All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jiri Slaby <jslaby@suse.cz>
To: stable@vger.kernel.org
Cc: Boqun Feng <boqun.feng@gmail.com>,
	Michael Ellerman <mpe@ellerman.id.au>,
	Jiri Slaby <jslaby@suse.cz>
Subject: [patch added to 3.12-stable] powerpc: Make value-returning atomics fully ordered
Date: Thu, 28 Jan 2016 11:52:13 +0100	[thread overview]
Message-ID: <1453978346-20237-22-git-send-email-jslaby@suse.cz> (raw)
In-Reply-To: <1453978346-20237-1-git-send-email-jslaby@suse.cz>

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

This patch has been added to the 3.12 stable tree. If you have any
objections, please let us know.

===============

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: Jiri Slaby <jslaby@suse.cz>
---
 arch/powerpc/include/asm/synch.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/include/asm/synch.h b/arch/powerpc/include/asm/synch.h
index e682a7143edb..c50868681f9e 100644
--- 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
-- 
2.7.0


  parent reply	other threads:[~2016-01-28 10:52 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-28 10:51 [patch added to 3.12-stable] ARM: 8158/1: LLVMLinux: use static inline in ARM ftrace.h Jiri Slaby
2016-01-28 10:51 ` [patch added to 3.12-stable] ARM: 8160/1: drop warning about return_address not using unwind tables Jiri Slaby
2016-01-28 11:07   ` Uwe Kleine-König
2016-01-28 10:51 ` [patch added to 3.12-stable] drm/radeon: cypress_dpm: Fix unused variable warning when CONFIG_ACPI=n Jiri Slaby
2016-01-28 10:51 ` [patch added to 3.12-stable] drm: radeon: ni_dpm: " Jiri Slaby
2016-01-28 10:51 ` [patch added to 3.12-stable] lkdtm: adjust recursion size to avoid warnings Jiri Slaby
2016-01-28 10:51 ` [patch added to 3.12-stable] RDMA/cxgb4: Fix gcc warning on 32-bit arch Jiri Slaby
2016-01-28 10:51 ` [patch added to 3.12-stable] mISDN: avoid arch specific __builtin_return_address call Jiri Slaby
2016-01-28 10:51 ` [patch added to 3.12-stable] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] ipv6/addrlabel: fix ip6addrlbl_get() Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] sctp: sctp should release assoc when sctp_make_abort_user return NULL in sctp_close Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] connector: bump skb->users before callback invocation Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] unix: properly account for FDs passed over unix sockets Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] bridge: Only call /sbin/bridge-stp for the initial network namespace Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] net: sctp: prevent writes to cookie_hmac_alg from accessing invalid memory Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] ipv6: tcp: add rcu locking in tcp_v6_send_synack() Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] tcp_yeah: don't set ssthresh below 2 Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] phonet: properly unshare skbs in phonet_rcv() Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] ipv6: update skb->csum when CE mark is propagated Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] team: Replace rcu_read_lock with a mutex in team_vlan_rx_kill_vid Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] xfrm: dst_entries_init() per-net dst_ops Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] powerpc/tm: Block signal return setting invalid MSR state Jiri Slaby
2016-01-28 10:52 ` Jiri Slaby [this message]
2016-01-28 10:52 ` [patch added to 3.12-stable] powerpc: Make {cmp}xchg* and their atomic_ versions fully ordered Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] scripts/recordmcount.pl: support data in text section on powerpc Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] arm64: KVM: Fix AArch32 to AArch64 register mapping Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] arm64: fix building without CONFIG_UID16 Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] arm64: Clear out any singlestep state on a ptrace detach operation Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] arm64: mm: ensure that the zero page is visible to the page table walker Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] parisc iommu: fix panic due to trying to allocate too large region Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] HID: core: Avoid uninitialized buffer access Jiri Slaby
2016-01-28 10:52 ` [patch added to 3.12-stable] mn10300: Select CONFIG_HAVE_UID16 to fix build failure Jiri Slaby

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1453978346-20237-22-git-send-email-jslaby@suse.cz \
    --to=jslaby@suse.cz \
    --cc=boqun.feng@gmail.com \
    --cc=mpe@ellerman.id.au \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.