Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, Jan 14, 2013 at 10:08:41PM -0500, Steven Rostedt wrote:
> On Mon, 2013-01-14 at 11:49 -0800, Paul E. McKenney wrote:
>
> > diff --git a/kernel/rcu.h b/kernel/rcu.h
> > index 7f8e759..b5d30c3 100644
> > --- a/kernel/rcu.h
> > +++ b/kernel/rcu.h
> > @@ -24,6 +24,7 @@
> > #define __LINUX_RCU_H
> >
> > #ifdef CONFIG_RCU_TRACE
> > +#include
> > #define RCU_TRACE(stmt) stmt
> > #else /* #ifdef CONFIG_RCU_TRACE */
> > #define RCU_TRACE(stmt)
> > @@ -118,4 +119,13 @@ int rcu_jiffies_till_stall_check(void);
> >
> > #endif /* #ifdef CONFIG_RCU_STALL_COMMON */
> >
> > +static inline u64 notrace rcu_trace_clock_local(void)
>
> "inline" now implies notrace. That is, we define inline to include
> notrace in the headers. The 'notrace' now is redundant.
Cool!!! I have removed the "notrace", having just barely resisted the
urge to add two more instances of "notrace", just for good measure. ;-)
Thanx, Paul
> -- Steve
>
> > +{
> > +#ifdef CONFIG_RCU_TRACE
> > + return trace_clock_local();
> > +#else /* #ifdef CONFIG_RCU_TRACE */
> > + return 0;
> > +#endif /* #else #ifdef CONFIG_RCU_TRACE */
> > +}
> > +
> > #endif /* __LINUX_RCU_H */
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, 2013-01-14 at 11:49 -0800, Paul E. McKenney wrote:
> diff --git a/kernel/rcu.h b/kernel/rcu.h
> index 7f8e759..b5d30c3 100644
> --- a/kernel/rcu.h
> +++ b/kernel/rcu.h
> @@ -24,6 +24,7 @@
> #define __LINUX_RCU_H
>
> #ifdef CONFIG_RCU_TRACE
> +#include
> #define RCU_TRACE(stmt) stmt
> #else /* #ifdef CONFIG_RCU_TRACE */
> #define RCU_TRACE(stmt)
> @@ -118,4 +119,13 @@ int rcu_jiffies_till_stall_check(void);
>
> #endif /* #ifdef CONFIG_RCU_STALL_COMMON */
>
> +static inline u64 notrace rcu_trace_clock_local(void)
"inline" now implies notrace. That is, we define inline to include
notrace in the headers. The 'notrace' now is redundant.
-- Steve
> +{
> +#ifdef CONFIG_RCU_TRACE
> + return trace_clock_local();
> +#else /* #ifdef CONFIG_RCU_TRACE */
> + return 0;
> +#endif /* #else #ifdef CONFIG_RCU_TRACE */
> +}
> +
> #endif /* __LINUX_RCU_H */
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, Jan 14, 2013 at 10:49:05AM -0800, Randy Dunlap wrote:
> On 01/14/13 09:32, Paul E. McKenney wrote:
> > On Mon, Jan 14, 2013 at 08:45:06AM -0800, Randy Dunlap wrote:
> >> On 01/07/13 19:53, Paul E. McKenney wrote:
> >>> On Mon, Jan 07, 2013 at 07:36:19PM -0500, Steven Rostedt wrote:
> On Mon, 2013-01-07 at 18:12 -0500, Steven Rostedt wrote:
> > On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote:
> >> Hi Paul,
> >>
> >> On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney"
> >> wrote:
> >>>
> >>> On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote:
>
> on i386 or x86_64:
>
> ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined!
> >>>
> >>> Hello, Randy,
> >>>
> >>> Did your build include the following, also pushed to -next in that
> >>> same
> >>> batch from -rcu? Including Steven Rostedt on CC for his take.
> >>
> >> That commit was certainly in next-20130107.
> >
> > Could be bad config dependencies.
>
> Paul,
>
> You need to also select TRACE_CLOCK if you are going to use it.
> >>>
> >>> Thank you, Steve!
> >>>
> >>> Randy, does the following patch help?
