netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH net-next-2.6 2/4] net: u64_stats_fetch_begin_bh() and u64_stats_fetch_retry_bh()
@ 2010-06-24 10:54 Eric Dumazet
  2010-06-29  6:37 ` David Miller
  0 siblings, 1 reply; 2+ messages in thread
From: Eric Dumazet @ 2010-06-24 10:54 UTC (permalink / raw)
  To: David Miller; +Cc: netdev

- Must disable preemption in case of 32bit UP in u64_stats_fetch_begin()
and u64_stats_fetch_retry()

- Add new u64_stats_fetch_begin_bh() and u64_stats_fetch_retry_bh() for
network usage, disabling BH on 32bit UP only.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
---
 include/linux/u64_stats_sync.h |   59 +++++++++++++++++++++++--------
 1 file changed, 44 insertions(+), 15 deletions(-)

diff --git a/include/linux/u64_stats_sync.h b/include/linux/u64_stats_sync.h
index 201d319..00c1592 100644
--- a/include/linux/u64_stats_sync.h
+++ b/include/linux/u64_stats_sync.h
@@ -27,6 +27,9 @@
  *    (On UP, there is no seqcount_t protection, a reader allowing interrupts could
  *     read partial values)
  *
+ * 7) For softirq uses, readers can use u64_stats_fetch_begin_bh() and
+ *    u64_stats_fetch_retry_bh() helpers
+ *
  * Usage :
  *
  * Stats producer (writer) should use following template granted it already got
@@ -58,54 +61,80 @@
  */
 #include <linux/seqlock.h>
 
-#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 struct u64_stats_sync {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	seqcount_t	seq;
+#endif
 };
 
 static void inline u64_stats_update_begin(struct u64_stats_sync *syncp)
 {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	write_seqcount_begin(&syncp->seq);
+#endif
 }
 
 static void inline u64_stats_update_end(struct u64_stats_sync *syncp)
 {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	write_seqcount_end(&syncp->seq);
+#endif
 }
 
 static unsigned int inline u64_stats_fetch_begin(const struct u64_stats_sync *syncp)
 {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	return read_seqcount_begin(&syncp->seq);
+#else
+#if BITS_PER_LONG==32
+	preempt_disable();
+#endif
+	return 0;
+#endif
 }
 
 static bool inline u64_stats_fetch_retry(const struct u64_stats_sync *syncp,
 					 unsigned int start)
 {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	return read_seqcount_retry(&syncp->seq, start);
-}
-
 #else
-struct u64_stats_sync {
-};
-
-static void inline u64_stats_update_begin(struct u64_stats_sync *syncp)
-{
-}
-
-static void inline u64_stats_update_end(struct u64_stats_sync *syncp)
-{
+#if BITS_PER_LONG==32
+	preempt_enable();
+#endif
+	return false;
+#endif
 }
 
-static unsigned int inline u64_stats_fetch_begin(const struct u64_stats_sync *syncp)
+/*
+ * In case softirq handlers can update u64 counters, readers can use following helpers
+ * - SMP 32bit arches use seqcount protection, irq safe.
+ * - UP 32bit must disable BH.
+ * - 64bit have no problem atomically reading u64 values, irq safe.
+ */
+static unsigned int inline u64_stats_fetch_begin_bh(const struct u64_stats_sync *syncp)
 {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
+	return read_seqcount_begin(&syncp->seq);
+#else
+#if BITS_PER_LONG==32
+	local_bh_disable();
+#endif
 	return 0;
+#endif
 }
 
-static bool inline u64_stats_fetch_retry(const struct u64_stats_sync *syncp,
+static bool inline u64_stats_fetch_retry_bh(const struct u64_stats_sync *syncp,
 					 unsigned int start)
 {
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
+	return read_seqcount_retry(&syncp->seq, start);
+#else
+#if BITS_PER_LONG==32
+	local_bh_enable();
+#endif
 	return false;
-}
 #endif
+}
 
 #endif /* _LINUX_U64_STATS_SYNC_H */



^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH net-next-2.6 2/4] net: u64_stats_fetch_begin_bh() and u64_stats_fetch_retry_bh()
  2010-06-24 10:54 [PATCH net-next-2.6 2/4] net: u64_stats_fetch_begin_bh() and u64_stats_fetch_retry_bh() Eric Dumazet
@ 2010-06-29  6:37 ` David Miller
  0 siblings, 0 replies; 2+ messages in thread
From: David Miller @ 2010-06-29  6:37 UTC (permalink / raw)
  To: eric.dumazet; +Cc: netdev

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 24 Jun 2010 12:54:06 +0200

> - Must disable preemption in case of 32bit UP in u64_stats_fetch_begin()
> and u64_stats_fetch_retry()
> 
> - Add new u64_stats_fetch_begin_bh() and u64_stats_fetch_retry_bh() for
> network usage, disabling BH on 32bit UP only.
> 
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>

Applied.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2010-06-29  6:37 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-06-24 10:54 [PATCH net-next-2.6 2/4] net: u64_stats_fetch_begin_bh() and u64_stats_fetch_retry_bh() Eric Dumazet
2010-06-29  6:37 ` David Miller

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).