This is a follow-up to PR/25874, with the identical test case. The compilation now causes another ICE: /scratch>g++ -v -O -fopenmp -c bug.ii Using built-in specs. Target: i686-pc-linux-gnu Configured with: /scratch/gompcc/configure --quiet --prefix=/scratch/ugccgomp --enable-languages=c++,fortran --with-gmp=/usr/local/appl/gmp-4.1.4 --enable-checking=release Thread model: posix gcc version 4.2.0-gomp-20050608-branch 20060126 (experimental) (merged 20060126) /scratch/ugccgomp/libexec/gcc/i686-pc-linux-gnu/4.2.0-gomp-20050608-branch/cc1plus -fpreprocessed bug.ii -quiet -dumpbase bug.ii -mtune=generic -auxbase bug -O -version -fopenmp -o /tmp/cchqvFNL.s GNU C++ version 4.2.0-gomp-20050608-branch 20060126 (experimental) (merged 20060126) (i686-pc-linux-gnu) compiled by GNU C version 4.2.0-gomp-20050608-branch 20060126 (experimental) (merged 20060126). GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072 Compiler executable checksum: ddc6509291ab6cf0d07599318fa2caeb Healpix_cxx/alm_powspec_tools.cc: In function 'void rotate_alm(Alm<xcomplex<T> >&, Alm<xcomplex<T> >&, Alm<xcomplex<T> >&, double, double, double) [with T = double]': Healpix_cxx/alm_powspec_tools.cc:476: internal compiler error: in copy_body_r, at tree-inline.c:630 Please submit a full bug report, with preprocessed source if appropriate. See <URL:http://gcc.gnu.org/bugs.html> for instructions. Unfortunately, I probably won't be able to reduce the test case until the weekend at least.
Created attachment 10762 [details] unreduced testcase
This looks like a duplicate of PR 25990 to me since I can trigger the bug also at -O0 with appropriate "--param ggc-min-expand" settings. If I change the parameter the ICE happens at different places.
Should be fixed with the patch for PR 25990.