gf2121 commented on PR #14203:
URL: https://github.com/apache/lucene/pull/14203#issuecomment-2659505128
Thanks for feedback!
And sorry for my poor english.. Do you mean something like this by `single
batch size of 16 of 32` ?
```
private static void readDelta16(IndexInput in, int count, int[] docIds)
throws IOException {
final int min = in.readVInt();
int k = 0;
for (int bound = count - 31; k < bound; k += 32) {
in.readInts(docIds, k, 16);
// Can be inlined to make offsets consistent so that loop get
auto-vectorized.
inner16(k, docIds, 16, min);
}
for (; k < count; k++) {
docIds[k] = Short.toUnsignedInt(in.readShort()) + min;
}
}
private void readInts24(IndexInput in, int count, int[] docIDs) throws
IOException {
int k = 0;
for (int bound = count - 31; k < bound; k += 32) {
in.readInts(scratch, k, 24);
shift(k, docIDs, scratch, 24);
// Can be inlined to make offsets consistent so that loop get
auto-vectorized.
remainder24(k, docIDs, scratch, 8, 16, 24);
}
readScalarInts24(in, count - k, docIDs, k);
}
```
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
