Hacker News new | ask | show | jobs
by Kipters 1118 days ago
You mean this IntVector[0], which I assume is the Java experimental API anthony88 was referring to, correct? If that operation being missing is a blocker, I feel there may be some middle ground other than implementing the whole thing in C++ (like adding it or fast tracking work on this API)

[0]: https://docs.oracle.com/en/java/javase/19/docs/api/jdk.incub...
1 comments
gergo_barany 1118 days ago
The compress operation being missing is not a blocker, since the compress operation is not missing: https://docs.oracle.com/en/java/javase/19/docs/api/jdk.incub...
link
Kipters 1117 days ago
I don't get the other user's point then
link
gergo_barany 1117 days ago
The user didn't have a point, they just made something up about an API that they know nothing about.
link
anonymoushn 1116 days ago
Hello, thanks for pointing out that I missed this one. It looks like the methods exposed may be sufficient for implementing a sort. I also looked for vpalignr and vpshufb which seem to be missing, but I'm open to the possibility that I have missed these as well.
link
gergo_barany 1115 days ago
vpalignr is called "slice", vpshufb is "rearrange". The latter is easy to find if you search the page linked above for "shuf". The former is a bit harder, but I found it by searching for "concat" while thinking about how it might be possible to express it.

If you need the vpshufb behavior of zeroing elements where the index is negative, I think you will need to build something out of multiple operations. The compiler is of course free to match those multiple operations to one target instruction, i.e., recognize that what you are trying to say is really a vpshufb. It can do this in the same way that it can match multiple operations like x + y * 8 + 12 to a single lea instruction.

link
anonymoushn 1115 days ago
Another way that rearrange is not vpshufb is that 0 in the 17th position means "get me the first byte (from a different 16-byte lane)" rather than "get me the 17th byte (from this 16-byte lane)" It would be very impressive if the compiler could take a lookup table of values designed to be used as the shuffle argument to rearrange and transform it into a lookup table of values designed to be used as the shuffle argument to vpshufb.
link