Skip to contentSkip to conversation

fix: bump tests thread limit to 256

PATCH In review Michael Jeanson via lttng-dev 1 participant 4 messages Dec 9, 2020 – Dec 11, 2020

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 fix: bump tests thread limit to 256

Machines with more than 128 CPUs are becomming more common, the proper fix here would be to dynamically allocate the array which we will do, but in the meantime bump the limit to 256 to fix the problem on a 160 CPUs ppc64el system where this was reported.

Cc Paul E. McKenney <[email protected]>
Change-Id Ib3cb5d8cb4515e6f626be33c2685fa38cb081782
---
 tests/common/api.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tests/common/api.h +1 -1
index 2b72ec5..b15e588 100644
@@ -108,7 +108,7 @@ static void spin_unlock(spinlock_t *sp)
 
 typedef pthread_t thread_id_t;
 
-#define NR_THREADS 128
+#define NR_THREADS 256
 
 #define __THREAD_ID_MAP_EMPTY ((thread_id_t) 0)
 #define __THREAD_ID_MAP_WAITING ((thread_id_t) 1)
Signature · 6 lines
2.29.2

_______________________________________________
lttng-dev mailing list
[email protected]
https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
#1 lore raw

Hi Paul,

Should I merge this temporary fix for liburcu tests, or should we go for dynamic allocation of the array right away instead ?

Thanks,

Mathieu

Quoted 27 lines · ----- On Dec 9, 2020, at 1:15 PM, Michael Jeanson [email protected] wrote:

Machines with more than 128 CPUs are becomming more common, the proper fix here would be to dynamically allocate the array which we will do, but in the meantime bump the limit to 256 to fix the problem on a 160 CPUs ppc64el system where this was reported.

Cc Paul E. McKenney <[email protected]>
Change-Id Ib3cb5d8cb4515e6f626be33c2685fa38cb081782

---
tests/common/api.h | 2 +-

1 file changed, 1 insertion(+), 1 deletion(-)
tests/common/api.h +1 -2
index 2b72ec5..b15e588 100644
@@ -108,7 +108,7 @@ static void spin_unlock(spinlock_t *sp)
 
 typedef pthread_t thread_id_t;
 
-#define NR_THREADS 128
+#define NR_THREADS 256
 
 #define __THREAD_ID_MAP_EMPTY ((thread_id_t) 0)
 #define __THREAD_ID_MAP_WAITING ((thread_id_t) 1)
--

2.29.2

Signature · 7 lines
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
_______________________________________________
lttng-dev mailing list
[email protected]
https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
#2 lore raw
Quoted 4 lines · Mathieu Desnoyers

Hi Paul,

Should I merge this temporary fix for liburcu tests, or should we go for dynamic allocation of the array right away instead ?

Getting something running now is a good thing. I have occasional access to a system that could use 512, though. (448 suffices, but powers of two and all that.)

Longer term, I agree with dynamic allocation.

Thanx, Paul

Quoted 26 lines

Thanks,

Mathieu

Quoted 20 lines · ----- On Dec 9, 2020, at 1:15 PM, Michael Jeanson [email protected] wrote:

Machines with more than 128 CPUs are becomming more common, the proper fix here would be to dynamically allocate the array which we will do, but in the meantime bump the limit to 256 to fix the problem on a 160 CPUs ppc64el system where this was reported.

Cc Paul E. McKenney <[email protected]>
Change-Id Ib3cb5d8cb4515e6f626be33c2685fa38cb081782

---
tests/common/api.h | 2 +-

1 file changed, 1 insertion(+), 1 deletion(-)
tests/common/api.h +0 -0
index 2b72ec5..b15e588 100644
@@ -108,7 +108,7 @@ static void spin_unlock(spinlock_t *sp)
 
 typedef pthread_t thread_id_t;
 
Quoted 12 lines
Quoted 7 lines
tests/common/api.h +1 -2
-#define NR_THREADS 128
+#define NR_THREADS 256
 
 #define __THREAD_ID_MAP_EMPTY ((thread_id_t) 0)
 #define __THREAD_ID_MAP_WAITING ((thread_id_t) 1)
--

2.29.2

Signature · 3 lines
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
[email protected]
https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

#3 lore raw
In reply to Mathieu Desnoyers via lttng-dev

Hi Michael,

After discussion with Paul, we can go for bumping the max nr cpu limit of this test to 4096.

1536 would cover all the current hardware Paul E. McKenney is aware of. 4096 would cover all hardware he has ever received a bug report on.

Can you try it out and submit an updated patch ?

Thanks,

Mathieu

Quoted 43 lines · ----- On Dec 9, 2020, at 1:29 PM, Mathieu Desnoyers…

Hi Paul,

Should I merge this temporary fix for liburcu tests, or should we go for dynamic allocation of the array right away instead ?

Thanks,

Mathieu

Quoted 26 lines · ----- On Dec 9, 2020, at 1:15 PM, Michael Jeanson [email protected] wrote:

Machines with more than 128 CPUs are becomming more common, the proper fix here would be to dynamically allocate the array which we will do, but in the meantime bump the limit to 256 to fix the problem on a 160 CPUs ppc64el system where this was reported.

Cc Paul E. McKenney <[email protected]>
Change-Id Ib3cb5d8cb4515e6f626be33c2685fa38cb081782

---
tests/common/api.h | 2 +-

1 file changed, 1 insertion(+), 1 deletion(-)
tests/common/api.h +1 -2
index 2b72ec5..b15e588 100644
@@ -108,7 +108,7 @@ static void spin_unlock(spinlock_t *sp)
 
 typedef pthread_t thread_id_t;
 
-#define NR_THREADS 128
+#define NR_THREADS 256
 
 #define __THREAD_ID_MAP_EMPTY ((thread_id_t) 0)
 #define __THREAD_ID_MAP_WAITING ((thread_id_t) 1)
--
Quoted 1 line

2.29.2

--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

Signature · 7 lines
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
_______________________________________________
lttng-dev mailing list
[email protected]
https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev