[tip:perf/core] tools/perf: Add required memory barriers

From: tip-bot for Peter Zijlstra
Date: Wed Nov 06 2013 - 08:20:55 EST

Next message: tip-bot for Preeti U Murthy: "[tip:sched/core] sched: Remove unnecessary iteration over sched domains to update nr_busy_cpus"
Previous message: tip-bot for Yan, Zheng: "[tip:perf/core] perf/x86/intel: Add Ivy Bridge-EP uncore IRP box support"
Next in thread: Vince Weaver: "Re: [tip:perf/core] tools/perf: Add required memory barriers"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Commit-ID: a94d342b9cb09edfe888ea972af0883b6a8d992b
Gitweb: http://git.kernel.org/tip/a94d342b9cb09edfe888ea972af0883b6a8d992b
Author: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
AuthorDate: Wed, 30 Oct 2013 11:42:46 +0100
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Wed, 6 Nov 2013 12:34:26 +0100

tools/perf: Add required memory barriers

To match patch bf378d341e48 ("perf: Fix perf ring buffer memory
ordering") change userspace to also adhere to the ordering outlined.

Signed-off-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Michael Neuling <mikey@xxxxxxxxxxx>
Cc: "Paul E. McKenney" <paulmck@xxxxxxxxxxxxxxxxxx>
Cc: james.hogan@xxxxxxxxxx
Cc: Vince Weaver <vince@xxxxxxxxxx>
Cc: Victor Kaplansky <VICTORK@xxxxxxxxxx>
Cc: Oleg Nesterov <oleg@xxxxxxxxxx>
Cc: Anton Blanchard <anton@xxxxxxxxx>
Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
Cc: Frederic Weisbecker <fweisbec@xxxxxxxxx>
Cc: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxx>
Cc: Michael Ellerman <michael@xxxxxxxxxxxxxx>
Link: http://lkml.kernel.org/r/20131030104246.GH16117@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
tools/perf/perf.h | 59 ++++++++++++++++++++++++++++++++++++++----------
tools/perf/tests/rdpmc.c | 2 --
tools/perf/util/evlist.h | 4 ++--
3 files changed, 49 insertions(+), 16 deletions(-)

diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index f61c230..6a587e84 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -4,6 +4,8 @@
#include <asm/unistd.h>

#if defined(__i386__)
+#define mb() asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
+#define wmb() asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
#define rmb() asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
#define cpu_relax() asm volatile("rep; nop" ::: "memory");
#define CPUINFO_PROC "model name"
@@ -13,6 +15,8 @@
#endif

#if defined(__x86_64__)
+#define mb() asm volatile("mfence" ::: "memory")
+#define wmb() asm volatile("sfence" ::: "memory")
#define rmb() asm volatile("lfence" ::: "memory")
#define cpu_relax() asm volatile("rep; nop" ::: "memory");
#define CPUINFO_PROC "model name"
@@ -23,45 +27,61 @@

#ifdef __powerpc__
#include "../../arch/powerpc/include/uapi/asm/unistd.h"
+#define mb() asm volatile ("sync" ::: "memory")
+#define wmb() asm volatile ("sync" ::: "memory")
#define rmb() asm volatile ("sync" ::: "memory")
-#define cpu_relax() asm volatile ("" ::: "memory");
#define CPUINFO_PROC "cpu"
#endif

#ifdef __s390__
+#define mb() asm volatile("bcr 15,0" ::: "memory")
+#define wmb() asm volatile("bcr 15,0" ::: "memory")
#define rmb() asm volatile("bcr 15,0" ::: "memory")
-#define cpu_relax() asm volatile("" ::: "memory");
#endif

#ifdef __sh__
#if defined(__SH4A__) || defined(__SH5__)
+# define mb() asm volatile("synco" ::: "memory")
+# define wmb() asm volatile("synco" ::: "memory")
# define rmb() asm volatile("synco" ::: "memory")
#else
+# define mb() asm volatile("" ::: "memory")
+# define wmb() asm volatile("" ::: "memory")
# define rmb() asm volatile("" ::: "memory")
#endif
-#define cpu_relax() asm volatile("" ::: "memory")
#define CPUINFO_PROC "cpu type"
#endif

#ifdef __hppa__
+#define mb() asm volatile("" ::: "memory")
+#define wmb() asm volatile("" ::: "memory")
#define rmb() asm volatile("" ::: "memory")
-#define cpu_relax() asm volatile("" ::: "memory");
#define CPUINFO_PROC "cpu"
#endif

