Re: Another (ESP?) scsi blk-mq problem on sparc64

[Jens Axboe <axboe@kernel.dk>]

[-- Attachment #1: Type: text/plain, Size: 1053 bytes --]

On 01/28/2015 11:53 PM, Meelis Roos wrote:
> On Mon, 24 Nov 2014, David Miller wrote:
>
>> From: mroos@linux.ee
>> Date: Tue, 25 Nov 2014 00:23:20 +0200 (EET)
>>
>>>>>>> Yes, that does look like the case.  Do you have a good trick on how
>>>>>>> to allocate a map for the highest possible cpu number without first
>>>>>>> iterating the cpu map?  I couldn't find something that looks like a
>>>>>>> highest_possible_cpu() helper.
>>>>>>
>>>>>> Honestly I think that num_posible_cpus() should return the max of
>>>>>> number of CPUs (weigt), and the highest numbered CPU. It's a pain in
>>>>>> the butt to handle this otherwise.
>>>>>
>>>>> Hear, hear!!!  That would make my life easier, and would make this sort
>>>>> of problem much less likely to occur!
>>>>
>>>> How about this one?
>>>
>>> It make the machine work.
>>
>> Thanks for testing!
>>
>
> What's the status of this fix? It is still not applied on yesterdays
> 3.19.0-rc6-00105-gc59c961 git...

Hmm, I thought commit a33c1ba29138 took care of it... Does the attached 
work?

-- 
Jens Axboe


[-- Attachment #2: map-sz.patch --]
[-- Type: text/x-patch, Size: 644 bytes --]

diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c
index 5f13f4d0bcce..527d315dc1a5 100644
--- a/block/blk-mq-cpumap.c
+++ b/block/blk-mq-cpumap.c
@@ -88,10 +88,11 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues)
 unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set)
 {
 	unsigned int *map;
+	size_t sz;
 
 	/* If cpus are offline, map them to first hctx */
-	map = kzalloc_node(sizeof(*map) * nr_cpu_ids, GFP_KERNEL,
-				set->numa_node);
+	sz = max_t(unsigned int, nr_cpu_ids, num_possible_cpus());
+	map = kzalloc_node(sizeof(*map) * sz, GFP_KERNEL, set->numa_node);
 	if (!map)
 		return NULL;