Re: ES45 and SMP?

["Michael L. Hitch" <mhitch%lightning.msu.montana.edu@localhost>]

On Wed, 10 Sep 2008, Christos Zoulas wrote:

> In article 
> <Pine.NEB.4.64.0809091602450.28649%lightning.msu.montana.edu@localhost>,
> Michael L. Hitch <mhitch%lightning.msu.montana.edu@localhost> wrote:
...
> >   Ah, progress...
> >
> >   The ES45 I have still had Tru64 on it, and when I booted it, I noticed
> > that it loaded at 0xfffffc0000430000.  NetBSD was loading at
> > 0xfffffc0000300000, so I build a kernel that loaded at 0xfffffc0000430000.
> > This kernel completed loading and began executing:
> >
> > 8962064+479936 [529536+351743]=0x9d8c10
> >
> > Entering netbsd-ES45 at 0xfffffc0000431130...
> >
> > NetBSD does not yet support system type 38 (???).
> >
> > panic: platform not supported
> >
> >
> >   Next step would be to add the system type support, which may be a bit
> > more than I know.
>
> Good progress! Does not look very hard from reading the code!

  Um, what code?  The ES45 apparently uses a different chipset than the
ES40 & friends.  It has 4 hoses instead of the 2 that the ES40 has.  The 
show config command doesn't show any useful information about the chipset 
(the ES40 and CS20 display the chipset).

ES40:
Core Logic
Cchip           DECchip 21272-CA Rev 9(C4)
Dchip           DECchip 21272-DA Rev 2
Pchip 0         DECchip 21272-EA Rev 2
Pchip 1         DECchip 21272-EA Rev 2
TIG             Rev 10

ES45:
Core Logic
Cchip           Rev 17
Dchip           Rev 17
PPchip 0        Rev 17
PPchip 1        Rev 17
TIG             Rev 2.6


  However, I thought I'd give it a try by adding a TITAN system id (38) 
and treating it the same as the dec_6600.  It actually sort of worked.  I
could boot up to trying to mount root (an install kernel would boot and 
run from the memory disk).  The integrated raid controller wasn't 
configured (an i2o DPT controller), so I had no disks.  Trying to mount an 
nfs root also failed.  It also only probed two pci hoses.  I've added
the iop controler, but it fails to reset [exactly like it did on sparc64, 
which I never saw a resolution to].

  Since Linux appears to support the ES45, I started looking at their 
support and found that the chipset is very similar to the 21272.  The
Pchips have two ports instead of one that the PCI bus hangs off of, and 
the Pchip registers had a slightly different layout.  After noting that
the tlbia offset into the Pchip was different, I fixed the tsc_dma.c code
to use the proper access with the TITAN system type, and my ethernet 
worked properly and I could then mount my nfs root.

  Further digging through the Linux support showed my how the 2nd port on 
the Pchips are accessed and a little bit of tweaking on tsp.c got me 
access to the other two PCI hoses.  [They have the fiber HBAs on them, but 
they are Emulex HP LP9000 which aren't supported, but at least they are 
now found.]

  I'm not sure if it's better to copy the tsp* and tsc* code to support 
the TITAN system, or to use the dec_6600 support with the relatively minor 
changes I've done to make it work (but cleaned up a bit).  I don't really 
want to have all that duplicated code that's almost the same.

  Here's the dmesg:

Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
    2006, 2007, 2008
    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 4.99.72 (GENERIC-$Revision: 1.324 $) #21: Thu Sep 11 11:42:46 MDT 2008
        