#ifdef __sparc__
+#ifdef __LP64__
+#define mb() asm volatile("ba,pt %%xcc, 1f\n" \
+ "membar #StoreLoad\n" \
+ "1:\n":::"memory")
+#else
+#define mb() asm volatile("":::"memory")
+#endif
+#define wmb() asm volatile("":::"memory")
#define rmb() asm volatile("":::"memory")
-#define cpu_relax() asm volatile("":::"memory")
#define CPUINFO_PROC "cpu"
#endif

#ifdef __alpha__
+#define mb() asm volatile("mb" ::: "memory")
+#define wmb() asm volatile("wmb" ::: "memory")
#define rmb() asm volatile("mb" ::: "memory")
-#define cpu_relax() asm volatile("" ::: "memory")
#define CPUINFO_PROC "cpu model"
#endif

#ifdef __ia64__
+#define mb() asm volatile ("mf" ::: "memory")
+#define wmb() asm volatile ("mf" ::: "memory")
#define rmb() asm volatile ("mf" ::: "memory")
#define cpu_relax() asm volatile ("hint @pause" ::: "memory")
#define CPUINFO_PROC "model name"
@@ -72,40 +92,55 @@
* Use the __kuser_memory_barrier helper in the CPU helper page. See
* arch/arm/kernel/entry-armv.S in the kernel source for details.
*/
+#define mb() ((void(*)(void))0xffff0fa0)()
+#define wmb() ((void(*)(void))0xffff0fa0)()
#define rmb() ((void(*)(void))0xffff0fa0)()
-#define cpu_relax() asm volatile("":::"memory")
#define CPUINFO_PROC "Processor"
#endif

#ifdef __aarch64__
-#define rmb() asm volatile("dmb ld" ::: "memory")
+#define mb() asm volatile("dmb ish" ::: "memory")
+#define wmb() asm volatile("dmb ishld" ::: "memory")
+#define rmb() asm volatile("dmb ishst" ::: "memory")
#define cpu_relax() asm volatile("yield" ::: "memory")
#endif

#ifdef __mips__
-#define rmb() asm volatile( \
+#define mb() asm volatile( \
".set mips2\n\t" \
"sync\n\t" \
".set mips0" \
: /* no output */ \
: /* no input */ \
: "memory")
-#define cpu_relax() asm volatile("" ::: "memory")
+#define wmb() mb()
+#define rmb() mb()
#define CPUINFO_PROC "cpu model"
#endif

#ifdef __arc__
+#define mb() asm volatile("" ::: "memory")
+#define wmb() asm volatile("" ::: "memory")
#define rmb() asm volatile("" ::: "memory")
-#define cpu_relax() rmb()
#define CPUINFO_PROC "Processor"
#endif

#ifdef __metag__
+#define mb() asm volatile("" ::: "memory")
+#define wmb() asm volatile("" ::: "memory")
#define rmb() asm volatile("" ::: "memory")
-#define cpu_relax() asm volatile("" ::: "memory")
#define CPUINFO_PROC "CPU"
#endif

+#define barrier() asm volatile ("" ::: "memory")
+
+#ifndef cpu_relax
+#define cpu_relax() barrier()
+#endif
+
+#define ACCESS_ONCE(x) (*(volatile typeof(x) *)&(x))
+
+
#include <time.h>
#include <unistd.h>
#include <sys/types.h>
diff --git a/tools/perf/tests/rdpmc.c b/tools/perf/tests/rdpmc.c
index ff94886..46649c2 100644
--- a/tools/perf/tests/rdpmc.c
+++ b/tools/perf/tests/rdpmc.c
@@ -9,8 +9,6 @@

#if defined(__x86_64__) || defined(__i386__)

-#define barrier() asm volatile("" ::: "memory")
-
static u64 rdpmc(unsigned int counter)
{
unsigned int low, high;
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index e99eaed..ecaa582 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -177,7 +177,7 @@ int perf_evlist__strerror_open(struct perf_evlist *evlist, int err, char *buf, s
static inline unsigned int perf_mmap__read_head(struct perf_mmap *mm)
{
struct perf_event_mmap_page *pc = mm->base;
- int head = pc->data_head;
+ int head = ACCESS_ONCE(pc->data_head);
rmb();
return head;
}
@@ -190,7 +190,7 @@ static inline void perf_mmap__write_tail(struct perf_mmap *md,
/*
* ensure all reads are done before we write the tail out.
*/
- /* mb(); */
+ mb();
pc->data_tail = tail;
}

--
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: tip-bot for Preeti U Murthy: "[tip:sched/core] sched: Remove unnecessary iteration over sched domains to update nr_busy_cpus"
Previous message: tip-bot for Yan, Zheng: "[tip:perf/core] perf/x86/intel: Add Ivy Bridge-EP uncore IRP box support"
Next in thread: Vince Weaver: "Re: [tip:perf/core] tools/perf: Add required memory barriers"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]