Moin.
With 2.6.18.x everything works fine.
But 2.16.19-rc does not boot if the laptop runs on battery _and_ lapic is
defined as boot parameter.
The kernel loads and starts, for a fraction of a second some messages appears,
then the screen goes blank and nothing more happens.
I'm unable to read the last message, screen blanking is to fast, but the
'picture' looks like that he stops near the messages where at the normal
boot demsg "Local APIC disabled by BIOS -- you can enable it with "lapic""
appears.
Netconsole is not started at this point, and the laptop doesn't have a
serial port.
dmesg on_battery without lpapic:
Linux version 2.6.19-rc4 (root@halso) (gcc version 4.1.1 (Gentoo 4.1.1)) #1 PREEMPT Tue Oct 31 21:42:25 CET 2006
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009f000 (usable)
BIOS-e820: 000000000009f000 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000d2000 - 00000000000d4000 (reserved)
BIOS-e820: 00000000000dc000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 000000003ff50000 (usable)
BIOS-e820: 000000003ff50000 - 000000003ff67000 (ACPI data)
BIOS-e820: 000000003ff67000 - 000000003ff79000 (ACPI NVS)
BIOS-e820: 000000003ff80000 - 0000000040000000 (reserved)
BIOS-e820: 00000000ff800000 - 0000000100000000 (reserved)
127MB HIGHMEM available.
896MB LOWMEM available.
Entering add_active_range(0, 0, 261968) 0 entries of 256 used
Zone PFN ranges:
DMA 0 -> 4096
Normal 4096 -> 229376
HighMem 229376 -> 261968
early_node_map[1] active PFN ranges
0: 0 -> 261968
On node 0 totalpages: 261968
DMA zone: 32 pages used for memmap
DMA zone: 0 pages reserved
DMA zone: 4064 pages, LIFO batch:0
Normal zone: 1760 pages used for memmap
Normal zone: 223520 pages, LIFO batch:31
HighMem zone: 254 pages used for memmap
HighMem zone: 32338 pages, LIFO batch:7
DMI present.
ACPI: RSDP (v002 IBM ) @ 0x000f6d70
ACPI: XSDT (v001 IBM TP-1R 0x00003210 LTP 0x00000000) @ 0x3ff5a6bd
ACPI: FADT (v003 IBM TP-1R 0x00003210 IBM 0x00000001) @ 0x3ff5a800
ACPI: SSDT (v001 IBM TP-1R 0x00003210 MSFT 0x0100000e) @ 0x3ff5a9b4
ACPI: ECDT (v001 IBM TP-1R 0x00003210 IBM 0x00000001) @ 0x3ff66ecc
ACPI: TCPA (v001 IBM TP-1R 0x00003210 PTL 0x00000001) @ 0x3ff66f1e
ACPI: BOOT (v001 IBM TP-1R 0x00003210 LTP 0x00000001) @ 0x3ff66fd8
ACPI: DSDT (v001 IBM TP-1R 0x00003210 MSFT 0x0100000e) @ 0x00000000
ACPI: PM-Timer IO Port: 0x1008
Allocating PCI resources starting at 50000000 (gap: 40000000:bf800000)
Detected 598.086 MHz processor.
