Robert:
Yes I am still alive and still working on Judy -- believe it or not.
I would be delighted to answer very technical questions about Judy.
Also hard to believe is Judy was released "-Wall" free. Compilers have
improved a lot.
64 bit machines and compilers were rare back in 2002. Obviously, the Judy
code wasnot tested in 64 bit with "-DSEARCH_LINEAR" defined. It was released
with with "-DSEARCH_BINARY" which means the bug at line 523 would of not be
caught.
Instead of "JU_COPY3_PINDEX_TO_LONG(i_ndex, P_leaf);" it should of
been"COPYINDEX(i_ndex, P_leaf);" on line 523.
It works in 32 bit because the only "NONNAT" Leaf is 3, whereas 64bit has
3,5,6,7.
If this does not work, feel free to email me and I will take the time to try it
out.
BTW, We have much faster Leaf search routines now. In 2002 the LINEAR
andBINARY searches were about the same speed. I expected the BINARY tobe
faster in the future. That is why it was released that way. However, in 15
yearsof testing, we have discovered a lot of things on how computers really
work.
Thank you for interest.
doug
Doug Baskins <[email protected]>
On Thursday, December 28, 2017, 4:40:45 PM GMT+7, Robert Gregory
<[email protected]> wrote:
<!--#yiv7431109212 _filtered #yiv7431109212 {font-family:"Cambria
Math";panose-1:2 4 5 3 5 4 6 3 2 4;} _filtered #yiv7431109212
{font-family:Calibri;panose-1:2 15 5 2 2 2 4 3 2 4;}#yiv7431109212
#yiv7431109212 p.yiv7431109212MsoNormal, #yiv7431109212
li.yiv7431109212MsoNormal, #yiv7431109212 div.yiv7431109212MsoNormal
{margin:0in;margin-bottom:.0001pt;font-size:11.0pt;font-family:"Calibri",
sans-serif;}#yiv7431109212 p.yiv7431109212MsoListParagraph, #yiv7431109212
li.yiv7431109212MsoListParagraph, #yiv7431109212
div.yiv7431109212MsoListParagraph
{margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:.5in;margin-bottom:.0001pt;font-size:11.0pt;font-family:"Calibri",
sans-serif;}#yiv7431109212 .yiv7431109212MsoChpDefault {} _filtered
#yiv7431109212 {margin:1.0in 1.0in 1.0in 1.0in;}#yiv7431109212
div.yiv7431109212WordSection1 {}#yiv7431109212 _filtered #yiv7431109212 {}
_filtered #yiv7431109212 {} _filtered #yiv7431109212 {} _filtered
#yiv7431109212 {} _filtered #yiv7431109212 {} _filtered #yiv7431109212 {}
_filtered #yiv7431109212 {} _filtered #yiv7431109212 {} _filtered
#yiv7431109212 {} _filtered #yiv7431109212 {}#yiv7431109212 ol
{margin-bottom:0in;}#yiv7431109212 ul {margin-bottom:0in;}-->
Two Questions –
- Are you still alive?
- Would you mind answering a highly targeted question regarding your 64-bit
implementation of Judy arrays?
If the answer to question 1 is “no”, no reply is expected and question 2 is
moot. If the answer to question 1 is “yes,” a reply to question #2 would be
appreciated. If the reply to question 2 is “no,” have a happy new year (albeit
free from further cheerful e-mail interruptions by me).
[My question deals with an issue in JudyPrivate.h not directly addressed by the
Wiki page version of the file. Using VC2013 (or Pelles C), lines 1586, 1593,
and 1600 reference defined code which raises warning 4293 (shift too large) due
to the fact that *something* which should be a Word_T (unsigned long long) is
being treated instead as an unsigned long. I’ve stared at this for a while and
don’t see anything obvious… but I may just be tired. In any event, everything
is fine with 32 bit… and the first insertion works on 64 bit; however,
subsequent insertions appear to work but indeed do not.]
-Rob Gregory
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Judy-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/judy-devel
