linux-mips
[Top] [All Lists]

Re: [PATCH 2/16] arch related Makefile update for lemote fulong mini-PC

To: Thiemo Seufer <ths@networkno.de>
Subject: Re: [PATCH 2/16] arch related Makefile update for lemote fulong mini-PC
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 16 Apr 2007 15:01:38 +0100
Cc: Zhang Fuxin <zhangfx@lemote.com>, tiansm@lemote.com, linux-mips@linux-mips.org
In-reply-to: <20070416124422.GB1402@networkno.de>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <11766507651736-git-send-email-tiansm@lemote.com> <11766507661317-git-send-email-tiansm@lemote.com> <11766507661726-git-send-email-tiansm@lemote.com> <20070415222847.GA1402@networkno.de> <4623387F.3070905@lemote.com> <20070416124422.GB1402@networkno.de>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mutt/1.4.2.2i
On Mon, Apr 16, 2007 at 01:44:22PM +0100, Thiemo Seufer wrote:

> Zhang Fuxin wrote:
> > >> +cflags-$(CONFIG_CPU_LOONGSON2)  += -march=r4600 -Wa,--trap
> > > 
> > > I wonder why this is r4600. I heard the Loongson2 is MIPS IV compatible,
> > > so r5000 / r8000 / r10000 would be better choices.
> > 
> > Presently Loongson-2E is nearly MIPS III compatible(with some
> > self-defined extensions), next version will be mips64 release2 compatible.
> 
> I see.
> 
> > -march=r4600 is inherited from loongson-1, -march=mips3 might be a
> > better choice.
> 
> Maybe. The 'mips3' maps to -march=r4000, it would assume more memory
> latency and a slower integer divider then -march=r4600.

I don't really have an issue with that since I see it as a temporary
solution until gcc and binutils know about Loongson 2 specifics.  Given
the bit I know about the Loognson 2 processor architecture I would not
expect a significant performance boost from trying different values for
-march with current toolchains.

  Ralf

<Prev in Thread] Current Thread [Next in Thread>