http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46166
Summary: optimization and/or removing an if(.false.) statement
leads to bad results
Product: gcc
Version: 4.6.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: fortran
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: sjbe...@comcast.net
Created attachment 22147
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22147
example program described in bug report
On Mac OS X 10.6.4, with gfortran from recent sources on the 4.6 trunk:
et:new.single.v2 sjbespa$ /volumes/dev/gfortran-4.6-trunk/bin/gfortran -v
Using built-in specs.
COLLECT_GCC=/volumes/dev/gfortran-4.6-trunk/bin/gfortran
COLLECT_LTO_WRAPPER=/Volumes/dev/gfortran-4.6-trunk/bin/../libexec/gcc/x86_64-apple-darwin10/4.6.0/lto-wrapper
Target: x86_64-apple-darwin10
Configured with: ../gfortran-4.6-src/configure
--prefix=/volumes/dev/gfortran-4.6-trunk --enable-languages=fortran
--host=x86_64-apple-darwin10 --build=x86_64-apple-darwin10
--target=x86_64-apple-darwin10
Thread model: posix
gcc version 4.6.0 20101013 (experimental) (GCC)
the following bug(s) have been observed:
(1) if the attached file is compiled as follows
sjbespa$ /volumes/dev/gfortran-4.6-trunk/bin/gfortran -O0 driver.f90
sjbespa$ ./a.out
and run, the program executes correctly. The output is shown in example 1.
(2) if the optimization level is changed to -O1 or higher:
sjbespa$ /volumes/dev/gfortran-4.6-trunk/bin/gfortran -O1 driver.f90
sjbespa$ ./a.out
the program doesn't run correctly as shown in example 2. It appears that the
variables fcurr and flast have invalid values.
(3) finally, if the if statement beginning at statement 251 is commented out,
the example program fails in the same manner as in (2) independent of the
optimization level.
**** example 1 ****
sjbespa$ /volumes/dev/gfortran-4.6-trunk/bin/gfortran -O0 driver.f90
sjbespa$ ./a.out
input data for constitutive model solver
rhosat = 40743665431535.492
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
rho 1000000.0000000000
plStrainInv 0.0000000000000000
plStrainInvLast 0.0000000000000000
plStrainInvRate 0.0000000000000000
dlamba 0.0000000000000000
fcurr -9999999.9999000002
flast -9999999.9999000002
215.99849824095310 0.0000000000000000 6.95322298958717986E-310
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
rho 1000000.0000000000
plStrainInv 0.0000000000000000
plStrainInvLast 0.0000000000000000
plStrainInvRate 0.0000000000000000
dlamba 0.0000000000000000
fcurr -9999999.9999000002
flast -9999999.9999000002
863.99399296381239 0.0000000000000000 6.95322298958717986E-310
**** example 2 *****
sjbespa$ /volumes/dev/gfortran-4.6-trunk/bin/gfortran -O1 driver.f90
sjbespa$ ./a.out
input data for constitutive model solver
rhosat = 40743665431535.492
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
rho 1000000.0000000000
plStrainInv 0.0000000000000000
plStrainInvLast 0.0000000000000000
plStrainInvRate 0.0000000000000000
dlamba 0.0000000000000000
fcurr -9999999.9999000002
flast -9999999.9999000002
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
rho 1005087.5000000000
plStrainInv 9.99999999999999955E-007
plStrainInvLast 0.0000000000000000
plStrainInvRate 9.99999999999999955E-007
dlamba 9.99999999999999955E-007
fcurr NaN
flast NaN
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
rho NaN
plStrainInv NaN
plStrainInvLast 0.0000000000000000
plStrainInvRate NaN
dlamba NaN
fcurr NaN
flast NaN
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
