RPCS3/llvm
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm.git synced 2024-11-24 20:30:06 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Add instruction selection for ffloor of vectors when SSE4.1 or AVX is enabled.

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163473 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Craig Topper 2012-09-08 17:42:27 +00:00
parent 46b4b112d2
commit 12fb5c667f
3 changed files with 79 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
lib/Target/X86/X86ISelLowering.cpp
View File

@ -946,6 +946,9 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM)
setOperationAction(ISD::FRINT, MVT::f64, Legal);
setOperationAction(ISD::FNEARBYINT, MVT::f64, Legal);
setOperationAction(ISD::FFLOOR, MVT::v4f32, Legal);
setOperationAction(ISD::FFLOOR, MVT::v2f64, Legal);
// FIXME: Do we need to handle scalar-to-vector here?
setOperationAction(ISD::MUL, MVT::v4i32, Legal);
@ -1023,6 +1026,7 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM)
setOperationAction(ISD::FMUL, MVT::v8f32, Legal);
setOperationAction(ISD::FDIV, MVT::v8f32, Legal);
setOperationAction(ISD::FSQRT, MVT::v8f32, Legal);
setOperationAction(ISD::FFLOOR, MVT::v8f32, Legal);
setOperationAction(ISD::FNEG, MVT::v8f32, Custom);
setOperationAction(ISD::FABS, MVT::v8f32, Custom);
@ -1031,6 +1035,7 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM)
setOperationAction(ISD::FMUL, MVT::v4f64, Legal);
setOperationAction(ISD::FDIV, MVT::v4f64, Legal);
setOperationAction(ISD::FSQRT, MVT::v4f64, Legal);
setOperationAction(ISD::FFLOOR, MVT::v4f64, Legal);
setOperationAction(ISD::FNEG, MVT::v4f64, Custom);
setOperationAction(ISD::FABS, MVT::v4f64, Custom);

56
lib/Target/X86/X86InstrSSE.td
View File

@ -6210,6 +6210,15 @@ let Predicates = [HasAVX] in {
(VROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x3))>;
def : Pat<(f64 (ftrunc FR64:$src)),
(VROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x3))>;
def : Pat<(v4f32 (ffloor VR128:$src)),
(VROUNDPSr VR128:$src, (i32 0x1))>;
def : Pat<(v2f64 (ffloor VR128:$src)),
(VROUNDPDr VR128:$src, (i32 0x1))>;
def : Pat<(v8f32 (ffloor VR256:$src)),
(VROUNDYPSr VR256:$src, (i32 0x1))>;
def : Pat<(v4f64 (ffloor VR256:$src)),
(VROUNDYPDr VR256:$src, (i32 0x1))>;
}
defm ROUND : sse41_fp_unop_rm<0x08, 0x09, "round", f128mem, VR128,
@ -6219,26 +6228,33 @@ let Constraints = "$src1 = $dst" in
defm ROUND : sse41_fp_binop_rm<0x0A, 0x0B, "round",
int_x86_sse41_round_ss, int_x86_sse41_round_sd>;
def : Pat<(ffloor FR32:$src),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x1))>;
def : Pat<(f64 (ffloor FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x1))>;
def : Pat<(f32 (fnearbyint FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0xC))>;
def : Pat<(f64 (fnearbyint FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0xC))>;
def : Pat<(f32 (fceil FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x2))>;
def : Pat<(f64 (fceil FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x2))>;
def : Pat<(f32 (frint FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x4))>;
def : Pat<(f64 (frint FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x4))>;
def : Pat<(f32 (ftrunc FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x3))>;
def : Pat<(f64 (ftrunc FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x3))>;
let Predicates = [UseSSE41] in {
def : Pat<(ffloor FR32:$src),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x1))>;
def : Pat<(f64 (ffloor FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x1))>;
def : Pat<(f32 (fnearbyint FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0xC))>;
def : Pat<(f64 (fnearbyint FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0xC))>;
def : Pat<(f32 (fceil FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x2))>;
def : Pat<(f64 (fceil FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x2))>;
def : Pat<(f32 (frint FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x4))>;
def : Pat<(f64 (frint FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x4))>;
def : Pat<(f32 (ftrunc FR32:$src)),
(ROUNDSSr (f32 (IMPLICIT_DEF)), FR32:$src, (i32 0x3))>;
def : Pat<(f64 (ftrunc FR64:$src)),
(ROUNDSDr (f64 (IMPLICIT_DEF)), FR64:$src, (i32 0x3))>;
def : Pat<(v4f32 (ffloor VR128:$src)),
(ROUNDPSr VR128:$src, (i32 0x1))>;
def : Pat<(v2f64 (ffloor VR128:$src)),
(ROUNDPDr VR128:$src, (i32 0x1))>;
}
//===----------------------------------------------------------------------===//
// SSE4.1 - Packed Bit Test

38
test/CodeGen/X86/vec_floor.ll Normal file
View File

@ -0,0 +1,38 @@
; RUN: llc < %s -mtriple=x86_64-apple-darwin -march=x86 -mcpu=corei7-avx | FileCheck %s
define <2 x double> @floor_v2f64(<2 x double> %p)
{
; CHECK: floor_v2f64
; CHECK: vroundpd
%t = call <2 x double> @llvm.floor.v2f64(<2 x double> %p)
ret <2 x double> %t
}
declare <2 x double> @llvm.floor.v2f64(<2 x double> %p)
define <4 x float> @floor_v4f32(<4 x float> %p)
{
; CHECK: floor_v4f32
; CHECK: vroundps
%t = call <4 x float> @llvm.floor.v4f32(<4 x float> %p)
ret <4 x float> %t
}
declare <4 x float> @llvm.floor.v4f32(<4 x float> %p)
define <4 x double> @floor_v4f64(<4 x double> %p)
{
; CHECK: floor_v4f64
; CHECK: vroundpd
%t = call <4 x double> @llvm.floor.v4f64(<4 x double> %p)
ret <4 x double> %t
}
declare <4 x double> @llvm.floor.v4f64(<4 x double> %p)
define <8 x float> @floor_v8f32(<8 x float> %p)
{
; CHECK: floor_v8f32
; CHECK: vroundps
%t = call <8 x float> @llvm.floor.v8f32(<8 x float> %p)
ret <8 x float> %t
}
declare <8 x float> @llvm.floor.v8f32(<8 x float> %p)