[gambit-list] Segmentation fault from simple (infinite) loop

Christian Jaeger christian at pflanze.mine.nu
Sun Feb 10 17:53:16 EST 2008


Bradley Lucier wrote:
> On Feb 10, 2008, at 4:53 PM, Christian Jaeger wrote:
> 
>> Note that I also never got a segfault with any variant when being  
>> compiled.

I only tried with v4.0.1 on PPC linux.

> 
> I got a segfault on 32-bit PPC MacOS with
> 
> ./configure 'CC=gcc -mcpu=7450 -g" --enaable-single-host
> 
> This is using gcc-4.2.1:

I can confirm that with 4.0 beta 21 on x86 I'm also getting the segfault 
with compiled code. (I'm now wondering why I've never run myself into 
this bug? Maybe I've never used println etc. in loops.)

It may be just luck, in one case I didn't get a segfault in the 
interpreter [for a very long time?] just because I was using the full 
filename (thus 4 chars longer) (with all .o* files removed, thus no 
change to which file is actually being loaded).

Christian.



More information about the Gambit-list mailing list