Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1319669

[PATCH 4.3 128/157] powerpc: Make value-returning atomics fully ordered

Show all headers | View raw

4.3-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Boqun Feng <boqun.feng@gmail.com>

commit 49e9cf3f0c04bf76ffa59242254110309554861d upstream.

According to memory-barriers.txt:

> Any atomic operation that modifies some state in memory and returns
> information about the state (old or new) implies an SMP-conditional
> general memory barrier (smp_mb()) on each side of the actual
> operation ...

Which mean these operations should be fully ordered. However on PPC,
PPC_ATOMIC_ENTRY_BARRIER is the barrier before the actual operation,
which is currently "lwsync" if SMP=y. The leading "lwsync" can not
guarantee fully ordered atomics, according to Paul Mckenney:

https://lkml.org/lkml/2015/10/14/970

To fix this, we define PPC_ATOMIC_ENTRY_BARRIER as "sync" to guarantee
the fully-ordered semantics.

This also makes futex atomics fully ordered, which can avoid possible
memory ordering problems if userspace code relies on futex system call
for fully ordered semantics.

Fixes: b97021f85517 ("powerpc: Fix atomic_xxx_return barrier semantics")
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/powerpc/include/asm/synch.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/arch/powerpc/include/asm/synch.h
+++ b/arch/powerpc/include/asm/synch.h
@@ -44,7 +44,7 @@ static inline void isync(void)
 	MAKE_LWSYNC_SECTION_ENTRY(97, __lwsync_fixup);
 #define PPC_ACQUIRE_BARRIER	 "\n" stringify_in_c(__PPC_ACQUIRE_BARRIER)
 #define PPC_RELEASE_BARRIER	 stringify_in_c(LWSYNC) "\n"
-#define PPC_ATOMIC_ENTRY_BARRIER "\n" stringify_in_c(LWSYNC) "\n"
+#define PPC_ATOMIC_ENTRY_BARRIER "\n" stringify_in_c(sync) "\n"
 #define PPC_ATOMIC_EXIT_BARRIER	 "\n" stringify_in_c(sync) "\n"
 #else
 #define PPC_ACQUIRE_BARRIER

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 4.3 000/157] 4.3.5-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 20:10 +0100
  [PATCH 4.3 001/157] x86 smpboot: Re-enable init_udelay=0 by default on modern CPUs Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 20:10 +0100
    Re: [PATCH 4.3 001/157] x86 smpboot: Re-enable init_udelay=0 by  default on modern CPUs Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 23:30 +0100
    Re: [PATCH 4.3 001/157] x86 smpboot: Re-enable init_udelay=0 by  default on modern CPUs Donald Parsons <dparsons@brightdsl.net> - 2016-01-27 23:30 +0100
  [PATCH 4.3 052/157] ALSA: seq: Fix race at timer setup and close Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 20:10 +0100
  [PATCH 4.3 046/157] ALSA: hda - Add mic mute hotkey quirk for Lenovo ThinkCentre AIO Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 20:10 +0100
  [PATCH 4.3 010/157] KVM: svm: unconditionally intercept #DB Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 20:10 +0100
  [PATCH 4.3 155/157] mn10300: Select CONFIG_HAVE_UID16 to fix build failure Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 20:10 +0100
  [PATCH 4.3 154/157] fix the regression from "direct-io: Fix negative return from dio read beyond eof" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:20 +0100
  [PATCH 4.3 126/157] powerpc/opal-irqchip: Fix deadlock introduced by "Fix double endian conversion" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:20 +0100
  [PATCH 4.3 135/157] arm64: cmpxchg_dbl: fix return value type Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:20 +0100
  [PATCH 4.3 134/157] arm64: bpf: fix mod-by-zero case Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:20 +0100
  [PATCH 4.3 044/157] ALSA: hda - Set SKL+ hda controller power at freeze() and thaw() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 128/157] powerpc: Make value-returning atomics fully ordered Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 157/157] arm64: KVM: Add workaround for Cortex-A57 erratum 834220 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 131/157] powerpc/module: Handle R_PPC64_ENTRY relocations Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 129/157] powerpc: Make {cmp}xchg* and their atomic_ versions fully ordered Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 133/157] arm64: bpf: fix div-by-zero case Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 156/157] arm64: restore bogomips information in /proc/cpuinfo Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 147/157] parisc iommu: fix panic due to trying to allocate too large region Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 132/157] recordmcount: arm64: Replace the ignored mcount call into nop Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 148/157] HID: wacom: Tie cached HID_DG_CONTACTCOUNT indices to report ID Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 150/157] HID: core: Avoid uninitialized buffer access Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 149/157] HID: wacom: Expect touch_max touches if HID_DG_CONTACTCOUNT not present Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 130/157] scripts/recordmcount.pl: support data in text section on powerpc Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  [PATCH 4.3 153/157] direct-io: Fix negative return from dio read beyond eof Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-01-27 21:30 +0100
  Re: [PATCH 4.3 000/157] 4.3.5-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2016-01-28 00:30 +0100
  Re: [PATCH 4.3 000/157] 4.3.5-stable review Guenter Roeck <linux@roeck-us.net> - 2016-01-28 03:20 +0100

csiph-web