OOLite locks up with Open-Source Radeon driver (Linux)

[Ijon Tichy]

Hi again,
detected another problem:

After kernel upgrade to current version 3.17, again I had troubles with recompilation of proprietary NVidia drivers. Being tired of this, reverted to "native" Nouveau drivers; unfortunately, it turned out, they are very slow compared to NVidia's ones. Then I replaced the adaptor with old Radeon 9600 - and I have to admit, that the quality of open-source "Gallium" drivers for Radeon is very good: there is only a few percent more CPU load while the "combat" screen is visible, and - surprisingly - on "Market prices" screen I have even slightly lower(!) load, than before (with NVidia on "closed" drivers); it's not full 100% anymore, but 90-92% instead.

Well the drivers seems to work fine, but there's a serious problem:

1. I use "virtual desktop". Formerly (with NVidia) I was using Xorg.conf, now it's set with xrandr:

xrandr --output DVI-0 --mode 1280x800 --panning 1680x1050

The problem is, that although formerly OOLite correctly detected "visible window" size (1280x800), and opened its window exactly using such size - now, working under Gallium driver it opens its window of the size of entire virtual screen (1680x1050), making part of action window invisible.

"You may resize the game window", surely you'll say. Well, that's the second - serious - problem:

2. When trying to move/resize game window - or even to move mouse cursor across it - it locks whole system completely. Even "power off" button doesn't work, which is suggesting, that the system may be completely hung up (the title theme music drums are repeated in a short loop like kinda machine gun ). There wasn't such behaviour formerly, with NVidia "closed" drivers.

No idea, are the drivers to blame, or OOLite - but I didn't notice such problem with any other program so far.

As for "window size" issue: if it's not possible to detect "visible window" size properly in each case, a command line option to set desired game window dimensions would be welcome (tried "oolite --help" but it didn't respond with anything, just started the game). Well even having this I'm afraid, that still the second problem may be an issue when trying to move mouse cursor over the smaller window - or just while trying to move game window elsewhere on the screen.

Slackware 14.1 32bit, X.Org X Server 1.14.3, kernel 3.17, OOLite 1.80

Output of my xrandr:

Screen 0: minimum 320 x 200, current 1680 x 1050, maximum 4096 x 4096
VGA-0 disconnected (normal left inverted right x axis y axis)
DVI-0 connected 1680x1050+0+0 (normal left inverted right x axis y axis) 473mm x 296mm panning 1680x1050+0+0
1680x1050 60.0 +
1280x1024 60.0
1440x900 59.9
1280x800 59.9*
1024x768 60.0
800x600 60.3 56.2
640x480 60.0
720x400 70.1
S-video disconnected (normal left inverted right x axis y axis)

Ah, and one more: I had the same problem as this guy, until I properly reverted to Open Source drivers. Obviously, that messages about "pulseaudio problems" are misleading. It would be worthy to improve problem detection and/or reporting.

[another_commander]

... a command line option to set desired game window dimensions would be welcome...

I'm afraid I can't assist with the actual problem as I am running the game on Windows, but at least I can help you with that. Add the following two lines inside your .GNUstepDefaults file and Oolite will start with the specified game window size the next time you run it (the numbers are just an example, set them to whatever dimensions you need):

Code: Select all

window_height = 620;
window_width = 1010;

[cim]

... a command line option to set desired game window dimensions would be welcome...

I'm afraid I can't assist with the actual problem as I am running the game on Windows, but at least I can help you with that. Add the following two lines inside your .GNUstepDefaults file and Oolite will start with the specified game window size the next time you run it (the numbers are just an example, set them to whatever dimensions you need):

Code: Select all

window_height = 620;
window_width = 1010;

The Linux .GNUstepDefaults file is in XML format. I don't know why GNUstep chose a different default format on the two platforms... so it might also look something like this

Code: Select all

	<key>window_height</key>
	<integer>800</integer>
	<key>window_width</key>
	<integer>1000</integer>

Obviously, that messages about "pulseaudio problems" are misleading.

