foul
New Member
Posts: 7
|
Post by foul on Feb 6, 2007 19:51:36 GMT -5
foul@Foulnux:~/pSX$ gdb pSX GNU gdb 6.4.90-debian Copyright (C) 2006 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i486-linux-gnu"...Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1".
(gdb) run Starting program: /home/foul/pSX/pSX [Thread debugging using libthread_db enabled] [New Thread -1233614224 (LWP 6385)] Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 buffer_time=341315us buffer_size=15052 offset=2205 period=21333us period_size=940 (dir=1) pad=0 WiseGroup.,Ltd MP-8866 Dual USB Joypad: id=958a1c8e js=/dev/input/js1 ff=(null) WiseGroup.,Ltd MP-8866 Dual USB Joypad: id=958a1c8f js=/dev/input/js0 ff=(null) sound: underrun sound: underrun sound: underrun sound: underrun sound: underrun
Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1233614224 (LWP 6385)] 0xb6b50e44 in _nv000087gl () from /usr/lib/libGLcore.so.1 (gdb)
|
|
|
Post by chrissibaer on Feb 6, 2007 23:34:17 GMT -5
Hi everyone,
i am happy to see, that pSX is being ported to Linux. Thank you very much, pSX_Author!
However it also segfaults for me. I've tested it on a SUSE10.2, with Nvidia drivers. The pSX window appears for a split second and then it segfaults.
Here is the backtrace: --- chris@darkstar:~/emus/pSX> gdb pSX GNU gdb 6.5 Copyright (C) 2006 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i586-suse-linux"...Using host libthread_db library "/lib/libthread_db.so.1".
(gdb) run Starting program: /home/chris/emus/pSX/pSX Failed to read a valid object file image from memory. [Thread debugging using libthread_db enabled] [New Thread -1233111312 (LWP 9019)] Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 buffer_time=341315us buffer_size=15052 offset=882 period=21333us period_size=940 (dir=1) pad=0
Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1233111312 (LWP 9019)] 0xb35b3ef0 in ?? () (gdb) bt #0 0xb35b3ef0 in ?? () #1 0xb7d98d17 in gdk_gl_get_proc_address () from /opt/gnome/lib/libgdkglext-x11-1.0.so.0 #2 0xb7d9d585 in gdk_gl_get_glXSwapIntervalSGI () from /opt/gnome/lib/libgdkglext-x11-1.0.so.0 #3 0x08109e25 in video::draw_frame () #4 0x080c3ba2 in psx_gpu_software::paint_frame () #5 0x080c4404 in psx_gpu_software::show_frame () #6 0x080d69b9 in idle_callback () #7 0xb75b4061 in g_source_is_destroyed () from /opt/gnome/lib/libglib-2.0.so.0 #8 0xb75b5de2 in g_main_context_dispatch () from /opt/gnome/lib/libglib-2.0.so.0 #9 0xb75b8e1f in g_main_context_prepare () from /opt/gnome/lib/libglib-2.0.so.0 #10 0xb75b91c9 in g_main_loop_run () from /opt/gnome/lib/libglib-2.0.so.0 #11 0xb7a16cd4 in gtk_main () from /opt/gnome/lib/libgtk-x11-2.0.so.0 #12 0x080d7c35 in do_main () #13 0x080d8288 in main () (gdb) quit The program is running. Exit anyway? (y or n) y ---
Hopefully this bt will be helpfull. Thanks again for porting.
Thanks, Christian
|
|
hanman
Full Member
Irvine "Super-Pimp" Kinneas
Posts: 142
|
Post by hanman on Feb 7, 2007 9:21:00 GMT -5
ok, here's my gdb output:
(gdb) run Starting program: /home/andy/pSX/pSX warning: Lowest section in system-supplied DSO at 0xffffe000 is .hash at ffffe0b4 [Thread debugging using libthread_db enabled] [New Thread 4135098960 (LWP 1816)]
(pSX:1816): Gtk-WARNING **: Locale not supported by C library. Using the fallback 'C' locale. Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
(pSX:1816): Gtk-WARNING **: Unable to locate theme engine in module_path: "qtengine",
(pSX:1816): Gtk-WARNING **: Unable to locate theme engine in module_path: "qtengine", buffer_time=341315us buffer_size=15052 offset=882 period=21333us period_size=940 (dir=1) pad=0
Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 4135098960 (LWP 1816)] 0xf4c26018 in ?? () (gdb) bt #0 0xf4c26018 in ?? () #1 0x0811334e in mips_base__dynamic_execute () at src/mips_base_asm.asm:501 #2 0x07ffffff in ?? () #3 0x000099b1 in ?? () #4 0x085c631c in ?? () #5 0xffd1ac98 in ?? () #6 0xffd1ac7c in ?? () #7 0x085c56d0 in ?? () #8 0x085b8630 in ?? () #9 0x00200000 in ?? () #10 0xf4c26018 in ?? () #11 0x0811156f in mips_base::dynamic_execute () Previous frame inner to this frame (corrupt stack?)
thanks for everything, pSX Arthor
|
|
|
Post by patrickp on Feb 7, 2007 17:18:40 GMT -5
hanman, I notice you're getting Gtk warnings - did you check that you have the packages myself and Ultima mentioned installed? Probably won't stop the crashes, but it may fix your Gtk warnings, and perhaps make the emulator run a bit better...
