Regression: ICE with 2025.1 (RANK intrinsic)

[OP1]

The following code used to work with the previous ICE version, and with 2025.1 it now leads to an ICE:

INTEGER, ALLOCATABLE ::  XS(:)
CALL USEINT(RANK(XS)) ! ICE
!CALL USEINT(INT(RANK(XS))) ! No error
CONTAINS
    SUBROUTINE USEINT(X)
        INTEGER :: X
    END SUBROUTINE USEINT
END

[andrew_4619]

https://community.intel.com/t5/Intel-Fortran-Compiler/2025-1-ICE-passing-shape/m-p/1678166#M175342

You have got me thinking that my ICE is the same, I was blaming SHAPE..... 

[Xiaoping_D_Intel]

A bug report has been opened for it and will be fixed in future product.