Yes. Unfortunately, if the game crashes in such a way that it isn't able to close its audio file handles first, but the OpenAL library survives long enough to notice that, those messages are printed by the library. There appears to be no way to turn them off despite them being completely useless.

Not sure about the crash problem - if you can try getting the window size to smaller than the screen size, that might help.

(Also could be a hardware problem - if it's an old card maybe it's damaged? Try running a graphics stress test / diagnostics program on it?)

[another_commander]

The Linux .GNUstepDefaults file is in XML format. I don't know why GNUstep chose a different default format on the two platforms...

You are right cim, it's XML format on Linux but OpenStep in Windows and it's not the GNUstep project who is responsible for that, it is... me. I deliberately changed the output format of .GNUstepDefaults in the Windows gnustep-base library to keep it more readable and because it was the format that we've had for this file in previous GNUstep versions. Sorry for potentially causing confusion.

[Ijon Tichy]

Not sure about the crash problem - if you can try getting the window size to smaller than the screen size, that might help.

(Also could be a hardware problem - if it's an old card maybe it's damaged? Try running a graphics stress test / diagnostics program on it?)

Yes - thanks, guys - the workaround allowed to start the game in the window of custom size, but - unfortunately - the "freezing problem" remains. It's even worse: I noticed, that any attempt to "pan" the window (I don't mean the game window, but the visible part of whole virtual screen) freezes the game and the system. I didn't even "touch" the game window with mouse cursor this time, just tried to move visible part - and it froze the machine again.

No, it's not graphics adaptor; it's old, but I owe it around 10 years, and it never failed before. Why should it fail now in such strange way (no other problems) - and with just one and only application?

[Norby]

I think this still can be driver related, you can try to determine if you test other games which use similar graphics libraries to reproduce your problems: I do not know exactly but maybe Extreme Tux Racer, Pingus and Supertuxcart.

I suggest to tell more about your compilation problem of proprietary NVidia drivers and if we can help then you will have a solution.

[Ijon Tichy]

I think this still can be driver related, you can try to determine if you test other games which use similar graphics libraries to reproduce your problems: I do not know exactly but maybe Extreme Tux Racer, Pingus and Supertuxcart.

Just tried Pingus - no such problem. Works flawlessly. Tried also Foobillard+ - this one also doesn't lock my machine up; I can play it. Obviously there is something in OOlite worthy of closer examination.

I suggest to tell more about your compilation problem of proprietary NVidia drivers and if we can help then you will have a solution.

Well usual "unable to build NVidia module" - the tail of the log:

/root/software/nvidia_driver/NVIDIA-Linux-x86-173.14.39-pkg1/usr/src/nv/nv-l
inux.h:218:8: warning: "NV_ACPI_WALK_NAMESPACE_ARGUMENT_COUNT" is not define
d [-Wundef]
#elif (NV_ACPI_WALK_NAMESPACE_ARGUMENT_COUNT == 7)
^
/root/software/nvidia_driver/NVIDIA-Linux-x86-173.14.39-pkg1/usr/src/nv/nv-l
inux.h:224:2: error: #error "NV_ACPI_WALK_NAMESPACE_ARGUMENT_COUNT value unr
ecognized!"
#error "NV_ACPI_WALK_NAMESPACE_ARGUMENT_COUNT value unrecognized!"
^
/root/software/nvidia_driver/NVIDIA-Linux-x86-173.14.39-pkg1/usr/src/nv/nv.c
: In function 'nv_kern_ioctl':
/root/software/nvidia_driver/NVIDIA-Linux-x86-173.14.39-pkg1/usr/src/nv/nv.c
:2646:26: warning: comparison between signed and unsigned integer expression
s [-Wsign-compare]
if (arg_size < (sizeof(*ci) * num_nv_devices))
^
make[4]: *** [/root/software/nvidia_driver/NVIDIA-Linux-x86-173.14.39-pkg1/u
sr/src/nv/nv.o] Error 1
make[3]: *** [_module_/root/software/nvidia_driver/NVIDIA-Linux-x86-173.14.3
9-pkg1/usr/src/nv] Error 2
make[2]: *** [sub-make] Error 2
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make[1]: *** [module] Error 1
make: *** [module] Error 2
-> Error.
ERROR: Unable to build the NVIDIA kernel module.
ERROR: Installation has failed. Please see the file
'/var/log/nvidia-installer.log' for details. You may find suggestions
on fixing installation problems in the README available on the Linux
driver download page at http://www.nvidia.com.

