Sorry - I meant to add as a test case.... New code on 64-bit bit
JVERSION
Engine: j701/2011-02-23/15:25
Library: 7.01.045
Platform: Win 64
Installer: unknown
InstallPath: c:/users/joe bogner/downloads/jsrc/build
?. 16 # 2 2 2 16
0 0 0 1 0 0 0 0 1 1 0 1 1 1 0 1 0 1 1 1 1 1 1 0 0 1 0 0 1 1 0 0 1 1 1 0 1 0 0 0
1 1 1 1 0 1 0 0 4 14 3 1 6 14 10 15 13 5 6 2 14 0 5 14
JVERSION
Engine: j803/2014-10-19-11:11:11
Library: 8.03.13
Platform: Win 32
Installer: J803 install
InstallPath: c:/users/joe bogner/downloads/j803_win32
?. 16 # 2 2 2 16
1 0 0 1 0 1 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 1 1 1 1 1 1 0
1 1 1 0 0 1 1 1 13 12 13 4 9 15 5 9 6 3 9 1 2 11 1 8
JVERSION
Engine: j803/2014-10-19-11:11:11
Library: 8.04.13
Qt IDE: 1.4.6/5.4.2
Platform: Win 64
Installer: J804 install
InstallPath: c:/users/joe bogner/j64-804
?. 16 # 2 2 2 16
0 1 0 1 1 0 0 1 1 1 0 0 0 0 1 0 0 1 0 1 1 0 0 0 0 0 1 0 0 1 0 1 0 0 0
0 0 0 1 0 1 1 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 1 1 0 1 1
On Thu, Dec 17, 2015 at 10:56 PM, Joe Bogner <[email protected]> wrote:
> looks like the problem likes in
> https://github.com/openj/core/blob/18fd23bbdc2f50770eb3047e978cd5e4e3b47039/vrand.c#L726
>
> It looks like it checks to see if the first number in the list is 2
> and if so, it uses the roll2 implementation
>
> RZ(w=vi(w)); m=*AV(w);
> if( 2==m)RZ(z=roll2 (w,&b));
>
> It also appears to be a 64-bit bug
>
> 32-bit:
>
> ?. 16 # 2 2 2 16
> 1 0 0 1 0 1 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 1 1
> 1 1 1 1 0 1 1 1 0 0 1 1 1 13 12 13 4 9 15 5 9 6 3 9 1 2 11 1 8
>
> 64-bit
> ?. 16 # 2 2 2 16
> 0 1 0 1 1 0 0 1 1 1 0 0 0 0 1 0 0 1 0 1 1 0 0 0 0 0 1 0 0 1 0 1 0 0 0
> 0 0 0 1 0 1 1 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 1 1 0 1 1
>
> The simplest 'fix' would be to disable the roll 2 optimization by
> commenting it out, but that may not be ideal since presumably it was
> put in place for a reason to speed up random booleans on large arrays.
> The next fix would be to check to see if all numbers in the ravel are
> 2 and only then use roll2. The last, probably more complicated fix
> would be to support 2 and other numbers in roll2 safely in 64-bit, but
> even that is pretty simple
>
>
> just copy and paste the block of shifts (since there is only 8), to
> add 8 more for 64-bit... need to do this for the random and fixed seed
> part
>
> }else for(j=0;j<q;++j){
> t=NEXT;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t;
>
> //start new code
> #if SY_64
> t>>=1;
> if(!(2==*v++&&2==*v++&&2==*v++&&2==*v++))R mark; *zv++=mk&t; t>>=1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark;
> *zv++ = mk&t; t >>= 1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark;
> *zv++ = mk&t; t >>= 1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark;
> *zv++ = mk&t; t >>= 1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark;
> *zv++ = mk&t; t >>= 1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark;
> *zv++ = mk&t; t >>= 1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark;
> *zv++ = mk&t; t >>= 1;
> if (!(2 == *v++ && 2 == *v++ && 2 == *v++ && 2 == *v++))R mark; *zv++ =
> mk&t;
> #endif
> }
>
> On Thu, Dec 17, 2015 at 10:29 PM, Linda A Alvord
> <[email protected]> wrote:
>> These are the results you should get in a fresh version of JHS or jqt. If
>> you use ?. you will insure repeated exact results in a session. In all cases
>> the length of the list is 64 elements so there should be enough 2's.
>>
>> ?. 16 # 2 2 2 16
>> 1 0 0 1 0 1 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 1 1 1 1 1
>> 1 0 1 1 1 0 0 1 1 1 13 12 13 4 9 15 5 9 6 3 9 1 2 11 1 8
>> ?. 16 # 3 2 2 16
>> 0 2 2 1 2 0 2 1 1 2 0 2 1 1 1 2 0 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 1 1 1
>> 1 1 1 0 1 1 1 0 0 1 13 11 13 12 13 4 9 15 5 9 6 3 9 1 2 11
>> ?. 16 # 16 2 2 2
>> 2 11 15 4 6 7 4 1 4 2 6 14 1 14 7 12 0 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 1
>> 1 1 1 1 1 0 1 1 1 0 0 1 1 1 1 0 1 0 1 1 1 1 0 1 1 1 0 1
>>
>> $?. 16 # 2 2 2 16
>> 64
>> $?. 16 # 3 2 2 16
>> 64
>> $?. 16 # 16 2 2 2
>> 64
>>
>> Linda
>>
>> -----Original Message-----
>> From: [email protected]
>> [mailto:[email protected]] On Behalf Of 'Pascal
>> Jasmin' via Programming
>> Sent: Thursday, December 17, 2015 12:41 PM
>> To: [email protected]
>> Subject: Re: [Jprogramming] bug in ?
>>
>> ?. 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
>> 2 2 2 2 2 2 2 2 2 2 2 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16
>> 0 0 0 0 1 1 0 0 0 1 1 1 0 0 0 0 1 0 0 0 0 1 1 1 1 1 1 0 1 1 1 0 1 0 1 1 1 1
>> 0 1 1 1 1 0 0 1 1 1 8 9 6 8 14 7 12 9 1 15 2 15 3 13 3
>>
>> ?. 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
>> 2 2 2 2 2 2 2 2 2 2 2 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16
>> 0 1 0 1 1 0 0 1 1 1 0 0 0 0 1 0 0 1 0 1 1 0 0 0 0 0 1 0 0 1 0 1 0 0 0 0 0 0
>> 1 0 1 1 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 1 1 0 1 1
>>
>>
>> one less leading 2 in first example. 47 leading 2s, correct result. 48
>> leading 2s and everything is 2.
>>
>>
>> ----- Original Message -----
>> From: Raul Miller <[email protected]>
>> To: Programming forum <[email protected]>
>> Sent: Thursday, December 17, 2015 12:28 PM
>> Subject: Re: [Jprogramming] bug in ?
>>
>> Consider
>> ?.2 2 2 32
>> 0 1 0 18
>>
>> Thanks,
>>
>> --
>> Raul
>>
>>
>> On Thu, Dec 17, 2015 at 12:13 PM, 'Pascal Jasmin' via Programming
>> <[email protected]> wrote:
>>>
>>>
>>>
>>> ?. 16 # 2 2 2 16
>>>
>>>
>>> result is same as
>>>
>>> ?. 16 # 2 2 2 2
>>>
>>> ignores the request for 0-15 range random numbers.
>>>
>>>
>>> ----- Original Message -----
>>> From: Raul Miller <[email protected]>
>>> To: Programming forum <[email protected]>
>>> Sent: Thursday, December 17, 2015 12:10 PM
>>> Subject: Re: [Jprogramming] bug in ?
>>>
>>> On Thu, Dec 17, 2015 at 11:49 AM, 'Pascal Jasmin' via Programming
>>> <[email protected]> wrote:
>>>> ?. 16 # 2 2 2 16
>>>> 0 1 0 1 1 0 0 1 1 1 0 0 0 0 1 0 0 1 0 1 1 0 0 0 0 0 1 0 0 1 0 1 0 0 0 0 0
>> 0 1 0 1 1 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 1 1 0 1 1
>>>>
>>>>
>>>> ?. 16 # 3 2 2 16
>>>> 0 2 1 0 2 1 1 1 2 1 0 2 1 2 1 2 1 0 1 1 1 1 1 1 0 0 1 0 0 1 1 0 0 1 1 1 0
>> 1 0 0 0 1 1 1 1 0 1 0 8 4 14 3 1 6 14 10 15 13 5 6 2 14 0 5
>>>>
>>>> The last 16 numbers in the first example should match the 2nd.
>>>
>>> Why do you say that?
>>>
>>> Thanks,
>>>
>>> --
>>> Raul
>>>
>>>
>>>
>>> ----------------------------------------------------------------------
>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>>
>>
>>> ----------------------------------------------------------------------
>>> For information about J forums see http://www.jsoftware.com/forums.htm
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
