Discussion:
Got a bug for you
(too old to reply)
Gene Heskett
2017-06-14 21:32:17 UTC
Permalink
Raw Message
Greetings everybody;

Playing around with lathe-pawn.ngc, trying to add the G96 CSS commands to
it so while it might not be spot on, it at least demo's that it works,
and I have that, although the rev ceiling in 1st gear gets in the way,
it does work.

But the canary died:

However, while I can re-run the code if it runs to completion, and it
will do it correctly a 2nd time, I can't stop it, and edit the code (or
do an x touchoff to get the nachine closer to the real centerline 0 )
and then reload the edited code as it claims there is not an Svalue in
the G96 line. Except it fusses about line 0. And the G96 is in line 6.

Shutting down LinuxCNC and then launching it again lets me reload the
code w/o any fussing. And it runs correctly.

It fusses about it after a touch-off, but I can hit r to run it, and it
runs with the machine moving in the correct space, but drawing the
backplot about 10cm inward from the white code scan trace in the
backplot.

This is:
/var/cache/apt/archives/linuxcnc-uspace-dev_1%3a2.8.0~pre1.3155.ga8df216_armhf.deb
/var/cache/apt/archives/linuxcnc-uspace_1%3a2.8.0~pre1.3155.ga8df216_armhf.deb

Thanks everybody, I gotta go cook, or go out, depends on what the missus
wants.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-06-15 09:33:14 UTC
Permalink
Raw Message
Post by Gene Heskett
Greetings everybody;
Playing around with lathe-pawn.ngc, trying to add the G96 CSS commands
to it so while it might not be spot on, it at least demo's that it
works, and I have that, although the rev ceiling in 1st gear gets in
the way, it does work.
However, while I can re-run the code if it runs to completion, and it
will do it correctly a 2nd time, I can't stop it, and edit the code
(or do an x touchoff to get the nachine closer to the real centerline
0 ) and then reload the edited code as it claims there is not an
Svalue in the G96 line. Except it fusses about line 0. And the G96 is
in line 6.
Shutting down LinuxCNC and then launching it again lets me reload the
code w/o any fussing. And it runs correctly.
It fusses about it after a touch-off, but I can hit r to run it, and
it runs with the machine moving in the correct space, but drawing the
backplot about 10cm inward from the white code scan trace in the
backplot.
/var/cache/apt/archives/linuxcnc-uspace-dev_1%3a2.8.0~pre1.3155.ga8df2
16_armhf.deb
/var/cache/apt/archives/linuxcnc-uspace_1%3a2.8.0~pre1.3155.ga8df216_a
rmhf.deb
And I just woke up, realizing that the above info is incorrect. While I
do have the above installed, what I am actually running is a git pull
that was in an image Bertho sent me, and which contains only a 2.8-pre
marking. The majority of the pull is carrying a date of May 27th, and
looking at the config.log, it was built for a newer kernel that turned
out to be a problem child with up-times of sub 20 minutes. So the
possibility of miss-matches are endless.

IOW call in the St Bernards until I can make a build for this kernel.

My apologies for the noise. I'll see what I can do about a rebuild with
Berthos PM help.
Post by Gene Heskett
Thanks everybody, I gotta go cook, or go out, depends on what the
missus wants.
Cheers, Gene Heskett
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-06-15 21:25:08 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Gene Heskett
Greetings everybody;
Playing around with lathe-pawn.ngc, trying to add the G96 CSS
commands to it so while it might not be spot on, it at least demo's
that it works, and I have that, although the rev ceiling in 1st gear
gets in the way, it does work.
However, while I can re-run the code if it runs to completion, and
it will do it correctly a 2nd time, I can't stop it, and edit the
code (or do an x touchoff to get the nachine closer to the real
centerline 0 ) and then reload the edited code as it claims there is
not an Svalue in the G96 line. Except it fusses about line 0. And
the G96 is in line 6.
Shutting down LinuxCNC and then launching it again lets me reload
the code w/o any fussing. And it runs correctly.
It fusses about it after a touch-off, but I can hit r to run it, and
it runs with the machine moving in the correct space, but drawing
the backplot about 10cm inward from the white code scan trace in the
backplot.
/var/cache/apt/archives/linuxcnc-uspace-dev_1%3a2.8.0~pre1.3155.ga8d
f2 16_armhf.deb
/var/cache/apt/archives/linuxcnc-uspace_1%3a2.8.0~pre1.3155.ga8df216
_a rmhf.deb
And I just woke up, realizing that the above info is incorrect. While
I do have the above installed, what I am actually running is a git
pull that was in an image Bertho sent me, and which contains only a
2.8-pre marking. The majority of the pull is carrying a date of May
27th, and looking at the config.log, it was built for a newer kernel
that turned out to be a problem child with up-times of sub 20 minutes.
So the possibility of miss-matches are endless.
IOW call in the St Bernards until I can make a build for this kernel.
My apologies for the noise. I'll see what I can do about a rebuild
with Berthos PM help.
Actually, a rebuild didn't seem to be 100% required. What I did was move
the new hm2_rpspi.so driver over the existing one from the deb's I
quoted. This makes lcnc run, where the original bails out trying to
ident the card and failing. And the problem persists.

On a stop or esc break, the code CAN be reloaded IF one selects F5 and
enters a G97 on the command line. A G97 2 lines about the G96 D700 S12,
has no effect, although a queue breaker statement between those lines
might help. Not yet tried. So the "bug" if thats what it is, is still
there.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Loading...