|
|
hanman
Full Member
Irvine "Super-Pimp" Kinneas
Posts: 142
|
Post by hanman on Feb 7, 2007 18:34:14 GMT -5
hanman, I notice you're getting Gtk warnings - did you check that you have the packages myself and Ultima mentioned installed? Probably won't stop the crashes, but it may fix your Gtk warnings, and perhaps make the emulator run a bit better... yeah, i have libgtkglext1 installed; i couldn't get it to run at all without that. i have pretty much the same setup at home as i do at work (except 64 vs. 32 bit), and it does fine at work. thanks anyway, patrickp.
|
|
|
Post by Ultima on Feb 7, 2007 18:50:47 GMT -5
When I see MIPS in the backtrace, for some reason, I think "processor issue" Odd question-marks in the backtrace, but I'm not sure if that'd be considered abnormal.
|
|
hanman
Full Member
Irvine "Super-Pimp" Kinneas
Posts: 142
|
Post by hanman on Feb 7, 2007 19:24:32 GMT -5
i was a little puzzled by it, myself. i'm certainly not running on MIPS hardware.
|
|
|
Post by Sune on Feb 7, 2007 20:31:18 GMT -5
My first thought was that it was referring to the emulated MIPS CPU.
If not then it's very mysterious indeed.
|
|
|
Post by Ultima on Feb 7, 2007 20:39:20 GMT -5
Probably is.
[talking-out-of-arse] Since the error does refer to some .asm file, another random guess I'd make is that it has something to do with the assembler code not... being... compatible? I'd think that odd though, as the x86-64 CPUs are supposed to be backwards compatible, are they not? Maybe it's just another Ultima theory gone down the drain xD [/talking-out-of-arse]
|
|
|
Post by pSX Author on Feb 8, 2007 9:04:56 GMT -5
ok, here's my gdb output: (gdb) run Starting program: /home/andy/pSX/pSX warning: Lowest section in system-supplied DSO at 0xffffe000 is .hash at ffffe0b4 [Thread debugging using libthread_db enabled] [New Thread 4135098960 (LWP 1816)]
(pSX:1816): Gtk-WARNING **: Locale not supported by C library. Using the fallback 'C' locale. Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Error: scsi status=251 sense=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
(pSX:1816): Gtk-WARNING **: Unable to locate theme engine in module_path: "qtengine",
(pSX:1816): Gtk-WARNING **: Unable to locate theme engine in module_path: "qtengine", buffer_time=341315us buffer_size=15052 offset=882 period=21333us period_size=940 (dir=1) pad=0
Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 4135098960 (LWP 1816)] 0xf4c26018 in ?? () (gdb) bt #0 0xf4c26018 in ?? () #1 0x0811334e in mips_base__dynamic_execute () at src/mips_base_asm.asm:501 #2 0x07ffffff in ?? () #3 0x000099b1 in ?? () #4 0x085c631c in ?? () #5 0xffd1ac98 in ?? () #6 0xffd1ac7c in ?? () #7 0x085c56d0 in ?? () #8 0x085b8630 in ?? () #9 0x00200000 in ?? () #10 0xf4c26018 in ?? () #11 0x0811156f in mips_base::dynamic_execute () Previous frame inner to this frame (corrupt stack?)
thanks for everything, pSX Arthor This crash is due to the CPU supporting the NX bit (currently pSX allocates code for the recompiler on the heap, but on newer CPUs the heap is not executable). I have not yet implemented a solution for this, but I will this weekend...
|
|
|
Post by pSX Author on Feb 8, 2007 9:07:28 GMT -5
Probably is. [talking-out-of-arse] Since the error does refer to some .asm file, another random guess I'd make is that it has something to do with the assembler code not... being... compatible? I'd think that odd though, as the x86-64 CPUs are supposed to be backwards compatible, are they not? Maybe it's just another Ultima theory gone down the drain xD [/talking-out-of-arse] The reason there is an .asm file in the callstack is that the interface between the C++ code and the recompiler generated code is written in assembler. But.. gdb is confused here anyway because the callstack makes no sense to it (the recompiler does not use a normal stack frame setup - in fact the "stack" for the recompiled code is the emulated r3000 stack... so its not suprising gdb doesn't know what is going on!).
|
|
foul
New Member
Posts: 7
|
Post by foul on Feb 8, 2007 9:32:15 GMT -5
tell us when a new version will be ready .. i'm impatient to test it
|
|
|
Post by pSX Author on Feb 8, 2007 10:10:11 GMT -5
tell us when a new version will be ready .. i'm impatient to test it At the weekend...
|
|
|
Post by patrickp on Feb 8, 2007 14:06:47 GMT -5
[dribble]You're not the only one, foul![/dribble]
|
|
hanman
Full Member
Irvine "Super-Pimp" Kinneas
Posts: 142
|
Post by hanman on Feb 9, 2007 15:46:17 GMT -5
This crash is due to the CPU supporting the NX bit (currently pSX allocates code for the recompiler on the heap, but on newer CPUs the heap is not executable). I have not yet implemented a solution for this, but I will this weekend... So, you're saying my rig has too many new fangled features? SWEET! ;D anyway, thanks for the update. i'm looking forward to this weekend!
|
|