RFR: 8318650: Optimized subword gather for x86 targets. [v11]

[Jatin Bhateja]

On Wed, 31 Jan 2024 21:29:08 GMT, Sandhya Viswanathan <sviswanathan at openjdk.org> wrote:

>> Jatin Bhateja has refreshed the contents of this pull request, and previous commits have been removed. Incremental views are not available.
>
> src/hotspot/cpu/x86/c2_MacroAssembler_x86.cpp line 1613:
> 
>> 1611:   vpand(xtmp, idx_vec, xtmp, vlen_enc);
>> 1612:   // Load double words from normalized indices.
>> 1613:   evpgatherdd(dst, gmask, Address(base, xtmp, scale), vlen_enc);
> 
> Considering the byte vector case, could we not do here directly:
> evpgatherdd(dst, gmask, Address(base, idx_vec, scale), vlen_enc);
> Then we dont need lines 1609-1611 and also 1616-1621 as well.

Gathering is happening at doubleword granularity, where as masks are applier over sub-word lanes.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/16354#discussion_r1475481601