linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2 0/2] pvclock: more code cleanup
@ 2016-06-15 12:46 Paolo Bonzini
  2016-06-15 12:46 ` [PATCH v2 1/2] pvclock: introduce seqcount-like API Paolo Bonzini
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: Paolo Bonzini @ 2016-06-15 12:46 UTC (permalink / raw)
  To: linux-kernel, kvm; +Cc: Andy Lutomirski, Ingo Molnar, x86, Minfei Huang

Patch 1 simplifies the pvclock.h API by moving seqcount logic into
two new functions pvclock_read_begin and pvclock_read_retry.
Patch 2 uses the new simplified API in the vDSO.

Paolo

Paolo Bonzini (2):
  pvclock: introduce seqcount-like API
  x86: vdso: use __pvclock_read_cycles

 arch/x86/entry/vdso/vclock_gettime.c | 25 +++++------------------
 arch/x86/include/asm/pvclock.h       | 39 +++++++++++++++++++++---------------
 arch/x86/kernel/pvclock.c            | 17 ++++++----------
 3 files changed, 34 insertions(+), 47 deletions(-)

-- 
1.8.3.1

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH v2 1/2] pvclock: introduce seqcount-like API
  2016-06-15 12:46 [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini
@ 2016-06-15 12:46 ` Paolo Bonzini
  2016-06-15 12:46 ` [PATCH v2 2/2] x86: vdso: use __pvclock_read_cycles Paolo Bonzini
  2016-06-27 14:16 ` [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini
  2 siblings, 0 replies; 4+ messages in thread
From: Paolo Bonzini @ 2016-06-15 12:46 UTC (permalink / raw)
  To: linux-kernel, kvm; +Cc: x86, Andy Lutomirski, Ingo Molnar, Minfei Huang

The version field in struct pvclock_vcpu_time_info basically implements
a seqcount.  Wrap it with the usual read_begin and read_retry functions,
and use these APIs instead of peppering the code with smp_rmb()s.
While at it, change it to the more pedantically correct virt_rmb().

With this change, __pvclock_read_cycles can be simplified noticeably.

Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Minfei Huang <mnghuan@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/entry/vdso/vclock_gettime.c |  9 ++-------
 arch/x86/include/asm/pvclock.h       | 39 +++++++++++++++++++++---------------
 arch/x86/kernel/pvclock.c            | 17 ++++++----------
 3 files changed, 31 insertions(+), 34 deletions(-)

diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c
index 2f02d23a05ef..db1e3b4c3693 100644
--- a/arch/x86/entry/vdso/vclock_gettime.c
+++ b/arch/x86/entry/vdso/vclock_gettime.c
@@ -123,9 +123,7 @@ static notrace cycle_t vread_pvclock(int *mode)
 	 */
 
 	do {
-		version = pvti->version;
-
-		smp_rmb();
+		version = pvclock_read_begin(pvti);
 
 		if (unlikely(!(pvti->flags & PVCLOCK_TSC_STABLE_BIT))) {
 			*mode = VCLOCK_NONE;
@@ -137,10 +135,7 @@ static notrace cycle_t vread_pvclock(int *mode)
 		pvti_tsc_shift = pvti->tsc_shift;
 		pvti_system_time = pvti->system_time;
 		pvti_tsc = pvti->tsc_timestamp;
-
-		/* Make sure that the version double-check is last. */
-		smp_rmb();
-	} while (unlikely((version & 1) || version != pvti->version));
+	} while (pvclock_read_retry(pvti, version));
 
 	delta = tsc - pvti_tsc;
 	ret = pvti_system_time +
diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h
index 7c1c89598688..0ee92db1e9f3 100644
--- a/arch/x86/include/asm/pvclock.h
+++ b/arch/x86/include/asm/pvclock.h
@@ -25,6 +25,24 @@ void pvclock_resume(void);
 
 void pvclock_touch_watchdogs(void);
 
