Revive Milan support

[Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>]

jun@ wrote:

> The Japan NetBSD Users' Group held booth
> at the Open Source Conference 2018 Osaka on Jan.27 2018:
> 
>  http://www.ospn.jp/osc2018-osaka/
>  https://www.facebook.com/events/398913583842597/
>  http://www.jp.NetBSD.org/
>  https://www.facebook.com/NetBSD.jp/
> 
> Presentation:
> 
> NetBSD/atari - Port history and ATARI Compatible "Millan" board support.
>  by tsutsui@
> 
>  https://speakerdeck.com/tsutsui/osc2018osaka
> 
>  https://de.wikipedia.org/wiki/Milan_(Computer)
>  http://mail-index.netbsd.org/port-atari/2000/08/28/0000.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091314.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091315.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091316.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091317.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091318.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091319.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091320.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091321.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091322.html
>  http://mail-index.netbsd.org/source-changes/2018/01/20/msg091324.html
>  http://mail-index.netbsd.org/source-changes/2018/01/28/msg091554.html
> 
> Booth:
>  https://twitter.com/tsutsuii/status/957068328366829569
> 
> Milan (Atari compatible) running NetBSD/atari 8.0_BETA 
>  https://twitter.com/tsutsuii/status/957062304398426112

As Ebihara-san wrote, I've been working to revive Milan support
since I received the board last December.
Now NetBSD/atari -current (and patched 8.0_BETA) on Milan
reaches single user.

I'm really glad to see Milan kernel actually working even after
major MI API changes, like the following threads :-)
 http://mail-index.netbsd.org/port-atari/2007/03/07/0000.html
 http://mail-index.netbsd.org/port-atari/2008/12/thread1.html#000126
 http://mail-index.netbsd.org/port-atari/2008/12/thread1.html#000184

I would like to thank Julian Coleman and Leo Weppelman for providing
their Milan board.

Current status:
- only 16MB RAM is works; see the original leo@'s post back in 2001:
  http://mail-index.netbsd.org/port-atari/2001/06/07/0000.html
- S3 Trio64V VGA console works
- 3c905-TX with ep(4) at pci and 3c509 with ep(4) at isa work
- bootloader also works
  (we should prepare link wdboot from sdboot for milan)
- something wrong around ISA interrupt and it hangs on
  heavy disk/network load
- no interrupt on secondary IDE (maybe firmware issue)

TODO:
- probe whole RAM on all four SIMMs
  (needs info about physcal address mappings)
- rework ISA interrupt dispatcher (how to defer IPLVM interrupts)
- test more PCI devices including bus master DMA

dmesg (patched 8.0_BETA):

---
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
    2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017
    The NetBSD Foundation, Inc.  All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
    The Regents of the University of California.  All rights reserved.

NetBSD 8.0_BETA (MILANP) #27: Mon Jan 29 02:50:54 JST 2018
	tsutsui@mirage:/r/work/netbsd-8/src/sys/arch/atari/compile/MILANP
Atari Milan (m68040 CPU/MMU/FPU)
total memory = 16384 KB
avail memory = 12480 KB
timecounter: Timecounters tick every 15.625 msec
mainbus0 (root)
clock0 at mainbus0: system hz 64 timer-A divisor 200/192
timecounter: Timecounter "clock" frequency 12288 Hz quality 100
ser0 at mainbus0: modem1 on 68901 MFP1 USART
nvr0 at mainbus0
isab0 at mainbus0
isa0 at isab0
ep0 at isa0 port 0x300-0x30f irq 5: 3Com 3C509 Ethernet
ep0: address 00:10:4b:07:fd:05, 8KB byte-wide FIFO, 1:1 Rx:Tx split
ep0: 10baseT (default 10baseT)
pckbc0 at isa0 port 0x60-0x64
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0 (mux ignored): console keyboard
isapnp0 at isa0 port 0x279
fdcisa0 at isa0 port 0x3f0-0x3f7 irq 6 drq 2
isapnp0: no ISA Plug 'n Play devices found
pcib0 at mainbus0
pci0 at pcib0
pci0: i/o space, memory space enabled
vendor 8086 product 122e (ISA bridge, revision 0x02) at pci0 dev 1 function 0 not configured
piixide0 at pci0 dev 1 function 1: Intel 82371FB IDE controller (PIIX) (rev. 0x02)
piixide0: bus-master DMA support present
piixide0: primary channel wired to compatibility mode
piixide0: primary channel interrupting at irq 14
atabus0 at piixide0 channel 0
piixide0: secondary channel wired to compatibility mode
piixide0: secondary channel interrupting at irq 15
atabus1 at piixide0 channel 1
ep1 at pci0 dev 4 function 0: 3Com 3c905-TX 10/100 Ethernet
ep1: interrupting at irq 11
ep1: address 00:60:97:59:f3:bb, 8KB word-wide FIFO, 3:5 Rx:Tx split
ukphy0 at ep1 phy 24: OUI 0x1000e8, model 0x0000, rev. 1
ukphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
vga0 at pci0 dev 5 function 0: vendor 5333 product 8811 (rev. 0x54)
wsdisplay0 at vga0 (kbdmux ignored): console (80x25, vt100 emulation), using wskbd0
wsdisplay0: screen 1-3 added (80x25, vt100 emulation)
drm at vga0 not configured
timecounter: Timecounter "clockinterrupt" frequency 64 Hz quality 0
atapibus0 at atabus0: 2 targets
cd0 at atapibus0 drive 1: <SONY    CD-RW  CRX160E, , 1.0g> cdrom removable
cd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 2 (Ultra/33)
wd0 at atabus0 drive 0
wd0: <FUJITSU MPB3043ATU>
wd0: drive supports 16-sector PIO transfers, LBA addressing
wd0: 4125 MB, 8940 cyl, 15 head, 63 sec, 512 bytes/sect x 8448300 sectors
wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 2 (Ultra/33)
wd0(piixide0:0:0): using PIO mode 4
cd0(piixide0:0:1): using PIO mode 4, DMA mode 2 (using DMA)
fdisa0 at fdcisa0 drive 0: 1.44MB, 80 cyl, 2 head, 18 sec
root on wd0a dumps on wd0b
root file system type: ffs
kern.module.path=/stand/atari/8.0/modules

---
Izumi Tsutsui