https://bugs.winehq.org/show_bug.cgi?id=53063
Bug ID: 53063 Summary: SPORE crashes at launch after 4a98b07c4bcc35a698448261478ba856c149cbea Product: Wine Version: 7.9 Hardware: x86-64 OS: Linux Status: NEW Keywords: regression Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: z.figura12@gmail.com CC: huw@codeweavers.com, zzhang@codeweavers.com Regression SHA1: 4a98b07c4bcc35a698448261478ba856c149cbea Distribution: ---
winedbg doesn't seem to be able to get a backtrace, but the crash is inside of glTexSubImage2D(), called from wined3d_create().
The application does seem to use D3DKMT APIs directly, but not that one, and it crashes before getting to that point.
I'm currently debugging this.
https://bugs.winehq.org/show_bug.cgi?id=53063
--- Comment #1 from Zeb Figura z.figura12@gmail.com --- I somehow managed to get a backtrace just by adding channels. It crashes in ld-linux.so, in tls_get_addr_tail() of all things. Somehow the_map is NULL here:
https://sourceware.org/git/?p=glibc.git;a=blob;f=elf/dl-tls.c;hb=glibc-2.33#...
Which suggests that somehow glibc data structures are being corrupted?
Full backtrace winedbg can get is:
=>0 0xf7f7e56f in ld-linux.so.2 (+0x1356f) (0x00000000) 1 0x756b269b in libllvm-13.so.1 (+0x8b769b) (0x7d76f980) 2 0x00000000 (0x7a30212c) 3 0x7587e100 in libllvm-13.so.1 (+0xa83100) (0x7587dfc0)
Vastly incomplete, but anyway the glTexSubImage2D() call is coming from match_fbo_tex_update().
https://bugs.winehq.org/show_bug.cgi?id=53063
--- Comment #2 from Zeb Figura z.figura12@gmail.com --- But on the other hand, D3DKMTOpenAdapterFromLuid() is only called from wined3d, so it doesn't seem like the game should be doing anything special here.
VK_ICD_FILENAMES=/usr/share/vulkan/icd.d/lvp_icd.i686.json makes it work. This is probably going to be painful to debug without a working debugger or valgrind.
https://bugs.winehq.org/show_bug.cgi?id=53063
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|SPORE crashes at launch |SPORE crashes at launch |after | |4a98b07c4bcc35a698448261478 | |ba856c149cbea |
https://bugs.winehq.org/show_bug.cgi?id=53063
--- Comment #3 from Zhiyi Zhang zzhang@codeweavers.com --- Should be fixed with 00eff14f33233a7aa2330d9242e1f3584cc3b4e2
https://bugs.winehq.org/show_bug.cgi?id=53063
--- Comment #4 from Zeb Figura z.figura12@gmail.com --- Unfortunately it's still crashing for me. Forcing lavapipe still works around it.
https://bugs.winehq.org/show_bug.cgi?id=53063
--- Comment #5 from Zhiyi Zhang zzhang@codeweavers.com --- Unfortunately, I can't reproduce the crash, even with 4a98b07.
https://bugs.winehq.org/show_bug.cgi?id=53063
Zeb Figura z.figura12@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |DUPLICATE
--- Comment #6 from Zeb Figura z.figura12@gmail.com --- Reducing the address space limit reserved by the preloader from 0x68000000 to 0x48000000 fixes it. So this is another instance of bug 52840.
*** This bug has been marked as a duplicate of bug 52840 ***
https://bugs.winehq.org/show_bug.cgi?id=53063
Alistair Leslie-Hughes leslie_alistair@hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #7 from Alistair Leslie-Hughes leslie_alistair@hotmail.com --- Closing duplicate.