Just a quick remark about "bc". It give a segmentation fault evertime it's envoked with "bc -l" and a division is made. Example: $ echo "2/3" | bc -l Segmentation fault $ Could de problem be related with the -O2 optimization ? -- Marco Ferra
Hi, On Sat, Jun 03, 2006 at 17:56:02 +0200, Marco Ferra wrote:
Just a quick remark about "bc". It give a segmentation fault evertime it's envoked with "bc -l" and a division is made. Example:
$ echo "2/3" | bc -l Segmentation fault $
Could de problem be related with the -O2 optimization ? Hardly.
Would you mind telling us which version of CRUX, bc and the bc port you're using? It's kinda hard to diagnose otherwise. Johannes -- Johannes Winkelmann mailto:jw@smts.ch Zurich, Switzerland http://jw.smts.ch
On 6/3/06, Johannes Winkelmann <jw@smts.ch> wrote:
$ echo "2/3" | bc -l Segmentation fault $
Could de problem be related with the -O2 optimization ? Hardly.
Would you mind telling us which version of CRUX, bc and the bc port you're using? It's kinda hard to diagnose otherwise.
I have just installed the 2.2 CRUX release from the CD (i686), only the core packages. bc version is 1.06. mferra
On 6/3/06, Johannes Winkelmann <jw@smts.ch> wrote:
$ echo "2/3" | bc -l Segmentation fault $
Could de problem be related with the -O2 optimization ? Hardly.
Would you mind telling us which version of CRUX, bc and the bc port you're using? It's kinda hard to diagnose otherwise.
I have just installed the 2.2 CRUX release from the CD (i686), only the core packages. bc version is 1.06. Oh, okay; The bc version on the CD is known to have problems when called in --mathlib mode. You can update to 1.06-3 from the ports tree, and
Hi Marco, On Sat, Jun 03, 2006 at 18:29:19 +0200, Marco Ferra wrote: those issues should go away. It seems we missed that on the release notes page, I'll add that there. HTH, Johannes -- Johannes Winkelmann mailto:jw@smts.ch Zurich, Switzerland http://jw.smts.ch
On 6/3/06, Johannes Winkelmann <jw@smts.ch> wrote: [...]
Oh, okay; The bc version on the CD is known to have problems when called in --mathlib mode. You can update to 1.06-3 from the ports tree, and those issues should go away.
It seems we missed that on the release notes page, I'll add that there.
Thanks for pointing me out the solution, no problems with the optimizations then. mferra
participants (2)
-
Johannes Winkelmann -
Marco Ferra