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

From: Sasha Levin
Date: Sat May 30 2015 - 00:12:13 EST

Next message: Sergey Senozhatsky: "Re: [PATCH] zram: clear disk io accounting when reset zram device"
Previous message: Mike Galbraith: "Re: [PATCH RESEND] sched: prefer an idle cpu vs an idle sibling for BALANCE_WAKE"
Next in thread: Waiman Long: "Re: [tip:locking/core] locking/pvqspinlock, x86: Implement the paravirt qspinlock call patching"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

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
[ 8693.503301] CPU: 9 PID: 0 Comm: swapper/9 Tainted: G D 4.1.0-rc5-next-20150529-sasha-00039-g7fd455d-dirty #2263
[ 8693.503335] ffffffffb6a1423a b6f92731d7a76ba3 ffff8802b349f918 ffffffffb6a1423a
[ 8693.503355] 0000000000000000 ffff8802b349f9a8 ffff8802b349f998 ffffffffad5c70ee
[ 8693.503375] ffffffffad2eb58e 0000000000000004 0000000000000086 1ffff1011953cbb4
[ 8693.503379] Call Trace:
[ 8693.503409] ? dump_stack (lib/dump_stack.c:52)
[ 8693.503426] dump_stack (lib/dump_stack.c:52)
[ 8693.503454] kasan_report_error (mm/kasan/report.c:132 mm/kasan/report.c:193)
[ 8693.503463] ? __pv_queued_spin_lock_slowpath (./arch/x86/include/asm/paravirt.h:730 kernel/locking/qspinlock.c:410)
[ 8693.503474] ? kasan_report_error (mm/kasan/report.c:186)
[ 8693.503488] ? trace_hardirqs_off_caller (./arch/x86/include/asm/current.h:14 kernel/locking/lockdep.c:2652)
[ 8693.503504] __asan_report_load8_noabort (mm/kasan/report.c:230 mm/kasan/report.c:251)
[ 8693.503517] ? __pv_queued_spin_lock_slowpath (./arch/x86/include/asm/paravirt.h:730 kernel/locking/qspinlock.c:410)
[ 8693.503526] __pv_queued_spin_lock_slowpath (./arch/x86/include/asm/paravirt.h:730 kernel/locking/qspinlock.c:410)
[ 8693.503541] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503557] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503566] ? trace_hardirqs_off_caller (./arch/x86/include/asm/current.h:14 kernel/locking/lockdep.c:2652)
[ 8693.503578] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503589] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503605] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.503614] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503623] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503631] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503639] ? async_page_fault (arch/x86/kernel/entry_64.S:1261)
[ 8693.503663] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503681] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.503691] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503699] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503730] ? trace_hardirqs_off_caller (./arch/x86/include/asm/current.h:14 kernel/locking/lockdep.c:2652)
[ 8693.503743] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503754] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503772] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.503784] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503794] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503802] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503814] ? async_page_fault (arch/x86/kernel/entry_64.S:1261)
[ 8693.503829] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503845] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.503854] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503863] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503872] ? trace_hardirqs_off_caller (./arch/x86/include/asm/current.h:14 kernel/locking/lockdep.c:2652)
[ 8693.503888] ? async_page_fault (arch/x86/kernel/entry_64.S:1261)
[ 8693.503897] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503907] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503922] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.503935] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503943] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.503962] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503970] ? async_page_fault (arch/x86/kernel/entry_64.S:1261)
[ 8693.503980] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.503994] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.504002] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504014] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504029] ? trace_hardirqs_off_caller (./arch/x86/include/asm/current.h:14 kernel/locking/lockdep.c:2652)
[ 8693.504042] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504052] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504064] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.504077] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504086] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504093] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504102] ? async_page_fault (arch/x86/kernel/entry_64.S:1261)
[ 8693.504112] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504126] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.504135] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504146] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504161] ? trace_hardirqs_off_caller (./arch/x86/include/asm/current.h:14 kernel/locking/lockdep.c:2652)
[ 8693.504172] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504185] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504201] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.504224] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504233] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504240] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504252] ? async_page_fault (arch/x86/kernel/entry_64.S:1261)
[ 8693.504263] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504276] ? native_iret (arch/x86/kernel/entry_64.S:806)
[ 8693.504295] ? error_sti (arch/x86/kernel/entry_64.S:1334)
[ 8693.504303] ? trace_hardirqs_off_thunk (arch/x86/lib/thunk_64.S:43)
[ 8693.504311] Memory state around the buggy address:
[ 8693.504320] ffffffffb9495800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[ 8693.504326] ffffffffb9495880: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[ 8693.504337] >ffffffffb9495900: 00 00 00 00 00 00 00 00 00 00 00 00 fa fa fa fa
[ 8693.504347] ^
[ 8693.504353] ffffffffb9495980: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[ 8693.504363] ffffffffb9495a00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Thanks,
Sasha
--
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/

Next message: Sergey Senozhatsky: "Re: [PATCH] zram: clear disk io accounting when reset zram device"
Previous message: Mike Galbraith: "Re: [PATCH RESEND] sched: prefer an idle cpu vs an idle sibling for BALANCE_WAKE"
Next in thread: Waiman Long: "Re: [tip:locking/core] locking/pvqspinlock, x86: Implement the paravirt qspinlock call patching"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]