The problem is, NVidia isn't willing to support their older products, they stopped to develop 173.14.39 driver, it worked a few months more thanks to the patch released by some good people (for kernel 3.14), but with latest kernel version 3.17 the patch stopped working.

[Ijon Tichy]

I did another try, and ran the game with "strace -o oolog oolite". Here's what have been written into log until lock-up - maybe it can tell you something?

execve("/usr/local/bin/oolite", ["oolite"], [/* 51 vars */]) = 0
brk(0) = 0x812f000
uname({sys="Linux", node="Ferrycy", ...}) = 0
access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=200846, ...}) = 0
mmap2(NULL, 200846, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb7791000
close(3) = 0
open("/lib/libtermcap.so.2", O_RDONLY|O_CLOEXEC) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0`\n\0\0004\0\0\0"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=10280, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7790000
mmap2(NULL, 13192, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0xb778c000
mmap2(0xb778f000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x2000) = 0xb778f000
close(3) = 0
open("/lib/libdl.so.2", O_RDONLY|O_CLOEXEC) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\260\n\0\0004\0\0\0"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=17505, ...}) = 0
mmap2(NULL, 16516, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0xb7787000
mmap2(0xb778a000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x2000) = 0xb778a000
close(3) = 0
open("/lib/libc.so.6", O_RDONLY|O_CLOEXEC) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\0\231\1\0004\0\0\0"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=1816558, ...}) = 0
mmap2(NULL, 1616412, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0xb75fc000
mmap2(0xb7781000, 12288, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x184000) = 0xb7781000
mmap2(0xb7784000, 10780, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0xb7784000
close(3) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb75fb000
set_thread_area({entry_number:-1 -> 6, base_addr:0xb75fb6c0, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}) = 0
mprotect(0xb7781000, 8192, PROT_READ) = 0
mprotect(0xb778a000, 4096, PROT_READ) = 0
mprotect(0xb77e4000, 4096, PROT_READ) = 0
munmap(0xb7791000, 200846) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
open("/dev/tty", O_RDWR|O_NONBLOCK|O_LARGEFILE) = 3
close(3) = 0
open("/usr/lib/locale/locale-archive", O_RDONLY|O_LARGEFILE|O_CLOEXEC) = -1 ENOENT (No such file or directory)
brk(0) = 0x812f000
brk(0x8130000) = 0x8130000
brk(0x8131000) = 0x8131000
open("/usr/share/locale/locale.alias", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=2502, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb77c2000
read(3, "# Locale name alias data base.\n#"..., 4096) = 2502
brk(0x8132000) = 0x8132000
brk(0x8133000) = 0x8133000
read(3, "", 4096) = 0
close(3) = 0
munmap(0xb77c2000, 4096) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
open("/dev/tty", O_RDWR|O_NONBLOCK|O_LARGEFILE) = 3
close(3) = 0
open("/usr/lib/locale/locale-archive", O_RDONLY|O_LARGEFILE|O_CLOEXEC) = -1 ENOENT (No such file or directory)
brk(0) = 0x812f000
brk(0x8130000) = 0x8130000
brk(0x8131000) = 0x8131000
open("/usr/share/locale/locale.alias", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=2502, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb77c2000
read(3, "# Locale name alias data base.\n#"..., 4096) = 2502
brk(0x8132000) = 0x8132000
brk(0x8133000) = 0x8133000
read(3, "", 4096) = 0
close(3) = 0
munmap(0xb77c2000, 4096) = 0
open("/usr/lib/locale/pl_PL/LC_IDENTIFICATION", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=346, ...}) = 0
mmap2(NULL, 346, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77c2000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_MEASUREMENT", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=28, ...}) = 0
