Re: PROBLEM: E6850 has an 8+ minute delay during boot

[Andrew Morton <akpm@linux-foundation.org>]

On Fri, 14 Dec 2007 04:10:31 -0500 "Arun Thomas" <arun.thomas@gmail.com> wrote:

> My Dell Vostro desktop w/ an Intel E6850 dual-core 3GHz CPU has an 8+
> minute delay during boot. The machine seems to run fine after it
> boots. The problem occurs on the Ubuntu gutsy 2.6.22-14 kernel,
> 2.6.23.9, and 2.6.24-rc5. I have upgraded to the most recent Dell BIOS
> version (1.0.8). I am running an SMP kernel currently, but the problem
> presents itself in uniprocessor mode (nosmp) also. I'm running the
> machine in 32-bit mode w/hyperthreading configured.
> 
> Here are a few interesting things in the dmesg output for the
> 2.6.24-rc5 kernel (smp). The full dmesg is also included farther down.
> 
> [   20.101501] CPU: After generic identify, caps: bfebfbff 20100000
> 00000000 00000000 0000e3fd 00000000 00000001 00000000
> [   20.101505] monitor/mwait feature present.
> [   20.101506] using mwait in idle threads.
> [   20.101508] CPU: L1 I cache: 32K, L1 D cache: 32K
> [   20.101509] CPU: L2 cache: 4096K
> [   20.101511] CPU: Physical Processor ID: 0
> [   20.101511] CPU: Processor Core ID: 0
> [   20.101513] CPU: After all inits, caps: bfebfbff 20100000 00000000
> 00003940 0000e3fd 00000000 00000001 00000000
> [   20.101517] Compat vDSO mapped to ffffe000.
> [   20.101525] Checking 'hlt' instruction... OK.
> [   20.117651] SMP alternatives: switching to UP code
> [   20.118574] ACPI: Core revision 20070126
> [   20.175154] CPU0: Intel(R) Core(TM)2 Duo CPU     E6850  @ 3.00GHz stepping 0b
> [   20.175163] SMP alternatives: switching to SMP code
> [   20.175437] Booting processor 1/1 eip 3000
> [   20.185881] Initializing CPU#1
> [  527.926587] Calibrating delay using timer specific routine..
> 200566.59 BogoMIPS (lpj=401133185)
> [  527.926592] CPU: After generic identify, caps: bfebfbff 20100000
> 00000000 00000000 0000e3fd 00000000 00000001 00000000
> [  527.926595] monitor/mwait feature present.
> [  527.926596] CPU: L1 I cache: 32K, L1 D cache: 32K
> [  527.926597] CPU: L2 cache: 4096K
> [  527.926599] CPU: Physical Processor ID: 0

Silly question: is that delay actually observeable by a human,
or is it just a leap in the printk timestamping?