No error messages using huge kernel on 12.1

brianL · 05-05-2008, 05:43 AM

This is not a problem, just a bit puzzling.
I'm not getting all those:

Code:

kobject_add failed for uhci_hcd with -EEXIST

that occurred with 12.0 using the huge kernel, and which can be avoided by switching to the generic. Boot up is fast enough, everything seems to run OK. I ran Eric's mkinitrd_command_generator.sh, and got different results to what I got with 12.0:
For 12.0:

Code:

mkinitrd -c -k XXXXX -m pata_via:ata_generic:mbcache:jbd:ext3 -f ext3 -r /dev/hda3

For 12.1:

Code:

mkinitrd -c -k 2.6.24.5-smp -m pata_via:pata_acpi:ata_generic:mbcache:jbd:ext3 -f ext3 -r /dev/hda3

pata_acpi added. Not sure whether to make an initrd and switch to the generic kernel or not.

gegechris99 · 05-05-2008, 06:29 AM

Why not give it a try?

As this is a different kernel than from 12.0, it's no surprise that a new module appear.

Just put two entries in your /etc/lilo.conf ("generic + initrd" and "Huge") so that if initrd is not adequate, you can fall back to the huge kernel.

brianL · 05-05-2008, 06:40 AM

Quote:

Originally Posted by gegechris99

Why not give it a try?

As this is a different kernel than from 12.0, it's no surprise that a new module appear.

Just put two entries in your /etc/lilo.conf ("generic + initrd" and "Huge") so that if initrd is not adequate, you can fall back to the huge kernel.

Yeah, that's what I was thinking of doing, to see if there's any difference in the boot up messages. The differences were very noticable in 12.0 with huge and generic. But, as it says in CHANGES_AND_HINTS.TXT, those error messages can be ignored. I will make an initrd though, and try the generic kernel.

brianL · 05-05-2008, 07:18 AM

Just done it. Running on the generic kernel now. No noticeable differences in booting. Edited lilo.conf as so:

Code:

# Linux bootable partition config begins
# Generic kernel
image = /boot/vmlinuz-generic-smp-2.6.24.5-smp
  initrd = /boot/initrd.gz
  root = /dev/hda3
  label = LinuxG
  read-only
image = /boot/vmlinuz
  root = /dev/hda3
  label = LinuxH
  read-only
# Linux bootable partition config ends

andrew.46 · 05-05-2008, 07:41 AM

I have tried both:

Quote:

Originally Posted by brianL

Just done it. Running on the generic kernel now. No noticeable differences in booting.

and to tell you the truth I found the huge kernel booted faster and ran a little better although I believe this cannot be true :-)

Andrew

brianL · 05-05-2008, 07:50 AM

Quote:

Originally Posted by andrew.46

I have tried both:
and to tell you the truth I found the huge kernel booted faster
Andrew

Yeah, perhaps a shade faster. I was reading the booting messages - or trying to - looking for any stand out differences like there was with 12.0.

brianL · 05-06-2008, 04:19 AM

Just to satisfy my curiosity (killed the cat, but I'm still here (who said "Unfortunately"?)), yesterday when I had the huge, then generic, kernels running, I did:

Code:

lsmod > mod_G.txt

then:

Code:

lsmod > mod_H.txt

then:

Code:

diff -y mod_G.txt mod_H.txt > mod_diff.txt

and got this:

Code:

