[PATCH v2] complete conversion set_bit() -> __cpumask_set_cpu() by 4aaca0e9cd

[PATCH v2] complete conversion set_bit() -> __cpumask_set_cpu() by 4aaca0e9cd

[Jan Beulich]

[-- Attachment #1: Type: text/plain, Size: 926 bytes --]

While converting to __cpumask_set_cpu() was correct, the first argument
passed should have been corrected to be "cpu" instead of "nr" at once.
The wrong construct results in problems on systems with relatively few
CPUs.

Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
---
v2: As Andrew points out the original code was wrong too, and hence
    the use of __cpumask_set_cpu) was correct. What was wrong was the
    use of "nr" as the bit position.

--- a/xen/common/softirq.c
+++ b/xen/common/softirq.c
@@ -106,7 +106,7 @@ void cpu_raise_softirq(unsigned int cpu,
     if ( !per_cpu(batching, this_cpu) || in_irq() )
         smp_send_event_check_cpu(cpu);
     else
-        __cpumask_set_cpu(nr, &per_cpu(batch_mask, this_cpu));
+        __cpumask_set_cpu(cpu, &per_cpu(batch_mask, this_cpu));
 }
 
 void cpu_raise_softirq_batch_begin(void)




[-- Attachment #2: cpumask-relax-fix.patch --]
[-- Type: text/plain, Size: 990 bytes --]

complete conversion set_bit() -> __cpumask_set_cpu() by 4aaca0e9cd

While converting to __cpumask_set_cpu() was correct, the first argument
passed should have been corrected to be "cpu" instead of "nr" at once.
The wrong construct results in problems on systems with relatively few
CPUs.

Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
---
v2: As Andrew points out the original code was wrong too, and hence
    the use of __cpumask_set_cpu) was correct. What was wrong was the
    use of "nr" as the bit position.

--- a/xen/common/softirq.c
+++ b/xen/common/softirq.c
@@ -106,7 +106,7 @@ void cpu_raise_softirq(unsigned int cpu,
     if ( !per_cpu(batching, this_cpu) || in_irq() )
         smp_send_event_check_cpu(cpu);
     else
-        __cpumask_set_cpu(nr, &per_cpu(batch_mask, this_cpu));
+        __cpumask_set_cpu(cpu, &per_cpu(batch_mask, this_cpu));
 }
 
 void cpu_raise_softirq_batch_begin(void)

[-- Attachment #3: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

Re: [PATCH v2] complete conversion set_bit() -> __cpumask_set_cpu() by 4aaca0e9cd

[Andrew Cooper]

On 23/02/15 13:47, Jan Beulich wrote:
> While converting to __cpumask_set_cpu() was correct, the first argument
> passed should have been corrected to be "cpu" instead of "nr" at once.
> The wrong construct results in problems on systems with relatively few
> CPUs.
>
> Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Looks better.

Reviewed-by: Andrew Cooper <andrew.cooper3@citirx.com>

> ---
> v2: As Andrew points out the original code was wrong too, and hence
>     the use of __cpumask_set_cpu) was correct. What was wrong was the
>     use of "nr" as the bit position.
>
> --- a/xen/common/softirq.c
> +++ b/xen/common/softirq.c
> @@ -106,7 +106,7 @@ void cpu_raise_softirq(unsigned int cpu,
>      if ( !per_cpu(batching, this_cpu) || in_irq() )
>          smp_send_event_check_cpu(cpu);
>      else
> -        __cpumask_set_cpu(nr, &per_cpu(batch_mask, this_cpu));
> +        __cpumask_set_cpu(cpu, &per_cpu(batch_mask, this_cpu));
>  }
>  
>  void cpu_raise_softirq_batch_begin(void)
>
>
>