user/sven/linux.git/kernel, branch v3.4.41

sched_clock: Prevent 64bit inatomicity on 32bit systems

2013-04-17T04:27:27Z

commit a1cbcaa9ea87b87a96b9fc465951dcf36e459ca2 upstream. The sched_clock_remote() implementation has the following inatomicity problem on 32bit systems when accessing the remote scd->clock, which is a 64bit value. CPU0 CPU1 sched_clock_local() sched_clock_remote(CPU0) ... remote_clock = scd[CPU0]->clock read_low32bit(scd[CPU0]->clock) cmpxchg64(scd->clock,...) read_high32bit(scd[CPU0]->clock) While the update of scd->clock is using an atomic64 mechanism, the readout on the remote cpu is not, which can cause completely bogus readouts. It is a quite rare problem, because it requires the update to hit the narrow race window between the low/high readout and the update must go across the 32bit boundary. The resulting misbehaviour is, that CPU1 will see the sched_clock on CPU1 ~4 seconds ahead of it's own and update CPU1s sched_clock value to this bogus timestamp. This stays that way due to the clamping implementation for about 4 seconds until the synchronization with CLOCK_MONOTONIC undoes the problem. The issue is hard to observe, because it might only result in a less accurate SCHED_OTHER timeslicing behaviour. To create observable damage on realtime scheduling classes, it is necessary that the bogus update of CPU1 sched_clock happens in the context of an realtime thread, which then gets charged 4 seconds of RT runtime, which results in the RT throttler mechanism to trigger and prevent scheduling of RT tasks for a little less than 4 seconds. So this is quite unlikely as well. The issue was quite hard to decode as the reproduction time is between 2 days and 3 weeks and intrusive tracing makes it less likely, but the following trace recorded with trace_clock=global, which uses sched_clock_local(), gave the final hint: -0 0d..30 400269.477150: hrtimer_cancel: hrtimer=0xf7061e80 -0 0d..30 400269.477151: hrtimer_start: hrtimer=0xf7061e80 ... irq/20-S-587 1d..32 400273.772118: sched_wakeup: comm= ... target_cpu=0 -0 0dN.30 400273.772118: hrtimer_cancel: hrtimer=0xf7061e80 What happens is that CPU0 goes idle and invokes sched_clock_idle_sleep_event() which invokes sched_clock_local() and CPU1 runs a remote wakeup for CPU0 at the same time, which invokes sched_remote_clock(). The time jump gets propagated to CPU0 via sched_remote_clock() and stays stale on both cores for ~4 seconds. There are only two other possibilities, which could cause a stale sched clock: 1) ktime_get() which reads out CLOCK_MONOTONIC returns a sporadic wrong value. 2) sched_clock() which reads the TSC returns a sporadic wrong value. #1 can be excluded because sched_clock would continue to increase for one jiffy and then go stale. #2 can be excluded because it would not make the clock jump forward. It would just result in a stale sched_clock for one jiffy. After quite some brain twisting and finding the same pattern on other traces, sched_clock_remote() remained the only place which could cause such a problem and as explained above it's indeed racy on 32bit systems. So while on 64bit systems the readout is atomic, we need to verify the remote readout on 32bit machines. We need to protect the local->clock readout in sched_clock_remote() on 32bit as well because an NMI could hit between the low and the high readout, call sched_clock_local() and modify local->clock. Thanks to Siegfried Wulsch for bearing with my debug requests and going through the tedious tasks of running a bunch of reproducer systems to generate the debug information which let me decode the issue. Reported-by: Siegfried Wulsch Acked-by: Peter Zijlstra Cc: Steven Rostedt Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1304051544160.21884@ionos Signed-off-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman

PM / reboot: call syscore_shutdown() after disable_nonboot_cpus()

2013-04-17T04:27:26Z

commit 6f389a8f1dd22a24f3d9afc2812b30d639e94625 upstream. As commit 40dc166c (PM / Core: Introduce struct syscore_ops for core subsystems PM) say, syscore_ops operations should be carried with one CPU on-line and interrupts disabled. However, after commit f96972f2d (kernel/sys.c: call disable_nonboot_cpus() in kernel_restart()), syscore_shutdown() is called before disable_nonboot_cpus(), so break the rules. We have a MIPS machine with a 8259A PIC, and there is an external timer (HPET) linked at 8259A. Since 8259A has been shutdown too early (by syscore_shutdown()), disable_nonboot_cpus() runs without timer interrupt, so it hangs and reboot fails. This patch call syscore_shutdown() a little later (after disable_nonboot_cpus()) to avoid reboot failure, this is the same way as poweroff does. For consistency, add disable_nonboot_cpus() to kernel_halt(). Signed-off-by: Huacai Chen Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman

tracing: Fix double free when function profile init failed

2013-04-17T04:27:26Z

commit 83e03b3fe4daffdebbb42151d5410d730ae50bd1 upstream. On the failure path, stat->start and stat->pages will refer same page. So it'll attempt to free the same page again and get kernel panic. Link: http://lkml.kernel.org/r/1364820385-32027-1-git-send-email-namhyung@kernel.org Signed-off-by: Namhyung Kim Cc: Frederic Weisbecker Cc: Namhyung Kim Signed-off-by: Steven Rostedt Signed-off-by: Greg Kroah-Hartman

panic: fix a possible deadlock in panic()

