RE: [RFC][PATCH] i386 x86-64 Eliminate Local APIC timer interrupt

From: Pallipadi, Venkatesh
Date: Fri Apr 29 2005 - 21:57:31 EST

>-----Original Message-----
>From: Andrew Morton [mailto:akpm@xxxxxxxx]
>Sent: Friday, April 29, 2005 7:33 PM
>To: Pallipadi, Venkatesh
>Cc: torvalds@xxxxxxxx; mingo@xxxxxxx;
>linux-kernel@xxxxxxxxxxxxxxx; Shah, Rajesh;
>johnstul@xxxxxxxxxx; ak@xxxxxxx; Mallick, Asit K
>Subject: Re: [RFC][PATCH] i386 x86-64 Eliminate Local APIC
>timer interrupt
>The patch (at least, as I merged it) goes into a ghastly death
>spiral early
>in boot.
>Serial console says:
>Initializing CPU#1
>Calibrating delay using timer specific routine.. 5615.95
>BogoMIPS (lpj=2807978)
>CPU: Trace cache: 12K uops, L1 D cache: 8K
>CPU: L2 cache: 512K
>CPU: Physical Processor ID: 0
>CPU1: Intel Pentium 4 (Northwood) stepping 07
>Total of 2 processors activated (11238.26 BogoMIPS).
>..TIMER: vector=0x31 pin1=2 pin2=-1
>checking TSC synchronization across 2 CPUs: passed.
>Brought up 2 CPUs
>Unable to handle kernel NULL pointer dereference
>which isn't very helpful.
>tty output is at
>which is also less that totally useful.
>There's waaaaaaaaay too much low-level x86 stuff happening at
>present. We
>need to settle it down, go more slowly, take more care and test things
>better, please. Next -mm has already been delayed by two days
>due to my
>having to chase down all the bugs people have been sending me.

I did test this patch on variety of systems and didn't see any failures.

Probably some other change in mm conflicting with this patch?
Is there way to get the all the patches in mm, so that I can try same
Kernel and reproduce this failure?

I agree though that this patch is very risky and needs some discussion
and lot of testing before it goes into base.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at