Module                  Size  Used by				Module                  Size  Used by
snd_seq_dummy           6660  0 				snd_seq_dummy           6660  0 
snd_seq_oss            32896  0 				snd_seq_oss            32896  0 
snd_seq_midi_event     10112  1 snd_seq_oss			snd_seq_midi_event     10112  1 snd_seq_oss
snd_seq                50256  5 snd_seq_dummy,snd_seq_oss,snd	snd_seq                50256  5 snd_seq_dummy,snd_seq_oss,snd
snd_pcm_oss            40352  0 				snd_pcm_oss            40352  0 
snd_mixer_oss          17920  1 snd_pcm_oss			snd_mixer_oss          17920  1 snd_pcm_oss
ipv6                  234724  12 				ipv6                  234724  12 
xt_limit                6144  2 				xt_limit                6144  2 
xt_pkttype              5632  1 				xt_pkttype              5632  1 
xt_tcpudp               6912  17 				xt_tcpudp               6912  17 
ipt_LOG                 9600  11 				ipt_LOG                 9600  11 
xt_state                5888  6 				xt_state                5888  6 
iptable_mangle          6144  0 				iptable_mangle          6144  0 
iptable_nat             9860  0 				iptable_nat             9860  0 
iptable_filter          6272  1 				iptable_filter          6272  1 
nf_conntrack_irc        8984  0 				nf_conntrack_irc        8984  0 
nf_nat_ftp              6656  0 				nf_nat_ftp              6656  0 
nf_nat                 19500  2 iptable_nat,nf_nat_ftp		nf_nat                 19500  2 iptable_nat,nf_nat_ftp
nf_conntrack_ftp       10912  1 nf_nat_ftp			nf_conntrack_ftp       10912  1 nf_nat_ftp
nf_conntrack_ipv4      17032  8 iptable_nat			nf_conntrack_ipv4      17032  8 iptable_nat
nf_conntrack           53440  7 xt_state,iptable_nat,nf_connt	nf_conntrack           53440  7 xt_state,iptable_nat,nf_connt
ip_tables              14788  3 iptable_mangle,iptable_nat,ip	ip_tables              14788  3 iptable_mangle,iptable_nat,ip
x_tables               14980  7 xt_limit,xt_pkttype,xt_tcpudp	x_tables               14980  7 xt_limit,xt_pkttype,xt_tcpudp
pcmcia                 35884  0 				pcmcia                 35884  0 
pcmcia_core            35988  1 pcmcia				pcmcia_core            35988  1 pcmcia
nls_iso8859_1           7808  1 			      <
nls_cp437               9472  1 			      <
vfat                   14592  1 			      <
fat                    49692  1 vfat			      <
lp                     13348  0 				lp                     13348  0 
fuse                   45588  3 				fuse                   45588  3 
snd_via82xx            26520  0 				snd_via82xx            26520  0 
snd_ac97_codec         98724  1 snd_via82xx			snd_ac97_codec         98724  1 snd_via82xx
ac97_bus                5760  1 snd_ac97_codec			ac97_bus                5760  1 snd_ac97_codec
snd_pcm                72068  3 snd_pcm_oss,snd_via82xx,snd_a	snd_pcm                72068  3 snd_pcm_oss,snd_via82xx,snd_a
snd_timer              22532  2 snd_seq,snd_pcm			snd_timer              22532  2 snd_seq,snd_pcm
snd_page_alloc         11528  2 snd_via82xx,snd_pcm		snd_page_alloc         11528  2 snd_via82xx,snd_pcm
snd_mpu401_uart        10496  1 snd_via82xx			snd_mpu401_uart        10496  1 snd_via82xx
snd_rawmidi            22560  1 snd_mpu401_uart			snd_rawmidi            22560  1 snd_mpu401_uart
snd_seq_device         10380  4 snd_seq_dummy,snd_seq_oss,snd <
snd                    47716  11 snd_seq_oss,snd_seq,snd_pcm_ <
psmouse                40336  0 				psmouse                40336  0 
button                 10000  0 			      |	snd_seq_device         10380  4 snd_seq_dummy,snd_seq_oss,snd
thermal                16540  0 				thermal                16540  0 
i2c_viapro             11156  0 			      <
via_rhine              24968  0 			      <
ns558                   7680  0 			      <
processor              32680  1 thermal			      <
rtc_cmos               11040  0 				rtc_cmos               11040  0 
via_agp                12032  1 			      |	button                 10000  0 
i2c_core               22528  1 i2c_viapro		      |	processor              32680  1 thermal
soundcore               9824  1 snd			      <
rtc_core               18696  1 rtc_cmos			rtc_core               18696  1 rtc_cmos
gameport               14472  3 snd_via82xx,ns558	      |	i2c_viapro             11156  0 
mii                     8448  1 via_rhine		      <
serio_raw               9092  0 			      <
rtc_lib                 6528  1 rtc_core			rtc_lib                 6528  1 rtc_core
agpgart                30664  1 via_agp			      |	snd                    47716  11 snd_seq_oss,snd_seq,snd_pcm_
							      >	via_agp                12032  1 
shpchp                 32788  0 				shpchp                 32788  0 
							      >	ns558                   7680  0 
							      >	serio_raw               9092  0 
							      >	agpgart                30664  1 via_agp
							      >	via_rhine              24968  0 
							      >	soundcore               9824  1 snd
evdev                  12672  4 				evdev                  12672  4 
ehci_hcd               35468  0 			      |	gameport               14472  3 snd_via82xx,ns558
uhci_hcd               25996  0 			      |	mii                     8448  1 via_rhine
parport_pc             27556  1 				parport_pc             27556  1 
parport                34632  2 lp,parport_pc			parport                34632  2 lp,parport_pc
ext3                  124808  1 			      <
jbd                    43924  1 ext3			      <
mbcache                10496  1 ext3			      <
ata_generic             8836  0 			      <
pata_acpi               8832  0 			      <
pata_via               12932  0 			      <

Interesting? Useful?

T3slider · 05-06-2008, 12:17 PM

brianL, interesting, but not terribly significant. Modules built into the kernel are not shown in `lsmod` output, and so though it may appear that the generic kernel is running more modules, this is not actually the case. A better approach would be to start the system up using each kernel, let it sit there for a few minutes, and then run `free` and `top` and see how much RAM is being used and the state of the CPU. That would prove definitively whether or not the generic kernel saves resources over the huge kernel.

brianL · 05-06-2008, 01:23 PM

Thanks for the explanation of the lsmod findings. I did try htop on both, but the differences in cpu and RAM and swap usage shown by that were very small.
The "tests" I did were just out of curiosity caused by the lack of those error messages, which were so obvious when using the huge kernel in 12.0, I wasn't intending any significant research. It gave me a chance to use lsmod and diff, anyway.