2013-04-12T16:38:47Z

commit 190320c3b6640d4104650f55ff69611e050ea06b upstream. panic_lock is meant to ensure that panic processing takes place only on one cpu; if any of the other cpus encounter a panic, they will spin waiting to be shut down. However, this causes a regression in this scenario: 1. Cpu 0 encounters a panic and acquires the panic_lock and proceeds with the panic processing. 2. There is an interrupt on cpu 0 that also encounters an error condition and invokes panic. 3. This second invocation fails to acquire the panic_lock and enters the infinite while loop in panic_smp_self_stop. Thus all panic processing is stopped, and the cpu is stuck for eternity in the while(1) inside panic_smp_self_stop. To address this, disable local interrupts with local_irq_disable before acquiring the panic_lock. This will prevent interrupt handlers from executing during the panic processing, thus avoiding this particular problem. Signed-off-by: Vikram Mulukutla Reviewed-by: Stephen Boyd Cc: Michael Holzheu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Cc: Paul Gortmaker Signed-off-by: Greg Kroah-Hartman

ftrace: Consistently restore trace function on sysctl enabling

2013-04-12T16:38:45Z

commit 5000c418840b309251c5887f0b56503aae30f84c upstream. If we reenable ftrace via syctl, we currently set ftrace_trace_function based on the previous simplistic algorithm. This is inconsistent with what update_ftrace_function does. So better call that helper instead. Link: http://lkml.kernel.org/r/5151D26F.1070702@siemens.com Signed-off-by: Jan Kiszka Signed-off-by: Steven Rostedt Signed-off-by: Greg Kroah-Hartman

tracing: Prevent buffer overwrite disabled for latency tracers

2013-04-05T17:04:37Z

commit 613f04a0f51e6e68ac6fe571ab79da3c0a5eb4da upstream. The latency tracers require the buffers to be in overwrite mode, otherwise they get screwed up. Force the buffers to stay in overwrite mode when latency tracers are enabled. Added a flag_changed() method to the tracer structure to allow the tracers to see what flags are being changed, and also be able to prevent the change from happing. [Backported for 3.4-stable. Re-added current_trace NULL checks; removed allocated_snapshot field; adapted to tracing_trace_options_write without trace_set_options.] Signed-off-by: Steven Rostedt Signed-off-by: Lingzhu Xiang Reviewed-by: CAI Qian Signed-off-by: Greg Kroah-Hartman

tracing: Protect tracer flags with trace_types_lock

2013-04-05T17:04:37Z

commit 69d34da2984c95b33ea21518227e1f9470f11d95 upstream. Seems that the tracer flags have never been protected from synchronous writes. Luckily, admins don't usually modify the tracing flags via two different tasks. But if scripts were to be used to modify them, then they could get corrupted. Move the trace_types_lock that protects against tracers changing to also protect the flags being set. [Backported for 3.4, 3.0-stable. Moved return to after unlock.] Signed-off-by: Steven Rostedt Signed-off-by: Lingzhu Xiang Reviewed-by: CAI Qian Signed-off-by: Greg Kroah-Hartman

kernel/signal.c: use __ARCH_HAS_SA_RESTORER instead of SA_RESTORER

2013-04-05T17:04:14Z

commit 522cff142d7d2f9230839c9e1f21a4d8bcc22a4a upstream. __ARCH_HAS_SA_RESTORER is the preferred conditional for use in 3.9 and later kernels, per Kees. Signed-off-by: Andrew Morton Cc: Emese Revfy Cc: Emese Revfy Cc: PaX Team Cc: Al Viro Cc: Oleg Nesterov Cc: "Eric W. Biederman" Cc: Serge Hallyn Cc: Julien Tinnes Signed-off-by: Linus Torvalds Cc: Ben Hutchings Signed-off-by: Greg Kroah-Hartman

nohz: Make tick_nohz_irq_exit() irq safe

2013-03-28T19:12:27Z

commit e5ab012c3271990e8457055c25cafddc1ae8aa6b upstream. As it stands, irq_exit() may or may not be called with irqs disabled, depending on __ARCH_IRQ_EXIT_IRQS_DISABLED that the arch can define. It makes tick_nohz_irq_exit() unsafe. For example two interrupts can race in tick_nohz_stop_sched_tick(): the inner most one computes the expiring time on top of the timer list, then it's interrupted right before reprogramming the clock. The new interrupt enqueues a new timer list timer, it reprogram the clock to take it into account and it exits. The CPUs resumes the inner most interrupt and performs the clock reprogramming without considering the new timer list timer. This regression has been introduced by: 280f06774afedf849f0b34248ed6aff57d0f6908 ("nohz: Separate out irq exit and idle loop dyntick logic") Let's fix it right now with the appropriate protections. A saner long term solution will be to remove __ARCH_IRQ_EXIT_IRQS_DISABLED and mandate that irq_exit() is called with interrupts disabled. Signed-off-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1361373336-11337-1-git-send-email-fweisbec@gmail.com Signed-off-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman Signed-off-by: Thomas Gleixner Signed-off-by: Lingzhu Xiang Reviewed-by: CAI Qian Signed-off-by: Greg Kroah-Hartman

clockevents: Don't allow dummy broadcast timers

2013-03-28T19:12:26Z

commit a7dc19b8652c862d5b7c4d2339bd3c428bd29c4a upstream. Currently tick_check_broadcast_device doesn't reject clock_event_devices with CLOCK_EVT_FEAT_DUMMY, and may select them in preference to real hardware if they have a higher rating value. In this situation, the dummy timer is responsible for broadcasting to itself, and the core clockevents code may attempt to call non-existent callbacks for programming the dummy, eventually leading to a panic. This patch makes tick_check_broadcast_device always reject dummy timers, preventing this problem. Signed-off-by: Mark Rutland Cc: linux-arm-kernel@lists.infradead.org Cc: Jon Medhurst (Tixy) Signed-off-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman