2.6.25.4-rt6: missing global_rt_runtime() in sched.c
PATCH 1/1 2.6.25.4-rt6: missing global_rt_runtime() in sched.c
I already sent this patch, but seems that nobody cared... the last 2
revisions of the -rt kernel don't compile properly.
Second patch to fix this.
Regards
Giovanni "ciaby" Civardi
diff -uNr linux-2.6.25.4-rt6.orig/kernel/sched.c
linux-2.6.25.4-rt6/kernel/sched.c
--- linux-2.6.25.4-rt6.orig/kernel/sched.c 2008-06-09
11:35:57.000000000 +0200
+++ linux-2.6.25.4-rt6/kernel/sched.c 2008-06-09 11:40:45.000000000 +0200
@@ -672,6 +672,13 @@
*/
#define RUNTIME_INF ((u64)~0ULL)
+static u64 global_rt_runtime(void)
+{
+ if (sysctl_sched_rt_period < 0)
+ return RUNTIME_INF;
+ return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC;
+}
+
/*
* We really dont want to do anything complex within switch_to()
* on PREEMPT_RT - this check enforces this.
A message that is not in the archive (sent off-list or not yet received): [email protected]
Probably because you don't use Group Scheduling:
--
#ifdef CONFIG_RT_GROUP_SCHED
init_task_group.rt_runtime = global_rt_runtime();
INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
init_tg_rt_entry(rq, &init_task_group,
&per_cpu(init_rt_rq, i),
&per_cpu(init_sched_rt_entity, i), i, 1);#endif
--
If you enable it, it tries to initialize the init_task_group.rt_runtime,
but the global_rt_runtime() function has been lost somewhere between -rt3
and -rt4. I sent the patch again, i hope is gonna be in the next -rt
release.
Cheers
Giovanni "ciaby" Civardi
Quoted 147 lines · Javier Sanz
Umm,
I think that -rt6 compile ok without you patch, for me, at least ...
Regards
J
2008/6/9 ciaby <[email protected]>:
Quoted 42 lines
I already sent this patch, but seems that nobody cared... the last 2
revisions of the -rt kernel don't compile properly.
Second patch to fix this.
Regards
Giovanni "ciaby" Civardi
diff -uNr
linux-2.6.25.4-rt6.orig/kernel/sched.c
linux-2.6.25.4-rt6/kernel/sched.c---
linux-2.6.25.4-rt6.orig/kernel/sched.c2008-06-09Quoted 5 lines
11:35:57.000000000 +0200
+++
linux-2.6.25.4-rt6/kernel/sched.c2008-06-09 11:40:45.000000000+0200
Quoted 64 lines
(quoted diff) +7
-1@@ -672,6 +672,13 @@ */ #define RUNTIME_INF ((u64)~0ULL) +static u64 global_rt_runtime(void) +{ + if (sysctl_sched_rt_period < 0) + return RUNTIME_INF; + return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC; +} + /* * We really dont want to do anything complex within switch_to() * on PREEMPT_RT - this check enforces this. --Quoted 1 line
To unsubscribe from this list: send the line "unsubscribe
linux-rt-users"
Quoted 7 lines
in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 7 lines · ciaby
I already sent this patch, but seems that nobody cared... the last 2 revisions of the -rt kernel don't compile properly. Second patch to fix this.
RegardsGiovanni "ciaby" Civardi
Thanks for the patch. It worked perfectly for me.
Quoted 13 lines · ciaby
I already sent this patch, but seems that nobody cared... the last 2 revisions of the -rt kernel don't compile properly. Second patch to fix this.
RegardsGiovanni "ciaby" Civardi
diff -uNr
linux-2.6.25.4-rt6.orig/kernel/sched.clinux-2.6.25.4-rt6/kernel/sched.c
---linux-2.6.25.4-rt6.orig/kernel/sched.c2008-06-09
11:35:57.000000000 +0200
+++linux-2.6.25.4-rt6/kernel/sched.c2008-06-09 11:40:45.000000000 +0200
Review line 679
Quoted 8 lines
(quoted diff) +5
-0@@ -672,6 +672,13 @@ */ #define RUNTIME_INF ((u64)~0ULL) +static u64 global_rt_runtime(void) +{ + if (sysctl_sched_rt_period < 0) + return RUNTIME_INF; + return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC;
Your patch has a couple of problems .. It's line wrapped, it's missing proper tabbing (got stripped maybe), and no Signed-off-by: ..
Daniel
PATCH 1/1 2.6.25.4-rt6: missing global_rt_runtime() in sched.c
In reply to ciabyNot-line-wrapped, with proper tabs :)
Cheers
Giovanni 'ciaby' Civardi
Quoted 4 lines · ciaby
Not-line-wrapped, with proper tabs :)
CheersGiovanni 'ciaby' Civardi
No patch tho included here tho ..
Daniel
Quoted 6 lines · ciaby
I already sent this patch, but seems that nobody cared... the last 2 revisions of the -rt kernel don't compile properly. Second patch to fix this.
Regards
I still don't have a "Signed-off-by:" for this patch. Please send it to me ASAP, otherwise it will miss the next release.
-- Steve
Quoted 29 lines
Giovanni "ciaby" Civardi
diff -uNr
linux-2.6.25.4-rt6.orig/kernel/sched.clinux-2.6.25.4-rt6/kernel/sched.c
---linux-2.6.25.4-rt6.orig/kernel/sched.c2008-06-09
11:35:57.000000000 +0200
+++linux-2.6.25.4-rt6/kernel/sched.c2008-06-09 11:40:45.000000000 +0200(quoted diff) +7
-1@@ -672,6 +672,13 @@ */ #define RUNTIME_INF ((u64)~0ULL) +static u64 global_rt_runtime(void) +{ + if (sysctl_sched_rt_period < 0) + return RUNTIME_INF; + return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC; +} + /* * We really dont want to do anything complex within switch_to() * on PREEMPT_RT - this check enforces this. --To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 5 lines · Steven Rostedt
Quoted 2 lines · ciaby
Quoted 8 lines
Quoted 4 lines
I already sent this patch, but seems that nobody cared... the last 2 revisions of the -rt kernel don't compile properly. Second patch to fix this.
RegardsI still don't have a "Signed-off-by:" for this patch. Please send it to me ASAP, otherwise it will miss the next release.
I'm posting out -rt7 now without this patch. I'm hoping to get your Signed-off-by for -rt8 which will come out soon (I want to get Gregory's patches testest ASAP too).
-- Steve