[PowerPC] Implement the vclz instructions for PWR8

Patch by Kit Barton.

Add the vector count leading zeros instruction for byte, halfword,
word, and doubleword sizes.  This is a fairly straightforward addition
after the changes made for vpopcnt:

 1. Add the correct definitions for the various instructions in
    PPCInstrAltivec.td
 2. Make the CTLZ operation legal on vector types when using P8Altivec
    in PPCISelLowering.cpp 

Test Plan

Created new test case in test/CodeGen/PowerPC/vec_clz.ll to check the
instructions are being generated when the CTLZ operation is used in
LLVM.

Check the encoding and decoding in test/MC/PowerPC/ppc_encoding_vmx.s
and test/Disassembler/PowerPC/ppc_encoding_vmx.txt respectively.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228301 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Bill Schmidt
2015-02-05 15:24:47 +00:00
parent 29b65fa1df
commit 202b6045bf
5 changed files with 91 additions and 4 deletions

View File

@@ -0,0 +1,40 @@
; Check the vctlz* instructions that were added in P8
; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr8 < %s | FileCheck %s
; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s
declare <16 x i8> @llvm.ctlz.v16i8(<16 x i8>) nounwind readnone
declare <8 x i16> @llvm.ctlz.v8i16(<8 x i16>) nounwind readnone
declare <4 x i32> @llvm.ctlz.v4i32(<4 x i32>) nounwind readnone
declare <2 x i64> @llvm.ctlz.v2i64(<2 x i64>) nounwind readnone
define <16 x i8> @test_v16i8(<16 x i8> %x) nounwind readnone {
%vcnt = tail call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %x)
ret <16 x i8> %vcnt
; CHECK: @test_v16i8
; CHECK: vclzb 2, 2
; CHECK: blr
}
define <8 x i16> @test_v8i16(<8 x i16> %x) nounwind readnone {
%vcnt = tail call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %x)
ret <8 x i16> %vcnt
; CHECK: @test_v8i16
; CHECK: vclzh 2, 2
; CHECK: blr
}
define <4 x i32> @test_v4i32(<4 x i32> %x) nounwind readnone {
%vcnt = tail call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %x)
ret <4 x i32> %vcnt
; CHECK: @test_v4i32
; CHECK: vclzw 2, 2
; CHECK: blr
}
define <2 x i64> @test_v2i64(<2 x i64> %x) nounwind readnone {
%vcnt = tail call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %x)
ret <2 x i64> %vcnt
; CHECK: @test_v2i64
; CHECK: vclzd 2, 2
; CHECK: blr
}