lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

itimers: remove the per-cpu-ish-ness

Browse Source 
Either we bounce once cacheline per cpu per tick, yielding n^2 bounces
or we just bounce a single..

Also, using per-cpu allocations for the thread-groups complicates the
per-cpu allocator in that its currently aimed to be a fixed sized
allocator and the only possible extention to that would be vmap based,
which is seriously constrained on 32 bit archs.

So making the per-cpu memory requirement depend on the number of
processes is an issue.

Lastly, it didn't deal with cpu-hotplug, although admittedly that might
be fixable.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Peter Zijlstra
2008-11-24 17:06:57 +01:00
committed by Ingo Molnar
parent ede6f5aea0
commit 490dea45d0
5 changed files with 46 additions and 107 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
kernel/sched_stats.h
View File

@@ -296,6 +296,7 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next)
static inline void account_group_user_time(struct task_struct *tsk,
cputime_t cputime)
{
struct task_cputime *times;
struct signal_struct *sig;
/* tsk == current, ensure it is safe to use ->signal */
@@ -303,13 +304,11 @@ static inline void account_group_user_time(struct task_struct *tsk,
return;
sig = tsk->signal;
if (sig->cputime.totals) {
struct task_cputime *times;
times = &sig->cputime.totals;
times = per_cpu_ptr(sig->cputime.totals, get_cpu());
times->utime = cputime_add(times->utime, cputime);
put_cpu_no_resched();
}
spin_lock(&times->lock);
times->utime = cputime_add(times->utime, cputime);
spin_unlock(&times->lock);
}
/**
@@ -325,6 +324,7 @@ static inline void account_group_user_time(struct task_struct *tsk,
static inline void account_group_system_time(struct task_struct *tsk,
cputime_t cputime)
{
struct task_cputime *times;
struct signal_struct *sig;
/* tsk == current, ensure it is safe to use ->signal */
@@ -332,13 +332,11 @@ static inline void account_group_system_time(struct task_struct *tsk,
return;
sig = tsk->signal;
if (sig->cputime.totals) {
struct task_cputime *times;
times = &sig->cputime.totals;
times = per_cpu_ptr(sig->cputime.totals, get_cpu());
times->stime = cputime_add(times->stime, cputime);
put_cpu_no_resched();
}
spin_lock(&times->lock);
times->stime = cputime_add(times->stime, cputime);
spin_unlock(&times->lock);
}
/**
@@ -354,6 +352,7 @@ static inline void account_group_system_time(struct task_struct *tsk,
static inline void account_group_exec_runtime(struct task_struct *tsk,
unsigned long long ns)
{
struct task_cputime *times;
struct signal_struct *sig;
sig = tsk->signal;
@@ -362,11 +361,9 @@ static inline void account_group_exec_runtime(struct task_struct *tsk,
if (unlikely(!sig))
return;
if (sig->cputime.totals) {
struct task_cputime *times;
times = &sig->cputime.totals;
times = per_cpu_ptr(sig->cputime.totals, get_cpu());
times->sum_exec_runtime += ns;
put_cpu_no_resched();
}
spin_lock(&times->lock);
times->sum_exec_runtime += ns;
spin_unlock(&times->lock);
}