vol error · brlcad · Zulip Chat Archive

1) ls actually doesnt work
2) this is not a consequence of creating the vol but when it cant find the file associated with the vol. It gives a warning at first then crashes

starseeker (Aug 21 2020 at 16:10):

So if you make a sphere in the database as well, does ls show that?

Sumagna Das (Aug 21 2020 at 16:11):

it shows the sph but not the vols yet the vols exist

starseeker (Aug 21 2020 at 16:11):

So that'd be the first question I'd try to debug - why does ls not see the vol when it does see the sph?

Sumagna Das (Aug 21 2020 at 16:12):

hmm

starseeker (Aug 21 2020 at 16:12):

src/libged/ls/ls.c

Sumagna Das (Aug 21 2020 at 16:13):

i saw it when i was creating the vol of the skelly but forgot to report it to you guys

Sumagna Das (Aug 21 2020 at 16:28):

Thread 1 "mged" received signal SIGSEGV, Segmentation fault.
0x00007ffff783d45a in rt_vol_plot (vhead=0x7fffffffc840, ip=0x7fffffffc890,
    UNUSED_ttol=0x555555916c30, UNUSED_tol=0x555555916c80,
    UNUSED_info=0x7fffffffc860)
    at /mnt/sda2/github/brlcad/src/librt/primitives/vol/vol.c:993
warning: Source file is more recent than executable.
993         v1 = VOL(vip, x, y, z);

Sumagna Das (Aug 21 2020 at 16:28):

this is what it says

starseeker (Aug 21 2020 at 16:35):

So something about the data is invalid. The question is how it got to that state - you'll need to work backwards. What is the VOL function or macro, what is vip, what should be in it, and how is it supposed to get into it? What happened instead, and why?

Sumagna Das (Aug 21 2020 at 17:35):

all i had to get into that state was move the .bw file associated to the VOL to somewhere else where the file doesnt think it is

starseeker (Aug 21 2020 at 17:38):

And everything worked before you did that?

Sumagna Das (Aug 21 2020 at 17:38):

yea

starseeker (Aug 21 2020 at 17:39):

That sounds like there's some missing validation in the listing and lookup code for vol objects.

Sumagna Das (Aug 21 2020 at 17:39):

might be

starseeker (Aug 21 2020 at 17:40):

I'm in the middle of something right now - I'll see if I can reproduce the issue later.

Sumagna Das (Aug 21 2020 at 17:40):

Sumagna Das (Aug 21 2020 at 17:41):

i moved the files back to where they should be and the SIGSEGV is gone but lsstill cant find the vols

Sumagna Das (Aug 21 2020 at 17:48):

starseeker said:

I'm in the middle of something right now - I'll see if I can reproduce the issue later.

i also cant see into the code where this issue is happening right now so i reported to you guys so that you can see whats happening and try to fix it

starseeker (Aug 21 2020 at 20:45):

r76876 should fix the crash - I can't reproduce the ls issue. What did you name your vol primitive?

Sumagna Das (Aug 22 2020 at 05:22):

starseeker said:

r76876 should fix the crash - I can't reproduce the ls issue. What did you name your vol primitive?

skel

Sumagna Das (Aug 22 2020 at 05:22):

i will see to that

Sumagna Das (Aug 24 2020 at 20:02):

The vol segmentation fault is gone in the most recent revision. Don't know why that happened in an earlier one

Sumagna Das (Aug 24 2020 at 20:02):

But the ls issue is still there. Might be a problem with my .g file

Sumagna Das (Aug 24 2020 at 20:02):

Will upload it tomorrow if you want to see it

starseeker (Aug 24 2020 at 20:04):

sure

Sumagna Das (Aug 25 2020 at 06:55):

skel.g
the .g file

Sumagna Das (Aug 25 2020 at 06:56):

all_layers.bw
the .bw file associated with the vol if you need it

starseeker (Aug 25 2020 at 11:47):

Oh - somehow, you had the "hidden" attribute set on it. Try:

mged> unhide skel
mged> ls

Sumagna Das (Aug 25 2020 at 14:22):

will try that

Sumagna Das (Aug 25 2020 at 14:26):

starseeker said:

Oh - somehow, you had the "hidden" attribute set on it. Try:

mged> unhide skel
mged> ls

i didnt set the hidden attribute on the skel vol then how did it happen, any ideas?

starseeker (Aug 25 2020 at 15:30):

did you accidentally type "hide" at some point?

Sumagna Das (Aug 25 2020 at 15:58):

starseeker said:

did you accidentally type "hide" at some point?

never

Sumagna Das (Aug 25 2020 at 15:59):

i dont even know about most of the commands in brlcad(going to try them out after my exams)

Sean (Aug 25 2020 at 21:00):

Sumagna Das said:

i dont know if its an error or not but it was a segmentation fault(core dumped)

Crashing is always an error ;)

Sean (Aug 25 2020 at 21:04):

That is bizarre that it got hidden. We have very little code that actually will mark an object hidden... my first guess would be to ask if that .g had been edited with your GUI app or with the coreInterface at some point -- maybe it does something with the hidden flag.

Sumagna Das (Aug 26 2020 at 03:29):

Sean said:

Sumagna Das said:

i dont know if its an error or not but it was a segmentation fault(core dumped)

Crashing is always an error :wink:

the seg fault got fixed in the recent revision

Sumagna Das (Aug 26 2020 at 03:29):

Sean said:

That is bizarre that it got hidden. We have very little code that actually will mark an object hidden... my first guess would be to ask if that .g had been edited with your GUI app or with the coreInterface at some point -- maybe it does something with the hidden flag.

i only opened tthe .g file in mged and nowhere else

starseeker (Aug 26 2020 at 12:39):

Only thing I can think is that when you had the segfault crash something odd may have happened to the .g file. Regardless, it hasn't re-occurred? And if you go through the same steps to recreate the file, it works cleanly and doesn't hide it?

Sean (Aug 26 2020 at 15:41):

yeah, if you would, see if you can reproduce the error

Sumagna Das (Aug 26 2020 at 18:43):

Just fyi, the installation of mged I used while creating the .g file is from an old revision (probably from 76k-76.1k range)

Sumagna Das (Aug 26 2020 at 18:44):

Anyways I will try to recreate the error if you want

Sean (Aug 26 2020 at 18:45):

that would be great. suggest trying to reproduce at all, and then if you can figure out a specific sequence of steps, even better.

Sumagna Das (Aug 27 2020 at 08:33):

i tried recreating the hiding error but couldnt recreate it

Sumagna Das (Aug 27 2020 at 08:34):

but the segfault is occuring anyways

Sumagna Das (Aug 27 2020 at 08:36):

is there a way to see command history for a file?

Sumagna Das (Aug 27 2020 at 08:36):

i dont remember using hide command but still would like to check to see if it was my fault or the segfault

starseeker (Aug 27 2020 at 11:57):

You're still getting a segfault in trunk, or you're using the older version?

Sumagna Das (Aug 27 2020 at 11:58):

the older version i think

Sumagna Das (Aug 27 2020 at 11:58):

Sumagna Das said:

Just fyi, the installation of mged I used while creating the .g file is from an old revision (probably from 76k-76.1k range)

i said so

starseeker (Aug 27 2020 at 11:59):

Yes, that was the revision you used to create it - I was making sure you are using the same version to do the testing now.

starseeker (Aug 27 2020 at 11:59):

There is some command history, but I don't believe it persists from one MGED session to the next by default

Sumagna Das (Aug 27 2020 at 12:00):

does the seg fault in older version matter now? its fixed in the recent one

starseeker (Aug 27 2020 at 12:00):

It doesn't matter, unless it was the cause of the hiding issue. If it was though, it may be hard to reproduce - corruption behavior is seldom deterministic.

Sumagna Das (Aug 27 2020 at 12:01):

i am not getting the hide issue right now even in the older version

Sumagna Das (Aug 27 2020 at 12:02):

yet i dont remember using hide when i was creating the vol

Sumagna Das (Aug 27 2020 at 12:02):

so still no idea how that happened

starseeker (Aug 27 2020 at 12:02):

OK. Worth checking, but if it wasn't that then I don't have any other immediate ideas.

Sean (Aug 28 2020 at 00:08):

thanks for checking @Sumagna Das

Sumagna Das (Aug 28 2020 at 06:49):

No problem

Last updated: Oct 09 2024 at 00:44 UTC