| freemangordon | arno11: why not simply uninstall qt HIM plugin until it is fixed? | 07:38 |
|---|---|---|
| dsc_ | arno11: him causes segfault? | 08:22 |
| dsc_ | with telegram? | 08:22 |
| freemangordon | dsc_: with all qt apps | 08:23 |
| freemangordon | see https://paste.debian.net/hidden/fa306ee7/ | 08:24 |
| freemangordon | and also see Wizzup's comments ^^^ | 08:24 |
| dsc_ | this is only since daedalus right? | 08:25 |
| freemangordon | afaik | 08:25 |
| dsc_ | im in VM @ chimaera, how to go get daedalus | 08:26 |
| * freemangordon wonders if libhimplatforminputcontextplugin.so symbols were installed when this backtrace was captured | 08:27 | |
| freemangordon | you can clone the VM and do dist-upgrade | 08:27 |
| dsc_ | yes alright | 08:27 |
| freemangordon | however, I saw no issue here (in my daedalus VM) | 08:27 |
| dsc_ | re symbols, doesnt seem stripped or optimized away | 08:28 |
| freemangordon | maybe somethiong shall be done to trigger the bug | 08:28 |
| freemangordon | ah, right, seems ok | 08:28 |
| freemangordon | dsc_: I would recommend to wait for arno11 to provide steps to trigger the bug | 08:29 |
| dsc_ | hehe yes | 08:29 |
| freemangordon | also, see what Wizzup said ^^^, the fix could be something trivial like checking for nullptr | 08:29 |
| freemangordon | or maybe there is a flaw in the logic - why do you need to call QXcb::findHildonIm() in plugin destructor if VKB is not visible (for example) | 08:31 |
| dsc_ | QHildonInputContext::~QHildonInputContext() { sendHildonCommand(HILDON_IM_HIDE); } | 08:32 |
| freemangordon | I admit I didn't look in the code, so feel free to ignore all my comments :) | 08:32 |
| dsc_ | yes they are good suggestions | 08:32 |
| freemangordon | yeah, but is it visible? | 08:32 |
| freemangordon | or simply do what Wizzup said | 08:32 |
| freemangordon | anyway, have to run, ttyl | 08:33 |
| dsc_ | im looking | 08:33 |
| dsc_ | arno11: I pushed a change | 09:19 |
| Wizzup | dsc_: let me build a new daedalus vm image | 10:48 |
| Wizzup | I'm pretty sure the xcb connection is invalidated and this causes the segfault | 10:49 |
| Wizzup | I think ti happens upon closing a program | 10:49 |
| dsc_ | Wizzup: thanks. I think I fixed it then. | 11:04 |
| Wizzup | we made it to the next/final round of this nlnet call, will follow up when we know if make it or not | 20:47 |
| Wizzup | probably in 1-2 months | 20:48 |
| sicelo | very good news! thanks for sharing that info | 21:09 |
| Wizzup | yeah, although various things in the proposal we kind of already did, like daedalus | 21:31 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!