All of lore.kernel.org
 help / color / mirror / Atom feed
* Status of the FB drivers on m68k
@ 2017-10-10 18:32 John Paul Adrian Glaubitz
  2017-10-11  7:24 ` Geert Uytterhoeven
  0 siblings, 1 reply; 7+ messages in thread
From: John Paul Adrian Glaubitz @ 2017-10-10 18:32 UTC (permalink / raw)
  To: Linux/m68k; +Cc: Debian m68k

Hi!

Does anyone of the kernel guys, especially Geert, know what the current status
of the framebuffer drivers on m68k is?

I am currently discussing the topic of usable graphics cards on Debian/m68k
on the German Amiga forum a1k.org since one user asked me [1] and he told
me that in the past, users could use CONFIG_FB_CYBER and CONFIG_FB_VIRGE
which both seem gone, sadly.

Also, the CONFIG_FB_PM2 driver is marked as broken [2].

Is there a chance to bring the CYBER and VIRGE drivers back? And maybe
get PM2 working on m68k?

Adrian

> [1] https://www.a1k.org/forum/showthread.php?p=1114311#41
> [2] https://github.com/torvalds/linux/blob/master/drivers/video/fbdev/Kconfig#L266

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of the FB drivers on m68k
  2017-10-10 18:32 Status of the FB drivers on m68k John Paul Adrian Glaubitz
@ 2017-10-11  7:24 ` Geert Uytterhoeven
  2017-10-11 11:09   ` Christian T. Steigies
  2017-10-11 12:01   ` John Paul Adrian Glaubitz
  0 siblings, 2 replies; 7+ messages in thread
From: Geert Uytterhoeven @ 2017-10-11  7:24 UTC (permalink / raw)
  To: John Paul Adrian Glaubitz; +Cc: Linux/m68k, Debian m68k

Hi Adrian,

On Tue, Oct 10, 2017 at 8:32 PM, John Paul Adrian Glaubitz
<glaubitz@physik.fu-berlin.de> wrote:
> Does anyone of the kernel guys, especially Geert, know what the current status
> of the framebuffer drivers on m68k is?
>
> I am currently discussing the topic of usable graphics cards on Debian/m68k
> on the German Amiga forum a1k.org since one user asked me [1] and he told
> me that in the past, users could use CONFIG_FB_CYBER and CONFIG_FB_VIRGE
> which both seem gone, sadly.

CyberVision 64, CyberVision 64/3D, and Retina/Z3 support were removed in
commit e019630e78e3482c ("[PATCH] remove broken video drivers").

All of them were marked broken before because they needed an update
for the new fbdev framework, cfr. e.g full-history-linux
commit 6764ca1a7c58dc5f07bd730000a4c40be4ef9f00
Author: Geert Uytterhoeven <geert@linux-m68k.org>
Date:   Sun Sep 28 20:02:47 2003 -0700

    [PATCH] Amiga Cybervision 64/3D frame buffer device is broken

    Amiga Cybervision 64/3D frame buffer device is broken (needs update to new
    fbdev framework)

> Also, the CONFIG_FB_PM2 driver is marked as broken [2].

>From full-history-linux:
commit a05cb1584100212ae8b20a8026dac1f814760158
Author: Geert Uytterhoeven <geert@linux-m68k.org>
Date:   Mon Apr 12 17:29:57 2004 -0700

    [PATCH] Pm2fb is broken on Amiga

    Permedia2: Mark pm2fb broken on Amiga, until somebody fixes it (pm2fb.c
    explicitly tests for CONFIG_PCI right now)

> Is there a chance to bring the CYBER and VIRGE drivers back? And maybe
> get PM2 working on m68k?

There's always a chance ;-)
You "just" need someone with the hardware, programming skills, and a
strong desire to get it working again.

For CyberVision 64, CyberVision 64/3D, and Retina/Z3, the amount of work
shouldn't be that big. Just revert commit e019630e78e3482c, and look how
other fbdev driver were converted to the new framework.

However, in the mean time another S3 Trio / Virge driver was added, so it
may make more sense to add Amiga support to the latter.  But as I expect
that driver to contain many PC-isms, and rely on the BIOS having initialized
the graphics chip first, reviving FB_CYBER and FB_VIRGE as a first step,
may be the best approach.

I expect fixing Permedia2 to be much harder...

