Skip to content
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

cyclone68k is causing invalid memory accesses #2

Open
meepingsnesroms opened this issue Dec 29, 2018 · 2 comments
Open

cyclone68k is causing invalid memory accesses #2

meepingsnesroms opened this issue Dec 29, 2018 · 2 comments

Comments

@meepingsnesroms
Copy link

meepingsnesroms commented Dec 29, 2018

Exact version of emu:
meepingsnesroms/Mu@adbd803

Crashes regardless of if compiled for RetroArch or Qt.

To trigger a crash turn the emulated device off then back on.

I have had issues like this before and attributed it to the compressed jump table.
Back then it crashed on boot with a invalid memory access.

For the Qt build some opcodes are renamed but for the libretro build its using the output of your newest version.

This is the config.h used to build it as well:
https://github.com/meepingsnesroms/Mu/blob/master/src/m68k/cyclone/config.h

It works fine with musashi(and always has) so I know its not an emulator side issue.

If there is any more information you need to know I will gladly provide it.

@notaz
Copy link
Owner

notaz commented Jan 1, 2019

There is not enough information to help you here (something like backtrace with register dump would be helpful), and I don't have time to figure out how to build/use your project, sorry. It looks like you have removed the whole thing anyway? Compressed jumptable was off in your config, so it's something else.

@meepingsnesroms
Copy link
Author

Since the lowest power ARM platform I am targeting is the New 3DS I removed cyclone because I think using the local memory handlers in musashi will be enough to get it up to speed, that is true, but when I find a bug I try to get it fixed regardless of it will benefit my personal emulation goals.

I was not saying compressed jumptable was causing this issue, I was saying I originally thought it was the cause of the SIGSEGVs, but it wasnt, turning it off just masked the issue.

Here is a register dump but the stack trace is lacking, it is compiled in debug mode, but I think the stack may be corrupted so it cant get a trace:

01-01 15:52:49.249  4267  4267 F DEBUG   : *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
01-01 15:52:49.249  4267  4267 F DEBUG   : LineageOS Version: '14.1-20181020-NIGHTLY-otus'
01-01 15:52:49.249  4267  4267 F DEBUG   : Build fingerprint: 'motorola/otus_retuaws/otus:5.0.2/LXC22.99-12/10:user/release-keys'
01-01 15:52:49.249  4267  4267 F DEBUG   : Revision: '0'
01-01 15:52:49.249  4267  4267 F DEBUG   : ABI: 'arm'
01-01 15:52:49.249  4267  4267 F DEBUG   : pid: 4199, tid: 4245, name: qtMainLoopThrea  >>> org.MeepingSnesRoms.Mu <<<
01-01 15:52:49.249  4267  4267 F DEBUG   : signal 11 (SIGSEGV), code 2 (SEGV_ACCERR), fault addr 0x9607f698
01-01 15:52:49.250  4267  4267 F DEBUG   :     r0 00000003  r1 9a8a01d0  r2 97c00000  r3 00000010
01-01 15:52:49.250  4267  4267 F DEBUG   :     r4 00000000  r5 00000100  r6 9a975010  r7 9a975010
01-01 15:52:49.250  4267  4267 F DEBUG   :     r8 00000078  r9 aabfc4b0  sl 9be9ec7c  fp 9a8ccf38
01-01 15:52:49.250  4267  4267 F DEBUG   :     ip 0003b2a0  sp 9607f614  lr 9607f698  pc 9607f698  cpsr 80070010
01-01 15:52:49.252  4267  4267 F DEBUG   : 
01-01 15:52:49.252  4267  4267 F DEBUG   : backtrace:
01-01 15:52:49.252  4267  4267 F DEBUG   :     #00 pc 000fd698  <anonymous:95f82000>

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants