Re: [tip:locking/core] locking/pvqspinlock, x86: Implement the paravirt qspinlock call patching

[Waiman Long]

On 05/30/2015 12:09 AM, Sasha Levin wrote:
> On 05/08/2015 09:27 AM, tip-bot for Peter Zijlstra (Intel) wrote:
> > Commit-ID:  f233f7f1581e78fd9b4023f2e7d8c1ed89020cc9
> > Gitweb:     http://git.kernel.org/tip/f233f7f1581e78fd9b4023f2e7d8c1ed89020cc9
> > Author:     Peter Zijlstra (Intel)<peterz@xxxxxxxxxxxxx>
> > AuthorDate: Fri, 24 Apr 2015 14:56:38 -0400
> > Committer:  Ingo Molnar<mingo@xxxxxxxxxx>
> > CommitDate: Fri, 8 May 2015 12:37:09 +0200
> >
> > locking/pvqspinlock, x86: Implement the paravirt qspinlock call patching
> >
> > We use the regular paravirt call patching to switch between:
> >
> >    native_queued_spin_lock_slowpath()	__pv_queued_spin_lock_slowpath()
> >    native_queued_spin_unlock()		__pv_queued_spin_unlock()
> >
> > We use a callee saved call for the unlock function which reduces the
> > i-cache footprint and allows 'inlining' of SPIN_UNLOCK functions
> > again.
> >
> > We further optimize the unlock path by patching the direct call with a
> > "movb $0,%arg1" if we are indeed using the native unlock code. This
> > makes the unlock code almost as fast as the !PARAVIRT case.
> >
> > This significantly lowers the overhead of having
> > CONFIG_PARAVIRT_SPINLOCKS enabled, even for native code.
> >
> > Signed-off-by: Peter Zijlstra (Intel)<peterz@xxxxxxxxxxxxx>
> > Signed-off-by: Waiman Long<Waiman.Long@xxxxxx>
> > Signed-off-by: Peter Zijlstra (Intel)<peterz@xxxxxxxxxxxxx>
> > Cc: Andrew Morton<akpm@xxxxxxxxxxxxxxxxxxxx>
> > Cc: Boris Ostrovsky<boris.ostrovsky@xxxxxxxxxx>
> > Cc: Borislav Petkov<bp@xxxxxxxxx>
> > Cc: Daniel J Blueman<daniel@xxxxxxxxxxxxx>
> > Cc: David Vrabel<david.vrabel@xxxxxxxxxx>
> > Cc: Douglas Hatch<doug.hatch@xxxxxx>
> > Cc: H. Peter Anvin<hpa@xxxxxxxxx>
> > Cc: Konrad Rzeszutek Wilk<konrad.wilk@xxxxxxxxxx>
> > Cc: Linus Torvalds<torvalds@xxxxxxxxxxxxxxxxxxxx>
> > Cc: Oleg Nesterov<oleg@xxxxxxxxxx>
> > Cc: Paolo Bonzini<paolo.bonzini@xxxxxxxxx>
> > Cc: Paul E. McKenney<paulmck@xxxxxxxxxxxxxxxxxx>
> > Cc: Peter Zijlstra<peterz@xxxxxxxxxxxxx>
> > Cc: Raghavendra K T<raghavendra.kt@xxxxxxxxxxxxxxxxxx>
> > Cc: Rik van Riel<riel@xxxxxxxxxx>
> > Cc: Scott J Norton<scott.norton@xxxxxx>
> > Cc: Thomas Gleixner<tglx@xxxxxxxxxxxxx>
> > Cc: virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
> > Cc: xen-devel@xxxxxxxxxxxxxxxxxxxx
> > Link: http://lkml.kernel.org/r/1429901803-29771-10-git-send-email-Waiman.Long@xxxxxx
> > Signed-off-by: Ingo Molnar<mingo@xxxxxxxxxx>
> Hey Peter,
>
> I'm seeing this on the latest -next kernel:
>
> [ 8693.503262] BUG: KASan: out of bounds access in __pv_queued_spin_lock_slowpath+0x84e/0x8c0 at addr ffffffffb9495950
> [ 8693.503271] Read of size 8 by task swapper/9/0
> [ 8693.503289] Address belongs to variable pv_lock_ops+0x10/0x240

I would like to clarify what the message means. pv_locks_ops + 0x10 
should be the pv_wait function pointer. Also the structure should be 
just 32 bytes in size and so what does the "/0x240" mean?

Cheers,
Longman
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/