Re: xsrc/58321: macppc 10.0 X.org server on Power Mac G4 + ATI Rage gets SIGSEGV

[Taylor R Campbell <riastradh%NetBSD.org@localhost>]

> Date: Sun, 9 Jun 2024 00:24:40 +0900 (JST)
> From: Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>
> 
> [   113.148] (EE) 0: X (xorg_backtrace+0x44) [0x19d5e44]
> [   113.148] (EE) 1: X (os_move_fd+0xcc) [0x19d0b44]
> [   113.149] (EE) 2: /usr/lib/libc.so.12 (__sigtramp_siginfo_2+0x0) [0xfd4c0f70]
> [   113.149] (EE) 3: /usr/lib/libc.so.12 (strcmp+0x18) [0xfd4c62a8]
> [   113.150] (EE) 4: X (xf86ProbeOutputModes+0x628) [0x18e5dc0]
> [   113.150] (EE) 5: X (xf86InitialConfiguration+0x108) [0x18e6468]
> [   113.150] (EE) 6: /usr/X11R7/lib/modules/drivers/r128_drv.so (R128PreInit+0xbb8) [0xfc5611e0]
> [   113.151] (EE) 7: X (InitOutput+0x994) [0x18b54c8]
> [   113.151] (EE) 8: X (dix_main+0x1f4) [0x182efbc]
> [   113.151] (EE) 
> [   113.152] (EE) Segmentation fault at address 0x0

This is a null pointer dereference, probably passing a null pointer to
strcmp in xf86ProbeOuptutModes or xf86SortModes.

Is there a core dump that you can open in gdb?  Can you print the
locals inside xf86ProbeOutputModes?

If not, can you add printfs to xf86ProbeOutputModes to see what's
going on?