RFR: 8289552: Make intrinsic conversions between bit representations of half precision values and floats [v6]

[Vladimir Kozlov]

On Thu, 1 Sep 2022 18:31:07 GMT, Smita Kamath <svkamath at openjdk.org> wrote:

>> 8289552: Make intrinsic conversions between bit representations of half precision values and floats
>
> Smita Kamath has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Addressed review comments, updated microbenchmark

Looks like it is not ready. My JDK builds failed:

Exception in thread "main" java.nio.file.NoSuchFileException: /workspace/closed/src/jdk.compiler/share/data/symbols/../../../../../../open/src/jdk.compiler/share/data/symbols/java.base-7.sym.txt
	at java.base/sun.nio.fs.UnixException.translateToIOException(UnixException.java:92)

Also some compilers complain about missing parentheses in lines 482, 487, 488, 489, 522:

src/hotspot/share/runtime/sharedRuntime.cpp:482:31: note: place parentheses around the '&' expression to silence this warning
  jint f_signif_bits = doppel & 0x007fffff | msb;

-------------

PR: https://git.openjdk.org/jdk/pull/9781