mmap2(NULL, 28, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77c1000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_TELEPHONE", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=54, ...}) = 0
mmap2(NULL, 54, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77c0000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_ADDRESS", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=132, ...}) = 0
mmap2(NULL, 132, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77bf000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_NAME", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=67, ...}) = 0
mmap2(NULL, 67, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77be000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_PAPER", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=39, ...}) = 0
mmap2(NULL, 39, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77bd000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_MESSAGES", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_MESSAGES/SYS_LC_MESSAGES", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=59, ...}) = 0
mmap2(NULL, 59, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77bc000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_MONETARY", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=295, ...}) = 0
mmap2(NULL, 295, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77bb000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_COLLATE", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=21055, ...}) = 0
mmap2(NULL, 21055, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77b5000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_TIME", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=2435, ...}) = 0
mmap2(NULL, 2435, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77b4000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_NUMERIC", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=59, ...}) = 0
mmap2(NULL, 59, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb77b3000
close(3) = 0
open("/usr/lib/locale/pl_PL/LC_CTYPE", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=221632, ...}) = 0
mmap2(NULL, 221632, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb75c4000
close(3) = 0
getuid32() = 1000
getgid32() = 1000
geteuid32() = 1000
getegid32() = 1000
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
time(NULL) = 1413983197
brk(0x8134000) = 0x8134000
open("/proc/meminfo", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb77b2000
read(3, "MemTotal: 2074344 kB\nMemF"..., 1024) = 1024
close(3) = 0
munmap(0xb77b2000, 4096) = 0
brk(0x8135000) = 0x8135000
rt_sigaction(SIGCHLD, {SIG_DFL, [], SA_RESTORER|SA_RESTART, 0xb762bd68}, {SIG_DFL, [], 0}, = 0
rt_sigaction(SIGCHLD, {SIG_DFL, [], SA_RESTORER|SA_RESTART, 0xb762bd68}, {SIG_DFL, [], SA_RESTORER|SA_RESTART, 0xb762bd68}, = 0
rt_sigaction(SIGINT, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, {SIG_DFL, [], 0}, = 0
rt_sigaction(SIGINT, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, = 0
rt_sigaction(SIGQUIT, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, {SIG_DFL, [], 0}, = 0
rt_sigaction(SIGQUIT, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
rt_sigaction(SIGQUIT, {SIG_IGN, [], SA_RESTORER, 0xb762bd68}, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, = 0
uname({sys="Linux", node="Ferrycy", ...}) = 0
brk(0x8136000) = 0x8136000
brk(0x8137000) = 0x8137000
brk(0x8138000) = 0x8138000
brk(0x8139000) = 0x8139000
stat64("/home/zb", {st_mode=S_IFDIR|0755, st_size=12288, ...}) = 0
stat64(".", {st_mode=S_IFDIR|0755, st_size=12288, ...}) = 0
getpid() = 11416
getppid() = 11413
brk(0x813a000) = 0x813a000
gettimeofday({1413983197, 369787}, NULL) = 0
brk(0x813b000) = 0x813b000
getpgrp() = 11413
rt_sigaction(SIGCHLD, {0x80868a0, [], SA_RESTORER|SA_RESTART, 0xb762bd68}, {SIG_DFL, [], SA_RESTORER|SA_RESTART, 0xb762bd68}, = 0
getrlimit(RLIMIT_NPROC, {rlim_cur=16178, rlim_max=16178}) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
open("/usr/local/bin/oolite", O_RDONLY|O_LARGEFILE) = 3
