[9fans] FS 4th Ed

[9fans] FS 4th Ed

[Lucio De Re]

I seem to recall a small fix to the FS kernel to speed up return
after command execution on the console.  Right now I'm watching a

	check pfile

print a filename every 5 seconds, approximately.  I've been waiting
since yesterday morning for it to complete :-(

I'd like not to get caught with it in future.

BTW, I'd like to remove some directory entries (in /usr/lucio/tmp)
that have peculiar names and, seemingly, no matching file entry.
I don't know how they occurred, nor have I been able to remove
them.

++L

Re: [9fans] FS 4th Ed

[Lucio De Re]

On Tue, Jul 22, 2003 at 03:09:25PM +0900, okamoto@granite.cias.osakafu-u.ac.jp wrote:
>
> > I seem to recall a small fix to the FS kernel to speed up return
> > after command execution on the console.  Right now I'm watching a
> >
> > 	check pfile
> >
> > print a filename every 5 seconds, approximately.  I've been waiting
> > since yesterday morning for it to complete :-(
>
>  ?
>
> Kenji

Does that mean disbelief?  :-)  :-)

I may well have an old kernel, but the symptoms are real.  Each
command also takes a while to return to the prompt and discards any
input until the prompt appears.

++L

Re: [9fans] FS 4th Ed

[okamoto]

> I seem to recall a small fix to the FS kernel to speed up return
> after command execution on the console.  Right now I'm watching a
> 
> 	check pfile
> 
> print a filename every 5 seconds, approximately.  I've been waiting
> since yesterday morning for it to complete :-(

 ☺

Kenji

Re: [9fans] FS 4th Ed

[okamoto]

> Does that mean disbelief?  :-)  :-)

It meant I have sympathy to you.

Kenji

Re: [9fans] FS 4th Ed

[Axel Belinfante]

Don't know if it is related, but:
When I set up my 3ed and/or 4ed file server printing to the console
was quite slow (typical example would be the 'help' command which
would be printed line by line) until I configured (and used?) an
external console (console=0 in plan9.ini). Made _big_ difference.

Axel.

> > > I seem to recall a small fix to the FS kernel to speed up return
> > > after command execution on the console.  Right now I'm watching a
> > >
> > > 	check pfile
> > >
> > > print a filename every 5 seconds, approximately.  I've been waiting
> > > since yesterday morning for it to complete :-(
> >
> >  ?
> >
> > Kenji
>
> Does that mean disbelief?  :-)  :-)
>
> I may well have an old kernel, but the symptoms are real.  Each
> command also takes a while to return to the prompt and discards any
> input until the prompt appears.
>
> ++L

Re: [9fans] FS 4th Ed

[Lucio De Re]

On Tue, Jul 22, 2003 at 09:30:44AM -0400, Dan Cross wrote:
>
> > I may well have an old kernel, but the symptoms are real.  Each
> > command also takes a while to return to the prompt and discards any
> > input until the prompt appears.
>
> Using a serial console (as opposed to the CGA console) the problem is
> greatly attenuated.
>
Sounds like there's a timeout waiting for non-existent serial port
activity?  It's a small kernel, I could probably find such a problem,
except I seem to recall a comment from jmk that suggested a permanent
solution.  Ah, then onset of senility...

++L

Re: [9fans] FS 4th Ed

[Dan Cross]

> I may well have an old kernel, but the symptoms are real.  Each
> command also takes a while to return to the prompt and discards any
> input until the prompt appears.

Using a serial console (as opposed to the CGA console) the problem is
greatly attenuated.

	- Dan C.

Re: [9fans] FS 4th Ed

[andrey mirtchovski]

On Tue, 22 Jul 2003, Lucio De Re wrote:

> BTW, I'd like to remove some directory entries (in /usr/lucio/tmp)
> that have peculiar names and, seemingly, no matching file entry.
> I don't know how they occurred, nor have I been able to remove
> them.
>

does it look like this:

plan9-2% ls -l
--rwxr-xr-x M 96 andrey andrey 0 Jul 16 22:44 b16.81F7-8202
plan9-2% rm *
rm: b16.81F7-8202: 'b16.81F7-8202' corrupted file entry
plan9-2% echo 'begone' > b16.81F7-8202
b16.81F7-8202: rc: can't open: file already exists
plan9-2%

if this is the case 'clri' would fix it on the fossil console. run
fossil/flchk on it and you may see something like:

# clri /snapshot/2003/0721/1500/usr/andrey/font/efont/b16/b16.81F7-8202

(with your corresponding file entry, of course)

if that's not the case, try the 'remove' command on the fossil console.

andrey