From: Jens Axboe <axboe@kernel.dk>
To: paulmck@linux.vnet.ibm.com
Cc: Christoph Hellwig <hch@infradead.org>,
David Miller <davem@davemloft.net>,
mroos@linux.ee, linux-scsi@vger.kernel.org,
sparclinux@vger.kernel.org
Subject: Re: Another (ESP?) scsi blk-mq problem on sparc64
Date: Mon, 24 Nov 2014 10:16:15 -0700 [thread overview]
Message-ID: <547367DF.5060706@kernel.dk> (raw)
In-Reply-To: <20141124162249.GG5050@linux.vnet.ibm.com>
[-- Attachment #1: Type: text/plain, Size: 1089 bytes --]
On 11/24/2014 09:22 AM, Paul E. McKenney wrote:
> On Mon, Nov 24, 2014 at 08:35:46AM -0700, Jens Axboe wrote:
>> On 11/24/2014 01:21 AM, Christoph Hellwig wrote:
>>> On Fri, Nov 21, 2014 at 02:56:00PM -0500, David Miller wrote:
>>>> I would suggest looking into the possibility that we allocate the memory
>>>> using the count of valid cpus, rather than the largest cpu number.
>>>>
>>>> That's a common error that runs into problems with discontiguous
>>>> cpu numbering like Sparc sometimes has.
>>>
>>> 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?
--
Jens Axboe
[-- Attachment #2: cpu-weight.patch --]
[-- Type: text/x-patch, Size: 862 bytes --]
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 0a9a6da21e74..db21f68aaef0 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -83,7 +83,13 @@ extern const struct cpumask *const cpu_active_mask;
#if NR_CPUS > 1
#define num_online_cpus() cpumask_weight(cpu_online_mask)
-#define num_possible_cpus() cpumask_weight(cpu_possible_mask)
+/*
+ * For platforms with discontig CPU numbering, the weight of the possible
+ * bitmask may be less than the highest numbered CPU. Return the max of
+ * the two, to prevent accidentical bugs.
+ */
+#define num_possible_cpus() \
+ max_t(unsigned int, cpumask_weight(cpu_possible_mask), nr_cpu_ids)
#define num_present_cpus() cpumask_weight(cpu_present_mask)
#define num_active_cpus() cpumask_weight(cpu_active_mask)
#define cpu_online(cpu) cpumask_test_cpu((cpu), cpu_online_mask)
next prev parent reply other threads:[~2014-11-24 17:16 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-14 11:32 Another (ESP?) scsi blk-mq problem on sparc64 Meelis Roos
2014-11-14 16:58 ` Christoph Hellwig
2014-11-14 17:01 ` Jens Axboe
2014-11-14 19:35 ` Meelis Roos
2014-11-14 19:42 ` Jens Axboe
2014-11-14 22:59 ` Meelis Roos
2014-11-14 23:29 ` Jens Axboe
2014-11-15 6:48 ` Meelis Roos
2014-11-15 15:31 ` Jens Axboe
2014-11-20 6:01 ` Christoph Hellwig
2014-11-21 19:56 ` David Miller
2014-11-24 8:21 ` Christoph Hellwig
2014-11-24 15:35 ` Jens Axboe
2014-11-24 16:22 ` Paul E. McKenney
2014-11-24 17:16 ` Jens Axboe [this message]
2014-11-24 17:31 ` Paul E. McKenney
2014-11-24 17:33 ` Jens Axboe
2014-11-24 17:44 ` Paul E. McKenney
2014-11-24 21:56 ` David Miller
2014-11-24 22:01 ` Jens Axboe
2014-11-24 22:09 ` David Miller
2014-11-24 22:20 ` Jens Axboe
2014-11-24 22:23 ` mroos
2014-11-24 22:28 ` David Miller
2015-01-29 7:53 ` Meelis Roos
2015-01-29 16:37 ` Jens Axboe
2015-09-04 8:33 ` Meelis Roos
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=547367DF.5060706@kernel.dk \
--to=axboe@kernel.dk \
--cc=davem@davemloft.net \
--cc=hch@infradead.org \
--cc=linux-scsi@vger.kernel.org \
--cc=mroos@linux.ee \
--cc=paulmck@linux.vnet.ibm.com \
--cc=sparclinux@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).