Wings 3D Development Forum
[fixed] erl_crash_dump:0.5 - Printable Version

+- Wings 3D Development Forum (https://www.wings3d.com/forum)
+-- Forum: Wings 3D (https://www.wings3d.com/forum/forumdisplay.php?fid=1)
+--- Forum: Bug Reports (https://www.wings3d.com/forum/forumdisplay.php?fid=12)
+--- Thread: [fixed] erl_crash_dump:0.5 (/showthread.php?tid=2734)



[fixed] erl_crash_dump:0.5 - Hank - 12-07-2018

OS/ Linux Mint
Steps:
I open one of my projects with UV map set and when I press tab [smooth preview] system is doing re-calculation for a while and then I get erl_crash.dump file [51,0 MB!] - new record Smile

the begining is:

=erl_crash_dump:0.5
Fri Dec 7 18:46:35 2018
Slogan: binary_alloc: Cannot reallocate 10993139671 bytes of memory (of type "binary").
System version: Erlang/OTP 20 [erts-9.3] [source] [64-bit] [smp:8:8] [ds:8:8:10] [async-threads:10] [hipe] [kernel-poll:false]
Compiled: Tue Mar 27 15:12:06 2018
Taints: erl_tracer,zlib
Atoms: 21272
Calling Thread: scheduler:1
=scheduler:1
Scheduler Sleep Info Flags:
Scheduler Sleep Info Aux Work:
Current Port:
Run Queue Max Length: 0
Run Queue High Length: 0
Run Queue Normal Length: 0
Run Queue Low Length: 0
Run Queue Port Length: 0
Run Queue Flags: OUT_OF_WORK | HALFTIME_OUT_OF_WORK | NONEMPTY | EXEC
Current Process: <0.67.0>
Current Process State: Running
Current Process Internal State: ACT_PRIO_NORMAL | USR_PRIO_NORMAL | PRQ_PRIO_NORMAL | ACTIVE | RUNNING | TRAP_EXIT | OFF_HEAP_MSGQ
Current Process Program counter: 0x00007fb96fe18bc8 (wings_draw_setup:dup3/3 + 24)
Current Process CP: 0x00007fb96fe12870 (wings_draw_setupConfusedetup_smooth_normals/4 + 172Cool
Current Process Limited Stack Trace:

If necessary I can share it via Dropbox (?)

BR, Hank

for Win [1.5.4] and [2.2.1] is similar:

=erl_crash_dump:0.3
Fri Dec 07 19:06:51 2018
Slogan: binary_alloc: Cannot reallocate 18446744073042657239 bytes of memory (of type "binary").
System version: Erlang/OTP 17 [erts-6.0] [64-bit] [smp:8:8] [async-threads:10]
Compiled: Sun Apr 6 19:42:02 2014
Taints:
Atoms: 15653
=memory
total: 12661196624
processes: 62484656
processes_used: 62484656
system: 12598711968
atom: 463441
atom_used: 441485
binary: 12580879424
code: 12853439
ets: 820104
=hash_table:atom_tab
size: 12853
used: 9046
objs: 15653
depth: 8
=index_table:atom_tab
size: 16384

BR, Hank


RE: erl_crash_dump:0.5 - Hank - 12-07-2018

+ attached file
I can't find what's so uniqe in this object Sad

Just press TAB and W3D will crash...

BR, Hank


RE: erl_crash_dump:0.5 - micheus - 12-07-2018

Yeah, weird.
The first log it's failing to allocate 10GB of memory, but the second log is still worst. It's far away that 10GB.

I'm going to try check your model at home tomorrow.
Do you was able to load it in 2.1.7?
I already saw a log like this reading a corrupted file. Did you tried to backup it an rename the Wings3D backup file: *.~wings (I think it's this way)?


RE: erl_crash_dump:0.5 - Hank - 12-08-2018

2.1.7 has the same result - crash after 5-10 sec.
and even 1.5.4 is not better here Sad
when I do the copy with .~wings extension, then I can't see it on the list for uplaod - thus can't confirm your advise.

BR, Hank


RE: erl_crash_dump:0.5 - micheus - 12-12-2018

Hi Hank
I got some time to check this and the problem is caused by your model that has two malformed faces:


So, you can fix it and solve the problem. Smile
Anyway, I create a workaround for that case in the code. If dgud accept it next time we don't get a hard crash in this situation.


RE: [fixed] erl_crash_dump:0.5 - Hank - 12-12-2018

Perfect!
Thank you for checking and processing.

BR, Hank


RE: [fixed] erl_crash_dump:0.5 - micheus - 12-17-2018

A fix is available in Fixes for v2.2.1 (Patches)


RE: [fixed] erl_crash_dump:0.5 - Hank - 01-22-2019

today I got the same problem with 2.2.7 (even the patches are up to date Undecided )

=erl_crash_dump:0.5
Tue Jan 22 21:02:22 2019
Slogan: eheap_alloc: Cannot allocate 4723591984 bytes of memory (of type "heap").
System version: Erlang/OTP 20 [erts-9.3] [source] [64-bit] [smp:8:8] [ds:8:8:10] [async-threads:10] [hipe] [kernel-poll:false]
Compiled: Tue Mar 27 15:12:06 2018
Taints: erl_tracer,zlib
Atoms: 23356
Calling Thread: scheduler:0
=scheduler:1
Scheduler Sleep Info Flags: SLEEPING | TSE_SLEEPING | WAITING
Scheduler Sleep Info Aux Work: THR_PRGR_LATER_OP
Current Port:
Run Queue Max Length: 0
Run Queue High Length: 0
Run Queue Normal Length: 0
Run Queue Low Length: 0
Run Queue Port Length: 0

log file over 6 GB and increasing...

BR, Hank


RE: [fixed] erl_crash_dump:0.5 - micheus - 01-22-2019

It looks like we got another situation which cause the same memory crash.
Can you try to isolate the part of the model that may causing that - as was made before?