+static __always_inline
+unsigned pvclock_read_begin(const struct pvclock_vcpu_time_info *src)
+{
+	unsigned version = src->version & ~1;
+	/* Make sure that the version is read before the data. */
+	virt_rmb();
+	return version;
+}
+
+static __always_inline
+bool pvclock_read_retry(const struct pvclock_vcpu_time_info *src,
+			unsigned version)
+{
+	/* Make sure that the version is re-read after the data. */
+	virt_rmb();
+	return version != src->version;
+}
+
 /*
  * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
  * yielding a 64-bit result.
@@ -69,23 +87,12 @@ static inline u64 pvclock_scale_delta(u64 delta, u32 mul_frac, int shift)
 }
 
 static __always_inline
-unsigned __pvclock_read_cycles(const struct pvclock_vcpu_time_info *src,
-			       cycle_t *cycles, u8 *flags)
+cycle_t __pvclock_read_cycles(const struct pvclock_vcpu_time_info *src)
 {
-	unsigned version;
-	cycle_t offset;
-	u64 delta;
-
-	version = src->version;
-	/* Make the latest version visible */
-	smp_rmb();
-
-	delta = rdtsc_ordered() - src->tsc_timestamp;
-	offset = pvclock_scale_delta(delta, src->tsc_to_system_mul,
-				   src->tsc_shift);
-	*cycles = src->system_time + offset;
-	*flags = src->flags;
-	return version;
+	u64 delta = rdtsc_ordered() - src->tsc_timestamp;
+	cycle_t offset = pvclock_scale_delta(delta, src->tsc_to_system_mul,
+					     src->tsc_shift);
+	return src->system_time + offset;
 }
 
 struct pvclock_vsyscall_time_info {
diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c
index 06c58ce46762..3599404e3089 100644
--- a/arch/x86/kernel/pvclock.c
+++ b/arch/x86/kernel/pvclock.c
@@ -64,14 +64,9 @@ u8 pvclock_read_flags(struct pvclock_vcpu_time_info *src)
 	u8 flags;
 
 	do {
-		version = src->version;
-		/* Make the latest version visible */
-		smp_rmb();
-
+		version = pvclock_read_begin(src);
 		flags = src->flags;
-		/* Make sure that the version double-check is last. */
-		smp_rmb();
-	} while ((src->version & 1) || version != src->version);
+	} while (pvclock_read_retry(src, version));
 
 	return flags & valid_flags;
 }
@@ -84,10 +79,10 @@ cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
 	u8 flags;
 
 	do {
-		version = __pvclock_read_cycles(src, &ret, &flags);
-		/* Make sure that the version double-check is last. */
-		smp_rmb();
-	} while ((src->version & 1) || version != src->version);
+		version = pvclock_read_begin(src);
+		ret = __pvclock_read_cycles(src);
+		flags = src->flags;
+	} while (pvclock_read_retry(src, version));
 
 	if (unlikely((flags & PVCLOCK_GUEST_STOPPED) != 0)) {
 		src->flags &= ~PVCLOCK_GUEST_STOPPED;
-- 
1.8.3.1

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* [PATCH v2 2/2] x86: vdso: use __pvclock_read_cycles
  2016-06-15 12:46 [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini
  2016-06-15 12:46 ` [PATCH v2 1/2] pvclock: introduce seqcount-like API Paolo Bonzini
@ 2016-06-15 12:46 ` Paolo Bonzini
  2016-06-27 14:16 ` [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini
  2 siblings, 0 replies; 4+ messages in thread
From: Paolo Bonzini @ 2016-06-15 12:46 UTC (permalink / raw)
  To: linux-kernel, kvm; +Cc: x86, Andy Lutomirski, Ingo Molnar, Minfei Huang

The new simplified __pvclock_read_cycles does the same computation
as vread_pvclock, except that (because it takes the pvclock_vcpu_time_info
pointer) it has to be moved inside the loop.  Since the loop is expected to
never roll, this makes no difference.

Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Minfei Huang <mnghuan@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
	If acked I'd like to take this through the KVM tree.

 arch/x86/entry/vdso/vclock_gettime.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c
index db1e3b4c3693..94d54d0defa7 100644
--- a/arch/x86/entry/vdso/vclock_gettime.c
+++ b/arch/x86/entry/vdso/vclock_gettime.c
@@ -96,9 +96,8 @@ static notrace cycle_t vread_pvclock(int *mode)
 {
 	const struct pvclock_vcpu_time_info *pvti = &get_pvti0()->pvti;
 	cycle_t ret;
-	u64 tsc, pvti_tsc;
-	u64 last, delta, pvti_system_time;
-	u32 version, pvti_tsc_to_system_mul, pvti_tsc_shift;
+	u64 last;
+	u32 version;
 
 	/*
 	 * Note: The kernel and hypervisor must guarantee that cpu ID
@@ -130,18 +129,9 @@ static notrace cycle_t vread_pvclock(int *mode)
 			return 0;
 		}
 
-		tsc = rdtsc_ordered();
-		pvti_tsc_to_system_mul = pvti->tsc_to_system_mul;
-		pvti_tsc_shift = pvti->tsc_shift;
-		pvti_system_time = pvti->system_time;
-		pvti_tsc = pvti->tsc_timestamp;
+		ret = __pvclock_read_cycles(pvti);
 	} while (pvclock_read_retry(pvti, version));
 
-	delta = tsc - pvti_tsc;
-	ret = pvti_system_time +
-		pvclock_scale_delta(delta, pvti_tsc_to_system_mul,
-				    pvti_tsc_shift);
-
 	/* refer to vread_tsc() comment for rationale */
 	last = gtod->cycle_last;
 
-- 
1.8.3.1

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH v2 0/2] pvclock: more code cleanup
  2016-06-15 12:46 [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini
  2016-06-15 12:46 ` [PATCH v2 1/2] pvclock: introduce seqcount-like API Paolo Bonzini
  2016-06-15 12:46 ` [PATCH v2 2/2] x86: vdso: use __pvclock_read_cycles Paolo Bonzini
@ 2016-06-27 14:16 ` Paolo Bonzini
  2 siblings, 0 replies; 4+ messages in thread
From: Paolo Bonzini @ 2016-06-27 14:16 UTC (permalink / raw)
  To: linux-kernel, kvm; +Cc: Andy Lutomirski, Ingo Molnar, x86, Minfei Huang



On 15/06/2016 14:46, Paolo Bonzini wrote:
> Patch 1 simplifies the pvclock.h API by moving seqcount logic into
> two new functions pvclock_read_begin and pvclock_read_retry.
> Patch 2 uses the new simplified API in the vDSO.

Andy, I've now benchmarked the patches.

Patch 2 introduces no meaningful difference, however patch 1 makes 
clock_gettime slower by about 3%.  I can get this back with:

diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h
index 0ee92db1e9f3..d019f0cc80ec 100644
--- a/arch/x86/include/asm/pvclock.h
+++ b/arch/x86/include/asm/pvclock.h
@@ -40,7 +40,7 @@ bool pvclock_read_retry(const struct pvclock_vcpu_time_info *src,
 {
 	/* Make sure that the version is re-read after the data. */
 	virt_rmb();
-	return version != src->version;
+	return unlikely(version != src->version);
 }
 
 /*


... which for whatever reason makes GCC inline more aggressively.
I'm going to send v3.

Paolo

> Paolo
> 
> Paolo Bonzini (2):
>   pvclock: introduce seqcount-like API
>   x86: vdso: use __pvclock_read_cycles
> 
>  arch/x86/entry/vdso/vclock_gettime.c | 25 +++++------------------
>  arch/x86/include/asm/pvclock.h       | 39 +++++++++++++++++++++---------------
>  arch/x86/kernel/pvclock.c            | 17 ++++++----------
>  3 files changed, 34 insertions(+), 47 deletions(-)
> 

^ permalink raw reply related	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2016-06-27 14:16 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-06-15 12:46 [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini
2016-06-15 12:46 ` [PATCH v2 1/2] pvclock: introduce seqcount-like API Paolo Bonzini
2016-06-15 12:46 ` [PATCH v2 2/2] x86: vdso: use __pvclock_read_cycles Paolo Bonzini
2016-06-27 14:16 ` [PATCH v2 0/2] pvclock: more code cleanup Paolo Bonzini

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).