that did the re-parent since
it will now show the tgid instead of thread id.
- A multi-threaded application that expects to find the specific
thread that encountered a SIGPIPE using the ->si_pid will now
break.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Acked-By: Roland McGrath <roland@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
For SEND_SIG_NOINFO, si_pid is currently set to the pid of sender
in sender's active pid namespace. But if the receiver is in a
Eg: when parent sends the 'pdeath_signal' to a child that is in
a descendant pid namespace, we should set si_pid 0.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Acked-By: Roland McGrath <roland@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'tracing-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (241 commits)
sched, trace: update trace_sched_wakeup()
tracing/ftrace: don't trace on early stage of a secondary cpu boot, v3
Revert "x86: disable X86_PTRACE_BTS"
ring-buffer: prevent false positive warning
ring-buffer: fix dangling commit race
ftrace: enable format arguments checking
x86, bts: memory accounting
x86, bts: add fork and exit handling
ftrace: introduce tracing_reset_online_cpus() helper
tracing: fix warnings in kernel/trace/trace_sched_switch.c
tracing: fix warning in kernel/trace/trace.c
tracing/ring-buffer: remove unused ring_buffer size
trace: fix task state printout
ftrace: add not to regex on filtering functions
trace: better use of stack_trace_enabled for boot up code
trace: add a way to enable or disable the stack tracer
x86: entry_64 - introduce FTRACE_ frame macro v2
tracing/ftrace: add the printk-msg-only option
tracing/ftrace: use preempt_enable_no_resched_notrace in ring_buffer_time_stamp()
x86, bts: correctly report invalid bts records
...
Fixed up trivial conflict in scripts/recordmcount.pl due to SH bits
being already partly merged by the SH merge.
Impact: API *CHANGE*. Must update all tracepoint users.
Add DEFINE_TRACE() to tracepoints to let them declare the tracepoint
structure in a single spot for all the kernel. It helps reducing memory
consumption, especially when declaring a lot of tracepoints, e.g. for
kmalloc tracing.
*API CHANGE WARNING*: now, DECLARE_TRACE() must be used in headers for
tracepoint declarations rather than DEFINE_TRACE(). This is the sane way
to do it. The name previously used was misleading.
Updates scheduler instrumentation to follow this API change.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Inaugurate copy-on-write credentials management. This uses RCU to manage the
credentials pointer in the task_struct with respect to accesses by other tasks.
A process may only modify its own credentials, and so does not need locking to
access or modify its own credentials.
A mutex (cred_replace_mutex) is added to the task_struct to control the effect
of PTRACE_ATTACHED on credential calculations, particularly with respect to
execve().
With this patch, the contents of an active credentials struct may not be
changed directly; rather a new set of credentials must be prepared, modified
and committed using something like the following sequence of events:
struct cred *new = prepare_creds();
int ret = blah(new);
if (ret < 0) {
abort_creds(new);
return ret;
}
return commit_creds(new);
There are some exceptions to this rule: the keyrings pointed to by the active
credentials may be instantiated - keyrings violate the COW rule as managing
COW keyrings is tricky, given that it is possible for a task to directly alter
the keys in a keyring in use by another task.
To help enforce this, various pointers to sets of credentials, such as those in
the task_struct, are declared const. The purpose of this is compile-time
discouragement of altering credentials through those pointers. Once a set of
credentials has been made public through one of these pointers, it may not be
modified, except under special circumstances:
(1) Its reference count may incremented and decremented.
(2) The keyrings to which it points may be modified, but not replaced.
The only safe way to modify anything else is to create a replacement and commit
using the functions described in Documentation/credentials.txt (which will be
added by a later patch).
This patch and the preceding patches have been tested with the LTP SELinux
testsuite.
This patch makes several logical sets of alteration:
(1) execve().
This now prepares and commits credentials in various places in the
security code rather than altering the current creds directly.
(2) Temporary credential overrides.
do_coredump() and sys_faccessat() now prepare their own credentials and
temporarily override the ones currently on the acting thread, whilst
preventing interference from other threads by holding cred_replace_mutex
on the thread being dumped.
This will be replaced in a future patch by something that hands down the
credentials directly to the functions being called, rather than altering
the task's objective credentials.
(3) LSM interface.
A number of functions have been changed, added or removed:
(*) security_capset_check(), ->capset_check()
(*) security_capset_set(), ->capset_set()
Removed in favour of security_capset().
(*) security_capset(), ->capset()
New. This is passed a pointer to the new creds, a pointer to the old
creds and the proposed capability sets. It should fill in the new
creds or return an error. All pointers, barring the pointer to the
new creds, are now const.
(*) security_bprm_apply_creds(), ->bprm_apply_creds()
Changed; now returns a value, which will cause the process to be
killed if it's an error.
(*) security_task_alloc(), ->task_alloc_security()
Removed in favour of security_prepare_creds().
(*) security_cred_free(), ->cred_free()
New. Free security data attached to cred->security.
(*) security_prepare_creds(), ->cred_prepare()
New. Duplicate any security data attached to cred->security.
(*) security_commit_creds(), ->cred_commit()
New. Apply any security effects for the upcoming installation of new
security by commit_creds().
(*) security_task_post_setuid(), ->task_post_setuid()
Removed in favour of security_task_fix_setuid().
(*) security_task_fix_setuid(), ->task_fix_setuid()
Fix up the proposed new credentials for setuid(). This is used by
cap_set_fix_setuid() to implicitly adjust capabilities in line with
setuid() changes. Changes are made to the new credentials, rather
than the task itself as in security_task_post_setuid().
(*) security_task_reparent_to_init(), ->task_reparent_to_init()
Removed. Instead the task being reparented to init is referred
directly to init's credentials.
NOTE! This results in the loss of some state: SELinux's osid no
longer records the sid of the thread that forked it.
(*) security_key_alloc(), ->key_alloc()
(*) security_key_permission(), ->key_permission()
Changed. These now take cred pointers rather than task pointers to
refer to the security context.
(4) sys_capset().
This has been simplified and uses less locking. The LSM functions it
calls have been merged.
(5) reparent_to_kthreadd().
This gives the current thread the same credentials as init by simply using
commit_thread() to point that way.
(6) __sigqueue_alloc() and switch_uid()
__sigqueue_alloc() can't stop the target task from changing its creds
beneath it, so this function gets a reference to the currently applicable
user_struct which it then passes into the sigqueue struct it returns if
successful.
switch_uid() is now called from commit_creds(), and possibly should be
folded into that. commit_creds() should take care of protecting
__sigqueue_alloc().
(7) [sg]et[ug]id() and co and [sg]et_current_groups.
The set functions now all use prepare_creds(), commit_creds() and
abort_creds() to build and check a new set of credentials before applying
it.
security_task_set[ug]id() is called inside the prepared section. This
guarantees that nothing else will affect the creds until we've finished.
The calling of set_dumpable() has been moved into commit_creds().
Much of the functionality of set_user() has been moved into
commit_creds().
The get functions all simply access the data directly.
(8) security_task_prctl() and cap_task_prctl().
security_task_prctl() has been modified to return -ENOSYS if it doesn't
want to handle a function, or otherwise return the return value directly
rather than through an argument.
Additionally, cap_task_prctl() now prepares a new set of credentials, even
if it doesn't end up using it.
(9) Keyrings.
A number of changes have been made to the keyrings code:
(a) switch_uid_keyring(), copy_keys(), exit_keys() and suid_keys() have
all been dropped and built in to the credentials functions directly.
They may want separating out again later.
(b) key_alloc() and search_process_keyrings() now take a cred pointer
rather than a task pointer to specify the security context.
(c) copy_creds() gives a new thread within the same thread group a new
thread keyring if its parent had one, otherwise it discards the thread
keyring.
(d) The authorisation key now points directly to the credentials to extend
the search into rather pointing to the task that carries them.
(e) Installing thread, process or session keyrings causes a new set of
credentials to be created, even though it's not strictly necessary for
process or session keyrings (they're shared).
(10) Usermode helper.
The usermode helper code now carries a cred struct pointer in its
subprocess_info struct instead of a new session keyring pointer. This set
of credentials is derived from init_cred and installed on the new process
after it has been cloned.
call_usermodehelper_setup() allocates the new credentials and
call_usermodehelper_freeinfo() discards them if they haven't been used. A
special cred function (prepare_usermodeinfo_creds()) is provided
specifically for call_usermodehelper_setup() to call.
call_usermodehelper_setkeys() adjusts the credentials to sport the
supplied keyring as the new session keyring.
(11) SELinux.
SELinux has a number of changes, in addition to those to support the LSM
interface changes mentioned above:
(a) selinux_setprocattr() no longer does its check for whether the
current ptracer can access processes with the new SID inside the lock
that covers getting the ptracer's SID. Whilst this lock ensures that
the check is done with the ptracer pinned, the result is only valid
until the lock is released, so there's no point doing it inside the
lock.
(12) is_single_threaded().
This function has been extracted from selinux_setprocattr() and put into
a file of its own in the lib/ directory as join_session_keyring() now
wants to use it too.
The code in SELinux just checked to see whether a task shared mm_structs
with other tasks (CLONE_VM), but that isn't good enough. We really want
to know if they're part of the same thread group (CLONE_THREAD).
(13) nfsd.
The NFS server daemon now has to use the COW credentials to set the
credentials it is going to use. It really needs to pass the credentials
down to the functions it calls, but it can't do that until other patches
in this series have been applied.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: James Morris <jmorris@namei.org>
Use RCU to access another task's creds and to release a task's own creds.
This means that it will be possible for the credentials of a task to be
replaced without another task (a) requiring a full lock to read them, and (b)
seeing deallocated memory.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
Separate the task security context from task_struct. At this point, the
security data is temporarily embedded in the task_struct with two pointers
pointing to it.