Built 1 zonelists. Total pages: 259922
Kernel command line: root=/dev/sda4 [email protected]/eth0,[email protected]/00:30:1B:B2:40:0C
netconsole: local port 4444
netconsole: local IP 217.7.64.201
netconsole: interface eth0
netconsole: remote port 6666
netconsole: remote IP 217.7.64.224
netconsole: remote ethernet address 00:30:1b:b2:40:0c
Local APIC disabled by BIOS -- you can enable it with "lapic"
mapped APIC to ffffd000 (01803000)
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Initializing CPU#0
CPU 0 irqstacks, hard=c0468000 soft=c0467000
PID hash table entries: 4096 (order: 12, 16384 bytes)
Console: colour VGA+ 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 1034196k/1047872k available (2498k kernel code, 13064k reserved, 772k data, 188k init, 130368k highmem)
virtual kernel memory layout:
fixmap : 0xfffaa000 - 0xfffff000 ( 340 kB)
pkmap : 0xff800000 - 0xffc00000 (4096 kB)
vmalloc : 0xf8800000 - 0xff7fe000 ( 111 MB)
lowmem : 0xc0000000 - 0xf8000000 ( 896 MB)
.init : 0xc0433000 - 0xc0462000 ( 188 kB)
.data : 0xc0370837 - 0xc0431994 ( 772 kB)
.text : 0xc0100000 - 0xc0370837 (2498 kB)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay using timer specific routine.. 1196.94 BogoMIPS (lpj=5984714)
Mount-cache hash table entries: 512
CPU: After generic identify, caps: a7e9f9bf 00000000 00000000 00000000 00000180 00000000 00000000
CPU: L1 I cache: 32K, L1 D cache: 32K
CPU: L2 cache: 1024K
CPU: After all inits, caps: a7e9f9bf 00000000 00000000 00000040 00000180 00000000 00000000
[..... and so on ....]
diff to on_ac with lapic:
--- dm_batt 2006-10-31 21:53:36.000000000 +0100
+++ dm_pwr 2006-10-31 22:15:42.000000000 +0100
@@ -37,17 +37,18 @@
ACPI: DSDT (v001 IBM TP-1R 0x00003210 MSFT 0x0100000e) @ 0x00000000
ACPI: PM-Timer IO Port: 0x1008
Allocating PCI resources starting at 50000000 (gap: 40000000:bf800000)
-Detected 598.086 MHz processor.
+Detected 1694.558 MHz processor.
Built 1 zonelists. Total pages: 259922
-Kernel command line: root=/dev/sda4 [email protected]/eth0,[email protected]/00:30:1B:B2:40:0C
+Kernel command line: root=/dev/sda4 lapic [email protected]/eth0,[email protected]/00:30:1B:B2:40:0C
netconsole: local port 4444
netconsole: local IP 217.7.64.201
netconsole: interface eth0
netconsole: remote port 6666
netconsole: remote IP 217.7.64.224
netconsole: remote ethernet address 00:30:1b:b2:40:0c
-Local APIC disabled by BIOS -- you can enable it with "lapic"
-mapped APIC to ffffd000 (01803000)
+Local APIC disabled by BIOS -- reenabling.
+Found and enabled local APIC!
+mapped APIC to ffffd000 (fee00000)
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Initializing CPU#0
@@ -56,7 +57,7 @@
Console: colour VGA+ 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
-Memory: 1034196k/1047872k available (2498k kernel code, 13064k reserved, 772k data, 188k init, 130368k highmem)
+Memory: 1034196k/1047872k available (2498k kernel code, 13060k reserved, 772k data, 188k init, 130368k highmem)
virtual kernel memory layout:
fixmap : 0xfffaa000 - 0xfffff000 ( 340 kB)
pkmap : 0xff800000 - 0xffc00000 (4096 kB)
@@ -66,12 +67,12 @@
.data : 0xc0370837 - 0xc0431994 ( 772 kB)
.text : 0xc0100000 - 0xc0370837 (2498 kB)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
-Calibrating delay using timer specific routine.. 1196.94 BogoMIPS (lpj=5984714)
+Calibrating delay using timer specific routine.. 3390.66 BogoMIPS (lpj=16953345)
Mount-cache hash table entries: 512
-CPU: After generic identify, caps: a7e9f9bf 00000000 00000000 00000000 00000180 00000000 00000000
+CPU: After generic identify, caps: a7e9fbbf 00000000 00000000 00000000 00000180 00000000 00000000
CPU: L1 I cache: 32K, L1 D cache: 32K
CPU: L2 cache: 1024K
-CPU: After all inits, caps: a7e9f9bf 00000000 00000000 00000040 00000180 00000000 00000000
+CPU: After all inits, caps: a7e9fbbf 00000000 00000000 00000040 00000180 00000000 00000000
[...]
Patches welcome;)
<earny>
On Tue, 2006-10-31 at 22:27 +0100, Ernst Herzberg wrote:
> With 2.6.18.x everything works fine.
