I wrote:
> I have now upgraded to 5.02.2, which gives me an entirely different
> show-stopper bug when using -prof -auto-all:
>
> ---begin compilation output
> [...output of succesfully compiling a bunch of library modules]
> ghc -c -package lang -package concurrent
> -i"c:\\BareBonesArrows\\arrowslib" -prof -auto-all -o TestST.o TestST.hs
> Prologue junk?: .globl _TestST_loopIntegralzut1r_entry
> _TestST_loopIntegralzut1r_entry:
> movl $4012,%eax
> call __alloca
> /APP
>
> make: *** [TestST.o] Error 1
>
> ---end compilation output
>
> [...]
>
> My platform is Windows 2000. I'll try it on Linux and see how it goes.
I tried this under Linux (debian, using ghc 5.02.2). It compiled OK,
but produced an executable that dumped core immediately on startup. :-(
For what it's worth, here is as much info. as I could glean from gdb:
Starting program: /home/frp/afrp/src/testST
Program received signal SIGSEGV, Segmentation fault.
0x80e6673 in EnterFunCCS ()
(gdb) where
#0 0x80e6673 in EnterFunCCS ()
#1 0x1 in ?? ()
#2 0x80f1d24 in stg_upd_frame_info ()
#3 0x811f220 in CCS_DONT_CARE ()
#4 0x2 in ?? ()
Cannot access memory at address 0x900065ff.
Has profiling support been sucessfully tested in ghc 5.02.2?
-antony
--
Antony Courtney
Grad. Student, Dept. of Computer Science, Yale University
[EMAIL PROTECTED] http://www.apocalypse.org/pub/u/antony
_______________________________________________
Glasgow-haskell-bugs mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
