03:01 <gbruno> [github] kintel pushed 7 modifications (Adjust angle of viewoptions tests to make it more likely to trigger rendering issues) https://github.com/openscad/openscad/commit/9888982892480d0e4fb9cbece6b8430de6598f65

03:01 <gbruno> [github] kintel opened pull request #4594 (Adjust angle of viewoptions tests) https://github.com/openscad/openscad/pull/4594

03:15 <gbruno> [github] kintel pushed 7 modifications (Adjust angle of viewoptions tests to make it more likely to trigger rendering issues) https://github.com/openscad/openscad/commit/60519751e0071f7fc65204c5545dc794d11d7b8d

03:15 <gbruno> [github] kintel synchronize pull request #4594 (Adjust angle of viewoptions tests) https://github.com/openscad/openscad/pull/4594

04:21 <gbruno> [github] kintel edited pull request #4594 (Adjust angle of viewoptions tests) https://github.com/openscad/openscad/pull/4594

04:28 <gbruno> [github] kintel opened issue #4595 (OpenCSG rendering with edges: Z buffer tearing) https://github.com/openscad/openscad/issues/4595

04:29 <gbruno> [github] kintel edited pull request #4594 (Adjust angle of viewoptions tests) https://github.com/openscad/openscad/pull/4594

04:35 <gbruno> [github] kintel pushed 1 modifications (Tag disabled tests with issue #4595) https://github.com/openscad/openscad/commit/9dea0cc8e47791186db77b05d1375c13addb08c2

04:35 <gbruno> [github] kintel synchronize pull request #4594 (Adjust angle of viewoptions tests) https://github.com/openscad/openscad/pull/4594

04:35 <gbruno> [github] kintel closed pull request #4594 (Adjust angle of viewoptions tests) https://github.com/openscad/openscad/pull/4594

04:35 <gbruno> [github] kintel pushed 7 modifications (Adjust angle of viewoptions tests (#4594) Related to issue #4595) https://github.com/openscad/openscad/commit/42d1df76b1e0ddc244ccc1f6f1027a642aa5273f

04:36 <kintel> If anyone starts experiencing a couple of test failures on Linux, it's probably because of https://github.com/openscad/openscad/pull/4594. If so, let me know (or just disable the failing tests as mentioned in the PR)

04:37 <kintel> peepsalot I summarized the rendering issues we discussed here: https://github.com/openscad/openscad/issues/4595

04:37 <kintel> Feel free to add any further info or ideas

04:48 <gbruno> [github] kintel pushed 6 modifications (Tmp: GLEW fixes for bundled OpenCSG) https://github.com/openscad/openscad/commit/411baa39709651284eb769e8a49a1215299b80a4

11:31 <Scopeuk> Someone had fun https://hackaday.com/2023/04/11/ibm-selectric-typewriters-finally-get-diy-typeballs/

12:25 <gbruno> [github] schiele synchronize pull request #4538 (implement new scale and translate parameter to rotate_extrude) https://github.com/openscad/openscad/pull/4538

15:11 <teepee> Scopeuk: crazy stuff, I saw the video from Emily Velasco a while ago

15:13 <guso78[m]> Teepee, IS there a function to hierchically Copy/free an AbatractNode?

15:14 <teepee> I don't think I've seen one

15:14 <guso78[m]> AbstractNode

15:14 <teepee> why copy?

15:14 <guso78[m]> Is the tree Freed implicitely?

15:16 <teepee> I think that happens on the next parsing

15:16 <guso78[m]> ITS about Python. When a node IS User twice, the Same memory geht's Freed twice ...

15:17 <teepee> hmm, I thought they use a shared pointer

15:17 <guso78[m]> On next parsing the Contest might BE completely different. Not sure If next parsing even hast a Chance

15:18 <guso78[m]> Hmm, cant Recall If ITS shared ptr or unique ptr. Need to Look IT Up First 😄

15:19 <guso78[m]> Still get very random segfaulta eapecially with my libfive datatype. Want to understand

15:33 <gbruno> [github] kintel pushed 1 modifications (PBuffer support enabled by default) https://github.com/openscad/openscad/commit/ede1b2b854a653eeb37768836532184b6adc44e8

16:08 <lf94> guso78[m]: libfive will segfault on odd uses of it, fyi

16:08 <lf94> Ive had the same happen

16:12 <guso78[m]> Lf94 i have backtraced IT. IT happens in mimalloc_free after libfive has executed. IT happens very often when Render result IS empty.

16:15 <Scopeuk> does it trigger if you disable mimaloc?

16:18 <guso78[m]> Sorry did Not try IT. ITS Worth trying. IS IT a cmake Option?

