Re: [PATCH] Fortran: default-initialization and functions returning derived type[PR85750]

LGTM! Thanks for the Patch. - Andre Andre Vehreschild * ve...@gmx.de Am 15. Mai 2025 22:36:19 schrieb Harald Anlauf : Dear all, the attached patch fixes missing default-initialization of function results of derived type that happens under some conditions, see PR. The logic when default initia

Re: [PATCH] Fortran: default-initialization and functions returning derived type[PR85750]

Hi Andre, Am 15.05.25 um 22:13 schrieb Andre Vehreschild: LGTM! it's great that you reviewed the patch, as I was touching original code next to yours... ;-) Thanks for the Patch. Pushed as r16-669-gd31ab498b12ebb. Thanks, Harald - Andre Andre Vehreschild * ve...@gmx.de Am 15. Mai 2025

[PATCH] Fortran: default-initialization and functions returning derived type[PR85750]

Dear all, the attached patch fixes missing default-initialization of function results of derived type that happens under some conditions, see PR. The logic when default initialization is to be applied is rather contorted, and reversing the order of two cases fixed the issue. Regtesting revealed

ICE in gfc_trans_call

--disable-multilib --enable-checking=release --enable-host-shared --with-pic --enable-languages=c,c++,fortran,jit,lto Thread model: posix Supported LTO compression algorithms: zlib zstd gcc version 16.0.0 20250515 (experimental) (GCC) $ gfortran -c ivs.F90 $ gfortran -c bmn.F90 bmn.F90:10:13: 10