[Bug target/88892] [8/9 Regression] Double-to-float conversion uses wrong rounding mode when followed by memcpy

[fw at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88892

Florian Weimer <fw at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code
            Summary|Double-to-float conversion  |[8/9 Regression]
                   |uses wrong rounding mode    |Double-to-float conversion
                   |when followed by memcpy     |uses wrong rounding mode
                   |                            |when followed by memcpy

--- Comment #1 from Florian Weimer <fw at gcc dot gnu.org> ---
Does not reproduce with gcc-7.3.1-6.fc27.ppc64le.