ioctl(3, SNDCTL_TMR_TIMEBASE or SNDRV_TIMER_IOCTL_NEXT_DEVICE or TCGETS, 0xbfc6fdb8) = -1 ENOTTY (Inappropriate ioctl for device)
_llseek(3, 0, [0], SEEK_CUR) = 0
read(3, "#!/bin/sh\n/opt/Oolite/oolite.app"..., 80) = 59
_llseek(3, 0, [0], SEEK_SET) = 0
getrlimit(RLIMIT_NOFILE, {rlim_cur=1024, rlim_max=4*1024}) = 0
fcntl64(255, F_GETFD) = -1 EBADF (Bad file descriptor)
dup2(3, 255) = 255
close(3) = 0
fcntl64(255, F_SETFD, FD_CLOEXEC) = 0
fcntl64(255, F_GETFL) = 0x8000 (flags O_RDONLY|O_LARGEFILE)
fstat64(255, {st_mode=S_IFREG|0755, st_size=59, ...}) = 0
_llseek(255, 0, [0], SEEK_CUR) = 0
brk(0x813c000) = 0x813c000
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
read(255, "#!/bin/sh\n/opt/Oolite/oolite.app"..., 59) = 59
open("/usr/lib/gconv/gconv-modules.cache", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib/gconv/gconv-modules", O_RDONLY|O_CLOEXEC) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=56302, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb77b2000
read(3, "# GNU libc iconv configuration.\n"..., 4096) = 4096
brk(0x813d000) = 0x813d000
brk(0x813e000) = 0x813e000
read(3, "lias\tCSISO141JUSIB1002//\tJUS_I.B"..., 4096) = 4096
brk(0x813f000) = 0x813f000
brk(0x8140000) = 0x8140000
read(3, "odule\t\tcost\nalias\tISO-IR-110//\t\t"..., 4096) = 4096
brk(0x8141000) = 0x8141000
brk(0x8142000) = 0x8142000
brk(0x8143000) = 0x8143000
read(3, "9-14//\nalias\tL8//\t\t\tISO-8859-14/"..., 4096) = 4096
brk(0x8144000) = 0x8144000
brk(0x8145000) = 0x8145000
brk(0x8146000) = 0x8146000
read(3, "CSEBCDICES//\t\tEBCDIC-ES//\nalias\t"..., 4096) = 4096
brk(0x8147000) = 0x8147000
brk(0x8148000) = 0x8148000
brk(0x8149000) = 0x8149000
read(3, "IBM284//\nalias\tEBCDIC-CP-ES//\t\tI"..., 4096) = 4096
brk(0x814a000) = 0x814a000
brk(0x814b000) = 0x814b000
brk(0x814c000) = 0x814c000
read(3, "ias\t864//\t\t\tIBM864//\nalias\tCSIBM"..., 4096) = 4096
brk(0x814d000) = 0x814d000
brk(0x814e000) = 0x814e000
brk(0x814f000) = 0x814f000
read(3, "lias\tCSIBM939//\t\tIBM939//\nmodule"..., 4096) = 4096
brk(0x8150000) = 0x8150000
brk(0x8151000) = 0x8151000
brk(0x8152000) = 0x8152000
read(3, "alias\tcsGB2312//\t\tEUC-CN//\nalias"..., 4096) = 4096
brk(0x8153000) = 0x8153000
brk(0x8154000) = 0x8154000
read(3, "//\t\tISO-2022-CN-EXT//\nmodule\tISO"..., 4096) = 4096
brk(0x8155000) = 0x8155000
brk(0x8156000) = 0x8156000
brk(0x8157000) = 0x8157000
read(3, "st\nalias\tISO-IR-55//\t\tISO_5428//"..., 4096) = 4096
brk(0x8158000) = 0x8158000
brk(0x8159000) = 0x8159000
read(3, "\tARMSCII-8//\t\tARMSCII-8\t1\n\n#\tfro"..., 4096) = 4096
brk(0x815a000) = 0x815a000
brk(0x815b000) = 0x815b000
brk(0x815c000) = 0x815c000
read(3, "2//\t\tIBM1112\t\t1\n\n#\tfrom\t\t\tto\t\t\tm"..., 4096) = 4096
brk(0x815d000) = 0x815d000
brk(0x815e000) = 0x815e000
brk(0x815f000) = 0x815f000
read(3, "module\tIBM9448//\t\tINTERNAL\t\tIBM9"..., 4096) = 3054
brk(0x8160000) = 0x8160000
brk(0x8161000) = 0x8161000
read(3, "", 4096) = 0
close(3) = 0
munmap(0xb77b2000, 4096) = 0
brk(0x8162000) = 0x8162000
open("/usr/lib/gconv/ISO8859-2.so", O_RDONLY|O_CLOEXEC) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\360\3\0\0004\0\0\0"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=11298, ...}) = 0
brk(0x8163000) = 0x8163000
mmap2(NULL, 12328, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0xb77af000
mmap2(0xb77b1000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1000) = 0xb77b1000
close(3) = 0
mprotect(0xb77b1000, 4096, PROT_READ) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], = 0
brk(0x8164000) = 0x8164000
rt_sigprocmask(SIG_BLOCK, [INT CHLD], [], = 0
_llseek(255, -8, [51], SEEK_CUR) = 0
clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0xb75fb728) = 11417
rt_sigprocmask(SIG_SETMASK, [], NULL, = 0
rt_sigprocmask(SIG_BLOCK, [CHLD], [], = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, = 0
rt_sigprocmask(SIG_BLOCK, [CHLD], [], = 0
rt_sigaction(SIGINT, {0x8083d70, [], SA_RESTORER, 0xb762bd68}, {SIG_DFL, [], SA_RESTORER, 0xb762bd68}, = 0
waitpid(-1,