>
> But 2.16.19-rc does not boot if the laptop runs on battery _and_ lapic
> is
> defined as boot parameter.
>
> The kernel loads and starts, for a fraction of a second some messages
> appears,
> then the screen goes blank and nothing more happens.
>
> I'm unable to read the last message, screen blanking is to fast, but
> the
> 'picture' looks like that he stops near the messages where at the
> normal
> boot demsg "Local APIC disabled by BIOS -- you can enable it with
> "lapic""
> appears.
Hi,
If you don't enable lapic, you will see cat /proc/interrupts with
XT-PIC.
if you try enable lapic, somehow IRQ routing should change
and if /proc/interrupts still the same, with IRQs on XT-PIC.
I think, lapic still not enable and you just get problems.
Unless you know that lapic works, you should try it enable because it a
trap that just give you a problem.
Some years ago (2002/3) was a very common bug kernel try enable
automatically lapic (when BIOS don't) and computer hangs .
--
S?rgio M.B.
On Tue, Oct 31, 2006 at 10:27:54PM +0100, Ernst Herzberg wrote:
>
> Moin.
>
> With 2.6.18.x everything works fine.
>
> But 2.16.19-rc does not boot if the laptop runs on battery _and_ lapic is
> defined as boot parameter.
>
> The kernel loads and starts, for a fraction of a second some messages appears,
> then the screen goes blank and nothing more happens.
>
> I'm unable to read the last message, screen blanking is to fast, but the
> 'picture' looks like that he stops near the messages where at the normal
> boot demsg "Local APIC disabled by BIOS -- you can enable it with "lapic""
> appears.
@Ingo:
Any ideas?
@Ernst:
Thanks for your report.
What model is your laptop?
Unless someone is able to spot the problem from your bug report, please
do the following process of git bisecting for finding what broke it:
# install git and cogito on your computer
# clone Linus' tree:
cg-clone \
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
# start bisecting:
cd linux-2.6
git bisect start
git bisect bad v2.6.19-rc1
git bisect good v2.6.18
# start round
cp /path/to/.config .
make oldconfig
make
# install kernel, boot, check whether it's good or bad, then:
git bisect [bad|good]
# start next round
After at about 12 reboots, you'll have found the guilty commit
("... is first bad commit").
More information on git bisecting:
man git-bisect
> Netconsole is not started at this point, and the laptop doesn't have a
> serial port.
>
> dmesg on_battery without lpapic:
>
> Linux version 2.6.19-rc4 (root@halso) (gcc version 4.1.1 (Gentoo 4.1.1)) #1 PREEMPT Tue Oct 31 21:42:25 CET 2006
> BIOS-provided physical RAM map:
> BIOS-e820: 0000000000000000 - 000000000009f000 (usable)
> BIOS-e820: 000000000009f000 - 00000000000a0000 (reserved)
> BIOS-e820: 00000000000d2000 - 00000000000d4000 (reserved)
> BIOS-e820: 00000000000dc000 - 0000000000100000 (reserved)
> BIOS-e820: 0000000000100000 - 000000003ff50000 (usable)
> BIOS-e820: 000000003ff50000 - 000000003ff67000 (ACPI data)
> BIOS-e820: 000000003ff67000 - 000000003ff79000 (ACPI NVS)
> BIOS-e820: 000000003ff80000 - 0000000040000000 (reserved)
> BIOS-e820: 00000000ff800000 - 0000000100000000 (reserved)
> 127MB HIGHMEM available.
> 896MB LOWMEM available.
> Entering add_active_range(0, 0, 261968) 0 entries of 256 used
> Zone PFN ranges:
> DMA 0 -> 4096
> Normal 4096 -> 229376
> HighMem 229376 -> 261968
> early_node_map[1] active PFN ranges
> 0: 0 -> 261968
> On node 0 totalpages: 261968
> DMA zone: 32 pages used for memmap
> DMA zone: 0 pages reserved
> DMA zone: 4064 pages, LIFO batch:0
> Normal zone: 1760 pages used for memmap
> Normal zone: 223520 pages, LIFO batch:31
> HighMem zone: 254 pages used for memmap
> HighMem zone: 32338 pages, LIFO batch:7
> DMI present.
