CentOS on wildfire
Robin Humble
rjh+axp at cita.utoronto.ca
Sun Aug 13 09:33:02 UTC 2006
Hi,
we're trying to get CentOS 4.3 up on our gs320 (32 ev67 731MHz,
64G ram) but have been struggling a bit for several days now.
are there any recent wildfire success stories out there? (apart from
the one boot in 2000)
non-SMP kernels seem more or less ok (+/- a few Oops's) under either
CentOS's UP kernel (2.6.9-34.0.2.EC) or latest kernel.org (2.6.17.8).
in SMP mode however, 2.6.9-34.0.2.ECsmp invariably stops after
Freeing unused kernel memory: 176k freed
the next line should probably be one of:
Red Hat nash version 4.2.1.6 starting
atkbd.c: keyboard reset failed on isa0060/serio1
but it's never seen...
I've tried a bunch of wildfire specific and cut-down and tweaked
.config variants of the CentOS 2.6.9 kernel for alpha but nothing seems
to help here - it doesn't get past that line.
2.6.17.8 SMP seems close to working though - except for a lost irq4 (see
the boot log at the URL below) which (naively) seems to come from the
serial handler.
the serial console is also a bit screwed up at this stage with each key
press being one behind what's visible on the serial console. the
suggested 'irqpoll' boot option makes no obvious difference.
however the good news is that once the machine is up in init 3 (and the
below log is from the only time so far that I've got it that far) then
the machine seems ok and usable over an ssh login. I have yet to stress
test it in any way.
below is some /proc info for 2.6.17.8. the serial boot log and the
.config is at http://www.cita.utoronto.ca/~rjh/alpha/
BTW, CentOS 4.3 (standard SMP kernel) works great so far on our es40s.
CentOS 4.4's gcc4/gfortran (when it's built) has OpenMP which will make
the machines just as useful as they were under Tru64. whoo! :-) and as
soon as we fix our es45 then we'll give CentOS a run on there too...
but anyway, any hints from you alpha experts for wildfire tweaks for
failing CentOS kernels or the irq4 problem in 2.6.17.8 would be very
much appreciated.
let me know if it's more generic or if I should be posting to lkml instead.
this gs320 a non-production machine until it gets more stable (or Tru64
licenses magically become cheaper), so any speculative patches or tweaks
and many reboots are just fine...
cheers,
robin
% cat /proc/cpuinfo
cpu : Alpha
cpu model : EV67
cpu variation : 7
cpu revision : 0
cpu serial number : SM02603478
system type : Wildfire
system variation : 0
system revision : 0
system serial number : G2B24Y
cycle frequency [Hz] : 730782000
timer frequency [Hz] : 1000.00
page size [bytes] : 8192
phys. address bits : 44
max. addr. space # : 255
BogoMIPS : 1489.60
kernel unaligned acc : 0 (pc=0,va=0)
user unaligned acc : 0 (pc=0,va=0)
platform string : AlphaServer GS320 6/731
cpus detected : 32
cpus active : 32
cpu active mask : 00000000ffffffff
L1 Icache : 64K, 2-way, 64b line
L1 Dcache : 64K, 2-way, 64b line
L2 cache : 4096K, 1-way, 64b line
L3 cache : n/a
% cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3 CPU4 CPU5 CPU6 CPU7 CPU8 CPU9 CPU10 CPU11 CPU12 CPU13 CPU14 CPU15 CPU16 CPU17 CPU18 CPU19 CPU20 CPU21 CPU22 CPU23 CPU24 CPU25 CPU26 CPU27 CPU28 CPU29 CPU30 CPU31
2: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 XT-PIC cascade
4: 1003926 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WILDFIRE serial
8: 1042388 1035826 1035216 1034424 1033923 1033362 1032847 1032440 1031385 1031825 1031713 1031617 1031237 1031466 1031404 1031349 1031302 1031256 1031214 1031171 1031127 1031089 1031051 1031017 1030990 1030957 1030928 1030898 1030867 1030840 1030811 1030780 RTC +timer
14: 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WILDFIRE +ide0
36: 16916 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WILDFIRE qla1280
40: 4417 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WILDFIRE eth0
296: 22137 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WILDFIRE qla1280
IPI: 1275 1297 1361 1358 1387 1392 1398 1413 1403 1405 1413 1429 1418 1396 1401 1418 1395 1399 1406 1415 1406 1409 1418 1412 1417 1414 1418 1410 1400 1409 1416 1414
ERR: 0
% free
total used free shared buffers cached
Mem: 66467080 145960 66321120 0 16968 49072
-/+ buffers/cache: 79920 66387160
Swap: 4194264 0 4194264
% lspci
00:01.0 SCSI storage controller: QLogic Corp. ISP1020 Fast-wide SCSI (rev 05)
00:02.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 02)
00:07.0 ISA bridge: ALi Corporation M1533/M1535 PCI to ISA Bridge [Aladdin IV/V/V+] (rev c3)
00:0f.0 IDE interface: ALi Corporation M5229 IDE (rev c1)
00:13.0 USB Controller: ALi Corporation USB 1.1 Controller (rev 03)
01:04.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 05)
01:05.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 05)
0002:03:01.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 05)
0002:04:04.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 05)
0002:04:05.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 05)
0008:06:02.0 SCSI storage controller: QLogic Corp. ISP1020 Fast-wide SCSI (rev 05)
0008:06:03.0 Fibre Channel: Emulex Corporation LP8000 Fibre Channel Host Adapter (rev 02)
0009:07:07.0 SCSI storage controller: Adaptec AHA-3960D / AIC-7899A U160/m (rev 01)
0009:07:07.1 SCSI storage controller: Adaptec AHA-3960D / AIC-7899A U160/m (rev 01)
000a:08:01.0 Network controller: Digital Equipment Corporation: Unknown device 0018 (rev 24)
More information about the axp-list
mailing list