Skip to contentSkip to conversation

serial console slow

DISCUSSION root 3 participants 3 messages Nov 9, 2001

lore
Share & tools
lore

root

author
#0 lore raw

Hi!

Quoted 8 lines

I tried to boot my kernel using the serial console, using the console=ttyS0,115200 (it does the same thing with 9600) ... it work great until :

Freeing unused kernel memory: 36k freed serial console detected. Disabling virtual terminals. console=/dev/ttyS0

Quoted 5 lines

At this point the output of the serial line slow down dramaticly ... almost to a halt ... I get 1 line every 30 seconds !!!

why is this slow down occuring ? the part which is 100% kernel is going fast and ok, but when it become console related ... it slows down ?

Serial just got its control signals (it is now *userland* writing, and userland honours them), plus userland is probably opening/closing serial all the time. Bad.
Pavel

Signature · 12 lines
Philips Velo 1: 1"x4"x8", 300gram, 60, 12MB, 40bogomips, linux, mutt,
details at http://atrey.karlin.mff.cuni.cz/~pavel/velo/index.html.




----- End forwarded message -----

-- 
--
I'm really [email protected]. 	   Pavel
Look at http://atrey.karlin.mff.cuni.cz/~pavel ;-).
#1 lore raw
Followup to:  <[email protected]>
By author:    root <[email protected]>
In newsgroup: linux.dev.kernel
Quoted 14 lines

Hi!

Quoted 11 lines

I tried to boot my kernel using the serial console, using the console=ttyS0,115200 (it does the same thing with 9600) ... it work great until :

Freeing unused kernel memory: 36k freed serial console detected. Disabling virtual terminals. console=/dev/ttyS0

At this point the output of the serial line slow down dramaticly ... almost to a halt ... I get 1 line every 30 seconds !!!

Quoted 7 lines
Quoted 2 lines

why is this slow down occuring ? the part which is 100% kernel is going fast and ok, but when it become console related ... it slows down ?

Serial just got its control signals (it is now *userland* writing, and userland honours them), plus userland is probably opening/closing serial all the time. Bad.
Pavel

I have had much better luck talking to /dev/console in userland. IMNSHO this should *always* work; anything else is broken. /dev/console currently *IS* broken to some degree, multi-console hasn't worked properly for a long time, and you don't get job control running of it because it isn't a tty, but I think it's a lot less broken than things like the above...

(And dammit, I really would like to see console=tty0 console=ttyS0 to actually give me both consoles -- in userland *and* in the kernel...)

-hpa

Signature · 3 lines
<[email protected]> at work, <[email protected]> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt	<[email protected]>
#2 lore raw
In reply to root

In article <[email protected]>,

Quoted 6 lines · H. Peter Anvin

I have had much better luck talking to /dev/console in userland. IMNSHO this should *always* work; anything else is broken. /dev/console currently *IS* broken to some degree, multi-console hasn't worked properly for a long time, and you don't get job control running of it because it isn't a tty, but I think it's a lot less broken than things like the above...

You don't have job control on /dev/console because it will never become a controlling tty automatically. But if you make it your controlling tty with the right ioctl() you'll have job control.

In fact the latest sysvinit does this for process spawned from the 'sysinit', 'bootwait' and 'wait' type lines in /etc/inittab and that works fine.

Quoted 2 lines

(And dammit, I really would like to see console=tty0 console=ttyS0 to actually give me both consoles -- in userland *and* in the kernel...)

That would be extremely c00l ;)

Mike.

Signature · 2 lines
"Only two things are infinite, the universe and human stupidity,
 and I'm not sure about the former" -- Albert Einstein.