linux-apus-devel

Linux/m68k merge status

[Geert U. <ge...@li...>]

Linux/m68k merge status report:

  - Sent to Linus/Alan (27%):

      o Ariadne2 (+ X-Surf) and Hydra updates
	  Documentation/Configure.help
	  drivers/net/8390.c
	  drivers/net/8390.h
	  drivers/net/Config.in
	  drivers/net/Makefile
	  drivers/net/Space.c
	  drivers/net/ariadne2.c
	  drivers/net/hydra.c
	  include/linux/zorro_ids.h

      o Amiga timer resource fix
	  arch/m68k/amiga/config.c

      o Amiga Y2K fixes
	  arch/m68k/amiga/config.c

      o Amiga builtin serial port update
	  drivers/char/amiserial.c

      o Add __ashldi3()
	  arch/m68k/lib/Makefile
	  arch/m68k/lib/ashldi3.c

      o loops_per_jiffie updates
	  arch/m68k/atari/debug.c
	  arch/m68k/kernel/setup.c
	  include/asm-m68k/delay.h

      o drivers/video/Config updates
	  drivers/video/Config.in

      o asm-m68k/movs.h updates
	  include/asm-m68k/movs.h

      o config.in updates
	  arch/m68k/config.in

      o We need an ioctl() for keyboard delay/repeat
	  drivers/char/vt.c

      o HP/300 and MVME Lance updates
	  drivers/net/7990.c
	  drivers/net/7990.h
	  drivers/net/mvme147.c

      o SCSI fixes
	  drivers/scsi/mac_NCR5380.c
	  drivers/scsi/mac_esp.c
	  drivers/scsi/sun3x_esp.c

      o Export symbols
	  arch/m68k/kernel/m68k_ksyms.c

      o Mac IDE
	  drivers/ide/macide.c

      o Misc small fixes
	  arch/m68k/Makefile
	  arch/m68k/amiga/amisound.c
	  arch/m68k/amiga/config.c
	  arch/m68k/kernel/m68k_defs.h
	  arch/m68k/kernel/time.c
	  arch/m68k/kernel/traps.c
	  arch/m68k/mm/fault.c
	  drivers/block/ataflop.c
	  drivers/char/Makefile
	  drivers/char/mem.c
	  drivers/video/retz3fb.c
	  include/asm-m68k/floppy.h
	  include/asm-m68k/io.h
	  include/asm-m68k/param.h
	  include/asm-m68k/serial.h
	  include/asm-m68k/traps.h

      o stifb: PA-RISC init order disturbing
	  drivers/video/fbmem.c


  - What's left (73%):

      o A2232 serial driver: waiting for Jukka Marin
	  Documentation/Configure.help
	  drivers/char/Config.in
	  drivers/char/Makefile
	  drivers/char/ser_a2232.c
	  drivers/char/ser_a2232.h
	  drivers/char/tty_io.c
	  include/linux/tty.h

      o amifb-off: call setup function, even if video=<fbdev>:off
	  drivers/video/fbmem.c

      o atari-fat: Atari-guys: is this OK?
	  fs/fat/inode.c
	  include/linux/msdos_fs.h

      o genrtc: still need to apply Richard Z's merge patch. The current genrtc
	seems to work on my Amiga, though
	  drivers/char/Makefile
	  drivers/char/genrtc.c
	  drivers/char/misc.c
	  drivers/char/q40rtc.c

      o ioext: Parallel port driver for GVP I/O Extender. Ditch it until
	someone writes a parport-compliant one?
	  drivers/char/ioext.h
	  drivers/char/plip_ioext.c

      o m68k-clgenfb: waiting for someone to test this
	  drivers/video/clgenfb.c

      o m68k-mac: waiting for fiat and/or new patches from Mac/m68k guys
	  arch/m68k/mac/config.c
	  arch/m68k/mac/debug.c
	  arch/m68k/mac/iop.c
	  arch/m68k/mac/macints.c
	  arch/m68k/mac/misc.c
	  arch/m68k/mac/psc.c
	  arch/m68k/mac/via.c
	  include/asm-m68k/mac_psc.h

      o m68k-mace: waiting for fiat and/or new patches from Mac/m68k guys
	  drivers/net/Space.c
	  drivers/net/macmace.c

      o What exactly is going on here? MM was never my strongest point.
	  arch/m68k/mm/init.c
	  arch/m68k/mm/memory.c
	  arch/m68k/mm/motorola.c

      o We don't have portops, or do we? (after I apply Richard Z's I/O patch,
        of course)
	  drivers/char/mem.c

      o No need to set the default arch to m68k in Linus' tree :-)
	  Makefile

      o Relics from the old serial drivers. Ditch them?
	  drivers/char/16c552.h
	  drivers/char/mc68681.h

      o Is this correct? Remove things protected by #idef 0/#endif?
	  include/asm-m68k/string.h

      o Mac stuff shared with PowerMac. Please sort this out with the PowerMac
	guys first!
	  drivers/input/keybdev.c
	  drivers/macintosh/adb.c
	  drivers/macintosh/adbhid.c
	  drivers/macintosh/via-cuda.c
	  drivers/macintosh/via-macii.c
	  drivers/macintosh/via-maciisi.c
	  include/linux/adb.h
	  include/linux/pmu.h

      o What exactly is going on here? MM was never my strongest point. And
	this is generic code, so I'd like to avoid it.
	  mm/bootmem.c
	  mm/page_alloc.c

      o What to do with this? Ideally we want both the normal and the Amiga
	53c7xx driver in the same kernel.
	  drivers/scsi/53c7xx.h

      o According the Richard Z this is an endianness bug. Got no response from
	linux-net.
	  drivers/net/ne.c

      o People with PC-style NVRAM (Atari), can you please check this? It's
	generic code.
	  drivers/char/nvram.c

      o This file is out of sync anyway (Hi APUS guys!)
	  arch/ppc/amiga/config.c

      o real_root_dev might be obsolete, according to Andreas. Ditch it?
	  drivers/block/rd.c
	  include/linux/devfs_fs_kernel.h
	  include/linux/fs.h
	  init/main.c

      o People with PC-style RTC (Atari and Q40), can you please check this?
        It's generic code.
	  drivers/char/rtc.c
	  include/asm-alpha/mc146818rtc.h
	  include/asm-arm/mc146818rtc.h
	  include/asm-i386/mc146818rtc.h
	  include/asm-m68k/mc146818rtc.h
	  include/asm-m68k/rtc.h
	  include/asm-mips64/mc146818rtc.h
	  include/asm-mips/mc146818rtc.h
	  include/asm-ppc/mc146818rtc.h
	  include/asm-sparc64/mc146818rtc.h
	  include/asm-sparc/mc146818rtc.h

      o Sun3 stuff shared with SPARC. Please sort this out with the SPARC guys
	first!
	  drivers/sbus/char/sunkbd.c
	  drivers/sbus/char/sunmouse.c
	  drivers/sbus/char/sunserial.c
	  drivers/sbus/char/zs.c

      o Last batch of Sun3 changes from Sam
	  arch/m68k/kernel/head.S
	  arch/m68k/kernel/setup.c
	  arch/m68k/kernel/traps.c
	  arch/m68k/sun3x/Makefile
	  arch/m68k/sun3x/config.c
	  arch/m68k/sun3x/dvma.c
	  arch/m68k/sun3x/ints.c
	  arch/m68k/sun3x/ints.h
	  arch/m68k/sun3x/prom.c
	  drivers/net/sun3lance.c
	  drivers/scsi/scsi.h
	  include/asm-m68k/dvma.h
	  include/asm-m68k/sbus.h
	  include/asm-m68k/sun3xprom.h

      o The PowerMac guys didn't react to this. Perhaps I should just check it
	in in their bitkeeper tree?
	  drivers/video/valkyriefb.c
	  drivers/video/valkyriefb.h


  - Other issues: please send me new serial drivers! :-)


Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

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

Re: Linux/m68k merge status

[Joerg D. <jo...@do...>]

On Thu, Jan 04, 2001 at 05:42:03PM +0100, Geert Uytterhoeven wrote:
>=20
>       o Relics from the old serial drivers. Ditch them?
> 	  drivers/char/16c552.h
> 	  drivers/char/mc68681.h

Leave them in for me, please. If I ever get around to having a usable
system again, I'll most probably write drivers for the multiface card
again.

>   - Other issues: please send me new serial drivers! :-)

Comes time, come drivers (How did you manage to get into this obcure
"holiday" state? ;-)

Bye,

Joerg

Re: Linux/m68k merge status

[Geert U. <ge...@li...>]

On Thu, 4 Jan 2001, Joerg Dorchain wrote:
> On Thu, Jan 04, 2001 at 05:42:03PM +0100, Geert Uytterhoeven wrote:
> > 
> >       o Relics from the old serial drivers. Ditch them?
> > 	  drivers/char/16c552.h
> > 	  drivers/char/mc68681.h
> 
> Leave them in for me, please. If I ever get around to having a usable
> system again, I'll most probably write drivers for the multiface card
> again.

You can always extract them from an older Linux/m68k source tree, if necessary
(`The Internet as backup').

> >   - Other issues: please send me new serial drivers! :-)
> 
> Comes time, come drivers (How did you manage to get into this obcure
> "holiday" state? ;-)

It's called: because you had to stay home for a total of 5.5 months last year
due to a skiing accident, you didn't have time to consume these last 15 days
from your vacation account yet :-)

Only one more to go :-( Next Monday I'll be back at work...

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

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

Re: Linux/m68k merge status

[Richard Z. <Ric...@st...>]

Hi,

>       o genrtc: still need to apply Richard Z's merge patch. The current genrtc
> 	seems to work on my Amiga, though

there should not be much difference in functionality for the Amiga, the 
merged driver is just a bit more complicated at the moment.