> ACPI: RSDP (v002 IBM ) @ 0x000f6d70
> ACPI: XSDT (v001 IBM TP-1R 0x00003210 LTP 0x00000000) @ 0x3ff5a6bd
> ACPI: FADT (v003 IBM TP-1R 0x00003210 IBM 0x00000001) @ 0x3ff5a800
> ACPI: SSDT (v001 IBM TP-1R 0x00003210 MSFT 0x0100000e) @ 0x3ff5a9b4
> ACPI: ECDT (v001 IBM TP-1R 0x00003210 IBM 0x00000001) @ 0x3ff66ecc
> ACPI: TCPA (v001 IBM TP-1R 0x00003210 PTL 0x00000001) @ 0x3ff66f1e
> ACPI: BOOT (v001 IBM TP-1R 0x00003210 LTP 0x00000001) @ 0x3ff66fd8
> ACPI: DSDT (v001 IBM TP-1R 0x00003210 MSFT 0x0100000e) @ 0x00000000
> ACPI: PM-Timer IO Port: 0x1008
> Allocating PCI resources starting at 50000000 (gap: 40000000:bf800000)
> Detected 598.086 MHz processor.
> Built 1 zonelists. Total pages: 259922
> Kernel command line: root=/dev/sda4 [email protected]/eth0,[email protected]/00:30:1B:B2:40:0C
> netconsole: local port 4444
> netconsole: local IP 217.7.64.201
> netconsole: interface eth0
> netconsole: remote port 6666
> netconsole: remote IP 217.7.64.224
> netconsole: remote ethernet address 00:30:1b:b2:40:0c
> Local APIC disabled by BIOS -- you can enable it with "lapic"
> mapped APIC to ffffd000 (01803000)
> Enabling fast FPU save and restore... done.
> Enabling unmasked SIMD FPU exception support... done.
> Initializing CPU#0
> CPU 0 irqstacks, hard=c0468000 soft=c0467000
> PID hash table entries: 4096 (order: 12, 16384 bytes)
> Console: colour VGA+ 80x25
> Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
> Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
> Memory: 1034196k/1047872k available (2498k kernel code, 13064k reserved, 772k data, 188k init, 130368k highmem)
> virtual kernel memory layout:
> fixmap : 0xfffaa000 - 0xfffff000 ( 340 kB)
> pkmap : 0xff800000 - 0xffc00000 (4096 kB)
> vmalloc : 0xf8800000 - 0xff7fe000 ( 111 MB)
> lowmem : 0xc0000000 - 0xf8000000 ( 896 MB)
> .init : 0xc0433000 - 0xc0462000 ( 188 kB)
> .data : 0xc0370837 - 0xc0431994 ( 772 kB)
> .text : 0xc0100000 - 0xc0370837 (2498 kB)
> Checking if this processor honours the WP bit even in supervisor mode... Ok.
> Calibrating delay using timer specific routine.. 1196.94 BogoMIPS (lpj=5984714)
> Mount-cache hash table entries: 512
> CPU: After generic identify, caps: a7e9f9bf 00000000 00000000 00000000 00000180 00000000 00000000
> CPU: L1 I cache: 32K, L1 D cache: 32K
> CPU: L2 cache: 1024K
> CPU: After all inits, caps: a7e9f9bf 00000000 00000000 00000040 00000180 00000000 00000000
> [..... and so on ....]
>
>
> diff to on_ac with lapic:
>
> --- dm_batt 2006-10-31 21:53:36.000000000 +0100
> +++ dm_pwr 2006-10-31 22:15:42.000000000 +0100
> @@ -37,17 +37,18 @@
> ACPI: DSDT (v001 IBM TP-1R 0x00003210 MSFT 0x0100000e) @ 0x00000000
> ACPI: PM-Timer IO Port: 0x1008
> Allocating PCI resources starting at 50000000 (gap: 40000000:bf800000)
> -Detected 598.086 MHz processor.
> +Detected 1694.558 MHz processor.
