Okay, I have a test case for the problem I reported before that I've
attached.
We have two pairs of identical machines:
- 2 Tyan S2882 Dual Processor 244 stepping 10
- 2 Tyan S2882-D Dual processor dual core Opteron 275 stepping 2
The attached code when compiled with the Portland Group Fortran compiler
with -O2 and run on either of the 244's will abort in random locations:
[EMAIL PROTECTED] rams.debug]$ pgf95 -O2 -o testatob testatob.f90
[EMAIL PROTECTED] rams.debug]$ ./testatob
checkatob abort n= 246500 , i= 4685 a(i)= 8712085.
b(i)= 8465585.
Abort
[EMAIL PROTECTED] rams.debug]$ ./testatob
checkatob abort n= 246500 , i= 145817 a(i)= 9592717.
b(i)= 8853217.
Abort
[EMAIL PROTECTED] rams.debug]$ time ./testatob
checkatob abort n= 246500 , i= 118169 a(i)= 9565069.
b(i)= 8825569.
Aborted
real 0m31.842s
user 0m16.476s
sys 0m0.060s
Haven't seen it run longer than 1 minute yet.
However, it runs fine on the 275's (or at least I haven't seen it crash
yet). It also runs fine on the 244's when compiled with -O1.
So, I guess this points to a hardware issue, but it may be a somewhat
generalized hardware issue. I'd love to hear reports on other
(particularly other Tyan S2882 dual 244's) systems.
--
Orion Poplawski
Technical Manager 303-415-9701 x222
NWRA/CoRA Division FAX: 303-415-9702
3380 Mitchell Lane [EMAIL PROTECTED]
Boulder, CO 80301 http://www.cora.nwra.com
!
!
SUBROUTINE ATOB(N,A,B)
DIMENSION A(*),B(*)
DO 100 I=1,N
B(I)=A(I)
100 CONTINUE
RETURN
END
!
! ******************************************************************
!
!
SUBROUTINE CHECKATOB(N,A,B)
DIMENSION A(*),B(*)
DO 100 I=1,N
IF (B(I).NE.A(I)) then
write(6,*),'checkatob abort n=',n,', i=',i,' a(i)=',a(i),' b(i)=',b(i)
call abort
end if
100 CONTINUE
RETURN
END
!
! ******************************************************************
program test
real, allocatable :: a(:)
allocate(a(11665401))
do i=1,11665401
a(i)=i
end do
do
call atob(246500,a(8460901),a(1))
call checkatob(246500,a(8460901),a(1))
call atob(246500,a(8707401),a(1))
call checkatob(246500,a(8707401),a(1))
call atob(246500,a(9446901),a(1))
call checkatob(246500,a(9446901),a(1))
end do
end
_______________________________________________
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
