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>
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>
1. SIGKILL can't be blocked, remove this check from sigkill_pending().
2. When ptrace_stop() sees sigkill_pending() == T, it can just return.
Kill "int killed" and simplify the code. This also is more correct,
the tracer shouldn't see us in TASK_TRACED if we are not going to
stop.
I strongly believe this code needs further changes. We should do the "was
this task killed" check unconditionally, currently it depends on
arch_ptrace_stop_needed(). On the other hand, sigkill_pending() isn't
very clever. If the task was killed tkill(SIGKILL), the signal can be
already dequeued if the caller is do_exit().
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>
Change the type of pid and tgid variables from int to the POSIX type
pid_t.
Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br>
Cc: 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>
In the switch to configurable HZ in 2.6, the treatment of the si_utime and
si_stime fields that are exposed to userland via the siginfo structure
looks to have been botched. As things stand, these fields report times in
units of HZ, so that userland gets information that varies depending on
the HZ that the kernel was configured with. This patch changes the
reported values to use USER_HZ units.
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Acked-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Thomas Gleixner <tglx@linutronix.de>
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>
fae5fa44f1 changed do_signal_stop() to check
SIGNAL_UNKILLABLE, this wasn't needed. If signal_group_exit() == F, the
signal sent to SIGNAL_UNKILLABLE task must be already filtered out by the
caller, get_signal_to_deliver(). And if signal_group_exit() == T we are
not going to stop.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
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>
dequeue_signal() checks SIGNAL_GROUP_EXIT before setting
SIGNAL_STOP_DEQUEUED. This was added by
788e05a67c a long ago to avoid the
coredump/SIGSTOP race.
Since then the related code was changed, and now this subtle check is both
incomplete and unneeded at the same time. It is incomplete because
nowadays exec() doesn't set SIGNAL_GROUP_EXIT, so in fact we should check
signal_group_exit() to avoid a similar race. Fortunately, we doesn't need
the check at all. The only function which relies on SIGNAL_STOP_DEQUEUED
is do_signal_stop(), and it ignores this flag if signal_group_exit() == T,
this covers the SIGNAL_GROUP_EXIT case.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
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>
With the recent changes collect_signal() always returns true. Change it
to return void and update the single caller.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Factor out sigdelset() calls and remove the "still_pending" variable.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
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>
collect_signal() checks sigismember(&list->signal, sig), this is not
needed. This "sig" was just found by next_signal(), so it must be valid.
We have a (completely broken) call to ->notifier in between, but it must
not play with sigpending->signal bits or unlock ->siglock.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
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>
The bug was reported and analysed by Mark McLoughlin <markmc@redhat.com>,
the patch is based on his and Roland's suggestions.
posix_timer_event() always rewrites the pre-allocated siginfo before sending
the signal. Most of the written info is the same all the time, but memset(0)
is very wrong. If ->sigq is queued we can race with collect_signal() which
can fail to find this siginfo looking at .si_signo, or copy_siginfo() can
copy the wrong .si_code/si_tid/etc.
In short, sys_timer_settime() can in fact stop the active timer, or the user
can receive the siginfo with the wrong .si_xxx values.
Move "memset(->info, 0)" from posix_timer_event() to alloc_posix_timer(),
change send_sigqueue() to set .si_overrun = 0 when ->sigq is not queued.
It would be nice to move the whole sigq->info initialization from send to
create path, but this is not easy to do without uglifying timer_create()
further.
As Roland rightly pointed out, we need more cleanups/fixes here, see the
"FIXME" comment in the patch. Hopefully this patch makes sense anyway, and
it can mask the most bad implications.
Reported-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Oliver Pinter <oliver.pntr@gmail.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: stable@kernel.org
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/posix-timers.c | 17 +++++++++++++----
kernel/signal.c | 1 +
2 files changed, 14 insertions(+), 4 deletions(-)