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of the FB drivers on m68k
  2017-10-11  7:24 ` Geert Uytterhoeven
@ 2017-10-11 11:09   ` Christian T. Steigies
  2017-10-11 12:01   ` John Paul Adrian Glaubitz
  1 sibling, 0 replies; 7+ messages in thread
From: Christian T. Steigies @ 2017-10-11 11:09 UTC (permalink / raw)
  To: Geert Uytterhoeven; +Cc: John Paul Adrian Glaubitz, Linux/m68k, Debian m68k

Moin,
On Wed, Oct 11, 2017 at 09:24:57AM +0200, Geert Uytterhoeven wrote:
> On Tue, Oct 10, 2017 at 8:32 PM, John Paul Adrian Glaubitz
> 
> > Is there a chance to bring the CYBER and VIRGE drivers back? And maybe
> > get PM2 working on m68k?
> 
> There's always a chance ;-)
> You "just" need someone with the hardware, programming skills, and a
> strong desire to get it working again.

Wouldn't it be a coincidence if I gave my CV64/3D to Adrian for repair.
Did I?
 
Christian

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of the FB drivers on m68k
  2017-10-11  7:24 ` Geert Uytterhoeven
  2017-10-11 11:09   ` Christian T. Steigies
@ 2017-10-11 12:01   ` John Paul Adrian Glaubitz
  2017-10-13 15:45     ` Michael Schmitz
  1 sibling, 1 reply; 7+ messages in thread
From: John Paul Adrian Glaubitz @ 2017-10-11 12:01 UTC (permalink / raw)
  To: Geert Uytterhoeven; +Cc: Linux/m68k, Debian m68k

Hi Geert!

On 10/11/2017 09:24 AM, Geert Uytterhoeven wrote:
>> I am currently discussing the topic of usable graphics cards on Debian/m68k
>> on the German Amiga forum a1k.org since one user asked me [1] and he told
>> me that in the past, users could use CONFIG_FB_CYBER and CONFIG_FB_VIRGE
>> which both seem gone, sadly.
> 
> CyberVision 64, CyberVision 64/3D, and Retina/Z3 support were removed in
> commit e019630e78e3482c ("[PATCH] remove broken video drivers").
> 
> All of them were marked broken before because they needed an update
> for the new fbdev framework, cfr. e.g full-history-linux
> commit 6764ca1a7c58dc5f07bd730000a4c40be4ef9f00
> Author: Geert Uytterhoeven <geert@linux-m68k.org>
> Date:   Sun Sep 28 20:02:47 2003 -0700
> 
>      [PATCH] Amiga Cybervision 64/3D frame buffer device is broken
> 
>      Amiga Cybervision 64/3D frame buffer device is broken (needs update to new
>      fbdev framework)

Ok, thanks.

>> Also, the CONFIG_FB_PM2 driver is marked as broken [2].
> 
>  From full-history-linux:
> commit a05cb1584100212ae8b20a8026dac1f814760158
> Author: Geert Uytterhoeven <geert@linux-m68k.org>
> Date:   Mon Apr 12 17:29:57 2004 -0700
> 
>      [PATCH] Pm2fb is broken on Amiga
> 
>      Permedia2: Mark pm2fb broken on Amiga, until somebody fixes it (pm2fb.c
>      explicitly tests for CONFIG_PCI right now)
> 
>> Is there a chance to bring the CYBER and VIRGE drivers back? And maybe
>> get PM2 working on m68k?
> 
> There's always a chance ;-)
> You "just" need someone with the hardware, programming skills, and a
> strong desire to get it working again.

So, I have already someone from Berlin who is going to give me exactly these
three boards for kernel development and testing \o/.

> For CyberVision 64, CyberVision 64/3D, and Retina/Z3, the amount of work
> shouldn't be that big. Just revert commit e019630e78e3482c, and look how
> other fbdev driver were converted to the new framework.

Alright. I hope you, Michael Schmitz and Michael Karcher will be around
to help :-). I will have my A4000 set up by next week, equipped with
a Buddha, an X-Surf100 and the various graphics cards (one at a time).

> However, in the mean time another S3 Trio / Virge driver was added, so it
> may make more sense to add Amiga support to the latter.  But as I expect
> that driver to contain many PC-isms, and rely on the BIOS having initialized
> the graphics chip first, reviving FB_CYBER and FB_VIRGE as a first step,
> may be the best approach.

Ok, we'll see.

> I expect fixing Permedia2 to be much harder...
Ok, I guess the next things to be done then are:

- Michael Schmitz' ESP driver fixes for the Cyberstorm SCSI
- Michael Karcher's xsurf100 driver
- fixing Cybervision 64, 64/3D, Retina Z3

Might also be an idea to have another Linux/m68k meeting to
get these things done :-).