======================
Yes, it looks like "torn" - nothing more after "waitpid(-1,".

[Lone_Wolf]

Ijon Tichy,

kernel + mesa and Xorg all have parts of the linux graphics subsystem.

Your kernel might just be to new compared to the other components.

can you also install mesa 10.3.1 & xorg 1.16.1 , or try with an older kernel ?

Oolite latest.log might also give some clues

[Ijon Tichy]

Ijon Tichy,

kernel + mesa and Xorg all have parts of the linux graphics subsystem.

Your kernel might just be to new compared to the other components.

Oh, please... my distro is Slackware 14.1, not 4.1.

Anyway: made two additional attempts, using kernels 3.16 and 3.10.32. You guessed the result? Yes, you're right: no change.

can you also install mesa 10.3.1 & xorg 1.16.1 , or try with an older kernel ?

Oolite latest.log might also give some clues

Maybe, but at the moment I cannot promise this - present setup works for me, so there's no need to change anything. Surely I won't compile entire Xorg infrastructure, sorry - too much time and possible problems (as I remember from the past), and - honestly - I doubt it'll remove the bug from the game.

So you're hoping the problem will go away by itself? No, it won't - it needs to be traced and eliminated.

[cim]

Could you post the Latest.log?

The strace seems to be of the Oolite wrapper script rather than the Oolite binary. It looks to be about as expected for the wrapper script, though.

I noticed, that any attempt to "pan" the window (I don't mean the game window, but the visible part of whole virtual screen) freezes the game and the system.

Okay, what if you set the desktop to be the same size as the screen? Does that stop the crashing, or does it just crash later when you try something else?

[Ijon Tichy]

Could you post the Latest.log?

Not much there:

Opening log for Oolite version 1.80 (x86-32) under Linux 3.10.32-std410-i586 at 2014-10-22 18:44:44 +0200.
1 processor detected.
Build options: OpenAL, new planets.

Note that the contents of the log file can be adjusted by editing logcontrol.plist.

