[cl-opengl-devel] division-by-zero error
(argh, forgot to cc the list!) sorry, should have included more details in the first mail. I'm running debian stable, using the closed-source nvidia drivers and built sbcl from source myself. ok well, using (sb-int:with-float-traps-masked (:invalid :divide-by-zero)) as suggested by Luís allows me to load cl-glut-examples, but the issue still occurs if i try to run an example (i.e. rb-hello) backtrace from there is = (SB-VM:SIGFPE-HANDLER #<unavailable argument> #.(SB-SYS:INT-SAP #XB79FB36C) #<unavailable argument>) 0] backtrace 0: (SB-VM:SIGFPE-HANDLER #<unavailable argument> #.(SB-SYS:INT-SAP #XB79FB36C) #<unavailable argument>) 1: (SB-VM:SIGFPE-HANDLER #<unavailable argument> #.(SB-SYS:INT-SAP #XB79FB36C) #<unavailable argument>)[:EXTERNAL] 2: ((FLET #:WITHOUT-INTERRUPTS-BODY-[INVOKE-INTERRUPTION]11)) 3: (SB-SYS:INVOKE-INTERRUPTION #<CLOSURE (FLET SB-UNIX::INTERRUPTION) {B79FB105}>) 4: ((FLET SB-UNIX::RUN-HANDLER) 8 #.(SB-SYS:INT-SAP #XB79FB36C) #.(SB-SYS:INT-SAP #XB79FB3EC)) 5: ("foreign function: call_into_lisp") 6: ("foreign function: funcall3") 7: ("foreign function: interrupt_handle_now") On Thu, Mar 6, 2008 at 2:30 PM, Bart Botta <00003b@gmail.com> wrote:
(oops, forgot to reply to the list, sorry about the extra copies)
Just as a guess, you might try deleting .fasl files in cl-opengl and cffi directories, if that doesn't help, will need more information:
On 3/5/08, James Baker <cycle.code.media@gmail.com> wrote:
"division-by-zero" issue, I'm using both cffi + cl-opengl from darcs and sbcl 1.0.15.
What OS/platform/graphics card drivers?
(SB-VM:SIGFPE-HANDLER #<unavailable argument> #.(SB-SYS:INT-SAP #XB79FA9EC) #<unavailable argument>) 0]
What does it say if you type backtrace at that prompt?
-- -b-
On 3/5/08, James Baker <cycle.code.media@gmail.com> wrote:
ok well, using (sb-int:with-float-traps-masked (:invalid :divide-by-zero)) as suggested by Luís allows me to load cl-glut-examples, but the issue still occurs if i try to run an example (i.e. rb-hello)
well, you could try adding the with-float-traps-masked when you call rb-hello as well, but seems like it would be better to try to figure out what is breaking first...
backtrace from there is =
'from there' = from calling rb-hello? Do you get the same backtrace when loading cl-glut-examples without the with-float-traps-masked?
7: ("foreign function: interrupt_handle_now")
hmm, not much useful there. It didn't go any farther than that? -- -b-
Sorry perhaps that wasnt very clear on my part. The backtrace was from calling rb-hello, and yes thats all there was up to 7:("foreign function:...") James On Thu, Mar 6, 2008 at 3:03 PM, Bart Botta <00003b@gmail.com> wrote:
On 3/5/08, James Baker <cycle.code.media@gmail.com> wrote:
ok well, using (sb-int:with-float-traps-masked (:invalid :divide-by-zero)) as suggested by Luís allows me to load cl-glut-examples, but the issue still occurs if i try to run an example (i.e. rb-hello)
well, you could try adding the with-float-traps-masked when you call rb-hello as well, but seems like it would be better to try to figure out what is breaking first...
backtrace from there is =
'from there' = from calling rb-hello? Do you get the same backtrace when loading cl-glut-examples without the with-float-traps-masked?
7: ("foreign function: interrupt_handle_now")
hmm, not much useful there. It didn't go any farther than that?
-- -b-
"James Baker" <cycle.code.media@gmail.com> writes:
ok well, using (sb-int:with-float-traps-masked (:invalid :divide-by-zero)) as suggested by Luís allows me to load cl-glut-examples, but the issue still occurs if i try to run an example (i.e. rb-hello)
Did you run the example using with-float-traps-masked as well? -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/
I just checked then and calling an example using with-float-traps-masked works, thanks very much. James On Thu, Mar 6, 2008 at 4:28 PM, Luis Oliveira <luismbo@gmail.com> wrote:
"James Baker" <cycle.code.media@gmail.com> writes:
ok well, using (sb-int:with-float-traps-masked (:invalid :divide-by-zero)) as suggested by Luís allows me to load cl-glut-examples, but the issue still occurs if i try to run an example (i.e. rb-hello)
Did you run the example using with-float-traps-masked as well?
--
Luís Oliveira http://student.dei.uc.pt/~lmoliv/
_______________________________________________ cl-opengl-devel mailing list cl-opengl-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/cl-opengl-devel
"James Baker" <cycle.code.media@gmail.com> writes:
I just checked then and calling an example using with-float-traps-masked works, thanks very much.
Maybe we should use this in cl-glut then. I'm guessing you're running on x86-64, is that right? -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/
I'm running 32bit debian on a x86-64 James On Thu, Mar 6, 2008 at 4:48 PM, Luis Oliveira <luismbo@gmail.com> wrote:
"James Baker" <cycle.code.media@gmail.com> writes:
I just checked then and calling an example using with-float-traps-masked works, thanks very much.
Maybe we should use this in cl-glut then. I'm guessing you're running on x86-64, is that right?
--
Luís Oliveira http://student.dei.uc.pt/~lmoliv/
_______________________________________________ cl-opengl-devel mailing list cl-opengl-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/cl-opengl-devel
"James Baker" <cycle.code.media@gmail.com> writes:
I'm running 32bit debian on a x86-64
Please let me know if the attached patch helps. -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/
Please let me know if the attached patch helps.
ok I applied the patch, cleaned out all the fasl's and did a (asdf:oos 'asdf:load-op 'cl-glut-examples) and no complaints. Running any of the examples without (sb-int:with-float-traps-masked (:invalid :divide-by-zero) (....)) still causes the divide-by-zero error. Thanks again, James
"James Baker" <cycle.code.media@gmail.com> writes:
ok I applied the patch, [...] still causes the divide-by-zero error.
I had a chance to try things out in my linux/x86-64 box today and pushed a patch that should work this time. -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/
participants (3)
-
Bart Botta -
James Baker -
Luis Oliveira