[Tcl-bugs] [ tcl-Bugs-2798543 ] bogus bignum exponent calculations

[SourceForge.net <no...@so...>]

Bugs item #2798543, was opened at 2009-05-29 11:53
Message generated for change (Comment added) made by andreas_kupries
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2798543&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 47. Bytecode Compiler
Group: current: 8.5.7
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Don Porter (dgp)
Assigned to: Don Porter (dgp)
Summary: bogus bignum exponent calculations

Initial Comment:

% string length [expr {10**2**15}]
32769
% string length [expr {10**2**16}]
17
% expr {10**2**16}
21936950640377856


----------------------------------------------------------------------

>Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-06-01 08:39

Message:
Ok. The optimized table lookup is Kevin's work, IIRC.

----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2009-05-30 08:01

Message:

Neither.  All appearances say it's
in TEBC's own "optimized table lookup"
implementation.

----------------------------------------------------------------------

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-05-29 14:54

Message:
Curiosity: Is that ugliness all in Tcl's use of libtommath, or in
libtommath itself ?


----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2009-05-29 14:36

Message:

More ugliness:

% expr {10**65553 == 10**17}
1


----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2009-05-29 12:35

Message:

Something wrong computing the value
of "base" at line 5963?  Computation is
done in a short int, but what happens
when l2 is so big that we're overflowing
that?


----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2009-05-29 12:25

Message:

Simpler demo:

% expr {10**65536}
21936950640377856

Something in the logic that chooses
among the many (!!) implementations of
exponent computation is borked.



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2798543&group_id=10894