18:44:45.068 [sound.initialization.error]: Failed to open default sound device
18:44:45.137 [display.mode.list.native]: X11 native resolution detected: 1680 x 1050
18:44:45.491 [joystick.init]: Number of joysticks detected: 0
18:44:45.493 [rendering.opengl.version]: OpenGL renderer version: 2.1.0 ("2.1 Mesa 10.2.4"). Vendor: "X.Org R300 Project". Renderer: "Gallium 0.4 on ATI RV350".
18:44:45.493 [rendering.opengl.extensions]: OpenGL extensions (143):
GL_ATI_texture_mirror_once, GL_ARB_depth_texture, GL_ARB_invalidate_subdata, GL_ARB_texture_env_add, GL_NV_texture_barrier, GL_EXT_pixel_buffer_object, GL_ARB_half_float_vertex, GL_IBM_multimode_draw_arrays,
18:44:45.576 [rendering.opengl.gpuSpecific]: Matched GPU configuration "ATI R300 family (X/Gallium3D)".
18:44:45.576 [rendering.opengl.shader.support]: Shaders are supported.
18:44:45.649 [searchPaths.dumpAll]: Unrestricted mode - resource paths:
/opt/Oolite/oolite.app/Resources
AddOns
18:44:45.689 [sound.initialization.error]: Failed to open default sound device
18:44:45.912 [shipData.load.begin]: Loading ship data.
18:44:46.081 [script.javascript.init]: JavaScript reset successful.
18:44:46.102 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/laserhits.ogg"
18:44:46.104 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/ecm.ogg"
18:44:46.106 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/missile.ogg"
18:44:46.107 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/guiclick.ogg"
18:44:46.108 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/beep.ogg"
18:44:46.109 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/trumble.ogg"
18:44:46.110 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/ecm.ogg"
18:44:46.112 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/warning.ogg"
18:44:46.113 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/hit.ogg"
18:44:46.114 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/boop.ogg"
18:44:46.115 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/hullbang.ogg"
18:44:46.116 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/laser.ogg"
18:44:46.118 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/buy.ogg"
18:44:46.120 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/breakpattern.ogg"
18:44:46.121 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/bigbang.ogg"
18:44:46.123 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/witchabort.ogg"
18:44:46.124 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/ecm.ogg"
18:44:46.125 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/trumblesqueal.ogg"
18:44:46.127 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/sell.ogg"
18:44:46.127 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/bigbang.ogg"
18:44:46.129 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/scoop.ogg"
18:44:46.129 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/hullbang.ogg"
18:44:46.131 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/afterburner1.ogg"
18:44:46.132 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/afterburner1.ogg"
18:44:46.138 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/trumble.ogg"
18:44:46.139 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Sounds/trumblesqueal.ogg"
18:44:48.048 [startup.complete]: ========== Loading complete in 3.15 seconds. ==========
18:44:48.423 [sound.load.error]: Failed to load sound "/opt/Oolite/oolite.app/Resources/Music/OoliteTheme.ogg"

It's describing an attempt with kernel 3.10.32 - yes, I had some sound-related problems.

Okay, what if you set the desktop to be the same size as the screen? Does that stop the crashing, or does it just crash later when you try something else?

Yes, it stopped crashing completely. I could move mouse cursor around, I could do whatever I wanted with game window - no problem.

So it seems, OOlite does *something* mysterious, which confuses Radeon R300 driver, and only when in "virtual screen" mode. I'm unable to create such effect with any other application so far.

[cim]

So it seems, OOlite does *something* mysterious, which confuses Radeon R300 driver, and only when in "virtual screen" mode. I'm unable to create such effect with any other application so far.

Well, at least that narrows it down. I'll try setting up a virtual desktop here when I have a bit more time to see if I can get the same error (I doubt it, since I don't have the same card or driver, but worth a go...)

[Lone_Wolf]

So you're hoping the problem will go away by itself? No, it won't - it needs to be traced and eliminated.

I agree, reason i asked you to try out combo of latest xorg / mesa / kernel is that i recently helped someone using a card that also runs on the R300 driver and had troubles with virtual desktop bigger then screen size also.

combo was kernel 3.16.6 , xorg 1.15 and mesa 10.2.4 .
After upgrading to xorg 1.16.1 & mesa 10.3.1 the problem was gone.
That user is currently trying to figure out whether the problem was with xorg or mesa, so a bugreport can be filed at the correct location.

[Ijon Tichy]

I don't like so much "improving" perfectly running system - but OK: I'll try tomorrow to upgrade Mesa and xserver (not whole Xorg). Maybe indeed it can help.