>       o We don't have portops, or do we? (after I apply Richard Z's I/O patch,
>         of course)
> 	  drivers/char/mem.c

we should have - ISA pnptools have been reported to work fine on
Q40.


>       o According the Richard Z this is an endianness bug. Got no response from
> 	linux-net.
> 	  drivers/net/ne.c

try again? I see no other solution.

>       o People with PC-style RTC (Atari and Q40), can you please check this?

nope, Q40 uses only this one:

> 	  include/asm-m68k/rtc.h

waiting for genrtc merge, may need small cleanups.

Bye
Richard

Re: Linux/m68k merge status

[Geert U. <ge...@li...>]

On Thu, 4 Jan 2001, Richard Zidlicky wrote:
> >       o According the Richard Z this is an endianness bug. Got no response from
> > 	linux-net.
> > 	  drivers/net/ne.c
> 
> try again? I see no other solution.

After looking at other 8390-based drivers, you convinced me :-)
Sent to Linus and Alan.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

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

Re: Linux/m68k merge status

[Roman Z. <zi...@fh...>]

Hi,

On Thu, 4 Jan 2001, Geert Uytterhoeven wrote:

>       o What exactly is going on here? MM was never my strongest point. And
> 	this is generic code, so I'd like to avoid it.
> 	  mm/bootmem.c
> 	  mm/page_alloc.c

These are needed to map multiple physical memory chunks into a single
memory zone. Actually we should use the zone stuff, but I didn't had the
time lately for this. On the other hand this part works, so I used the
holiday for affs, which needs it more.

>       o This file is out of sync anyway (Hi APUS guys!)
> 	  arch/ppc/amiga/config.c

I'm not at home yet, I look at this next week, when I'm going to update
the apus tree.

>       o real_root_dev might be obsolete, according to Andreas. Ditch it?
> 	  drivers/block/rd.c
> 	  include/linux/devfs_fs_kernel.h
> 	  include/linux/fs.h
> 	  init/main.c

I think the main problem was the direct cast, if you use to_kdev_t(), it
should be easier to get the patch accepted.

bye, Roman

Re: Linux/m68k merge status

[Andreas S. <sc...@su...>]

Geert Uytterhoeven <ge...@li...> writes:

|>       o What exactly is going on here? MM was never my strongest point.
|> 	  arch/m68k/mm/init.c
|> 	  arch/m68k/mm/memory.c
|> 	  arch/m68k/mm/motorola.c

Just to get the statistics right.  And to make stram on Atari working.

|>       o People with PC-style NVRAM (Atari), can you please check this? It's
|> 	generic code.
|> 	  drivers/char/nvram.c

Since the driver has been written by Roman Hodek, this should be ok.

|>       o People with PC-style RTC (Atari and Q40), can you please check this?
|>         It's generic code.
|> 	  drivers/char/rtc.c
|> 	  include/asm-alpha/mc146818rtc.h
|> 	  include/asm-arm/mc146818rtc.h
|> 	  include/asm-i386/mc146818rtc.h
|> 	  include/asm-m68k/mc146818rtc.h
|> 	  include/asm-m68k/rtc.h
|> 	  include/asm-mips64/mc146818rtc.h
|> 	  include/asm-mips/mc146818rtc.h
|> 	  include/asm-ppc/mc146818rtc.h
|> 	  include/asm-sparc64/mc146818rtc.h
|> 	  include/asm-sparc/mc146818rtc.h

I think this should be negotiated with the author.

Andreas.

-- 
Andreas Schwab                                  "And now for something
SuSE Labs                                        completely different."
And...@su...
SuSE GmbH, Schanzäckerstr. 10, D-90443 Nürnberg

Re: Linux/m68k merge status

[Geert U. <ge...@li...>]

On 5 Jan 2001, Andreas Schwab wrote:
> |>       o People with PC-style NVRAM (Atari), can you please check this? It's
> |> 	generic code.
> |> 	  drivers/char/nvram.c
> 
> Since the driver has been written by Roman Hodek, this should be ok.

I believe you :-) But since the PC nvram.c changed as well, I'm not sure that I
did the merge correctly.

> |>       o People with PC-style RTC (Atari and Q40), can you please check this?
> |>         It's generic code.
> |> 	  drivers/char/rtc.c
> |> 	  include/asm-alpha/mc146818rtc.h
> |> 	  include/asm-arm/mc146818rtc.h
> |> 	  include/asm-i386/mc146818rtc.h
> |> 	  include/asm-m68k/mc146818rtc.h
> |> 	  include/asm-m68k/rtc.h
> |> 	  include/asm-mips64/mc146818rtc.h
> |> 	  include/asm-mips/mc146818rtc.h
> |> 	  include/asm-ppc/mc146818rtc.h
> |> 	  include/asm-sparc64/mc146818rtc.h
> |> 	  include/asm-sparc/mc146818rtc.h
> 
> I think this should be negotiated with the author.

Same here. I had to resolve lots of conflicts.

If anyone can tell me that nvram and rtc stuff are OK, I'll do my best to sort
it out with the respective maintainers.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

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