> >>>
> >>> Thanx, Paul
> >>
> >> Hi Paul,
> >>
> >> This patch fixed the Jan. 7 linux-next build errors, but there is
> >> still some similar problem with current linux-next (0114 and late
> >> last week as well).
> >>
> >> ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined!
> >
> > Ah, my bad...
> >
> > Does the following (rough and untested) patch help?
>
> Yes, it helps, but there is still the same problem in
> rcu_torture_reader() -- after that it builds OK.
One of those days, I guess... Thank you for fixing it up.
Please see below for a more civilized fix.
Thanx, Paul
diff --git a/kernel/rcu.h b/kernel/rcu.h
index 7f8e759..b5d30c3 100644
--- a/kernel/rcu.h
+++ b/kernel/rcu.h
@@ -24,6 +24,7 @@
#define __LINUX_RCU_H
#ifdef CONFIG_RCU_TRACE
+#include
#define RCU_TRACE(stmt) stmt
#else /* #ifdef CONFIG_RCU_TRACE */
#define RCU_TRACE(stmt)
@@ -118,4 +119,13 @@ int rcu_jiffies_till_stall_check(void);
#endif /* #ifdef CONFIG_RCU_STALL_COMMON */
+static inline u64 notrace rcu_trace_clock_local(void)
+{
+#ifdef CONFIG_RCU_TRACE
+ return trace_clock_local();
+#else /* #ifdef CONFIG_RCU_TRACE */
+ return 0;
+#endif /* #else #ifdef CONFIG_RCU_TRACE */
+}
+
#endif /* __LINUX_RCU_H */
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index cd4c35d..f28f501 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -46,8 +46,9 @@
#include
#include
#include
-#include
#include
+#include
+#include "rcu.h"
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Paul E. McKenney and Josh Triplett
");
@@ -1051,7 +1052,7 @@ static void rcu_torture_timer(unsigned long unused)
idx = cur_ops->readlock();
completed = cur_ops->completed();
- ts = trace_clock_local();
+ ts = rcu_trace_clock_local();
p = rcu_dereference_check(rcu_torture_current,
rcu_read_lock_bh_held() ||
rcu_read_lock_sched_held() ||
@@ -1122,7 +1123,7 @@ rcu_torture_reader(void *arg)
}
idx = cur_ops->readlock();
completed = cur_ops->completed();
- ts = trace_clock_local();
+ ts = rcu_trace_clock_local();
p = rcu_dereference_check(rcu_torture_current,
rcu_read_lock_bh_held() ||
rcu_read_lock_sched_held() ||
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On 01/14/13 09:32, Paul E. McKenney wrote: > On Mon, Jan 14, 2013 at 08:45:06AM -0800, Randy Dunlap wrote: >> On 01/07/13 19:53, Paul E. McKenney wrote: >>> On Mon, Jan 07, 2013 at 07:36:19PM -0500, Steven Rostedt wrote: On Mon, 2013-01-07 at 18:12 -0500, Steven Rostedt wrote: > On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: >> Hi Paul, >> >> On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" >> wrote: >>> >>> On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: on i386 or x86_64: ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! >>> >>> Hello, Randy, >>> >>> Did your build include the following, also pushed to -next in that same >>> batch from -rcu? Including Steven Rostedt on CC for his take. >> >> That commit was certainly in next-20130107. > > Could be bad config dependencies. Paul, You need to also select TRACE_CLOCK if you are going to use it. >>> >>> Thank you, Steve! >>> >>> Randy, does the following patch help? >>> >>> Thanx, Paul >> >> Hi Paul, >> >> This patch fixed the Jan. 7 linux-next build errors, but there is >> still some similar problem with current linux-next (0114 and late >> last week as well). >> >> ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > Ah, my bad... > > Does the following (rough and untested) patch help? Yes, it helps, but there is still the same problem in rcu_torture_reader() -- after that it builds OK. Thanks. > Thanx, Paul > > > > diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c > index cd4c35d..36f0106 100644 > --- a/kernel/rcutorture.c > +++ b/kernel/rcutorture.c > @@ -1047,11 +1047,13 @@ static void rcu_torture_timer(unsigned long unused) > static DEFINE_SPINLOCK(rand_lock); > struct rcu_torture *p; > int pipe_count; > - unsigned long long ts; > + unsigned long long ts = 0; > > idx = cur_ops->readlock(); > completed = cur_ops->completed(); > +#ifdef CONFIG_RCU_TRACE > ts = trace_clock_local(); > +#endif /* #ifdef CONFIG_RCU_TRACE */ > p = rcu_dereference_check(rcu_torture_current, > rcu_read_lock_bh_held() || > rcu_read_lock_sched_held() || > > -- -- ~Randy -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, Jan 14, 2013 at 08:45:06AM -0800, Randy Dunlap wrote: > On 