[debian-non-standard] Ok, now it rocks...

[Tomasz Chmielewski]

Martin Steigerwald schrieb:

> 1) Fast boot times compared to the VIA box. Internet is available as soon 
> as my laptop completed resuming from hibernation (TuxOnIce). This is how 
> I wanted it. I could have optimized the boot times of the VIA box as 
> well. It had an ext2 mit fsck fix option and thus bootet twice, but even 
> when I changed to ext3 I think it would have been slower. That compact 
> flash card did not do IDE, but thats not that much of an issue I think. I 
> believe that the "BIOS" stuff on the ASUS is *way* faster. I wouldn't be 
> surprised if the WL-500g is already booting the Linux kernel while the 
> VIA box still does POST processing in the BIOS for seconds.

It doesn't have any BIOS (BIOS is x86-only world, I think).

It has a CFE bootloader, and kernel executes a ~second after the device 
is powered on.
Compare it to an imaginary PC without BIOS, which starts with GRUB or 
LILO immediately after you power it on.


This is what you see if you have a serial cable:


CFE version 1.0.37 for BCM947XX (32bit,SP,LE)
Build Date: �| 10�� 12 22:21:19 CST 2006 (root at localhost.localdomain)
Copyright (C) 2000,2001,2002,2003 Broadcom Corporation.

Initializing Arena
Initializing Devices.
et0: Broadcom BCM47xx 10/100 Mbps Ethernet Controller 3.90.7.0
rndis0: Broadcom USB RNDIS Network Adapter (P-t-P)
CPU type 0x29006: 264MHz
Total memory: 33554432 KBytes

Total memory used by CFE:  0x80800000 - 0x8089AF40 (634688)
Initialized Data:          0x808313D0 - 0x80833790 (9152)
BSS Area:                  0x80833790 - 0x80834F40 (6064)
Local Heap:                0x80834F40 - 0x80898F40 (409600)
Stack Area:                0x80898F40 - 0x8089AF40 (8192)
Text (code) segment:       0x80800000 - 0x808313D0 (201680)
Boot area (physical):      0x0089B000 - 0x008DB000
Relocation Factor:         I:00000000 - D:00000000

Device eth0:  hwaddr 00-0E-A6-F1-ED-3C, ipaddr 192.168.1.1, mask 
255.255.255.0
         gateway not set, nameserver not set
Null Rescue Flag.
Loader:raw Filesys:raw Dev:flash0.os File: Options:(null)
Loading: .. 3740 bytes read
Entry at 0x80001000
Closing network.
Starting program at 0x80001000
Linux version 2.6.19.2 (tch at tomek) (gcc version 4.1.2) #9 Tue Apr 3 
21:30:54 CEST 2007
CPU revision is: 00029006
(...)


> 2) It works without swap space for my purpose.

At least until you want to compile something bigger than helloworld.c ;)


> Three hints I have for now:
> 
> 1) I used commit=120 as ext3 mount option to reduce. No good idea. I good 
> journal commit errors and after such one the box couldn't even 
> open /bin/ls anymore. Ext3 was good and after a reboot it worked again 
> for some time. Same with commit=30. Anyway according to vmstat 1 in my 
> configuration the box just does not write to the USB stick during usual 
> ADSL routing operation

Hmm, strange you had filesystem corruption with commit=120.
Technically, it should work; a drawback being that you might loose the 
last 120 seconds of changes.
Although, I didn't test it personally.

Theoretically, there is a better filesystem than ext3: logfs[1]. It even 
has transparent compression, which is quite cool - USB sticks are still 
very expensive when compared to hard disks (price/capacity ratio).

It's not in the mainline kernel, although the author wanted to push it 
upstream some day (someone might ping him and ask him if he still wants 
to do it).
It worked for me, although was pretty slow.
The bad news: it's not possible to boot from this filesystem directly.

(...)


> At last I wonder why this doesn't work:
> 
> # DSL
> auto dsl-provider
> iface dsl-provider inet ppp
> pre-up /sbin/ifconfig eth0 up # line maintained by pppoeconf
> post-up /usr/sbin/ntpdate-debian
> provider dsl-provider
> 
> Time is not correct. I suspect some timing problem, i.e. ntpdate-debian is 
> executed before ppp0 is really ready.

Yep, ADSL sometimes needs a couple of seconds to connect (and ppp starts 
in the background), so it won't work reliably.



[1] http://logfs.org


-- 
Tomasz Chmielewski
http://wpkg.org