Re: No cache control on ppc??
>Okay, that was the intel world.
>
>I haven't found anything similar in powerpc kernels, so I assume there is
>nothing like this. Is that correct?
>If so, is that a hardware restriction?
>Does the hardware do this automagically?
>
>At least on my iBook, lspci claims that theen Rage Mobility is connected via
>AGP 2x, so this could be relevant.
>xine can play DVDs fine on a PII-450, so why shouldn't it on a G3-600...
AFAIK (but my knowledge of x86 is limited), MTRR are also used to enable
write combining. There is no such setting on the chipset used by Apple,
though newer revisions of UniNorth do support write-combining, it is
automatically enabled by Apple's firmware on the AGP bus, but I don't
think this feature is available on the iBook (well, maybe, you can
check that in the device tree, in the pci@f0000000 node, if it has a
write-combining property via /proc/device-tree).
Did you try pure decoding of DVD without display ? How many fps do you
get that way ?
You may improve performances using DMA to blit the YUV frames from memory
to the video cards. Michel Danzer wrote some code to enable that feature
along with DRI on r128s (including the Rage M3 of your iBook). I don't
know the exact state of this, especially I can't tell if he already
improved it to send all buffers using a single kernel call, but even his
first implementation should help.
Ben.
------------------ RFC822 Header Follows ------------------
From: <benh@kernel.crashing.org>
To: Siggi Langauf <langausd@fachschaft.informatik.uni-stuttgart.de>
Cc: <debian-powerpc@lists.debian.org>
Subject: Re: No cache control on ppc??
Date: Sat, 12 Jan 2002 20:27:01 +0100
Message-Id: <20020112192701.22703@mailhost.mipsys.com>
In-Reply-To:
<Pine.LNX.4.21.0201121753510.11140-100000@darkwing.informatik.uni-stuttgart.d
e>
References:
<Pine.LNX.4.21.0201121753510.11140-100000@darkwing.informatik.uni-stuttgart.d
e>
X-Mailer: CTM PowerMail 3.1.1 <http://www.ctmdev.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
-----------------------------------------------------------
Reply to: