This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

-fdefault-integer-8 again!-(


Hi everybody!

I have had the bad idea to regtest gfortran at revision 142276 on
powerpc-apple-darwin9 with -fdefault-integer-8. The summary is:

# of expected passes		106767
# of unexpected failures	967
# of expected failures		168
# of unsupported tests		2848

(-m32 and -m64). The most annoying problem I have seen so far is
an ICE for gfortran.dg/alloc_comp_constructor_1.f90 (pr34143,
just closed as fixed by Paul Thomas):

FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O0  execution test
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O0  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O1  execution test
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O1  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O2  (internal compiler error)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O2  (test for excess errors)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O2  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer  (internal compiler error)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer  (test for excess errors)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer -funroll-loops  (internal compiler error)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer -funroll-loops  (test for excess errors)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer -funroll-loops  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer -funroll-all-loops -finline-functions  (internal compiler error)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer -funroll-all-loops -finline-functions  (test for excess errors)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -fomit-frame-pointer -funroll-all-loops -finline-functions  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -g  (internal compiler error)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -g  (test for excess errors)
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -O3 -g  scan-tree-dump-times original "builtin_free" 21
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -Os  execution test
FAIL: gfortran.dg/alloc_comp_constructor_1.f90  -Os  scan-tree-dump-times original "builtin_free" 21

The ICE occurs at -O2 and above and is

[karma] f90/bug% gfc -O2 -fdefault-integer-8 /opt/gcc/_gcc_clean/gcc/testsuite/gfortran.dg/alloc_comp_constructor_1.f90
/opt/gcc/_gcc_clean/gcc/testsuite/gfortran.dg/alloc_comp_constructor_1.f90: In function 'test_constructor':
/opt/gcc/_gcc_clean/gcc/testsuite/gfortran.dg/alloc_comp_constructor_1.f90:9: internal compiler error: in operand_subword, at emit-rtl.c:1364

Should I reopen pr34143 or open a new pr?

The test gfortran.fortran-torture/execute/equiv_5.f is also timed out:
the changes done by Thomas Koenig at revision 130279 do not
take into account dummy arguments passed as constants (for instance
INTEGER*4 I should have stayed INTEGER I). A similar problem
(without timing problem) is for gfortran.dg/no_range_check_1.f90
where the constants should be typed, and for gfortran.dg/nearest_1.f90
where we should have integer(4) ix, iy.

What is the best way to proceed knowing that I'll have very little
time in the coming week?

Cheers

Dominique

PS I have posted in pr27766#10 two obvious fixes for -fbounds-check.
Are they OK?


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]