svn commit: r225892 - head/sys/mips/mips

[Jayachandran C.]

On Thu, Oct 13, 2011 at 9:28 PM, Jayachandran C. <jchandra at freebsd.org> wrote:
> Hi Adrian,
>
> On Tue, Oct 4, 2011 at 9:48 PM, Adrian Chadd <adrian at freebsd.org> wrote:
>> On 4 October 2011 23:48, Adrian Chadd <adrian at freebsd.org> wrote:
>>> Hi all,
>>>
>>> I've tried jc's patch. The hand-wavy, brief summary when tested on my
>>> hostap (mips):
>>>
>>> * when doing single-stream, one way TCP tests (where it thus needs
>>> TX/RX traffic to occur), I get 100% CPU utilisation - 50% interrupt,
>>> 50% system, but the total interrupt rate isn't too high. It's much
>>> higher than without his patch. I'll go digging later to see what's
>>> going on.
>>
>> This turns out to be my fault. I seem to have somehow messed up
>> something in one of my more recent commits.
>> I'm quite certain that it won't have any impact on the results, as I
>> never see the interrupt latency issue with bidirectional traffic.
>>
>> The UDP results are correct though - jc's stuff hasn't entirely fixed
>> it. I'll do some more testing tomorrow morning and report what I find.
>
> I'm thinking of checking in this patch so that it gets wider testing.
> Did you get a chance to do the tests?
>
> The patch does not cause any further problems  for me on XLR/XLP - I
> have been testing it for a few days.  But I don't really have a
> testcase for the issue you see.

If there are no objections, I'm planning to check this in.

 I can keep the two if(!busy) blocks in cpu_idle() commented if you
prefer, and that should give the same behavior as you are seeing now.

JC.