* [PATCH 1/3] route_hazptr: Add dependency to hazptr
2018-11-24 9:34 [PATCH 0/3] Patches for Section Deferred Processing Junchang Wang
@ 2018-11-24 9:34 ` Junchang Wang
2018-11-24 9:34 ` [PATCH 2/3] route_hazptr: Fix a memory leak bug Junchang Wang
` (2 subsequent siblings)
3 siblings, 0 replies; 6+ messages in thread
From: Junchang Wang @ 2018-11-24 9:34 UTC (permalink / raw)
To: paulmck; +Cc: perfbook, Junchang Wang
Target route_hazptr depends on hazptr.c and hazptr.h. So add this dependency to Makefile.
Signed-off-by: Junchang Wang <junchangwang@gmail.com>
---
CodeSamples/defer/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/CodeSamples/defer/Makefile b/CodeSamples/defer/Makefile
index 5a08c1f..270c151 100644
--- a/CodeSamples/defer/Makefile
+++ b/CodeSamples/defer/Makefile
@@ -123,7 +123,7 @@ rcu_sig: rcu_sig.c rcu_sig.h ../api.h rcutorture.h
rcu_ts: rcu_ts.c rcu_ts.h ../api.h rcutorture.h
cc $(GCC_ARGS) -o rcu_ts -DTEST rcu_ts.c -lpthread
-route_hazptr: route_hazptr.c ../api.h ../lib/random.h ../lib/random.c routetorture.h
+route_hazptr: route_hazptr.c hazptr.c hazptr.h ../api.h ../lib/random.h ../lib/random.c routetorture.h
cc $(GCC_ARGS) -o route_hazptr route_hazptr.c hazptr.c ../lib/random.c -lpthread
route_rcu: route_rcu.c ../api.h ../lib/random.h ../lib/random.c routetorture.h
--
2.7.4
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH 2/3] route_hazptr: Fix a memory leak bug
2018-11-24 9:34 [PATCH 0/3] Patches for Section Deferred Processing Junchang Wang
2018-11-24 9:34 ` [PATCH 1/3] route_hazptr: Add dependency to hazptr Junchang Wang
@ 2018-11-24 9:34 ` Junchang Wang
2018-11-24 9:34 ` [PATCH 3/3] defer: Fix a typo Junchang Wang
2018-11-24 16:47 ` [PATCH 0/3] Patches for Section Deferred Processing Paul E. McKenney
3 siblings, 0 replies; 6+ messages in thread
From: Junchang Wang @ 2018-11-24 9:34 UTC (permalink / raw)
To: paulmck; +Cc: perfbook, Junchang Wang
Global pointer gplist is used to record hazard pointer array once it's
allocated. This can avoid allocating and freeing hazptr array each time
function hazptr_scan() is invoked. This patch fixes a memory leak bug and
helps achieve the design goal.
Signed-off-by: Junchang Wang <junchangwang@gmail.com>
---
CodeSamples/defer/hazptr.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/CodeSamples/defer/hazptr.c b/CodeSamples/defer/hazptr.c
index 91df310..061ca2c 100644
--- a/CodeSamples/defer/hazptr.c
+++ b/CodeSamples/defer/hazptr.c
@@ -53,6 +53,9 @@ void hazptr_thread_exit(void)
hazptr_scan();
poll(NULL, 0, 1);
}
+
+ if (gplist != NULL)
+ free(gplist);
}
void hazptr_reinitialize()
@@ -103,6 +106,7 @@ void hazptr_scan()
fprintf(stderr, "hazptr_scan: out of memory\n");
exit(EXIT_FAILURE);
}
+ gplist = plist;
}
/*
--
2.7.4
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 3/3] defer: Fix a typo
2018-11-24 9:34 [PATCH 0/3] Patches for Section Deferred Processing Junchang Wang
2018-11-24 9:34 ` [PATCH 1/3] route_hazptr: Add dependency to hazptr Junchang Wang
2018-11-24 9:34 ` [PATCH 2/3] route_hazptr: Fix a memory leak bug Junchang Wang
@ 2018-11-24 9:34 ` Junchang Wang
2018-11-24 16:47 ` [PATCH 0/3] Patches for Section Deferred Processing Paul E. McKenney
3 siblings, 0 replies; 6+ messages in thread
From: Junchang Wang @ 2018-11-24 9:34 UTC (permalink / raw)
To: paulmck; +Cc: perfbook, Junchang Wang
Signed-off-by: Junchang Wang <junchangwang@gmail.com>
---
defer/hazptr.tex | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/defer/hazptr.tex b/defer/hazptr.tex
index 346146f..09c5f21 100644
--- a/defer/hazptr.tex
+++ b/defer/hazptr.tex
@@ -91,7 +91,7 @@ recorded a hazard pointer for the data element.
It might be inefficient in some sense, but the fact is that
such restarting is absolutely required for correctness.
To see this, consider a hazard-pointer-protected linked list
- containing elements~A, B, and~C that is subjecte to the
+ containing elements~A, B, and~C that is subjected to the
following sequence of events:
\begin{enumerate}
--
2.7.4
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH 0/3] Patches for Section Deferred Processing
2018-11-24 9:34 [PATCH 0/3] Patches for Section Deferred Processing Junchang Wang
` (2 preceding siblings ...)
2018-11-24 9:34 ` [PATCH 3/3] defer: Fix a typo Junchang Wang
@ 2018-11-24 16:47 ` Paul E. McKenney
2018-11-25 2:18 ` Junchang Wang
3 siblings, 1 reply; 6+ messages in thread
From: Paul E. McKenney @ 2018-11-24 16:47 UTC (permalink / raw)
To: Junchang Wang; +Cc: perfbook
On Sat, Nov 24, 2018 at 05:34:51PM +0800, Junchang Wang wrote:
> Hi Paul,
>
> This is the patch sets for Section Deferred Processing. Please take a look.
