Date: Fri, 04 May 2012 11:00:06 -0700 From: Navdeep Parhar <np@FreeBSD.org> To: Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl> Cc: freebsd-hackers@freebsd.org Subject: Re: CPU selection for ithreads on 8.3 Message-ID: <4FA41926.8090904@FreeBSD.org> In-Reply-To: <alpine.BSF.2.00.1205041903530.9968@wojtek.tensor.gdynia.pl> References: <4FA1A421.6020601@FreeBSD.org> <alpine.BSF.2.00.1205041903530.9968@wojtek.tensor.gdynia.pl>
next in thread | previous in thread | raw e-mail | index | archive | help
On 05/04/12 10:05, Wojciech Puchar wrote:
>> cxgbc0}
>> 12 root -68 - 0K 816K WAIT 7 0:31 0.00% intr{irq280: cxgbc0}
>>
>> Back in the day there used to be code in cxgb to bind different
>> interrupts to different CPUs but it was removed because the kernel
>> distributed them across CPUs anyway. So what changed? This appears 8.3
>> specific. I don't see it on head and I don't have a 9 system readily
>
> maybe stupid question but isn't it better to execute same fragment of
> code repetitively on same CPU core and don't mess with cache trashing?
>
> Of course as long as one core can cope with it.
That, and cache trashing is not that big an issue if the threads are
working on their own sets of data. This is precisely the problem that
network stack parallelization tries to solve.
Navdeep
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4FA41926.8090904>