Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 29 Sep 2013 16:43:28 +0200
From:      Warner Losh <imp@bsdimp.com>
To:        Adrian Chadd <adrian@freebsd.org>
Cc:        "freebsd-mips@freebsd.org" <freebsd-mips@freebsd.org>
Subject:   Re: [rfc] mips74k/ar9344 support
Message-ID:  <945A50E4-685D-4CC7-88B5-89278744E389@bsdimp.com>
In-Reply-To: <CAJ-VmokBGYYZ1BdghEhYatVBbgH-pNKde-HQXCOfyuR35SrSAA@mail.gmail.com>
References:  <CAJ-Vmo=pJfPryJ8eEKs7xRMsBUxC=xc2DtZfqsOfLKuXRhredA@mail.gmail.com> <CAJ-VmokBGYYZ1BdghEhYatVBbgH-pNKde-HQXCOfyuR35SrSAA@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
I'd be tempted to use 'sll zero, 3' instead of '.word 0xc0' and define a =
macro for it. Why 'sll zero,3' instead of ehb? It allows us to use it on =
older MIPS platforms as well... (because it will assemble in other =
modes, not for binutils issues)...

Speaking of which, I'd also change the last 'nop' on the HAZARD_DELAY =
and ITLBNOPFIX to be 'sll zero, 3' as well. This will be an appropriate =
nop on older kit, as well as 'fail safe' when someone starts to cope =
with newer compilers.  Ditto COP0_SYNC.

On Sep 29, 2013, at 4:37 PM, Adrian Chadd wrote:

> grr, wrong list.
>=20
> also, try this:
>=20
> =
http://people.freebsd.org/~adrian/mips/20130929-mips74k-2.diff<http://peop=
le.freebsd.org/~adrian/mips/20130929-mips74k-1.diff>
>=20
>=20
>=20
> -adrian
>=20
> ---------- Forwarded message ----------
> From: Adrian Chadd <adrian@freebsd.org>
> Date: 29 September 2013 04:38
> Subject: [rfc] mips74k/ar9344 support
> To: "freebsd-wireless@freebsd.org" <freebsd-wireless@freebsd.org>
>=20
>=20
> Hi!
>=20
> Here's an initial hacked up patch to get the AR9344 to mountroot> .
>=20
> (Which isn't true - it panics due to missing PHY setup on if_arge; =
I'll fix
> that up soon.)
>=20
> http://people.freebsd.org/~adrian/mips/20130929-mips74k-1.diff
>=20
> The specific changes:
>=20
> * add CPU_MIPS24KC and CPU_MIPS74KC
> * mips32r2 CPUs require an EHB as a hazard for things, rather than =
NOPs
> * mips74k cores have a different CCA (cache coherency attributes) than =
the
> default - I lifted this from the netbsd mips support.
>=20
> now, the ar71xx CPUs are all currently marked as CPU_MIPS4KC. I'll =
follow
> this up with a commit to change that. There may be things we can =
optimise
> in the CPU_MIPS24KC case.
>=20
> Warner and I have started talking about how to properly fix all of the
> hazard handling. We'll discuss that later.
>=20
> I'd like to commit this to -HEAD so people wishing to hack on mips74k
> platforms can do stuff. Does anyone have any issues with the above?
>=20
> Thanks,
>=20
>=20
>=20
> -adrian
> _______________________________________________
> freebsd-mips@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-mips
> To unsubscribe, send any mail to =
"freebsd-mips-unsubscribe@freebsd.org"




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?945A50E4-685D-4CC7-88B5-89278744E389>