16:30 <Scopeuk> I think you can do it at build time, USE_MIMALLOC to OFF

16:31 <Scopeuk> sorry for delay, was trying to remember and look up if it was possible by environmental variable

16:35 <guso78[m]> No issue. You are never too slow as i have to Look after the Kids in parallel 🤣

16:37 <Scopeuk> if you can get them to read the mimalloc docs, they'll be asleep for hours :P

16:41 <guso78[m]> Need to Work in that. They are Just 1,5 and 3 🤣

17:03 <kintel> guso78[m] It's also important to always link mimalloc _before_ any other libraries

17:05 <guso78[m]> Ahh, interesting info. Thwnk you!

18:17 <gbruno> [github] kintel pushed 2 modifications (Doc and warnings) https://github.com/openscad/openscad/commit/d2e230b3a123f86a9c48551926b0ff74245257e6

18:17 <gbruno> [github] kintel synchronize pull request #4576 (Refactor offscreen context management) https://github.com/openscad/openscad/pull/4576

18:32 <gbruno> [github] kintel pushed 14 modifications (uint32_t harmonization) https://github.com/openscad/openscad/commit/88c89608887a382e2735c80080a8d7ae51e13759

18:32 <gbruno> [github] kintel synchronize pull request #4576 (Refactor offscreen context management) https://github.com/openscad/openscad/pull/4576

18:43 <gbruno> [github] kintel pushed 8 modifications (Review and cleanup) https://github.com/openscad/openscad/commit/9074646f9aabac842f9864b07b86c2b3de69786d

18:43 <gbruno> [github] kintel synchronize pull request #4576 (Refactor offscreen context management) https://github.com/openscad/openscad/pull/4576

18:47 <gbruno> [github] kintel pushed 8 modifications (Review and cleanup) https://github.com/openscad/openscad/commit/d4c67698d5878bccc491b1107e28ae256116634f

18:47 <gbruno> [github] kintel synchronize pull request #4576 (Refactor offscreen context management) https://github.com/openscad/openscad/pull/4576

18:52 <kintel> teepee Since you tend to get involved in submodule discussions: I'm considering adding OpenCSG as a submodule, to be able to do some more heavy work on it. I feel the whole EGL/glew issue, followed by OpenGL modernization, will effectively kill existing OpenCSG maintenance as it requires significant rework.

18:52 <kintel> For the time being, we can stay compatible with both, smth. like this: https://github.com/openscad/openscad/tree/kintel-opencsg-submodule-only

18:52 <kintel> ..but at some point, #ifdef hell will break us if we don't make a clean switch : /

18:52 <kintel> Any ideas around this?

19:08 <guso7857> kintel, i compiled now with mimalloc disabled and my testcase which created a segfault in 100% of the cases is fine now.

19:09 <guso7857> i will continue exploring without mimalloc for the moment to eventually catch a segfault

19:09 <guso7857> you are right. in my local Cmakelists, python stuff is definitely the 1st :)   but bcrypt is 2nd ... mimalloc is somewhere ...

19:10 <guso78> will grant 1st link lib place to other modules in trade of not sefaulting anymore :)

19:12 <kintel> mimalloc comes from "add_subdirectory(submodules)"

19:14 <kintel> I have a patch somewhere adding some comments. feel free to steal my comment as it may be a while before my branch lands: https://github.com/openscad/openscad/blob/kintel-opencsg-submodule/CMakeLists.txt#L182

19:14 <teepee> kintel: the submodule stuff mostly get very difficult to work with once it really gets recursive dragging in heaps of stuff we don't even need

19:14 <teepee> I don't see that so much with opencsg

19:15 <kintel> true, I'm also not planning on using OpenCSG's build system :)

19:15 <kintel> Challenge may just be distro packaging

19:15 <guso78> ok, i have to move place :)

19:15 <teepee> uff, it's still qmake, so I'm not exactly sad if that goes away

19:44 <gbruno> [github] kintel pushed 1 modifications (macOS build fix) https://github.com/openscad/openscad/commit/d5f3e3db5afc99c6431b2390b73ef3fce1662a2d

19:44 <gbruno> [github] kintel synchronize pull request #4576 (Refactor offscreen context management) https://github.com/openscad/openscad/pull/4576

23:42 <gbruno> [github] kintel pushed 2 modifications (Fixed codeql warning) https://github.com/openscad/openscad/commit/26eac114943c03576a4f4980363355a0f444d845

23:42 <gbruno> [github] kintel synchronize pull request #4576 (Refactor offscreen context management) https://github.com/openscad/openscad/pull/4576