2.6.0-test6[-mm4] boot failure on alpha

2.6.0-test6[-mm4] boot failure on alpha

[Nathan]

[-- Attachment #1: Type: text/plain, Size: 3067 bytes --]


I'm having a problem getting 2.6.0-test6 and/or 2.6.0-test6-mm4 to boot
on my AS2100.  It boots and runs fine with 2.4.21.  The following is what
I get when I boot -test6-mm4 (although it stops at the same point with
vanilla -test6).  Any suggestions, comments, or requests for further
information are welcome.


aboot: starting kernel boot/vmlinuz-2.6.0-test6-mm4 with arguments ro root=/dev/sda1 console=ttyS0,9600n8
Linux version 2.6.0-test6-mm4 (root@buddha) (gcc version 3.3.2 20030908 (Debian prerelease)) #1 SMP Mon Oct 6 18:14:06 CDT 2003
Booting on Sable using machine vector Sable from SRM
Major Options: SMP LEGACY_START VERBOSE_MCHECK MAGIC_SYSRQ 
Command line: ro root=/dev/sda1 console=ttyS0,9600n8
memcluster 0, usage 1, start        0, end      256
memcluster 1, usage 0, start      256, end    65527
memcluster 2, usage 1, start    65527, end    65536
freeing pages 256:384
freeing pages 811:65527
reserving pages 811:812
t2_init_arch: enabling SG TLB, IOCSR was 0xfe000084230a8190
SMP: 4 CPUs probed -- cpu_present_mask = f
On node 0 totalpages: 65527
  DMA zone: 65527 pages, LIFO batch:8
  Normal zone: 0 pages, LIFO batch:1
  HighMem zone: 0 pages, LIFO batch:1
Building zonelist for node : 0
Kernel command line: ro root=/dev/sda1 console=ttyS0,9600n8
PID hash table entries: 2048 (order 11: 32768 bytes)
Using epoch = 1952
Turning on RTC interrupts.
Console: colour VGA+ 80x25
Memory: 512960k/524216k available (2056k kernel code, 8840k reserved, 301k data, 440k init)
Calibrating delay loop... 545.60 BogoMIPS
Dentry cache hash table entries: 65536 (order: 6, 524288 bytes)
Inode-cache hash table entries: 32768 (order: 5, 262144 bytes)
Mount-cache hash table entries: 512 (order: 0, 8192 bytes)
POSIX conformance testing by UNIFIX
per-CPU timeslice cutoff: 374.48 usecs.
task migration cache decay timeout: 0 msecs.
SMP starting up secondaries.
Starting migration thread for cpu 0
Bringing up 1
Calibrating delay loop... 547.16 BogoMIPS
CPU 1 IS NOW UP!
Starting migration thread for cpu 1
Bringing up 2
Calibrating delay loop... 547.16 BogoMIPS
CPU 2 IS NOW UP!
Starting migration thread for cpu 2
Bringing up 3
Calibrating delay loop... 549.72 BogoMIPS
CPU 3 IS NOW UP!
Starting migration thread for cpu 3
CPUS done 4
SMP: Total of 4 processors activated (2195.89 BogoMIPS).
NET: Registered protocol family 16
EISA bus registered
PCI: Bus 1, bridge: 0000:00:07.0
  IO window: 9000-9fff
  MEM window: 01100000-011fffff
  PREFETCH window: disabled.
PCI: Bus 2, bridge: 0000:00:08.0

----
When I boot with 2.4.21, the PCI lines look like:

PCI: Bus 1, bridge: Digital Equipment Corporation DECchip 21050
  IO window: 9000-9fff
  MEM window: 01100000-011fffff
  PREFETCH window: disabled.
PCI: Bus 2, bridge: Digital Equipment Corporation DECchip 21152
  IO window: a000-afff
  MEM window: 01200000-012fffff
  PREFETCH window: 01300000-013fffff


-- 
Nathan Poznick <kraken@drunkmonkey.org>

Debt is a prolific mother of folly and of crime. - Benjamin Disraeli


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.0-test6[-mm4] boot failure on alpha

[Ivan Kokshaysky]

On Mon, Oct 06, 2003 at 06:57:59PM -0500, Nathan wrote:
> I'm having a problem getting 2.6.0-test6 and/or 2.6.0-test6-mm4 to boot
> on my AS2100.  It boots and runs fine with 2.4.21.  The following is what
> I get when I boot -test6-mm4 (although it stops at the same point with
> vanilla -test6).  Any suggestions, comments, or requests for further
> information are welcome.

Can you reproduce that with "nosmp" on boot command line?

Ivan.

Re: 2.6.0-test6[-mm4] boot failure on alpha

[Nathan Poznick]

Thus spake Ivan Kokshaysky:
> Can you reproduce that with "nosmp" on boot command line?

With nosmp, it still appears to hang in the same place:

SMP mode deactivated. 
Starting migration thread for cpu 0
CPUS done 0
SMP: Total of 1 processors activated (548.29 BogoMIPS).
NET: Registered protocol family 16
EISA bus registered
PCI: Bus 1, bridge: 0000:00:07.0
  IO window: 9000-9fff
  MEM window: 01100000-011fffff
  PREFETCH window: disabled.
PCI: Bus 2, bridge: 0000:00:08.0


-- 
Nathan Poznick <kraken@drunkmonkey.org>

"This man must have led a very full and active life..." "'Cause he's
got a squirrel in his stomach!" -Dr. Raeburn/Crow (as Dr. Raeburn).
#206