lkml.org 
[lkml]   [2009]   [Jul]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH] generic-ipi: make struct call_function_data lockless
On Fri, 24 Jul 2009 17:50:16 +0800
Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> wrote:

> This patch can remove spinlock from struct call_function_data, the
> reasons are below:
>
> 1: add a new interface for cpumask named cpumask_test_and_clear_cpu(),
> it can atomically test and clear specific cpu, we can use it instead
> of cpumask_test_cpu() and cpumask_clear_cpu() and no need data->lock
> to protect those in generic_smp_call_function_interrupt().
>
> 2: in smp_call_function_many(), after csd_lock() return, the current's
> cfd_data is deleted from call_function list, so it not have race
> between other cpus, then cfs_data is only used in
> smp_call_function_many() that must disable preemption and not from
> a hardware interrupthandler or from a bottom half handler to call,
> only the correspond cpu can use it, so it not have race in current
> cpu, no need cfs_data->lock to protect it.
>
> 3: after 1 and 2, cfs_data->lock is only use to protect cfs_data->refs in
> generic_smp_call_function_interrupt(), so we can define cfs_data->refs
> to atomic_t, and no need cfs_data->lock any more.

Looks good to me. One tiny cleanup:

--- a/kernel/smp.c~generic-ipi-make-struct-call_function_data-lockless-cleanup
+++ a/kernel/smp.c
@@ -193,7 +193,7 @@ void generic_smp_call_function_interrupt

data->csd.func(data->csd.info);

- refs = atomic_sub_return(1, &data->refs);
+ refs = atomic_dec_return(&data->refs);
WARN_ON(refs < 0);
if (!refs) {
spin_lock(&call_function.lock);
_



\
 
 \ /
  Last update: 2009-07-28 00:03    [W:0.132 / U:0.344 seconds]
©2003-2014 Jasper Spaans. Advertise on this site