Adrian

-- 
  .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
   `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of the FB drivers on m68k
  2017-10-11 12:01   ` John Paul Adrian Glaubitz
@ 2017-10-13 15:45     ` Michael Schmitz
  2017-10-14  2:23       ` Finn Thain
  0 siblings, 1 reply; 7+ messages in thread
From: Michael Schmitz @ 2017-10-13 15:45 UTC (permalink / raw)
  To: John Paul Adrian Glaubitz, Geert Uytterhoeven; +Cc: Linux/m68k, Debian m68k

Hi Adrian,

>> For CyberVision 64, CyberVision 64/3D, and Retina/Z3, the amount of work
>> shouldn't be that big. Just revert commit e019630e78e3482c, and look how
>> other fbdev driver were converted to the new framework.
> 
> Alright. I hope you, Michael Schmitz and Michael Karcher will be around

Take a look at commit a100501212f2e26bb6d70bfb5c55eefd90e22b65 for the
initial conversion please (use -b -w to cut down on whitespace changes
shown). Warning: lots of formatting changes make this quite hard to follow.

I'm sure I looked at changes to amifb.c to figure out how the fbdev
interface had changed. Haven't found that in current git history but it
should be in the full history tree.

> Ok, I guess the next things to be done then are:
> 
> - Michael Schmitz' ESP driver fixes for the Cyberstorm SCSI

Finn's fixes actually, and no guarantee what I have in mind will
actually work.

> - Michael Karcher's xsurf100 driver
> - fixing Cybervision 64, 64/3D, Retina Z3
> 
> Might also be an idea to have another Linux/m68k meeting to
> get these things done :-).

I won't be back in Europe before July next...

Cheers,

	Michael

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of the FB drivers on m68k
  2017-10-13 15:45     ` Michael Schmitz
@ 2017-10-14  2:23       ` Finn Thain
  2017-10-14  6:40         ` John Paul Adrian Glaubitz
  0 siblings, 1 reply; 7+ messages in thread
From: Finn Thain @ 2017-10-14  2:23 UTC (permalink / raw)
  To: Michael Schmitz
  Cc: John Paul Adrian Glaubitz, Geert Uytterhoeven, Linux/m68k, Debian m68k

On Sat, 14 Oct 2017, Michael Schmitz wrote:

> > - Michael Schmitz' ESP driver fixes for the Cyberstorm SCSI
> 
> Finn's fixes actually, and no guarantee what I have in mind will 
> actually work.
> 

Just to clarify, Michael and I did reach a solution (i.e. don't attempt 
2-byte DMA transfers) after we kicked around various theories for years, 
and tested some of them with Stan's help, and fixed some mac_esp and 
esp_scsi bugs along the way. But someone still has to write the code for 
zorro_esp. The code is easily adapted from mac_esp, now that mac_esp works 
properly, since you only have to implement the the transfer_size < 
FIFO_SIZE case or even just the write != 0 case. I can help with this 
patch if need be.

-- 

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of the FB drivers on m68k
  2017-10-14  2:23       ` Finn Thain
@ 2017-10-14  6:40         ` John Paul Adrian Glaubitz
  0 siblings, 0 replies; 7+ messages in thread
From: John Paul Adrian Glaubitz @ 2017-10-14  6:40 UTC (permalink / raw)
  To: Finn Thain; +Cc: Michael Schmitz, Geert Uytterhoeven, Linux/m68k, Debian m68k



> On Oct 14, 2017, at 4:23 AM, Finn Thain <fthain@telegraphics.com.au> wrote:
> 
> On Sat, 14 Oct 2017, Michael Schmitz wrote:
> 
>>> - Michael Schmitz' ESP driver fixes for the Cyberstorm SCSI
>> 
>> Finn's fixes actually, and no guarantee what I have in mind will 
>> actually work.
>> 
> 
> Just to clarify, Michael and I did reach a solution (i.e. don't attempt 
> 2-byte DMA transfers) after we kicked around various theories for years, 
> and tested some of them with Stan's help, and fixed some mac_esp and 
> esp_scsi bugs along the way. But someone still has to write the code for 
> zorro_esp. The code is easily adapted from mac_esp, now that mac_esp works 
> properly, since you only have to implement the the transfer_size < 
> FIFO_SIZE case or even just the write != 0 case. I can help with this 
> patch if need be.

Fantastic.

FWIW, I moved the Amiga 4000 back into my office yesterday but I didn’t get to hook it up yet, this will happen on Monday.

I have also lots of test hardware now:

 * Cybervision 64
 * Cybervision 64/3D
 * Retina Z-3
 * 2x X-Surf100
 * 3x Buddha IDE Controller

The Buddha will be used for the libata porting by Bartolomej (I hope I spelled his name correctly) from Samsung.

Adrian

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2017-10-14  6:40 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-10-10 18:32 Status of the FB drivers on m68k John Paul Adrian Glaubitz
2017-10-11  7:24 ` Geert Uytterhoeven
2017-10-11 11:09   ` Christian T. Steigies
2017-10-11 12:01   ` John Paul Adrian Glaubitz
2017-10-13 15:45     ` Michael Schmitz
2017-10-14  2:23       ` Finn Thain
2017-10-14  6:40         ` John Paul Adrian Glaubitz

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.