Apologize to re-opening this old thread. I tried Joe's patch on the current J source and it seems working. But it also crashed when running g640 in the test suite. I can get a simple repeatable expression that caused crash.
?@:(5e3&$)"0 ] 2 but 1e3 or 2e3 is ok. Any idea? Чт, 17 дек 2015, Joe Bogner написал(а): > 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 -- regards, ==================================================== GPG key 1024D/4434BAB3 2008-08-24 gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 gpg --keyserver subkeys.pgp.net --armor --export 4434BAB3 ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
