mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* 2.6.18-rc1 fails to boot on Ultra 5
@ 2006-07-09 10:40 Mikael Pettersson
  2006-07-09 18:44 ` David Miller
  2006-07-13  6:24 ` David Miller
  0 siblings, 2 replies; 10+ messages in thread
From: Mikael Pettersson @ 2006-07-09 10:40 UTC (permalink / raw)
  To: davem; +Cc: linux-kernel, sparclinux

2.6.17-git7 was the last development kernel to boot on my U5.
2.6.17-git8 to -git15 all hang immediately after being loaded.
2.6.17-git16 to 2.6.18-rc1 partially boot but crash and burn in
different places depending on kernel configuration: my standard
config got alignment exceptions in the floppy driver followed by
(sabre?) PCI errors and a hang; a minimal kernel gets further but
fails "su" probe and then oopses hard when the /dev/hda partition
table is about to be printed.

I'll try to capture the kernel messages soon, but I don't have
a null-modem serial cable to the U5 yet, and my attempts to use
a digital camera only resulted in blurry pictures of the screen :-(

/Mikael

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: 2.6.18-rc1 fails to boot on Ultra 5
@ 2006-07-13 12:18 Mikael Pettersson
  2006-07-13 13:38 ` Jima
                   ` (2 more replies)
  0 siblings, 3 replies; 10+ messages in thread
From: Mikael Pettersson @ 2006-07-13 12:18 UTC (permalink / raw)
  To: davem, mikpe; +Cc: linux-kernel, sparclinux

On Wed, 12 Jul 2006 23:24:19 -0700 (PDT), David Miller wrote:
>From: Mikael Pettersson <mikpe@it.uu.se>
>Date: Sun, 9 Jul 2006 12:40:58 +0200 (MEST)
>
>> 2.6.17-git16 to 2.6.18-rc1 partially boot but crash and burn in
>> different places depending on kernel configuration: my standard
>> config got alignment exceptions in the floppy driver followed by
>> (sabre?) PCI errors and a hang; a minimal kernel gets further but
>> fails "su" probe and then oopses hard when the /dev/hda partition
>> table is about to be printed.
>
>Several bugs operating here, all of which should be addressed by this
>patch below.  I was able to bootup my ultra5 fully, use the keyboard
>to login, etc. etc.

Confirmed. With this patch my Ultra5 boots fine again. Thanks.

(I still need that patch to repair PCI mmap() for X11, but
that's a different issue.)

>Let me know if you have any remaining problems.

No actual problems, but I noticed some differences in the
kernel log:

--- dmesg-2.6.17	2006-07-08 15:14:41.000000000 +0200
+++ dmesg-2.6.18-rc1	2006-07-13 13:30:34.000000000 +0200
...
 Kernel command line: ro root=LABEL=/
 PID hash table entries: 1024 (order: 10, 8192 bytes)
+start_kernel(): bug: interrupts were enabled early
 Console: colour dummy device 80x25
 Dentry cache hash table entries: 32768 (order: 5, 262144 bytes)

Something to worry about or not?

...
 Console: switching to colour frame buffer device 144x56
 atyfb: fb0: ATY Mach64 frame buffer device on PCI
 rtc_init: no PC rtc found
-su0 at 0x000001fff13062f8 (irq = 5,7ea) is a 16550A
-su1 at 0x000001fff13083f8 (irq = 9,7e9) is a 16550A
-ttyS0 at MMIO 0x1fff1400000 (irq = 6681504) is a SAB82532 V3.2
-ttyS1 at MMIO 0x1fff1400040 (irq = 6681504) is a SAB82532 V3.2
-isa bounce pool size: 16 pages
+se@14,400000: ttyS1 at MMIO 0x1fff1400000 (irq = 5) is a SAB82532 V3.2
 Floppy drive(s): fd0 is 1.44M
 FDC 0 is a National Semiconductor PC87306
 RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize

So ttyS0 became ttyS1, and the serial port at 0x1fff1400040
disappeared. OTOH, my Ultra5 only has a single serial port
connector so perhaps the old kernel was wrong in reporting
two serial ports.

/Mikael

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: 2.6.18-rc1 fails to boot on Ultra 5
@ 2006-07-14 11:28 Mikael Pettersson
  2006-07-14 12:49 ` Jima
  0 siblings, 1 reply; 10+ messages in thread
From: Mikael Pettersson @ 2006-07-14 11:28 UTC (permalink / raw)
  To: davem, mikpe; +Cc: linux-kernel, sparclinux

On Thu, 13 Jul 2006 16:04:01 -0700 (PDT), David Miller wrote:
>From: Mikael Pettersson <mikpe@it.uu.se>
>Date: Thu, 13 Jul 2006 14:18:33 +0200 (MEST)
>
>> No actual problems, but I noticed some differences in the
>> kernel log:
>
>This should take care of all 4 issues.  The typo in the sunsab
>driver causing the first port to not show up was good for a
>laugh. :)
>
>Let me know if it works for you too.

Works fine. The bug() is gone and I have both ttyS0 and ttyS1
again. Thanks.

(What I wrote previously about my Ultra5 only having one serial
port connector was wrong; I'm so used to DB9 connectors that I
mistook the DB25 connector for something else.)

/Mikael

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2006-07-14 12:49 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-07-09 10:40 2.6.18-rc1 fails to boot on Ultra 5 Mikael Pettersson
2006-07-09 18:44 ` David Miller
2006-07-13  6:24 ` David Miller
2006-07-13 12:18 Mikael Pettersson
2006-07-13 13:38 ` Jima
2006-07-13 20:23   ` David Miller
2006-07-13 20:17 ` David Miller
2006-07-13 23:04 ` David Miller
2006-07-14 11:28 Mikael Pettersson
2006-07-14 12:49 ` Jima

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®