MOL future plans

[Bill Fink]

On Tue, 9 Apr 2002, samuel@ibrium.se wrote:

> On Mon, Apr 08, 2002 at 10:48:26PM -0400, Bill Fink wrote:
> > On Fri, 5 Apr 2002, Samuel wrote:
> > 
> > > I'm open for suggestions though. What is the most 
> > > pressing issue?
> > 
> > Hi Samuel,
> > 
> > Well since no one mentioned my most pressing MOL concern, it would be
> > to make MOL SMP safe, so I could finally use the other processor on my
> > dual 500 MHz G4.
> 
> Actually, what happens at the moment if you try to start the
> rsync version of MOL on a SMP kernel? MOL is not completely SMP
> safe yet, so it won't be stable (will probably panic the kernel
> after some time), but I think it ought to be possible to at least
> run 'startmol --test'. It would be nice to get this verified though.

OK.  Running a 2.4.18-ben0 SMP kernel with the latest rsync of MOL, I ran
the "startmol --test" which ran successfully (output attached).  Then feeling
brave, or perhaps being foolhardy, and since I now have ext3 file systems,
I did a few syncs and then actually started up MOL.  I didn't do a lot, just
opened and closed a few folders, checked "About This Computer", and last went
to www.moviefone.com and watched a QuickTime movie trailer.  No problems were
noticed with this simple testing.

Could you elaborate a little more on what the SMP unsafe conditions might
be and if there are any workarounds, like avoiding doing certain things in
Linux while you're in MOL?

						-Thanks

						-Bill

P.S.  I could also send you the full mol-logfile from my SMP MOL session
      if that would be helpful at all.



Script started on Tue Apr  9 18:50:23 2002

gwiz% startmol --test
Copyright (C) 1997-2001 Samuel Rydh <samuel@ibrium.se>
Version 0.9.61 
Configuration file: /etc/molrc
Library directory: /usr/lib/mol
Session 0. Lockfile '/var/lock/mol-0'
Debugger nub disabled
The session save/restore feature is disabled
OF device tree: oftrees/oftree.nw
Timebase frequency: 24.90 MHz
256MB RAM mapped at 0x40000000
Running in PowerPC G3 mode
Using Linux keycodes
Using nvram-image 'nvram/nvram.nw'
Ethernet Interface (port 1) 'sheep-<eth0>' @ FE:FD:DE:AD:BE:EF
<*> 
<*>  ******************************
<*>  *    Testing performance     *
<*>  ******************************
<*> 
<*>  nop calibration     200.00 %
<*> --------- USER MODE ----------
<*>  li                  200.00 %
<*>  lwz r3,0(0)         106.22 %
<*>  lfd fr3,0(0)        106.25 %
<*>  stfd fr3,0(0)        34.36 %
<*>  b +4                 99.99 %
<*>  mftb                 35.42 %
<*>  mfctr                53.13 %
<*>  mtlr                 53.13 %
<*>  mflr                 53.13 %
<*>  isync                25.76 %
<*>  mfcr                 53.13 %
<*>  mtcrf 1,r3           52.47 %
<*> --------- PRIVILEGED MODE ----
<*>  mfmsr                 0.93 %
<*>  mfmsr r3 . mtmsr r3   0.37 %
<*>  tlbie r3              0.94 %
<*>  tlbsync               0.93 %
<*>  mfsr                  0.89 %
<*>  mtsr                  0.15 %
<*>  mfsrin                0.82 %
<*>  mtsrin                0.15 %
<*>  mtsrr1                0.90 %
<*>  mfsrr1                0.92 %
<*>  mfdec                 0.86 %
<*>  mtdec                 0.66 %
<*>  mtdbatu 0,r3          0.16 %
<*>  mtdbatl 0,r3          0.16 %
<*>  mtibatu 0,r3          0.16 %
<*>  mtibatl 0,r3          0.16 %
<*>  mfspr r3, HID0        0.22 %
<*>  mtspr HID0,r3         0.24 %
<*>  mfspr r3, HID1        0.22 %
<*>  mfspr r3, DBAR        0.22 %
<*> 
<*>  ******************************
<*>  *    Self-test successful    *
<*>  ******************************
<*> 
cleaning up...
Terminating threads...
DONE
mount: only root can do that
gwiz% exit

Script done on Tue Apr  9 18:50:43 2002