SE 28.0.1 hotfix 13 freezing using 100% of CPU

[rowbearto]

SE 28.0.1 linux x64 QT5 w/hotfix 13 freezes often on me using 100% of CPU

It seems to happen when doing live errors on python.

I have attached a log where I ran gdb and took a backtrace, continued, took another backtrace, repeat a few times, so you can see where it is stuck

[Clark]

Does the live errors for python (Pyint) ever show the errors?

Any idea about how many errors there are?

[rowbearto]

In that particular case it was frozen too long I had to stop it.

But in other cases when it froze for a long time I did see it showing many many errors.

[Clark]

I add a timeout to the error output.

Try installing this v28.0.1 installer: se_28000100_linux64qt5-2024-02-21-rte.tar.gz

https://www.dropbox.com/scl/fo/m97ry01qm7mmkav6t7fix/h?rlkey=g3zpq3kk236udh9byubbeniuc&dl=0

[rowbearto]

Thanks Clark! I will install this soon.

[rowbearto]

Hi Clark:

How long is the timeout you have added?

Haven't installed it yet, will do so soon.

[Clark]

2 seconds

[rowbearto]

Do I need to apply the latest hotfix to this install?

[Clark]

No. This has all the hot fixes already in it. No new hot fixes have come out since this build.

[rowbearto]

OK I suppose then that when new hotfixes come out I should apply them?

[rowbearto]

Got a core dump today on this install version. I think it is unrelated to the changes. Just before the crash I got a "Error Opening File", clicked OK then got that dialog again, then clicked OK again and it crashed. I've had other situations in the past where it kept coming back with "Error Opening File" every time I clicked OK.

I also notice another core dump from Feb 13.

Please PM me an email address where I can send a link to these core dumps.

EDIT: I think the core may happen when the file to reload has been restarted (a log file), so that the new size of the file is smaller than the old size, but I'm not 100% sure. I may have also previously chose "Never reload selected", but then double-clicked the file in the smart open window, don't remember exactly but it is likely.

[rowbearto]

Here's part of the backtrace of the most recent crash with the special installer you gave me with the RTE timeout:

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x000000000093e441 in hStreamMarkerAllocFindList(hfile_t*, slickedit::SEArray<int>&, long long, long long, long long, int, cmEdBlock*, long long, int) ()
[Current thread is 1 (Thread 0x7fc9a757f980 (LWP 14098))]
[?2004h(gdb) thread a        
[?2004l
[?2004h(gdb)
[?2004l
[?2004h(gdb) bt
[?2004l
#0  0x000000000093e441 in hStreamMarkerAllocFindList(hfile_t*, slickedit::SEArray<int>&, long long, long long, long long, int, cmEdBlock*, long long, int) ()
#1  0x0000000000898399 in refreshtext(bool, int, long long, int, int, int, long long*, fileview_t&, int&, int&, int, bool) [clone .constprop.203] ()
#2  0x000000000089912d in refreshwindow(int, int*, QPainter*, QRect*, bool) [clone .constprop.202] ()
#3  0x000000000089accb in se_refresh(int) ()
#4  0x00000000009813cb in vsCtlSlickCExecuteEvent ()
#5  0x0000000000981c78 in vsQTSlickCExecuteMousePressEvent ()

[Clark]

PM sent

[Clark]

Support got an email that might be from you but it looked a bit shady. Please email support from something like a gmail account so we don't get a fancy looking HTML email that look dangerous to open or click on links.

[rowbearto]

Please see the last PM I sent.