My shop box has no users!

[gene heskett]

On Thursday, February 03, 2011 06:34:43 pm Reinhold Rumberger did opine:

> On Donnerstag 03 Februar 2011, 23:31:22, gene heskett wrote:
> > On Thursday, February 03, 2011 05:07:30 pm reini did opine:
> <snip>
> 
> > Now what?  This all, videowise, worked until I updated the box about a
> > week or 10 days ago.  The end of the Xorg.0.log:
> > 
> > (II) VESA: driver for VESA chipsets: vesa
> > (II) Primary Device is: PCI 01 at 00:00:0
> > (EE) VESA: Kernel modesetting driver in use, refusing to load
> 
> <snip>
> 
> This is due to all that new modesetting stuff happening in the kernel.
> 
> You want the nomodeset option:
> http://ubuntuforums.org/showthread.php?t=1613132
> 
> From what I read that option will be removed at some point in time, so
> I'm not sure what will happen then...
> 
>   --Reinhold

In that event, all the users of emc, and there are hundreds (that we know 
of) scattered about this planet, will be screwed.  This is a case of none 
of the 'improved' drivers being usable, and the vendor drivers are deadly 
because they get their performance largely by hogging the IRQ's, often for 
hundreds of milliseconds at a time, deadly on a machine that is expected to 
output the next step command to the motor drivers in a loop with a sub 50 
microsecond cycle time.  You cannot tell a stepper motor to come to a 
complete halt from 450 rpm, wait for 300 milliseconds, and resume that 450 
rpm when the video card decides to let go so the real application can do 
its thing.  Unable to get back to speed, it will maybe take 1/8th step and 
lock up.  If it happened to all 4 motors at once, it would just be a 
stalled machine requiring all axises to have their zero points reset, the 
speed slowed some, and the whole program restarted from the top.  That may, 
and likely will waste several hours getting back to the point where it 
stalled on a previous run, but if it only happens to one of the motors that 
was moving the cutting bit out of the way so it can then traverse to the 
next cutting point, that traverse which is supposed to be done with 
clearance, may well plow the bit crossways through the workpiece a quarter 
inch deep.  That will wreck the part of course, and quite likely break a 
$20+ solid carbide milling cutter into 2 pieces worth only the raw price of 
the weight of the carbide itself.  Perhaps 50 cents?

Anyway, that once again was the magic potion, fixing it up nicely, but the 
picked resolution of 1024x768 is quite a bit less than that monitors native 
resolution of about 1650x1280.  Could this be alleviated by adding an 
option  = edid in the device section of xorg.conf?  Or even a vga=798 or 
some such on the kernel line?  The monitor is an lcd, a recent AOC made 
unit.  From the Xorg.0.log, it looks like edid is at least read:
(II) VESA(0): EDID (in hex):
(II) VESA(0):   00ffffffffffff0005e3311841c80000
(II) VESA(0):   1a140103682917782aeed1a555489b26
(II) VESA(0):   125054bfee00310a614c010101010101
(II) VESA(0):   010101010101662156aa51001e30468f
(II) VESA(0):   33009dea1000001e662150b051001b30
(II) VESA(0):   4070360022130000001e000000fd0037
(II) VESA(0):   4b1e5109000a202020202020000000fc
(II) VESA(0):   00383331570a2020202020202020003e
(II) VESA(0): EDID vendor "AOC", prod id 6193
(II) VESA(0): Using EDID range info for horizontal sync
(II) VESA(0): Using EDID range info for vertical refresh

but I have no clue how to decode that... :(

Or is this just the limits of the vesa driver?  The Xorg.0.log contains 
about 50k of its looking at all available modes before picking the one it 
ends up using.  It seems like there should be a way to nudge it in the 
right direction.  OTOH, the machine is running beautifully, with instant 
response to the keyboard motion keys and can once again 'do good work'  ;-)

Thanks Reinhold.  A bunch!


-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
<http://tinyurl.com/ddg5bz>
C'est magnifique, mais ce n'est pas la guerre!
	[It is magnificent, but it is not war]
		-- Pierre Bosquet, witnessing the charge of the Light 
Brigade