01/07/13 19:53, Paul E. McKenney wrote: > > On Mon, Jan 07, 2013 at 07:36:19PM -0500, Steven Rostedt wrote: > >> On Mon, 2013-01-07 at 18:12 -0500, Steven Rostedt wrote: > >>> On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: > Hi Paul, > > On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" > wrote: > > > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: > >> > >> on i386 or x86_64: > >> > >> ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > > > Hello, Randy, > > > > Did your build include the following, also pushed to -next in that same > > batch from -rcu? Including Steven Rostedt on CC for his take. > > That commit was certainly in next-20130107. > >>> > >>> Could be bad config dependencies. > >> > >> Paul, > >> > >> You need to also select TRACE_CLOCK if you are going to use it. > > > > Thank you, Steve! > > > > Randy, does the following patch help? > > > > Thanx, Paul > > Hi Paul, > > This patch fixed the Jan. 7 linux-next build errors, but there is > still some similar problem with current linux-next (0114 and late > last week as well). > > ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! Ah, my bad... Does the following (rough and untested) patch help? Thanx, Paul diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index cd4c35d..36f0106 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -1047,11 +1047,13 @@ static void rcu_torture_timer(unsigned long unused) static DEFINE_SPINLOCK(rand_lock); struct rcu_torture *p; int pipe_count; - unsigned long long ts; + unsigned long long ts = 0; idx = cur_ops->readlock(); completed = cur_ops->completed(); +#ifdef CONFIG_RCU_TRACE ts = trace_clock_local(); +#endif /* #ifdef CONFIG_RCU_TRACE */ p = rcu_dereference_check(rcu_torture_current, rcu_read_lock_bh_held() || rcu_read_lock_sched_held() || -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On 01/07/13 19:53, Paul E. McKenney wrote: > On Mon, Jan 07, 2013 at 07:36:19PM -0500, Steven Rostedt wrote: >> On Mon, 2013-01-07 at 18:12 -0500, Steven Rostedt wrote: >>> On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: Hi Paul, On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" wrote: > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: >> >> on i386 or x86_64: >> >> ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > Hello, Randy, > > Did your build include the following, also pushed to -next in that same > batch from -rcu? Including Steven Rostedt on CC for his take. That commit was certainly in next-20130107. >>> >>> Could be bad config dependencies. >> >> Paul, >> >> You need to also select TRACE_CLOCK if you are going to use it. > > Thank you, Steve! > > Randy, does the following patch help? Yes, that's good. Acked-by: Randy Dunlap Thanks. > Thanx, Paul > > > > diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug > index ce75d3b..b0fe7bd 100644 > --- a/lib/Kconfig.debug > +++ b/lib/Kconfig.debug > @@ -1010,6 +1010,7 @@ config RCU_CPU_STALL_INFO > config RCU_TRACE > bool "Enable tracing for RCU" > depends on DEBUG_KERNEL > + select TRACE_CLOCK > help > This option provides tracing in RCU which presents stats > in debugfs for debugging RCU implementation. > > -- -- ~Randy -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, Jan 07, 2013 at 07:36:19PM -0500, Steven Rostedt wrote: > On Mon, 2013-01-07 at 18:12 -0500, Steven Rostedt wrote: > > On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: > > > Hi Paul, > > > > > > On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" > > > wrote: > > > > > > > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: > > > > > > > > > > on i386 or x86_64: > > > > > > > > > > ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > > > > > > > Hello, Randy, > > > > > > > > Did your build include the following, also pushed to -next in that same > > > > batch from -rcu? Including Steven Rostedt on CC for his take. > > > > > > That commit was certainly in next-20130107. > > > > Could be bad config dependencies. > > Paul, > > You need to also select TRACE_CLOCK if you are going to use it. Thank you, Steve! Randy, does the following patch help? Thanx, Paul diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index ce75d3b..b0fe7bd 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1010,6 +1010,7 @@ config RCU_CPU_STALL_INFO config RCU_TRACE bool "Enable tracing for RCU" depends on DEBUG_KERNEL + select TRACE_CLOCK help This option provides tracing in RCU which presents stats in debugfs for debugging RCU implementation. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, 2013-01-07 at 18:12 -0500, Steven Rostedt wrote: > On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: > > Hi Paul, > > > > On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" > > wrote: > > > > > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: > > > > > > > > on i386 or x86_64: > > > > > > > > ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > > > > > Hello, Randy, > > > > > > Did your build include the following, also pushed to -next in that same > > > batch from -rcu? Including Steven Rostedt on CC for his take. > > > > That commit was certainly in next-20130107. > > > > Could be bad config dependencies. > Paul, You need to also select TRACE_CLOCK if you are going to use it. -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Mon, Jan 07, 2013 at 06:31:41PM -0500, Steven Rostedt wrote: > On Tue, 2013-01-08 at 10:24 +1100, Stephen Rothwell wrote: > > > > Stephen, can you send me the config that gave you that error. > > > > It was Randy and the config was in the mail the Paul cc'd you on. > > > > Ah, I didn't read that far. I'm use to just cutting off Paul's emails > short anyway. He never trims them ;-) And if I had trimmed this one, I would have had to resend anyway. ;-) Thanx, Paul -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
On Tue, 2013-01-08 at 10:24 +1100, Stephen Rothwell wrote: > > Stephen, can you send me the config that gave you that error. > > It was Randy and the config was in the mail the Paul cc'd you on. > Ah, I didn't read that far. I'm use to just cutting off Paul's emails short anyway. He never trims them ;-) -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
Hi Steve, On Mon, 07 Jan 2013 18:12:07 -0500 Steven Rostedt wrote: > > On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: > > > > On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" > > wrote: > > > > > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: > > > > > > > > on i386 or x86_64: > > > > > > > > ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > > > > > Hello, Randy, > > > > > > Did your build include the following, also pushed to -next in that same > > > batch from -rcu? Including Steven Rostedt on CC for his take. > > > > That commit was certainly in next-20130107. > > > > Could be bad config dependencies. > > Stephen, can you send me the config that gave you that error. It was Randy and the config was in the mail the Paul cc'd you on. -- Cheers, Stephen Rothwells...@canb.auug.org.au pgpBghO3P44Vi.pgp Description: PGP signature
Re: linux-next: Tree for Jan 7 (rcutorture)
On Tue, 2013-01-08 at 09:59 +1100, Stephen Rothwell wrote: > Hi Paul, > > On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" > wrote: > > > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: > > > > > > on i386 or x86_64: > > > > > > ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > > > Hello, Randy, > > > > Did your build include the following, also pushed to -next in that same > > batch from -rcu? Including Steven Rostedt on CC for his take. > > That commit was certainly in next-20130107. > Could be bad config dependencies. Stephen, can you send me the config that gave you that error. Thanks, -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: linux-next: Tree for Jan 7 (rcutorture)
Hi Paul, On Mon, 7 Jan 2013 14:16:27 -0800 "Paul E. McKenney" wrote: > > On Mon, Jan 07, 2013 at 11:42:36AM -0800, Randy Dunlap wrote: > > > > on i386 or x86_64: > > > > ERROR: "trace_clock_local" [kernel/rcutorture.ko] undefined! > > Hello, Randy, > > Did your build include the following, also pushed to -next in that same > batch from -rcu? Including Steven Rostedt on CC for his take. That commit was certainly in next-20130107. > > > tracing: Export trace_clock_local() > > The rcutorture tests need to be able to trace the time of the > beginning of an RCU read-side critical section, and thus need access > to trace_clock_local(). This commit therefore adds a the needed > EXPORT_SYMBOL_GPL(). > > Signed-off-by: Paul E. McKenney > Reviewed-by: Josh Triplett -- Cheers, Stephen Rothwells...@canb.auug.org.au pgpyspIZk3A7u.pgp Description: PGP signature