> Built 1 zonelists. Total pages: 259922
> -Kernel command line: root=/dev/sda4 [email protected]/eth0,[email protected]/00:30:1B:B2:40:0C
> +Kernel command line: root=/dev/sda4 lapic [email protected]/eth0,[email protected]/00:30:1B:B2:40:0C
> netconsole: local port 4444
> netconsole: local IP 217.7.64.201
> netconsole: interface eth0
> netconsole: remote port 6666
> netconsole: remote IP 217.7.64.224
> netconsole: remote ethernet address 00:30:1b:b2:40:0c
> -Local APIC disabled by BIOS -- you can enable it with "lapic"
> -mapped APIC to ffffd000 (01803000)
> +Local APIC disabled by BIOS -- reenabling.
> +Found and enabled local APIC!
> +mapped APIC to ffffd000 (fee00000)
> Enabling fast FPU save and restore... done.
> Enabling unmasked SIMD FPU exception support... done.
> Initializing CPU#0
> @@ -56,7 +57,7 @@
> Console: colour VGA+ 80x25
> Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
> Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
> -Memory: 1034196k/1047872k available (2498k kernel code, 13064k reserved, 772k data, 188k init, 130368k highmem)
> +Memory: 1034196k/1047872k available (2498k kernel code, 13060k reserved, 772k data, 188k init, 130368k highmem)
> virtual kernel memory layout:
> fixmap : 0xfffaa000 - 0xfffff000 ( 340 kB)
> pkmap : 0xff800000 - 0xffc00000 (4096 kB)
> @@ -66,12 +67,12 @@
> .data : 0xc0370837 - 0xc0431994 ( 772 kB)
> .text : 0xc0100000 - 0xc0370837 (2498 kB)
> Checking if this processor honours the WP bit even in supervisor mode... Ok.
> -Calibrating delay using timer specific routine.. 1196.94 BogoMIPS (lpj=5984714)
> +Calibrating delay using timer specific routine.. 3390.66 BogoMIPS (lpj=16953345)
> Mount-cache hash table entries: 512
> -CPU: After generic identify, caps: a7e9f9bf 00000000 00000000 00000000 00000180 00000000 00000000
> +CPU: After generic identify, caps: a7e9fbbf 00000000 00000000 00000000 00000180 00000000 00000000
> CPU: L1 I cache: 32K, L1 D cache: 32K
> CPU: L2 cache: 1024K
> -CPU: After all inits, caps: a7e9f9bf 00000000 00000000 00000040 00000180 00000000 00000000
> +CPU: After all inits, caps: a7e9fbbf 00000000 00000000 00000040 00000180 00000000 00000000
> [...]
>
> Patches welcome;)
>
> <earny>
>
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Wed, 2006-11-01 at 00:27 +0000, Sergio Monteiro Basto wrote:
> On Tue, 2006-10-31 at 22:27 +0100, Ernst Herzberg wrote:
> > With 2.6.18.x everything works fine.
> >
> > But 2.16.19-rc does not boot if the laptop runs on battery _and_ lapic
> > is
> > defined as boot parameter.
> >
> > The kernel loads and starts, for a fraction of a second some messages
> > appears,
> > then the screen goes blank and nothing more happens.
> >
> > I'm unable to read the last message, screen blanking is to fast, but
> > the
> > 'picture' looks like that he stops near the messages where at the
> > normal
> > boot demsg "Local APIC disabled by BIOS -- you can enable it with
> > "lapic""
> > appears.
>
> Hi,
> If you don't enable lapic, you will see cat /proc/interrupts with
> XT-PIC.
> if you try enable lapic, somehow IRQ routing should change
> and if /proc/interrupts still the same, with IRQs on XT-PIC.
> I think, lapic still not enable and you just get problems.
> Unless you know that lapic works, you should try it enable because it a
> trap that just give you a problem.
> Some years ago (2002/3) was a very common bug kernel try enable
> automatically lapic (when BIOS don't) and computer hangs .
>
>
s/you should try/you shouldn't try/
--
S?rgio M.B.
On Wednesday 01 November 2006 02:04, Adrian Bunk wrote:
> @Ingo:
> Any ideas?
>
>
> @Ernst:
> Thanks for your report.
