Pushed as r16-3481. Thanks for your help in getting it over the line!
Paul On Sun, 31 Aug 2025 at 12:45, Paul Richard Thomas < paul.richard.tho...@gmail.com> wrote: > Hi Harald and Jerry, > > As it happens, I had done the regtesting, caught the failures of > value_optional_[1-2].f90 but failed to update the .diff for submission. Mea > culpa for not checking the submission. > > More importantly, I forgot to deallocate the parameterized components - > see the attachment. Unlike the dereferencing error, this might have sat > undetected for a long time > > This version has been regtested :-) I will push to mainline later today > unless there are any objections. > > Thanks > > Paul > > > On Sat, 30 Aug 2025 at 18:34, Harald Anlauf <anl...@gmx.de> wrote: > >> Am 30.08.25 um 18:22 schrieb Jerry D: >> > On 8/30/25 9:16 AM, Harald Anlauf wrote: >> >> On 8/30/25 18:04, Jerry D wrote: >> >>> On 8/30/25 8:04 AM, Paul Richard Thomas wrote: >> >>>> Hi All, >> >>>> >> >>>> This patch is only a temporary fix because the chunks in trans- >> >>>> array.cc are representation dependent. As a whole, the patch is so >> >>>> straightforward that the ChangeLog serves as an explanation. >> >>>> >> >>>> Regtests with FC32/x86_64 - OK for mainline? >> >>>> >> >>>> Paul >> >>> >> >>> Yes, OK and thanks for the fix. >> >>> >> >>> Regards, >> >>> >> >>> Jerry >> >>> >> >> >> >> Actually - and obviously - it is NOT OK. It fails here on >> >> >> >> gfortran.dg/value_optional_1.f90 >> >> gfortran.dg/value_optional_2.f90 >> >> >> >> because of the new non-caught NULL-pointer dereference here: >> >> >> >> diff --git a/gcc/fortran/trans-expr.cc b/gcc/fortran/trans-expr.cc >> >> index 69952b33eaa..771d2c24fa9 100644 >> >> --- a/gcc/fortran/trans-expr.cc >> >> +++ b/gcc/fortran/trans-expr.cc >> >> @@ -6520,6 +6520,18 @@ conv_dummy_value (gfc_se * parmse, gfc_expr * >> >> e, gfc_symbol * fsym, >> >> >> >> gcc_assert (fsym && fsym->attr.value && !fsym->attr.dimension); >> >> >> >> + if (e->ts.type == BT_DERIVED && e->ts.u.derived->attr.pdt_type) >> >> >> >> Note that e == NULL for a missing actual argument, so you need to >> >> check for this case. >> >> >> >> Thanks, >> >> Harald >> >> >> > >> > oops, reviewed and did not apply and test. my bad. >> >> Yes, happens all the time ... ;-) >> >> (Seems that Paul short-cut the regtesting.) >> >> I just happen to know these testcases very well. >> >> > Jerry >> > >> >>
