Re: [PATCH] sched: try to fix mm leak when freeing sched groups

[Hillf Danton <dhillf@gmail.com>]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=UTF-8, Size: 2381 bytes --]

On Thu, Apr 21, 2011 at 9:44 PM, Peter Zijlstra <peterz@infradead.org> wrote:
> On Thu, 2011-04-21 at 21:37 +0800, Hillf Danton wrote:
>> When freeing sched groups, there is a sg (== sched_group_nodes[i]) not
>> got freed.
>> The relevant code is reshuffled to be clearer, and correct.
>
> The code doesn't even appear to exist, what kernel are you patching?
>
Hi Peter

It is linux-2.6.39-rc1.tar.bz2 downloaded at kernel.org, and
          >> --- a/kernel/sched.c  2011-03-30 03:09:48.000000000 +0800
is the time stamp, right?

thanks
          Hillf

>> Signed-off-by: Hillf Danton <dhillf@gmail.com>
>> ---
>>
>> --- a/kernel/sched.c  2011-03-30 03:09:48.000000000 +0800
>> +++ b/kernel/sched.c  2011-04-21 21:09:00.000000000 +0800
>> @@ -7106,7 +7106,7 @@ static void free_sched_groups(const stru
>>                       continue;
>>
>>               for (i = 0; i < nr_node_ids; i++) {
>> -                     struct sched_group *oldsg, *sg = sched_group_nodes[i];
>> +                     struct sched_group *sg = sched_group_nodes[i];
>>
>>                       cpumask_and(nodemask, cpumask_of_node(i), cpu_map);
>>                       if (cpumask_empty(nodemask))
>> @@ -7114,13 +7114,11 @@ static void free_sched_groups(const stru
>>
>>                       if (sg == NULL)
>>                               continue;
>> -                     sg = sg->next;
>> -next_sg:
>> -                     oldsg = sg;
>> -                     sg = sg->next;
>> -                     kfree(oldsg);
>> -                     if (oldsg != sched_group_nodes[i])
>> -                             goto next_sg;
>> +                     do {
>> +                             struct sched_group *next = sg->next;
>> +                             kfree(sg);
>> +                             sg = next;
>> +                     } while (sg != sched_group_nodes[i]);
>>               }
>>               kfree(sched_group_nodes);
>>               sched_group_nodes_bycpu[cpu] = NULL;
>
>
>
>
ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