AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:
CFG:
PT:0
S:andy.sharp@lsi.com
RQ:
SSV:mhbs.lsil.com
NSV:
SSH:
R:<Maxim.Kozlovsky@lsi.com>,<Bill.Fisher@lsi.com>,<Rendell.Fong@lsi.com>,<Brian.Stark@lsi.com>
MAID:2
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/LSI/INBOX	0	861DA0537719934884B3D30A2666FECC94294859@cosmail02.lsi.com
X-Sylpheed-End-Special-Headers: 1
Date: Thu, 17 Sep 2009 18:43:42 -0700
From: Andrew Sharp <andy.sharp@lsi.com>
To: "Kozlovsky, Maxim" <Maxim.Kozlovsky@lsi.com>
Cc: "Fisher, Bill" <Bill.Fisher@lsi.com>, "Fong, Rendell"
 <Rendell.Fong@lsi.com>, "Stark, Brian" <Brian.Stark@lsi.com>
Subject: Re: ACPU Thread Discussion Meeting
Message-ID: <20090917184342.2d1a4e2c@ripper.onstor.net>
In-Reply-To: <861DA0537719934884B3D30A2666FECC94294859@cosmail02.lsi.com>
References: <4AB2CD55.8040508@lsi.com>
	<861DA0537719934884B3D30A2666FECC9429484A@cosmail02.lsi.com>
	<4AB2D494.60905@lsi.com>
	<861DA0537719934884B3D30A2666FECC94294859@cosmail02.lsi.com>
Organization: LSI
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Thu, 17 Sep 2009 18:44:38 -0600 "Kozlovsky, Maxim"
<Maxim.Kozlovsky@lsi.com> wrote:

> 
> 
> >-----Original Message-----
> >From: William Fisher [mailto:bfisher@lsi.com]
> >Sent: Thursday, September 17, 2009 5:30 PM
> >To: Kozlovsky, Maxim
> >Cc: Fisher, Bill; Sharp, Andy; Fong, Rendell; Stark, Brian
> >Subject: Re: ACPU Thread Discussion Meeting
> >
> >Kozlovsky, Maxim wrote:
> >> Hello,
> >>
> >> There is no need for a meeting at this point, I'll write up some
> >> design
> >when I start working on it, which will happen after there is a
> >schedule. Then we'll have the design review meeting if it will not
> >be possible to resolve by e-mail.
> >>
> >ok, but I need something for the NCPU ASAP. I will have to hack
> >something out to start testing the TPL code with
> >the NETBIOS service.
> 
> [MK] 
> If you need it ASAP -
> The NCPU and ACPU threads should be just normal Linux kernel threads.
> Using fs thread api for either NCPU thread or for ACPU thread is not
> appropriate. Both NCPU and ACPU threads should process the input
> queues, and go to sleep if there is nothing to do in either of them.
> Whenever the event is added to the empty queue, there should be a
> wakeup. Some appropriate synchronization should be in place to avoid
> lost wakeups. There is not much else to do here. There is a worker
> thread API in Linux that may be adoptable to this purpose, but it may
> be too heavy weight, I am not familiar enough with it to tell right
> away.

Just use the standard event/queue handling API in the kernel.  It will
do all of the above mentioned things for you.
