-
Notifications
You must be signed in to change notification settings - Fork 62
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Whdload .lha Files mode crashes first run on Nvidia Shield (Android 12, arm 32 bit) #513
Comments
Huh, that is what libretro VFS was supposed to solve.. Please show the debug level verbose log of the core trying to create the WHDLoad helper image directory. Delete the old remnants first if there are any, and please also show what the |
Thank you! I don't think I've felt this much newbie embarrassment, nor had an issue solved this quickly and easily. I didn't know and should had checked whether Whdload uses any files and delete them. After deleted everything named WHD* in the system and saves directory, Retroarch crashed after the first attempt to load an .lha game. I am guessing it happened when the core re-created the Whdload files? Regardless, it now works when starting Retroarch a second time. Tried a couple of them, and they all seem to work now. Closing this one as solved. |
Better than nothing sure, but it should not crash at the first run.. Which means I'm still interested in the debug verbose log when that happens. |
I hope this helps:
|
Thanks, but unfortunately it only shows that everything went fine.. The next step after that is copying Please try to delete only Also after the first crash, does it crash again if you change the "WHDLoad Splash Screen" core option? Is there And please change the title since it does not represent what is happening here. |
I don't have one. As I understand, it's been removed for some timed now.
Crash on first run. Not on subsequent runs. The contents of the two prefs files are identical except for
There's no change in behaviour regardless what I set the Splash Screen option to, neither if I change on the first run nor after the first crash. |
Thanks. I guess that narrows it down enough that it is related to the prefs file. Shame that I can't replicate it no matter what. I can at least try adding logging to the related stages. Deleting only the Kickstart files that are shown at the end of the crash log should cause nothing then at the next run? |
Since I myself cannot reproduce it on x64 linux, are you certain this is a bug in the core and not something externally (Retroarch, Android, Android 12, the Nvidia Shield) casuing it?
I'll try that one out. |
That is what I'm wanting to find out. I pushed some cleanups that will also show more logging, so please post the crash log after deleting prefs once the version is available. |
Will do as soon as I'm back. I'm going away for a week. (never mind my previous lenghty post, the error with Whdload on my x64 linux system seems unrelated and I need to do some more trouble shooting before I can confirm it's not due to a compilation error) |
Sorry for the late update. I've been trying the same thing again with two more recent firmware updates to the nVidia shield that were released. But there's no difference in the output in the error logs compared to by earlier logdump. Hence I'm almost certain this is due to something crashing due to Android. Unfortunately I'm not proficient enough in using adb to know where to start if I would want to see the Android log files. |
You could try again @veganvelociraptor , there was a VFS fix right now. |
Also pretty sure it is not Android (12) related since everything works fine with my phone. |
Yep, it's certainly specific to the Nvidia Shield, or possibly the Android TV variant of Android. I don't have these issues on my phone either. What's strange is that I only have crash issues with two cores: UAE and VICE. |
Hmm, VICE does not extract anything on any run like PUAE does on first WHDLoad LHA run.. The cores do share a lot of similar libretro-side code, since maintaining them is more sane that way. So what the heck is the common denominator with the cores then..? I'd like to see a crashing debug log from VICE too. |
I finally learned how to use adb debug to get a log of the crashes. Give me the upcoming weekend to try to post something usable. I'll also try and see if Lakka on a Raspberry Pi crashes, just to exclude that it's the ARM CPU being the cause. A pity there is no NVidia Shield or Android TV Emulator (AFAIK?). |
Attempting to open a whdload .lha game when using Files mode on my Nvidia Shield, results in a CLI prompt being shown instead of the game actually loading.
The games work when switching to HDF mode instead. As the bug is not present on my x64 linux distro, I'm guessing this is specific to either the Nvidia Shield, Android 12 and/or the Android build of this core.
(using the latest nightly armeabi-v7a build of Retroarch and the PUAE core)
The text was updated successfully, but these errors were encountered: