Re: G5 powermacs

[Michael <macallan%netbsd.org@localhost>]

Hello,

On Sun, 4 Jun 2017 22:14:35 +0100
Sevan Janiyan <venture37%geeklan.co.uk@localhost> wrote:

> putting the latest netbsd kernel (uncompressed) and running
> boot first-boot/@0:3,\netbsd
> loads the kernel and fails with
> 
> load-size=a3d1cc adler32=f645da8a
> Loading ELF
> 
> CLAIM failed
>  ok

I finally got around to mess with my G5. Went much further than expected:

boot enet:,netbsd.g5
...
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 8.99.1 (GLENDEVERON) #1: Wed Jun 28 11:00:52 EDT 2017
        ml@blackbush:/stuff/build/obj/sys/arch/macppc/compile/GLENDEVERON
total memory = 2048 MB
oea_startup: failed to allocate DEAD ZONE: error=12
avail memory = 1958 MB
timecounter: Timecounters tick every 10.000 msec
found U3/U4 HT PIC at f8040000
OpenPIC Version 1.2: Supports 4 CPUs and 124 interrupt sources.
bootpath: 
mainbus0 (root)
cpu0 at mainbus0: 970MP (Revision 1.1), ID 0 (primary)
cpu0: HID0 , powersave: -1
cpu1 at mainbus0: ID 1
cpu1: processor off-line; multiprocessor support not present in kernel
uninorth0 at mainbus0
pci0 at uninorth0 bus 10
pci0: i/o space, memory space enabled
genfb0 at pci0 dev 0 function 0found EDID property...
console_instance: ffb3d980
package: ffa17948
: vendor 1002 product 7240 (rev. 0x00)
genfb0: framebuffer at 0x98008000, size 1920x1080, depth 8, stride 2048
wsdisplay0 at genfb0 kbdmux 1: console (default, vt100 emulation)
wsmux1: connecting to wsdisplay0
drm at genfb0 not configured
ibmcpc0 at mainbus0
Mapping in config space @ pa 0xf2000000, size: 0x02800000
pci1 at ibmcpc0 bus 4
pci1: i/o space, memory space enabled
pci2 at ibmcpc0 bus 5
pci2: i/o space, memory space enabled
bge0 at pci2 dev 4 function 0: Broadcom BCM5780 Gigabit Ethernet
bge0: interrupting at irq 66
bge0: HW config 00000015, 00000015, 00000000, 00000000 00000000
bge0: ASIC BCM5714 B3 (0x8003), Ethernet address 00:14:51:63:47:77
bge0: setting short Tx thresholds
brgphy0 at bge0 phy 1: BCM5780 1000BASE-T/X media interface, rev. 0
brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto
bge1 at pci2 dev 4 function 1: Broadcom BCM5780 Gigabit Ethernet
bge1: interrupting at irq 67
bge1: NVRAM lock timedout!
bge1: HW config 00000014, 00000014, 00000000, 00000000 00000000
bge1: ASIC BCM5714 B3 (0x8003), Ethernet address 00:14:51:63:47:78
bge1: setting short Tx thresholds
brgphy1 at bge1 phy 1: BCM5780 1000BASE-T/X media interface, rev. 0
brgphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto
pci3 at ibmcpc0 bus 6
pci3: i/o space, memory space enabled
pci4 at ibmcpc0 bus 7
pci4: i/o space, memory space enabled
pci5 at ibmcpc0 bus 8
pci5: i/o space, memory space enabled
pci6 at ibmcpc0 bus 9
pci6: i/o space, memory space enabled
pci7 at ibmcpc0 bus 2
pci7: i/o space, memory space enabled
vendor 106b product 0051 (ethernet network) at pci7 dev 15 function 0 not configured
pci8 at ibmcpc0 bus 1
pci8: i/o space, memory space enabled
obio0 at pci8 dev 7 function 0: addr 0x80000000
zsc0 at obio0 irq 23,24
zstty0 at zsc0 channel 0
zstty1 at zsc0 channel 1
i2s at obio0 offset 0x0 not configured
ohci0 at pci8 dev 11 function 0: vendor 1033 product 0035 (rev. 0x43)
ohci0: interrupting at irq 70
ohci0: OHCI version 1.0
usb0 at ohci0: USB revision 1.0
ohci1 at pci8 dev 11 function 1: vendor 1033 product 0035 (rev. 0x43)
ohci1: interrupting at irq 70
ohci1: OHCI version 1.0
usb1 at ohci1: USB revision 1.0
ehci0 at pci8 dev 11 function 2: vendor 1033 product 00e0 (rev. 0x04)
ehci0: interrupting at irq 70
ehci0: EHCI version 1.0
ehci0: companion controllers, 3 ports each: ohci0 ohci1
usb2 at ehci0: USB revision 2.0
pci9 at ibmcpc0 bus 3
pci9: i/o space, memory space enabled
svwsata0 at pci9 dev 12 function 0: ServerWorks K2 SATA Controller (rev. 0x00)
svwsata0: bus-master DMA support present
svwsata0: using irq 10 for native-PCI interrupt
svwsata0: primary channel wired to native-PCI mode
atabus0 at svwsata0 channel 0
svwsata0: secondary channel wired to native-PCI mode
atabus1 at svwsata0 channel 1
svwsata0: secondary channel wired to compatibility mode
atabus2 at svwsata0 channel 2
svwsata0: secondary channel configured to compatibility mode
atabus3 at svwsata0 channel 3
wdc0 at pci9 dev 13 function 0: interrupting at irq 38
atabus4 at wdc0 channel 0
vendor 106b product 0052 (IEEE1394 serial bus, OpenHCI) at pci9 dev 14 function 0 not configured
smu0 at mainbus0 mbox 0x860c gpio 0x62 irq 53smuiic0 at smu0iic0 at smuiic0: I2C bus
smusat0 at smuiic0
system-clock at smuiic0 address 0xd4 not configured

smuiic1 at smu0iic1 at smuiic1: I2C bus


vmmask 7f80000, addr 1000 schedmask 7f800000 highmask 7f800000
timecounter: Timecounter "clockinterrupt" frequency 100 Hz quality 0
timecounter: Timecounter "mftb" frequency 33333333 Hz quality 100
uhub0 at usb0: vendor 1033 (0x1033) OHCI root hub (0000), class 9/0,
rev 1.00/1.00, addr 1
uhub0: 3 ports with 3 removable, self powered
uhub1 at usb1: vendor 1033 (0x1033) OHCI root hub (0000), class 9/0, rev 1.00/1.00, addr 1
uhub1: 2 ports with 2 removable, self powered
uhub2 at usb2: vendor 1033 (0x1033) EHCI root hub (0000), class 9/0, rev 2.00/1.00, addr 1
uhub2: 5 ports with 5 removable, self powered
svwsata0 port 0: device present, speed: 1.5Gb/s
wd0 at atabus0 drive 0
wd0: <SAMSUNG HD153WI>
wd0: drive supports 16-sector PIO transfers, LBA48 addressing
wd0: 1397 GB, 2907021 cyl, 16 head, 63 sec, 512 bytes/sect x 2930277168 sectors
wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
wd0(svwsata0:0:0): using PIO mode 4, Ultra-DMA mode 6 (Ultra/133) (using DMA)
ehci0: handing over full speed device on port 1 to ohci0
svwsata0 port 1: PHY offline
atapibus0 at atabus4: 2 targets
cd0 at atapibus0 drive 0: <HL-DT-ST DVD-RW GWA-4165B, 76DB8EB84B89, C006> cdrom removable
cd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66)
cd0(wdc0:0:0): using PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66) (using DMA)
WARNING: 2 errors while detecting hardware; check system log.
boot device: cpu0
md0: internal 2048 KB image area
root on bge0
nfs_boot: trying DHCP/BOOTP
uhub3 at uhub0 port 1: Mitsumi Electric (0x5ac) Hub in Apple Extended USB Keyboard (0x1003), class 9/0, rev 1.10/4.10, addr 2
uhub3: 3 ports with 2 removable, bus powered
uhidev0 at uhub3 port 3 configuration 1 interface 0
uhidev0: Mitsumi Electric (0x5ac) Apple Extended USB Keyboard (0x20b), rev 1.10/4.10, addr 3, iclass 3/1
ukbd0 at uhidev0: 8 modifier keys, 6 key codes
wskbd0 at ukbd0: console keyboard, using wsdisplay0
uhidev1 at uhub3 port 3 configuration 1 interface 1
uhidev1: Mitsumi Electric (0x5ac) Apple Extended USB Keyboard (0x20b), rev 1.10/4.10, addr 3, iclass 3/0
uhidev1: 3 report ids
uhid0 at uhidev1 reportid 2: input=1, output=0, feature=0
uhid1 at uhidev1 reportid 3: input=3, output=0, feature=0
nfs_boot: DHCP next-server: 192.168.0.44
nfs_boot: my_name=glendeveron
nfs_boot: my_addr=192.168.0.24
nfs_boot: my_mask=255.255.255.0
nfs_boot: gateway=192.168.0.1
uhidev2 at uhub3 port 1 configuration 1 interface 0
uhidev2: Logitech (0x46d) Trackball (0xc404), rev 1.10/2.20, addr 4, iclass 3/1
ums0 at uhidev2: 3 buttons and Z dir
wsmouse0 at ums0 mux 0
root on branca:/stuff/roots/macppc
kern.module.path=/stand/macppc/8.99.1/modules
wsdisplay0: screen 1 added (default, vt100 emulation)
wsdisplay0: screen 2 added (default, vt100 emulation)
wsdisplay0: screen 3 added (default, vt100 emulation)
wsdisplay0: screen 4 added (default, vt100 emulation)

Some comments on that:
- yes, there's assorted, not necessarily related, debug spam in there
- the machine has 4GB RAM, it's capped at 2GB because the address space
  layout in the lower 4GB is just like on 32bit Macs - 2GB RAM, the
  rest PCI space, ROMs etc. so in 32bit mode the rest of the RAM is
  inaccessible
- it made it all the way to login:, although very slow ( I guess I
  blame nfs ) and with all fans at full blast - evidently we don't poke
  the SMU at the right time or in the right way.
- genfb Just Works(tm), I'll check if there are uncommitted local
  changes. I wouldn't be surprised, on 32bit machines we BAT-map all
  useful PCI space with paddr == vaddr, on the G5 we can't really do
  that.
- SMP support wasn't even attempted, probably works the same way as
  with G4s and OpenPIC.
- reading from the disk throws occasional DMA errors, no idea if that's
  the driver or G5 bus_space/bus_dma - I get no errors with OSX or
  FreeBSD
- the clock is running too fast, guess we get the decrementer frequency
  wrong

have fun
Michael