Phil Carmody on Wed, 10 Dec 2003 02:52:12 +0100


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

-fPIC joy, lexcmp woes.


After much confusion, it appears that without the patch it alwas fails to
build, and with the patch it builds cleanly, so -fPIC was the cure. However...


*** ../src/test/64/objets       Tue Jun 24 13:42:05 2003
--- gp.out      Tue Dec  9 19:35:45 2003
***************
*** 54,60 ****
  ? imag(2+3*I)
  3
  ? lex([1,3],[1,3,5])
! -1
  ? max(2,3)
  3
  ? min(2,3)
--- 54,60 ----
  ? imag(2+3*I)
  3
  ? lex([1,3],[1,3,5])
! 4294967295
  ? max(2,3)
  3
  ? min(2,3)
***************


I looked at basemath/gen2.c, however, I found myself a tad confused.


There's no way (unless the compiler's completely b0rked) that 0xffffffff can be
returned from any of the lexcmp functions. So maybe it's in the client of that
function that the error occurs. Alas without getting interactive access to the
machine, I can't muck around and find out.

Maybe someone else had access to an Athlon64/gcc/linux combo and can try to
reproduce it?


Also, while playing around locally, I noticed this:
<<<
(03:07) gp > lex([1,3,5],[1,3])
-1
(03:11) gp > lex([1,3],[1,3,5])
-1
>>>

!?!? Surely that can't be right?
(that's CVS nightly build on the Alpha).


Phil

=====
When inserting a CD, hold down shift to stop the AutoRun feature
In the Device Manager, disable the SbcpHid device.
http://www.cs.princeton.edu/~jhalderm/cd3/

__________________________________
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing.
http://photos.yahoo.com/