[Bug target/126933] Inlined SSE2 float -> bf16 conversions should be improved
"hjl.tools at gmail dot com via Gcc-bugs" <[email protected]>
| Newsgroups | gmane.comp.gcc.bugs |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126933
H.J. Lu <hjl.tools at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|FAIL: |Inlined SSE2 float -> bf16
|gcc.target/i386/truncsfbf-2 |conversions should be
|.c execution test |improved
--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> ---
The following SSE2 float -> bf16 conversions failed:
float -> bf16 failed: 10, 1.000000e-08
float -> bf16 failed: 11, -1.000000e-08
float -> bf16 failed: 14, 0.000000e+00
float -> bf16 failed: 15, -0.000000e+00
float -> bf16 of 0.000000e+00 should be 0.000000e+00, not 9.918e-39.