[Bug c/101128] New: Decimal floating point fails on ARM with gcc-10/11 but works with gcc-6/7/8/9

[Bug c/101128] New: Decimal floating point fails on ARM with gcc-10/11 but works with gcc-6/7/8/9

[caparson at uwaterloo dot ca]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101128

            Bug ID: 101128
           Summary: Decimal floating point fails on ARM with gcc-10/11 but
                    works with gcc-6/7/8/9
           Product: gcc
           Version: 10.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: caparson at uwaterloo dot ca
  Target Milestone: ---

Created attachment 51036
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51036&action=edit
A short snippet that reproduces the bug on ARM machines

Repro:
compile the following program (also attached) on an ARM machine with gcc-10 or
gcc-11.

  void fred() {
      1.0DL;
  }

You will receive the following error:

  gcc-10 test.c
  test.c: In function 'fred':
  test.c:2:5: error: decimal floating-point not supported for this target
      2 |     1.0DL;
        |     ^~~~~

The two ARM machines we tested this on are Huawei TaiShan 2280 V2 and Raspberry
Pi 4 Model B Rev 1.

Huawei
$ uname -a
Linux algol 5.4.0-74-generic #83-Ubuntu SMP Sat May 8 02:34:05 UTC 2021 aarch64
aarch64 aarch64 GNU/Linux

Raspberrypi
$ uname -a
Linux cfapi1 5.4.0-1036-raspi #39-Ubuntu SMP PREEMPT Wed May 12 17:37:51 UTC
2021 aarch64 aarch64 aarch64 GNU/Linux

If you compile with gcc-6 to gcc-9 this compiles without issue.

[Bug target/101128] Decimal floating point fails on ARM with gcc-10/11 but works with gcc-6/7/8/9

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101128

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ABI
             Target|                            |aarch64
          Component|c                           |target

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
The patches to support DFP for aarch64 were never accepted upstream ...
And the last I heard ARM was supposed to handle the ABI but never did ....

[Bug target/101128] DFP support for aarch64 is missing

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101128

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |DUPLICATE

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Dup of bug 96253.

*** This bug has been marked as a duplicate of bug 96253 ***

[Bug target/101128] DFP support for aarch64 is missing

[joseph at codesourcery dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101128

--- Comment #3 from joseph at codesourcery dot com <joseph at codesourcery dot com> ---
Bug 91985 was the bug that DFP constants were wrongly accepted by older 
GCC versions on architectures lacking DFP support.