[BUG] "bxt_cap_skip_non_gameplay_frames 1" crashes when demo(s) enqueued for playback #93
Labels
bug
Something isn't working
Comments
khanghugo
added a commit
to khanghugo/bxt-rs
that referenced
this issue
Jan 29, 2024
|
Thanks for the detailed report. I pushed out a fix #94 and it doesn't crash on my end (anymore). |
|
Oh wow, I was trying to reproduce this earlier and I had crash. It had nothing to do with this apparently (probably Wayland Linux shenanigan but I thought it did hence the #94 ). So, I cannot reproduce this error and the PR I just mentioned earlier MIGHT fix something, which I am positively believing that it will not fix anything. I am testing on Linux and it showed no problem. However, when I test with WINE, it seems to crash when demo changes through That's my early report. Let me see what is really going on. |
|
Maybe platform difference because I cannot reproduce it in Half Life with your report through WINE. Though Counter-Strike 1.6 crashes even without injection so I guess I cannot fix this unless someone else can chip in. Just to clarify why #94 exists again, I was trying reproduce this and I got crash (twice). So I came up with that fix and it doesn't crash. I then tried the non patched version and it doesn't crash. Very odd yes. |
|
I tried the DLL produced for #94 and the issue went away. I then retried the DLL from latest master like you described and the issue came back, so #94 does indeed appear to fix the issue, at least on my end. FWIW - my setup specifically is run with wine32on64 on an M2 Mac, so I definitely would not be surprised if the issue were specific to my setup 🙃 |
|
Spoke too soon - the issue is indeed fixed as described, but now introducing |
|
That is definitely odd. If possible, you can run the debug build and see if there is any backtrace from bxt-rs. The issue you are having is different now. For me, this works perfectly normal on my machine(s) (the one I was testing with has not been in used for weeks so I guess it could be why it crashed at first). I am curious why you would run wine for the game. Is it possible for you to run the game natively? I could be possible that the game might work better that way since it barely works for me when I try Counter-Strike. Anyways, I assume you are running on legacy version? (pre anniversary) |
|
Sorry, I think you'll have to clarify what you mean by 'backtrace' - I just downloaded the latest bxt-rs-Windows-debug artifact and injected that, then did I specifically am running games from GoldSrc Package 2.4 rather than directly from Steam, so not using wine is not a possibility for me (nor is running any GoldSrc game natively, since as I mentioned previously I'm on an M2 Mac). I'll dump the bxt-rs.log output here in case you're able to glean something from it I can't: |
|
Hey, have you resolved this? This still seems to behave normally for me. |
|
No, I have not resolved this. I am still waiting for additional guidance on how to debug this further. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Bug found with latest bxt-rs-Windows-release artifact built off master
Reproduction steps:
bxt_cap_skip_non_gameplay_frames 1.playdemoor viabxt_play_run/bxt_play_folder.Game crashes after freezing briefly on an ostensible OOB frame from the demo.
Same crash can be reproduced via
bxt_play_run/bxt_play_folderthenbxt_cap_skip_non_gameplay_frames 1during demo playback, but not viaplaydemoon a single demo thenbxt_cap_skip_non_gameplay_frames 1while it plays back, indicating the issue is with demos being queued to play rather than demos playing outright.Note that
bxt_cap_startdoes not need to be executed in order to reproduce the crash.When using
bxt_play_run/bxt_play_folder, bxt-rs.log shows DEBUG lines for demos being queued for playback, but nothing after that.The text was updated successfully, but these errors were encountered: