dri-devel.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Dave Airlie <airlied@gmail.com>
To: Alex Deucher <alexdeucher@gmail.com>
Cc: "Christian König" <ckoenig.leichtzumerken@gmail.com>,
	"Michel Dänzer" <michel@daenzer.net>,
	"Maling list - DRI developers" <dri-devel@lists.freedesktop.org>,
	"Rui Salvaterra" <rsalvaterra@gmail.com>,
	"debian-powerpc@lists.debian.org"
	<debian-powerpc@lists.debian.org>,
	"John Paul Adrian Glaubitz" <glaubitz@physik.fu-berlin.de>,
	"Karoly Balogh (Charlie/SGR)" <charlie@scenergy.dfmk.hu>
Subject: Re: [RFC] Remove AGP support from Radeon/Nouveau/TTM
Date: Wed, 13 May 2020 06:12:41 +1000	[thread overview]
Message-ID: <CAPM=9tw+Hw4Mkihtf9_+8T_OCbDpUmK07PeWvkf9aqGF9WQmNA@mail.gmail.com> (raw)
In-Reply-To: <CADnq5_Otu2vVpXysmt7icsVgJ_OyR9VJBEdF0JxTcr8mwWN3TQ@mail.gmail.com>

On Wed, 13 May 2020 at 04:21, Alex Deucher <alexdeucher@gmail.com> wrote:
>
> On Tue, May 12, 2020 at 1:02 PM Rui Salvaterra <rsalvaterra@gmail.com> wrote:
> >
> > On Tue, 12 May 2020 at 17:38, Daniel Vetter <daniel@ffwll.ch> wrote:
> > >
> > > Otherwise all agree, agp is a mighty mess and essentially just
> > > crapshot outside of x86. It kinda worked for the much more static
> > > allocations for dri1, but with in-kernel memory managers all the cache
> > > flushing issues showed up big time and it all fell to pieces. Plus a
> > > lot of these host chipset back then where designed for the rather
> > > static windows gpu managers, so even on x86 the coherency issues for
> > > agp mode when used together with ttm or something else really dynamic
> > > is pretty bad because the hw just doesn't really cope and has all
> > > kinds of flushing troubles and races. I think the later agp chipsets
> > > were better.
> >
> > That was rather insightful, thanks. I was starting to doubt my own
> > memory, as I was almost sure I never had any hangs with AGP on PowerPC
> > before KMS was a thing. But even on x86, I distinctly remember never
> > being able to get sideband addressing working with any AGP cards, my
> > system would randomly hang too.
> > I'm starting to believe AGP was shoehorned into PCI the same way VLB
> > was shoehorned into ISA (and for the same reason). History repeats
> > itself… :)
>
> Pre-KMS, the kernel just allocated a static relatively small (e.g., 8
> MB) AGP buffer which never changed.  In that case, things were
> somewhat more reliable.

This is why the AGP hw on Macs has issues I believe. It was designed
and only tested around the one static early allocation, I'm not sure
OSX ever did dynamic.

When it went dynamic I think the AGP bits had some problems with
coherency of the GART tables that we never figured out.

Dave.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2020-05-12 20:12 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-11 19:55 [RFC] Remove AGP support from Radeon/Nouveau/TTM John Paul Adrian Glaubitz
2020-05-11 20:05 ` Alex Deucher
2020-05-11 20:25   ` Rui Salvaterra
2020-05-11 20:51     ` Alex Deucher
2020-05-12  7:57       ` Michel Dänzer
2020-05-12  9:21         ` Rui Salvaterra
2020-05-12  9:40           ` Karoly Balogh (Charlie/SGR)
2020-05-12 13:22             ` Alex Deucher
2020-05-12 16:37               ` Daniel Vetter
2020-05-12 17:02                 ` Rui Salvaterra
2020-05-12 18:21                   ` Alex Deucher
2020-05-12 20:12                     ` Dave Airlie [this message]
2020-05-13  7:26                       ` Christian König
2020-05-12 18:22                 ` Alex Deucher
2020-05-13  7:19                   ` Daniel Vetter
2020-05-13  7:55                     ` Christian König
2020-05-13 10:25                       ` Daniel Vetter
2020-05-13  9:28                     ` Rui Salvaterra
2020-05-13 10:26                       ` Michel Dänzer
2020-05-13 10:29                         ` Daniel Vetter
2020-05-13 10:32                           ` Michel Dänzer
2020-05-13 10:39                         ` Rui Salvaterra
2020-05-13 10:57                           ` Michel Dänzer
2020-05-13 11:07                             ` Rui Salvaterra
2020-05-13 13:44                               ` Christian Zigotzky
2020-05-13 14:01                                 ` John Paul Adrian Glaubitz
2020-05-13 14:01                                 ` Rui Salvaterra
2020-05-13 20:19                                   ` Bertrand Dekoninck
2020-05-13 20:36                                     ` Rui Salvaterra
2020-05-11 20:41   ` John Paul Adrian Glaubitz
2020-05-11 20:46     ` Alex Deucher
2020-05-11 21:50       ` John Paul Adrian Glaubitz
2020-05-12  7:43       ` Christian König
2020-05-11 20:12 ` Christian König
2020-05-11 20:24   ` John Paul Adrian Glaubitz
2020-05-11 20:48     ` Gerhard Pircher
2020-05-12  5:04 ` David VANTYGHEM
2020-05-12  9:20   ` John Paul Adrian Glaubitz
  -- strict thread matches above, loose matches on Subject: below --
2020-05-11 17:17 Christian König
2020-05-11 20:14 ` Al Dunsmuir
2020-05-11 20:27   ` Alex Deucher
2020-05-11 20:43     ` Dave Airlie
2020-05-11 20:56       ` Al Dunsmuir
2020-05-12  8:11         ` Christian König
2020-05-11 20:59       ` Emil Velikov
2020-05-11 20:46     ` Al Dunsmuir

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAPM=9tw+Hw4Mkihtf9_+8T_OCbDpUmK07PeWvkf9aqGF9WQmNA@mail.gmail.com' \
    --to=airlied@gmail.com \
    --cc=alexdeucher@gmail.com \
    --cc=charlie@scenergy.dfmk.hu \
    --cc=ckoenig.leichtzumerken@gmail.com \
    --cc=debian-powerpc@lists.debian.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=glaubitz@physik.fu-berlin.de \
    --cc=michel@daenzer.net \
    --cc=rsalvaterra@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).