[Bug libfortran/110759] New: IEEE Fortran change broke RISC-V linux build

[Bug libfortran/110759] New: IEEE Fortran change broke RISC-V linux build

[patrick at rivosinc dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

            Bug ID: 110759
           Summary: IEEE Fortran change broke RISC-V linux build
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libfortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: patrick at rivosinc dot com
  Target Milestone: ---

Using https://github.com/riscv-collab/riscv-gnu-toolchain with tip-of-tree GCC
fails to build linux target.

Failures:

2023-07-20T14:14:06.1566362Z
../../../../gcc/libgfortran/ieee/ieee_arithmetic.F90:563:617:
2023-07-20T14:14:06.1566776Z 
2023-07-20T14:14:06.1567818Z   563 |   IEEE_COMPARISON(QUIET,EQ)
2023-07-20T14:14:06.1568423Z       |                                           
                                                                               
                                                                               
                                                                               
                                                                               
                                                                               
                                                                               
                                                                               
             1
2023-07-20T14:14:06.1569478Z Error: Syntax error in PROCEDURE statement at (1)
2023-07-20T14:14:06.1576233Z
../../../../gcc/libgfortran/ieee/ieee_arithmetic.F90:564:617:
2023-07-20T14:14:06.1576587Z 
2023-07-20T14:14:06.1577039Z   564 |   IEEE_COMPARISON(QUIET,GE)
2023-07-20T14:14:06.1577606Z      

Appears to be caused by
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=dca2874897ec58ea1c22a9c2161f112fff07cfb2

Affects rv32gc and rv64gc

Known failure: c5bd0e5870aed178b7f82e7b94f59a383e7c5b4f
Known success: 49bed11d96cf727de7e6ed35f065a4df29f6c589

[Bug libfortran/110759] IEEE Fortran change broke RISC-V linux build

[patrick at rivosinc dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #1 from Patrick O'Neill <patrick at rivosinc dot com> ---
Created attachment 55593
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=55593&action=edit
Failing build log

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |14.0
            Summary|IEEE Fortran change broke   |[14 Regression] IEEE
                   |RISC-V linux build          |Fortran change broke RISC-V
                   |                            |linux build
           Keywords|                            |build

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I don't see how it could have broke riscv only ...

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[patrick at rivosinc dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #3 from Patrick O'Neill <patrick at rivosinc dot com> ---
It may have broken other targets - I can only confirm with builds for RISCV so
I didn't want to speculate too much

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
What exact target? Is this the elf target or the linux target?


I just bootstrapped on x86_64-linux-gnu and it works.

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[patrick at rivosinc dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #5 from Patrick O'Neill <patrick at rivosinc dot com> ---
Linux target rv32gc-ilp32d, rv64gc-lp64d.
Newlib still builds successfully.

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2023-07-21

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I see the problem now.
 537 #ifdef HAVE_GFC_REAL_16
 538 #  define EXPAND_INTER_MACRO_16(TYPE,OP)
_gfortran_ieee_/**/TYPE/**/_/**/OP/**/_16
 539 #else
 540 #  define EXPAND_INTER_MACRO_16(TYPE,OP)
 541 #endif
 542 

...
 552       EXPAND_INTER_MACRO_16(TYPE,OP) , \


The comma should be part of EXPAND_INTER_MACRO_16 macro instead.

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
             Target|                            |!=x86_64
           Assignee|unassigned at gcc dot gnu.org      |pinskia at gcc dot gnu.org

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I will commit a fix in a hour or two; just going to build to make sure it
works.

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[patrick at rivosinc dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #8 from Patrick O'Neill <patrick at rivosinc dot com> ---
Awesome, thank you!

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #9 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The trunk branch has been updated by Andrew Pinski <pinskia@gcc.gnu.org>:

https://gcc.gnu.org/g:f32518726ee8e836d12d49aec8432679fcc42503

commit r14-2695-gf32518726ee8e836d12d49aec8432679fcc42503
Author: Andrew Pinski <apinski@marvell.com>
Date:   Fri Jul 21 02:26:09 2023 +0000

    libfortran: Fix build for targets that don't have 10byte or 16 byte
floating point

    So the problem here is EXPAND_INTER_MACRO_16 expands to nothing if 16 byte
FP does not
    exist but we still add a comma after it and that causes a build failure.
    The same is true for EXPAND_INTER_MACRO_10 too.

    Committed as obvious after a bootstrap and test on x86_64-linux-gnu and
aarch64-linux-gnu.

    libgfortran/ChangeLog:

            PR libfortran/110759
            * ieee/ieee_arithmetic.F90
            (COMP_INTERFACE): Remove the comma after EXPAND_INTER_MACRO_16
            and EXPAND_INTER_MACRO_10.
            (EXPAND_INTER_MACRO_16): Add comma here if 16 byte fp exist.
            (EXPAND_INTER_MACRO_10): Likewise.

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #10 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Fixed.

[Bug libfortran/110759] [14 Regression] IEEE Fortran change broke RISC-V linux build

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110759

--- Comment #11 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
Thanks Andrew for fixing it, my mistake. Apologies to everyone.