linux-stable.git/arch/x86/kernel/pvclock.c, branch linux-4.7.y

pvclock: Get rid of __pvclock_read_cycles in function pvclock_read_flags

2016-06-27T13:12:15+00:00

There is a generic function __pvclock_read_cycles to be used to get both
flags and cycles. For function pvclock_read_flags, it's useless to get
cycles value. To make this function be more effective, get this variable
flags directly in function.

Signed-off-by: Minfei Huang 
Signed-off-by: Paolo Bonzini

pvclock: Add CPU barriers to get correct version value

2016-06-27T13:12:14+00:00

Protocol for the "version" fields is: hypervisor raises it (making it
uneven) before it starts updating the fields and raises it again (making
it even) when it is done.  Thus the guest can make sure the time values
it got are consistent by checking the version before and after reading
them.

Add CPU barries after getting version value just like what function
vread_pvclock does, because all of callees in this function is inline.

Fixes: 502dfeff239e8313bfbe906ca0a1a6827ac8481b
Cc: stable@vger.kernel.org
Signed-off-by: Minfei Huang 
Signed-off-by: Paolo Bonzini

x86/vdso: Remove pvclock fixmap machinery

2015-12-11T07:56:03+00:00

Signed-off-by: Andy Lutomirski 
Reviewed-by: Paolo Bonzini 
Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Denys Vlasenko 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: linux-mm@kvack.org
Link: http://lkml.kernel.org/r/4933029991103ae44672c82b97a20035f5c1fe4f.1449702533.git.luto@kernel.org
Signed-off-by: Ingo Molnar

x86: pvclock: Really remove the sched notifier for cross-cpu migrations

2015-04-27T13:49:30+00:00

This reverts commits 0a4e6be9ca17c54817cf814b4b5aa60478c6df27
and 80f7fdb1c7f0f9266421f823964fd1962681f6ce.

The task migration notifier was originally introduced in order to support
the pvclock vsyscall with non-synchronized TSC, but KVM only supports it
with synchronized TSC.  Hence, on KVM the race condition is only needed
due to a bad implementation on the host side, and even then it's so rare
that it's mostly theoretical.

As far as KVM is concerned it's possible to fix the host, avoiding the
additional complexity in the vDSO and the (re)introduction of the task
migration notifier.

Xen, on the other hand, hasn't yet implemented vsyscall support at
all, so we do not care about its plans for non-synchronized TSC.

Reported-by: Peter Zijlstra 
Suggested-by: Marcelo Tosatti 
Signed-off-by: Paolo Bonzini

x86: kvm: Revert "remove sched notifier for cross-cpu migrations"

2015-03-23T23:22:48+00:00

The following point:

    2. per-CPU pvclock time info is updated if the
       underlying CPU changes.

Is not true anymore since "KVM: x86: update pvclock area conditionally,
on cpu migration".

Add task migration notification back.

Problem noticed by Andy Lutomirski.

Signed-off-by: Marcelo Tosatti 
CC: stable@kernel.org # 3.11+

hung_task: add method to reset detector

2013-11-06T07:49:02+00:00

In certain occasions it is possible for a hung task detector
positive to be false: continuation from a paused VM, for example.

Add a method to reset detection, similar as is done
with other kernel watchdogs.

Acked-by: Don Zickus 
Acked-by: Paolo Bonzini 
Signed-off-by: Marcelo Tosatti 
Signed-off-by: Gleb Natapov

pvclock: detect watchdog reset at pvclock read

2013-11-06T07:48:43+00:00

Implement reset of kernel watchdogs at pvclock read time. This avoids
adding special code to every watchdog.

This is possible for watchdogs which measure time based on sched_clock() or
ktime_get() variants.

Suggested by Don Zickus.

Acked-by: Don Zickus 
Acked-by: Paolo Bonzini 
Signed-off-by: Marcelo Tosatti 
Signed-off-by: Gleb Natapov

remove sched notifier for cross-cpu migrations

2013-07-18T10:29:30+00:00

Linux as a guest on KVM hypervisor, the only user of the pvclock
vsyscall interface, does not require notification on task migration
because:

1. cpu ID number maps 1:1 to per-CPU pvclock time info.
2. per-CPU pvclock time info is updated if the
   underlying CPU changes.
3. that version is increased whenever underlying CPU
   changes.

Which is sufficient to guarantee nanoseconds counter
is calculated properly.

Signed-off-by: Marcelo Tosatti 
Acked-by: Peter Zijlstra 
Signed-off-by: Gleb Natapov

x86/kvm: Fix pvclock vsyscall fixmap

2013-02-28T06:50:11+00:00

The physical memory fixmapped for the pvclock clock_gettime vsyscall
was allocated, and thus is not a kernel symbol. __pa() is the proper
method to use in this case.

Fixes the crash below when booting a next-20130204+ smp guest on a
3.8-rc5+ KVM host.

[    0.666410] udevd[97]: starting version 175
[    0.674043] udevd[97]: udevd:[97]: segfault at ffffffffff5fd020
     ip 00007fff069e277f sp 00007fff068c9ef8 error d

Acked-by: Marcelo Tosatti 
Signed-off-by: Peter Hurley 
Signed-off-by: Gleb Natapov

x86: pvclock: generic pvclock vsyscall initialization

2012-11-28T01:29:09+00:00

Originally from Jeremy Fitzhardinge.

Introduce generic, non hypervisor specific, pvclock initialization
routines.

Signed-off-by: Marcelo Tosatti