https://sourceware.org/bugzilla/show_bug.cgi?id=28230
--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> --- The current assembly codes expect and work with 10-byte .tfloat output. Changing .tfloat leads to hard to detect bugs. There is no strong reason to change it. .tfloat output with hex input is different since padding was added to the beginning, not end. Reduce its size to 10 bytes to match .tfloat is OK. -- You are receiving this mail because: You are on the CC list for the bug.
