jpountz commented on a change in pull request #973: LUCENE-9027: Use SIMD
instructions to decode postings.
URL: https://github.com/apache/lucene-solr/pull/973#discussion_r347371932
##########
File path:
lucene/core/src/java/org/apache/lucene/store/ByteBufferIndexInput.java
##########
@@ -107,6 +118,28 @@ public final void readBytes(byte[] b, int offset, int
len) throws IOException {
}
}
+ @Override
+ public void readLELongs(long[] dst, int offset, int length) throws
IOException {
+ if (curLongBufferViews == null) {
+ // Lazy init to not make pay for memory and initialization cost if you
don't need to read arrays of longs
Review comment:
> under the hood the CPU must still do unaligned long decodes (which I think
modern X86-64 are good at)? Maybe add a comment about why this trick is
worthwhile?
This is mostly a workaround for the lack of ByteBuffer#getLongs(long[]).
Avoiding the LongBuffer view would require to call ByteBuffer#getLong in a
loop, which seems to have some per-long overhead according to the
microbenchmarks I ran. I'll leave a comment.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
