Trouble on SuSE 6.2

Yeah I’ve gotten SDL to compile on SuSE 6.3 but it was actually 0.11.2 I
think. I can’t test it out now, I replaced the SuSE computer with Corel
Linux. No Corel-bashing please, it was for my sister.

Sam Lantinga wrote:> > I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end

everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g). What can be the reason for that? Are
there any special configure-switches I should use?

This is a known bug in SuSE Linux 6.2 Get any updates, or upgrade to 6.3


-= aaron p. matthews
-= rival entertainment
-= http://www.Nayzak.com/~jerryma/rival

I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end
everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g). What can be the reason for that? Are
there any special configure-switches I should use?
Here is some strace-output for testthread:

semget(IPC_PRIVATE, 1, IPC_CREAT|0x180|0600) = 3
semctl(3, 0, SETVAL, 0xbffff63c) = 0
semop(0x3, 0x2, 0, 0x4003dec8) = 0
brk(0x804d000) = 0x804d000
pipe([4, 5]) = 0
clone() = 19063
write(5, “\0\0\0\0\5\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0”…, 148) = 148
rt_sigprocmask(SIG_SETMASK, NULL, [RT_0], 8) = 0
write(5, “\200>\n@\0\0\0\0\330\365\377\277\20R\3@\0\360\37\0\0\0”…, 148) = 148
rt_sigprocmask(SIG_SETMASK, NULL, [RT_0], 8) = 0
rt_sigsuspend([] <unfinished …>
— SIGRT_0 (Real-Time Signal 0) —
<… rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call)
sigreturn() = ? (mask now [])
— SIGSEGV (Speicherzugriffsfehler) —
rt_sigaction(SIGSEGV, {SIG_DFL}, {0x4001c570, [], SA_RESTART|0x4000000}, 8) = 0
write(2, "Fatal signal: “, 14Fatal signal: ) = 14
write(2, “Segmentation Fault”, 18Segmentation Fault) = 18
write(2, " (SDL Parachute Deployed)\n”, 26 (SDL Parachute Deployed)
) = 26
write(5, “\200>\n@\2\0\0\0\365\377\377\377\f\363\377\277\266\217”…, 148) = 148
rt_sigprocmask(SIG_SETMASK, NULL, [SEGV RT_0], 8) = 0
rt_sigsuspend([SEGV] <unfinished …>
— SIGRT_0 (Real-Time Signal 0) —
<… rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call)
sigreturn() = ? (mask now [SEGV])
wait4(19063, NULL, __WCLONE, NULL) = 19063
_exit(-11) = ?
stefan at pico:/temp/SDL-1.0.0/test >–
/E-Mail: @Stefan_Tomanek | ICQ: 1177934 | Ask for CA-signed PGP-Key/
/Spielen unter Linux: http://spiele.freepage.de/linux-zocker/ /
/“We are M$ of Borg. We will add your technology to our own crap. /
/Your company will be bought. Open Source is futile!” /

I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end
everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g). What can be the reason for that? Are
there any special configure-switches I should use?

This is a known bug in SuSE Linux 6.2 Get any updates, or upgrade to 6.3

-Sam Lantinga				(slouken at devolution.com)

Lead Programmer, Loki Entertainment Software–
“Any sufficiently advanced bug is indistinguishable from a feature”
– Rich Kulawiec

At 02:45 PM 12/12/99 -0800, you wrote:

I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end
everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g). What can be the reason for that? Are
there any special configure-switches I should use?

This is a known bug in SuSE Linux 6.2 Get any updates, or upgrade to 6.3

Where did you hear that? Im not trying to disprove you or anything, its
just that I was going to get suse 6.2 or 6.3 soon.

-Mongoose WPI student majoring in Computer Science

At 02:45 PM 12/12/99 -0800, you wrote:

I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end
everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g). What can be the reason for that? Are
there any special configure-switches I should use?

This is a known bug in SuSE Linux 6.2 Get any updates, or upgrade to 6.3

Where did you hear that? Im not trying to disprove you or anything, its
just that I was going to get suse 6.2 or 6.3 soon.

We found out the hard way with Myth II: Soulblighter.

-Sam Lantinga				(slouken at devolution.com)

Lead Programmer, Loki Entertainment Software–
“Any sufficiently advanced bug is indistinguishable from a feature”
– Rich Kulawiec

Stefan Tomanek wrote:

I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end
everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g).
hmm… i don’t really the reason, but i also experienced bad problems on
SuSE 6.2 (not SDL) and threads :

if i’ve ever used an dlopen(), pthread_create() crashes with SIGSEGV.
i had to rewrite my program so it links my plugin libs into the binary
instead of loading them with dlopen(), until i’ve found out what’s the
reason.
(curious: xmms also uses dlopen() to load the plugins and it also uses
pthreads).

perhaps there’s are problem with the libc. (perhaps different versions
get
linked together or something like that)

e.w.

Enrico Weigelt wrote:

I can?t use SDL 1.0 on my SuSE-Box. Everything compiles well, end
everything runs, except the testhread-Program and any other Program that
uses Multithreading (Xtheater e.g).
hmm… i don’t really the reason, but i also experienced bad problems on
SuSE 6.2 (not SDL) and threads :

if i’ve ever used an dlopen(), pthread_create() crashes with SIGSEGV.
i had to rewrite my program so it links my plugin libs into the binary
instead of loading them with dlopen(), until i’ve found out what’s the
reason.
(curious: xmms also uses dlopen() to load the plugins and it also uses
pthreads).

perhaps there’s are problem with the libc. (perhaps different versions
get
linked together or something like that)

We had this problem in Mozilla: the libdl stuff in glibc 2.0 is not
threadsafe. The only way to make it work is to get glibc 2.1. Sorry.
glibc 2.1 is a requirement for Mozilla, and even though it might work
with glibc 2.0 (I think it will link properly), it is crashing very
often.

Maybe that XMMS does the dlopen() and dlsym() before starting any
threads.

Even though I think the right solution is not to start threads at all in
the first place… :wink:

(look at the bottom of http://www3.sympatico.ca/pphaneuf/ for more info)–
Pierre Phaneuf
Systems Exorcist

Pierre Phaneuf wrote:

Even though I think the right solution is not to start threads at all in
the first place… :wink:

(look at the bottom of http://www3.sympatico.ca/pphaneuf/ for more info)

Before I forget that Windows ever existed, I think there is a thing
called “fibers” in Win32 (maybe only on NT) that actually are what most
thread users are actually looking for. The plan is this: you keep the
number of threads always less than or equal to the number of CPUs in the
system, then use fibers to do what you would usually do with threads.–
Pierre Phaneuf
Systems Exorcist