>
> The first patch fixes a dependency issue in compiling route_hazptr.c.
>
> The second patch fixes a memory leak bug in hazptr.c.
>
> The last one fixes a typo in writing.
>
>
> Thanks,
> --Junchang
Good catches, applied, thank you!
On the memory-leak bug, I removed the "if" due to POSIX specifying
that free() is a no-op when given a NULL pointer.
http://pubs.opengroup.org/onlinepubs/009695399/functions/free.html
Please see below for the updated commit, and please let me know if I
messed anything up.
Thanx, Paul
> Junchang Wang (3):
> route_hazptr: Add dependency to hazptr
> route_hazptr: Fix a memory leak bug
> defer: Fix a typo
>
> CodeSamples/defer/Makefile | 2 +-
> CodeSamples/defer/hazptr.c | 4 ++++
> defer/hazptr.tex | 2 +-
> 3 files changed, 6 insertions(+), 2 deletions(-)
------------------------------------------------------------------------
commit b8c1647f53ddedf10bf2973371f6f685dfd31d62
Author: Junchang Wang <junchangwang@gmail.com>
Date: Sat Nov 24 17:34:53 2018 +0800
route_hazptr: Fix a memory leak bug
Global pointer gplist is used to record hazard pointer array once it's
allocated. This can avoid allocating and freeing hazptr array each time
function hazptr_scan() is invoked. This patch fixes a memory leak bug and
helps achieve the design goal.
Signed-off-by: Junchang Wang <junchangwang@gmail.com>
[ paulmck: Remove redundant "if" because free() ignores NULL argument. ]
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
diff --git a/CodeSamples/defer/hazptr.c b/CodeSamples/defer/hazptr.c
index 91df310e24a2..fdb387e59ea1 100644
--- a/CodeSamples/defer/hazptr.c
+++ b/CodeSamples/defer/hazptr.c
@@ -53,6 +53,8 @@ void hazptr_thread_exit(void)
hazptr_scan();
poll(NULL, 0, 1);
}
+
+ free(gplist);
}
void hazptr_reinitialize()
@@ -103,6 +105,7 @@ void hazptr_scan()
fprintf(stderr, "hazptr_scan: out of memory\n");
exit(EXIT_FAILURE);
}
+ gplist = plist;
}
/*
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH 0/3] Patches for Section Deferred Processing
2018-11-24 16:47 ` [PATCH 0/3] Patches for Section Deferred Processing Paul E. McKenney
@ 2018-11-25 2:18 ` Junchang Wang
0 siblings, 0 replies; 6+ messages in thread
From: Junchang Wang @ 2018-11-25 2:18 UTC (permalink / raw)
To: Paul McKenney; +Cc: perfbook
On Sun, Nov 25, 2018 at 12:47 AM Paul E. McKenney <paulmck@linux.ibm.com> wrote:
>
> On Sat, Nov 24, 2018 at 05:34:51PM +0800, Junchang Wang wrote:
> > Hi Paul,
> >
> > This is the patch sets for Section Deferred Processing. Please take a look.
> >
> > The first patch fixes a dependency issue in compiling route_hazptr.c.
> >
> > The second patch fixes a memory leak bug in hazptr.c.
> >
> > The last one fixes a typo in writing.
> >
> >
> > Thanks,
> > --Junchang
>
> Good catches, applied, thank you!
>
> On the memory-leak bug, I removed the "if" due to POSIX specifying
> that free() is a no-op when given a NULL pointer.
>
> http://pubs.opengroup.org/onlinepubs/009695399/functions/free.html
>
> Please see below for the updated commit, and please let me know if I
> messed anything up.
>
That's right. Thanks a lot for the comment and link :-).
Thanks,
--Junchang
> Thanx, Paul
>
> > Junchang Wang (3):
> > route_hazptr: Add dependency to hazptr
> > route_hazptr: Fix a memory leak bug
> > defer: Fix a typo
> >
> > CodeSamples/defer/Makefile | 2 +-
> > CodeSamples/defer/hazptr.c | 4 ++++
> > defer/hazptr.tex | 2 +-
> > 3 files changed, 6 insertions(+), 2 deletions(-)
>
> ------------------------------------------------------------------------
>
> commit b8c1647f53ddedf10bf2973371f6f685dfd31d62
> Author: Junchang Wang <junchangwang@gmail.com>
> Date: Sat Nov 24 17:34:53 2018 +0800
>
> route_hazptr: Fix a memory leak bug
>
> Global pointer gplist is used to record hazard pointer array once it's
> allocated. This can avoid allocating and freeing hazptr array each time
> function hazptr_scan() is invoked. This patch fixes a memory leak bug and
> helps achieve the design goal.
>
> Signed-off-by: Junchang Wang <junchangwang@gmail.com>
> [ paulmck: Remove redundant "if" because free() ignores NULL argument. ]
> Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
>
> diff --git a/CodeSamples/defer/hazptr.c b/CodeSamples/defer/hazptr.c
> index 91df310e24a2..fdb387e59ea1 100644
> --- a/CodeSamples/defer/hazptr.c
> +++ b/CodeSamples/defer/hazptr.c
> @@ -53,6 +53,8 @@ void hazptr_thread_exit(void)
> hazptr_scan();
> poll(NULL, 0, 1);
> }
> +
> + free(gplist);
> }
>
> void hazptr_reinitialize()
> @@ -103,6 +105,7 @@ void hazptr_scan()
> fprintf(stderr, "hazptr_scan: out of memory\n");
> exit(EXIT_FAILURE);
> }
> + gplist = plist;
> }
>
> /*
>
^ permalink raw reply [flat|nested] 6+ messages in thread