#10 Judy-am "make check" runs 8+hours

closed-fixed
nobody
None
5
2004-06-06
2003-06-22
Allen Smith
No

On an IRIX 6.5 system, 2 270 MHZ IP30 R12000
processors, 2560 Mbytes of RAM, with admittedly one
other (group of) processes (apache webserver) running,
a "make check" on Judy-am (after an apparently
successful compilation) gets to the Checkit script
running Judy1LCheck -n 300000 -B20... and sits there.
It's possible that it might have eventually completed,
but after I came back after _8 hours_ and it hadn't,
I concluded it had gone into a loop or otherwise was
never going to complete (no corefile, even when
manually interrupted, BTW). Compliation flags:

--prefix=/server/people/staff/easmith
--libdir=/server/people/staff/easmith/lib32
--disable-shared
--datadir=/server/people/staff/easmith/lib
--sysconfdir=/server/people/staff/easmith/lib

Environment variables of interest:

CC: cc -mips4 -n32 -L/server/people/staff/easmith/lib32
-L/usr/lib32/mips4 -L/usr/lib32 -L/lib32
-Wl,-rpath,/server/people/staff/easmith/lib32:/usr/lib32/mips4:/usr/lib32:/lib32
-Wl,-multigot -woff 1110

CFLAGS: -I${HOME}/include -O3
-OPT:fast_sqrt=OFF:Olimit=0 -TENV:X=1
-LNO:opt=1:ou_further=3
-OPT:fold_unsafe_relops=OFF:div_split=OFF:IEEE_arithmetic=1:roundoff=0
-LNO:vintr=OFF

Discussion

    • status: open --> closed-fixed
     
  • Logged In: YES
    user_id=558029

    Fixed in version 0.1.7 available at <http://judy.sourceforge/downloads>
    Problem was not reproducable, nor ever complained about again.
    If occurs again in above or later version please re-submit.