mhitch%net4.msu.montana.edu@localhost:/home/mhitch/NetBSD-current/OBJ/alphaev56/home/mhitch/NetBSD-current/src/sys/arch/alpha/compile/ES45
AlphaServer ES45 Model 2, 1000MHz, s/n 4224JSPZA0
8192 byte page size, 2 processors.
total memory = 8192 MB
(3056 KB reserved for PROM, 8189 MB used by NetBSD)
avail memory = 8067 MB
timecounter: Timecounters tick every 0.976 msec
mainbus0 (root)
cpu0 at mainbus0: ID 0 (primary), 21264C-5
cpu0: Architecture extensions: 1307<PAT,MVI,CIX,FIX,BWX>
cpu1 at mainbus0: ID 1, 21264C-5
cpu1: processor off-line; multiprocessor support not present in kernel
tsc0 at mainbus0: 21272 Core Logic Chipset, Cchip rev 0
tsc0: 8 Dchips, 2 memory buses of 32 bytes
tsc0: arrays present: 4096MB (split), 4096MB (split), 0MB, 0MB, Dchip 0 rev 1
tsp0 at tsc0
pci0 at tsp0 bus 0
pci0: i/o space, memory space enabled, rd/line, rd/mult, wr/inv ok
sio0 at pci0 dev 7 function 0: Acer Labs M1533 PCI-ISA Bridge (rev. 0xc3)
ppb0 at pci0 dev 8 function 0: Digital Equipment DC21154 PCI-PCI Bridge (rev. 
0x02)
pci1 at ppb0 bus 2
pci1: i/o space, memory space enabled, rd/line, wr/inv ok
Distributed Processing Technology Memory Controller (miscellaneous memory, 
revision 0x02) at pci1 dev 3 function 0 not configured
iop0 at pci1 dev 4 function 0: I2O adapter
iop0: reset rejected, status 0x0
vga0 at pci0 dev 11 function 0: 3D Labs GLINT Permedia 3 (rev. 0x01)
wsdisplay0 at vga0 kbdmux 1
wsmux1: connecting to wsdisplay0
drm at vga0 not configured
Compaq product 0xa0f7 (PCI Hot-Plug system, revision 0x12) at pci0 dev 12 
function 0 not configured
aceride0 at pci0 dev 16 function 0
aceride0: Acer Labs M5229 UDMA IDE Controller (rev. 0xc1)
aceride0: bus-master DMA support present
aceride0: primary channel configured to compatibility mode
aceride0: primary channel interrupting at isa irq 14
atabus0 at aceride0 channel 0
aceride0: secondary channel configured to compatibility mode
aceride0: secondary channel interrupting at isa irq 15
atabus1 at aceride0 channel 1
isa0 at sio0
com0 at isa0 port 0x3f8-0x3ff irq 4: ns16550a, working fifo
com0: console
com1 at isa0 port 0x2f8-0x2ff irq 3: ns16550a, working fifo
pckbc0 at isa0 port 0x60-0x64
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0 mux 1
wskbd0: connecting to wsdisplay0
pms0 at pckbc0 (aux slot)
pckbc0: using irq 12 for aux slot
wsmouse0 at pms0 mux 0
attimer0 at isa0 port 0x40-0x43: AT Timer
pcppi0 at isa0 port 0x61
midi0 at pcppi0: PC speaker (CPU-intensive output)
spkr0 at pcppi0
isabeep0 at pcppi0
fdc0 at isa0 port 0x3f0-0x3f7 irq 6 drq 2
mcclock0 at isa0 port 0x70-0x71: mc146818 compatible time-of-day clock
attimer0: attached to pcppi0
tsp1 at tsc0
pci2 at tsp1 bus 0
pci2: i/o space, memory space enabled, rd/line, rd/mult, wr/inv ok
ppb1 at pci2 dev 1 function 0: Intel S21152BA,S21154AE/BE PCI-PCI Bridge (rev. 
0x00)
pci3 at ppb1 bus 2
pci3: i/o space, memory space enabled, rd/line, wr/inv ok
fxp0 at pci3 dev 4 function 0: i82559 Ethernet, rev 8
fxp0: interrupting at dec 6600 irq 28
fxp0: Ethernet address 00:08:02:3e:f8:98
inphy0 at fxp0 phy 1: i82555 10/100 media interface, rev. 4
inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
fxp1 at pci3 dev 5 function 0: i82559 Ethernet, rev 8
fxp1: interrupting at dec 6600 irq 29
fxp1: Ethernet address 00:08:02:3e:f8:99
inphy1 at fxp1 phy 1: i82555 10/100 media interface, rev. 4
inphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
Compaq product 0xa0f7 (PCI Hot-Plug system, revision 0x12) at pci2 dev 6 
function 0 not configured
tsp2 at tsc0
pci4 at tsp2 bus 0
pci4: i/o space, memory space enabled, rd/line, rd/mult, wr/inv ok
Emulex LP9000 FibreChannel adapter (Fiber Channel serial bus, revision 0x01) at 
pci4 dev 1 function 0 not configured
tsp3 at tsc0
pci5 at tsp3 bus 0
pci5: i/o space, memory space enabled, rd/line, rd/mult, wr/inv ok
Emulex LP9000 FibreChannel adapter (Fiber Channel serial bus, revision 0x01) at 
pci5 dev 2 function 0 not configured
Compaq product 0xa0f7 (PCI Hot-Plug system, revision 0x12) at pci5 dev 6 
function 0 not configured
timecounter: Timecounter "clockinterrupt" frequency 1024 Hz quality 0
timecounter: Timecounter "PCC" frequency 999962752 Hz quality 1000
fd0 at fdc0 drive 0: 1.44MB, 80 cyl, 2 head, 18 sec
atapibus0 at atabus0: 2 targets
cd0 at atapibus0 drive 0: <Compaq  CRD-8402B, 2000/05/10, 1.03> cdrom removable
cd0: 32-bit data port
cd0: drive supports PIO mode 4, DMA mode 2
cd0(aceride0:0:0): using PIO mode 4, DMA mode 2 (using DMA)
Kernelized RAIDframe activated
root device (default fxp1): 
dump device: 
file system (default generic): 
root on fxp1
root on xxx.xxx.xxx.xxx:/home/export/alpha
root file system type: nfs
init path (default /sbin/init): 
init: trying /sbin/init

--
Michael L. Hitch                        mhitch%montana.edu@localhost
Computer Consultant
Information Technology Center
Montana State University        Bozeman, MT     USA