Re: msk(4) require to sync status buffer

[KIYOHARA Takashi <kiyohara%kk.iij4u.or.jp@localhost>]

Hi! tsutsui-san,

I apologize to a very slow reply.


From: Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>
Date: Tue, 26 Feb 2008 23:57:08 +0900

> Anyway, I don't think it's a good idea to use a pointer
> against DMA descriptor in that way because we should reduce
> accesses against DMA descs to avoid cache issue.
> 
> I'd suggest to rewrite the while() loop in msk_intr(),
> how about this one?


I tried variously today.  And, neither patch seems to be also appropriate.
For instance, I logged in other hosts with ssh(1) and executed 'ls -R /'.
The response died after it had operated for a while.  This msk(4) was not
able to be communicated after this.  X-<

Synchronous timing seems not to be correct yet.

Thanks,
--
kiyohara