> What model is your laptop?
It's a Thinkpad R50p
> Unless someone is able to spot the problem from your bug report, please
> do the following process of git bisecting for finding what broke it:
> ....
> After at about 12 reboots, ...
Will try, if i get a little spare time ;-)
Thanks,
<earny>
On Wed, Nov 01, 2006 at 03:05:33AM +0100, Ernst Herzberg wrote:
> On Wednesday 01 November 2006 02:04, Adrian Bunk wrote:
>
> > @Ingo:
> > Any ideas?
> >
> >
> > @Ernst:
> > Thanks for your report.
> > What model is your laptop?
>
> It's a Thinkpad R50p
Does anyone own a Thinkpad 2.6.19-rc did not break? ;-)
We have a (most likely unrelated) problem after resume reported by
owners of three different Thinkpad models.
> > Unless someone is able to spot the problem from your bug report, please
> > do the following process of git bisecting for finding what broke it:
> > ....
> > After at about 12 reboots, ...
>
> Will try, if i get a little spare time ;-)
Thanks a lot!
> Thanks,
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Wednesday 01 November 2006 03:34, Adrian Bunk wrote:
> On Wed, Nov 01, 2006 at 03:05:33AM +0100, Ernst Herzberg wrote:
> > On Wednesday 01 November 2006 02:04, Adrian Bunk wrote:
> > > @Ingo:
> > > Any ideas?
> > >
> > >
> > > @Ernst:
> > > Thanks for your report.
> > > What model is your laptop?
> >
> > It's a Thinkpad R50p
>
> Does anyone own a Thinkpad 2.6.19-rc did not break? ;-)
>
> We have a (most likely unrelated) problem after resume reported by
> owners of three different Thinkpad models.
>
> > > Unless someone is able to spot the problem from your bug report, please
> > > do the following process of git bisecting for finding what broke it:
> > > ....
> > > After at about 12 reboots, ...
> >
> > Will try, if i get a little spare time ;-)
>
> Thanks a lot!
halso:~/bisect/linux-2.6 # git bisect good
1fbbac4bcb03033d325c71fc7273aa0b9c1d9a03 is first bad commit
commit 1fbbac4bcb03033d325c71fc7273aa0b9c1d9a03
Author: Russell King <[email protected]>
Date: Sat Sep 16 21:09:41 2006 +0100
[SERIAL] serial_cs: convert multi-port table to quirk table
- rename multi_id table to serial_quirk / quirks[]
- use named initialisers
- store a pointer to the quirk table in the serial_info structure
so we can use the quirk table entry later.
- apply multi-port quirk after the multi-port guessing code,
but only if it's != -1.
Signed-off-by: Russell King <[email protected]>
:040000 040000 0e673263987d236da67e6668100335f3223626ea
6a32d673832850311983f072189fbaa09855f875 M drivers
(if there is no time, don't sleep:)
First feeling: unrelatet. But if i compile 2.6.19-rc4 without cardbus support,
the laptop boots on_battery and with lapic!
Need first a little sleep....
Thanks
<earny>
PS: _EXACT_ 12 kernels :-)
On Tuesday 31 October 2006 16:27, Ernst Herzberg wrote:
> With 2.6.18.x everything works fine.
>
> But 2.16.19-rc does not boot if the laptop runs on battery _and_ lapic is
> defined as boot parameter.
> Local APIC disabled by BIOS -- you can enable it with "lapic"
The BIOS is advising you here that it is a bad idea to enable the LAPIC on this system.
So why are you using the "lapic" boot parameter?
If you are running an CONFIG_SMP kernel with LAPIC enabled and deep C-states
(such as are available on Thinkpads when on battery mode), you will
run into the following bug:
http://bugzilla.kernel.org/show_bug.cgi?id=7376
As this is not new, the mystery is really why you see no problems in 2.6.18.
Perhaps you can forward the contents of /proc/interrupts for 2.6.18 with "lapic"
and we can see if the timer and the LOC are in sync or not when on battery?
/proc/acpi/processor/*/power will also tell us about the C-states --
dump it for both AC and battery mode.
thanks,
-Len