Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 08 Sep 2014 11:30:18 +0200
From:      =?windows-1252?Q?Jean-S=E9bastien_P=E9dron?= <dumbbell@FreeBSD.org>
To:        Konstantin Belousov <kostikbel@gmail.com>
Cc:        freebsd-x11@FreeBSD.org
Subject:   Re: [CFT] i915: add HW context support (for Mesa 9.2+)
Message-ID:  <540D772A.7010609@FreeBSD.org>
In-Reply-To: <20140908090130.GY2737@kib.kiev.ua>
References:  <540C1CA8.603@FreeBSD.org> <20140908090130.GY2737@kib.kiev.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--io1sTQo4EfbNogqHJg9qlbxb7nTivaOR3
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

On 08.09.2014 11:01, Konstantin Belousov wrote:
> Please do not add dependency on OFED to the drm2.

Thanks for the review, Konstantin!

I should have commentted the OFED part a bit more in my CFT and on the
wiki. If I played with OFED, it's not just for idr. Using a Linux API
wrapper such as the one available in sys/ofed is very handy to greatly
reduce the diff with Linux original code as well as the time taken to
port new code.

I'm not particularily happy with the dependency to OFED, and in the
current patch, this is mostly to have i915 hardware context ready to
test quickly. I planned to send a separate email to discuss the
possibility to move this Linux wrapper in a central place (ie. out of
OFED) so that it's usable by other drivers without depending on OFED itse=
lf.

IMHO, using such a wrapper is very valuable (if not required) if we want
to catch up with Linux in the DRM area.

I'll send that mail to freebsd-arch@ in a few minutes. Do you mind if we
move this discussion there?

> What is the revision which you used for import of i915_gem_context.c ?

I used Linux 3.8 (tag: v3.8).

> There were some very subtle semantical changes in the global gtt
> mappings functions, esp. WRT interaction with lowmem conditions.  I am
> not sure if it is applicable, and it would get me some time to really
> remember the details there.

I don't know if it's related, but so far, I received one report from
danfe@ (-CURRENT/i386) talking about:
  intel_do_flush_locked failed: Input/output error (Xorg.log)

and:
  error: [drm:pid12:i915_hangcheck_elapsed] *ERROR* Hangcheck timer
  elapsed... GPU hung (dmesg)

Though, I don't know if it's related to hardware context. I can't test
the patch myself.

> The i915 reset rework looks fine, I have it imported in my WIP branch,
> but I think it will be merged without troubles.
>=20
> Why do you need the changes to struct eb_objects ? If this is due to
> OFED, please do not.

Yes, it's because of OFED: FreeBSD's LIST_HEAD and Linux' LIST_HEAD
conflict.

--=20
Jean-S=E9bastien P=E9dron


--io1sTQo4EfbNogqHJg9qlbxb7nTivaOR3
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
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=CQXy
-----END PGP SIGNATURE-----

--io1sTQo4EfbNogqHJg9qlbxb7nTivaOR3--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?540D772A.7010609>