[Patch, Fortran] PR 88009: [9 Regression] ICE in find_intrinsic_vtab, at fortran/class.c:2761

[Steve Kargl]

On Sat, Jan 05, 2019 at 11:39:08AM -0800, Steve Kargl wrote:
> On Sat, Jan 05, 2019 at 03:35:21PM +0100, Janus Weil wrote:
> > Am Sa., 5. Jan. 2019 um 14:37 Uhr schrieb Thomas Koenig <tkoenig@netcologne.de>:
> > >
> > > Hi Janus,
> > >
> > > > the attached patch fixes PR 88009, an ICE-on-invalid regression caused
> > > > by one of my earlier commits. Apart from adding some extra checks to
> > > > avoid ICEs, it also uses the 'artificial' attribute to suppress bogus
> > > > errors (see comment #3) and does some minor cleanup in
> > > > resolve_fl_variable.
> > > >
> > > > Regtests cleanly on x86_64-linux-gnu. Ok for trunk?
> > >
> > >
> > > the patch looks good, OK for trunk.
> > 
> > thanks for the quick review, Thomas! Committed as r267598.
> > 
> 
> I still need to verify, but it appears that this commit 
> is causing a regression on x86_64-*-freebsd.
> 
> FAIL: gfortran.dg/class_alias.f90   -Os  execution test
> 
> Reverting parts of commits and rebuilding and testing takes
> about an hour or more, verification won't be available for
> bit.
> 

More info from build log.


Executing on host: /safe/sgk/gcc/objx/gcc/testsuite/gfortran/../../gfortran -B/safe/sgk/gcc/objx/gcc/testsuite/gfortran/../../ -B/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/ /safe/sgk/gcc/gccx/gcc/testsuite/gfortran.dg/class_alias.f90    -fno-diagnostics-show-caret -fno-diagnostics-show-line-numbers -fdiagnostics-color=never    -Os  -fdump-tree-original  -B/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libatomic/.libs -B/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs  -lm  -o ./class_alias.exe    (timeout = 300)
spawn -ignore SIGHUP /safe/sgk/gcc/objx/gcc/testsuite/gfortran/../../gfortran -B/safe/sgk/gcc/objx/gcc/testsuite/gfortran/../../ -B/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/ /safe/sgk/gcc/gccx/gcc/testsuite/gfortran.dg/class_alias.f90 -fno-diagnostics-show-caret -fno-diagnostics-show-line-numbers -fdiagnostics-color=never -Os -fdump-tree-original -B/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libatomic/.libs -B/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs -L/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs -lm -o ./class_alias.exe
PASS: gfortran.dg/class_alias.f90   -Os  (test for excess errors)
Setting LD_LIBRARY_PATH to .:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libatomic/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs:/safe/sgk/gcc/objx/gcc:.:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libgfortran/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libatomic/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs:/safe/sgk/gcc/objx/x86_64-unknown-freebsd13.0/./libquadmath/.libs:/safe/sgk/gcc/objx/gcc
Execution timeout is: 300
spawn [open ...]
STOP 2
FAIL: gfortran.dg/class_alias.f90   -Os  execution test

STOP 2 occurs at the end of testcase, so I suspect finalization is messed up.
-- 
Steve
20170425 https://www.youtube.com/watch?v=VWUpyCsUKR4
20161221 https://www.youtube.com/watch?v=IbCHE-hONow