Alex Crichton d6ce632b5b aarch64: Specialize constant vector shifts (#5976) ...

* aarch64: Specialize constant vector shifts

This commit adds special lowering rules for
vector-shifts-by-constant-amounts to use dedicated instructions which
cuts down on the codegen here quite a bit for constant values.

* Fix codegen for 0-shift-rights

* Special-case zero left-shifts as well

* Remove left-shift special case

2023-03-13 22:37:59 +00:00

..

alias

cranelift: Remove brz and brnz (#5630)

2023-01-30 20:34:56 +00:00

cfg

cranelift: Remove brz and brnz (#5630)

2023-01-30 20:34:56 +00:00

dce

cranelift: Remove brz and brnz (#5630)

2023-01-30 20:34:56 +00:00

domtree

Reuse the DominatorTree postorder travesal in BlockLoweringOrder (#5843)

2023-02-23 22:05:20 +00:00

egraph

Remove the Cranelift vselect instruction (#5918)

2023-03-08 00:42:05 +00:00

isa

aarch64: Specialize constant vector shifts (#5976)

2023-03-13 22:37:59 +00:00

legalizer

cranelift: Remove brz and brnz (#5630)

2023-01-30 20:34:56 +00:00

licm

Inline jump tables in parsed br_table instructions (#5755)

2023-02-09 14:24:04 -08:00

parser

Inline jump tables in parsed br_table instructions (#5755)

2023-02-09 14:24:04 -08:00

runtests

aarch64: Specialize constant vector shifts (#5976)

2023-03-13 22:37:59 +00:00

simple_gvn

cranelift: Remove brz and brnz (#5630)

2023-01-30 20:34:56 +00:00

simple_preopt

cranelift: Remove brz and brnz (#5630)

2023-01-30 20:34:56 +00:00

verifier

Change CLIF shuffle to validate lane indices (#5995)

2023-03-13 14:24:11 +00:00

wasm

x64: Improve memory support in {insert,extract}lane (#5982)

2023-03-13 19:30:44 +00:00