Re: [PATCH][2.6.7-rc3-mm1] perfctr cpumask cleanup

[Paul Jackson <pj@sgi.com>]

Or ... pushing the point further ... one _could_ remove the old_mask as
well.  However I think this makes the code less clear, even if it does
save a stack copy of a cpumask_t.  So I'm of mixed feelings on this
patch, edging slightly toward negative.  Same utter lack of testing as
the previous patch.

Signed-off-by: Paul Jackson <pj@sgi.com>

Index: 2.6.7-rc3-mm1/drivers/perfctr/virtual.c
===================================================================
--- 2.6.7-rc3-mm1.orig/drivers/perfctr/virtual.c	2004-06-09 15:38:32.000000000 -0700
+++ 2.6.7-rc3-mm1/drivers/perfctr/virtual.c	2004-06-09 15:53:06.000000000 -0700
@@ -403,14 +403,14 @@
 		return -EFAULT;
 
 	if (control.cpu_control.nractrs || control.cpu_control.nrictrs) {
-		cpumask_t old_mask, new_mask;
+		cpumask_t new_mask;
 
-		old_mask = tsk->cpus_allowed;
-		cpus_andnot(new_mask, old_mask, perfctr_cpus_forbidden_mask);
+		cpus_andnot(new_mask, tsk->cpus_allowed,
+						perfctr_cpus_forbidden_mask);
 
 		if (cpus_empty(new_mask))
 			return -EINVAL;
-		if (!cpus_equal(new_mask, old_mask))
+		if (!cpus_equal(tsk->cpus_allowed, new_mask))
 			set_cpus_allowed(tsk, new_mask);
 	}
 


-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj@sgi.com> 1.650.933.1373