Note that the Alpha arch is altered as it refers to (E)UID and (E)GID in
entry.S via asm-offsets.
With comment fixes Signed-off-by: Marc Dionne <marc.c.dionne@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
Wrap access to task credentials so that they can be separated more easily from
the task_struct during the introduction of COW creds.
Change most current->(|e|s|fs)[ug]id to current_(|e|s|fs)[ug]id().
Change some task->e?[ug]id to task_e?[ug]id(). In some places it makes more
sense to use RCU directly rather than a convenient wrapper; these will be
addressed by later patches.
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: James Morris <jmorris@namei.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-audit@redhat.com
Cc: containers@lists.linux-foundation.org
Cc: linux-mm@kvack.org
Signed-off-by: James Morris <jmorris@namei.org>
Currently "kill <sig> -1" kills processes in all namespaces and breaks the
isolation of namespaces. Earlier attempt to fix this was discussed at:
http://lkml.org/lkml/2008/7/23/148
As suggested by Oleg Nesterov in that thread, use "task_pid_vnr() > 1"
check since task_pid_vnr() returns 0 if process is outside the caller's
namespace.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Acked-by: Eric W. Biederman <ebiederm@xmission.com>
Tested-by: Daniel Hokka Zakrisson <daniel@hozac.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Instrument the scheduler activity (sched_switch, migration, wakeups,
wait for a task, signal delivery) and process/thread
creation/destruction (fork, exit, kthread stop). Actually, kthread
creation is not instrumented in this patch because it is architecture
dependent. It allows to connect tracers such as ftrace which detects
scheduling latencies, good/bad scheduler decisions. Tools like LTTng can
export this scheduler information along with instrumentation of the rest
of the kernel activity to perform post-mortem analysis on the scheduler
activity.
About the performance impact of tracepoints (which is comparable to
markers), even without immediate values optimizations, tests done by
Hideo Aoki on ia64 show no regression. His test case was using hackbench
on a kernel where scheduler instrumentation (about 5 events in code
scheduler code) was added. See the "Tracepoints" patch header for
performance result detail.
Changelog :
- Change instrumentation location and parameter to match ftrace
instrumentation, previously done with kernel markers.
[ mingo@elte.hu: conflict resolutions ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: 'Peter Zijlstra' <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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>
I outwitted myself again in commit 2b2a1ff64a,
and broke the SA_NOCLDWAIT behavior so it leaks zombies. This fixes it.
Reported-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Roland McGrath <roland@redhat.com>
* 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
posix-timers: fix posix_timer_event() vs dequeue_signal() race
posix-timers: do_schedule_next_timer: fix the setting of ->si_overrun
This defines a new hook tracehook_force_sigpending() that lets tracing
code decide to force TIF_SIGPENDING on in recalc_sigpending().
This is not used yet, so it compiles away to nothing for now. It lays the
groundwork for new tracing code that can interrupt a task synthetically
without actually sending a signal.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This moves the ptrace logic in task death (exit_notify) into tracehook.h
inlines. Some code is rearranged slightly to make things nicer. There is
no change, only cleanup.
There is one hook called with the tasklist_lock write-locked, as ptrace
needs. There is also a new hook called after exit_state changes and
without locks. This is a better place for tracing work to be in the
future, since it doesn't delay the whole system with locking.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This defines the tracehook_notify_jctl() hook to formalize the ptrace
effects on the job control notifications. There is no change, only
cleanup.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This defines the tracehook_get_signal() hook to allow tracing code to slip
in before normal signal dequeuing. This lays the groundwork for new
tracing features that can inject synthetic signals outside the normal
queue or control the disposition of delivered signals. The calling
convention lets tracehook_get_signal() decide both exactly what will
happen and what signal number to report in the handler/exit.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This defines tracehook_consider_fatal_signal() has a fine-grained hook for
deciding to skip the special cases for a fatal signal, as ptrace does.
There is no change, only cleanup.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This defines tracehook_consider_ignored_signal() has a fine-grained hook
for deciding to prevent the normal short-circuit of sending an ignored
signal, as ptrace does. There is no change, only cleanup.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The ptrace_notify() function should not be called by any modules. It was
only ever exported to be called by binfmt exec functions. But that is no
longer necessary since fs/exec.c deals with that generically now. There
should be no calls to ptrace_notify() from outside the core kernel.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This function operated on a pid_t to kill a task, which is no longer valid
in a containerized system.
It has finally lost all its users and we can safely remove it from the
tree.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move mm->core_waiters into "struct core_state" allocated on stack. This
shrinks mm_struct a little bit and allows further changes.
This patch mostly does s/core_waiters/core_state. The only essential
change is that coredump_wait() must clear mm->core_state before return.
The coredump_wait()'s path is uglified and .text grows by 30 bytes, this
is fixed by the next patch.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
