Skip to content

[LoongArch] Pre-commit tests for shuffle visiting same lane. NFC #151633

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: users/zhaoqi5/opt-extractelement-idx
Choose a base branch
from
Open

[LoongArch] Pre-commit tests for shuffle visiting same lane. NFC #151633

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
113 changes: 93 additions & 20 deletions llvm/test/CodeGen/LoongArch/lasx/shuffle-as-permute-and-shuffle.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,14 +17,25 @@ define <32 x i8> @shuffle_v32i8(<32 x i8> %a) {
ret <32 x i8> %shuffle
}

define <32 x i8> @shuffle_v32i8_same_lane(<32 x i8> %a) {
; CHECK-LABEL: shuffle_v32i8_same_lane:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI1_0)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI1_0)
; CHECK-NEXT: xvshuf.h $xr1, $xr0, $xr0
; CHECK-NEXT: xvori.b $xr0, $xr1, 0
; CHECK-NEXT: ret
%shuffle = shufflevector <32 x i8> %a, <32 x i8> poison, <32 x i32> <i32 14, i32 15, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
ret <32 x i8> %shuffle
}

define <16 x i16> @shuffle_v16i16(<16 x i16> %a) {
; CHECK-LABEL: shuffle_v16i16:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI1_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI1_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI1_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI1_1)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI2_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI2_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI2_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI2_1)
; CHECK-NEXT: xvpermi.d $xr3, $xr0, 78
; CHECK-NEXT: xvshuf.d $xr2, $xr0, $xr3
; CHECK-NEXT: xvshuf.w $xr1, $xr2, $xr0
Expand All @@ -34,13 +45,25 @@ define <16 x i16> @shuffle_v16i16(<16 x i16> %a) {
ret <16 x i16> %shuffle
}

define <16 x i16> @shuffle_v16i16_same_lane(<16 x i16> %a) {
; CHECK-LABEL: shuffle_v16i16_same_lane:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI3_0)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI3_0)
; CHECK-NEXT: xvshuf.h $xr1, $xr0, $xr0
; CHECK-NEXT: xvori.b $xr0, $xr1, 0
; CHECK-NEXT: ret
%shuffle = shufflevector <16 x i16> %a, <16 x i16> poison, <16 x i32> <i32 6, i32 7, i32 0, i32 5, i32 2, i32 3, i32 6, i32 5, i32 8, i32 9, i32 10, i32 13, i32 12, i32 15, i32 13, i32 15>
ret <16 x i16> %shuffle
}

define <8 x i32> @shuffle_v8i32(<8 x i32> %a) {
; CHECK-LABEL: shuffle_v8i32:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI2_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI2_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI2_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI2_1)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI4_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI4_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI4_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI4_1)
; CHECK-NEXT: xvpermi.d $xr3, $xr0, 78
; CHECK-NEXT: xvshuf.d $xr2, $xr0, $xr3
; CHECK-NEXT: xvshuf.d $xr1, $xr2, $xr0
Expand All @@ -50,13 +73,25 @@ define <8 x i32> @shuffle_v8i32(<8 x i32> %a) {
ret <8 x i32> %shuffle
}

define <8 x i32> @shuffle_v8i32_same_lane(<8 x i32> %a) {
; CHECK-LABEL: shuffle_v8i32_same_lane:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI5_0)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI5_0)
; CHECK-NEXT: xvshuf.d $xr1, $xr0, $xr0
; CHECK-NEXT: xvori.b $xr0, $xr1, 0
; CHECK-NEXT: ret
%shuffle = shufflevector <8 x i32> %a, <8 x i32> poison, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 4, i32 5, i32 6, i32 7>
ret <8 x i32> %shuffle
}

define <4 x i64> @shuffle_v4i64(<4 x i64> %a) {
; CHECK-LABEL: shuffle_v4i64:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI3_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI3_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI3_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI3_1)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI6_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI6_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI6_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI6_1)
; CHECK-NEXT: xvpermi.d $xr3, $xr0, 78
; CHECK-NEXT: xvshuf.d $xr2, $xr0, $xr3
; CHECK-NEXT: xvshuf.d $xr1, $xr2, $xr0
Expand All @@ -66,13 +101,25 @@ define <4 x i64> @shuffle_v4i64(<4 x i64> %a) {
ret <4 x i64> %shuffle
}

define <4 x i64> @shuffle_v4i64_same_lane(<4 x i64> %a) {
; CHECK-LABEL: shuffle_v4i64_same_lane:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI7_0)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI7_0)
; CHECK-NEXT: xvshuf.d $xr1, $xr0, $xr0
; CHECK-NEXT: xvori.b $xr0, $xr1, 0
; CHECK-NEXT: ret
%shuffle = shufflevector <4 x i64> %a, <4 x i64> poison, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
ret <4 x i64> %shuffle
}

define <8 x float> @shuffle_v8f32(<8 x float> %a) {
; CHECK-LABEL: shuffle_v8f32:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI4_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI4_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI4_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI4_1)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI8_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI8_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI8_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI8_1)
; CHECK-NEXT: xvpermi.d $xr3, $xr0, 78
; CHECK-NEXT: xvshuf.d $xr2, $xr0, $xr3
; CHECK-NEXT: xvshuf.d $xr1, $xr2, $xr0
Expand All @@ -82,13 +129,26 @@ define <8 x float> @shuffle_v8f32(<8 x float> %a) {
ret <8 x float> %shuffle
}

define <8 x float> @shuffle_v8f32_same_lane(<8 x float> %a) {
; CHECK-LABEL: shuffle_v8f32_same_lane:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI9_0)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI9_0)
; CHECK-NEXT: xvpermi.d $xr0, $xr0, 68
; CHECK-NEXT: xvshuf.w $xr1, $xr0, $xr0
; CHECK-NEXT: xvori.b $xr0, $xr1, 0
; CHECK-NEXT: ret
%shuffle = shufflevector <8 x float> %a, <8 x float> poison, <8 x i32> <i32 3, i32 2, i32 0, i32 2, i32 3, i32 1, i32 2, i32 3>
ret <8 x float> %shuffle
}

define <4 x double> @shuffle_v4f64(<4 x double> %a) {
; CHECK-LABEL: shuffle_v4f64:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI5_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI5_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI5_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI5_1)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI10_0)
; CHECK-NEXT: xvld $xr2, $a0, %pc_lo12(.LCPI10_0)
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI10_1)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI10_1)
; CHECK-NEXT: xvpermi.d $xr3, $xr0, 78
; CHECK-NEXT: xvshuf.d $xr2, $xr0, $xr3
; CHECK-NEXT: xvshuf.d $xr1, $xr2, $xr0
Expand All @@ -97,3 +157,16 @@ define <4 x double> @shuffle_v4f64(<4 x double> %a) {
%shuffle = shufflevector <4 x double> %a, <4 x double> poison, <4 x i32> <i32 3, i32 1, i32 2, i32 0>
ret <4 x double> %shuffle
}

define <4 x double> @shuffle_v4f64_same_lane(<4 x double> %a) {
; CHECK-LABEL: shuffle_v4f64_same_lane:
; CHECK: # %bb.0:
; CHECK-NEXT: pcalau12i $a0, %pc_hi20(.LCPI11_0)
; CHECK-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI11_0)
; CHECK-NEXT: xvpermi.d $xr0, $xr0, 78
; CHECK-NEXT: xvshuf.d $xr1, $xr0, $xr0
; CHECK-NEXT: xvori.b $xr0, $xr1, 0
Comment on lines +164 to +168
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seems a single xvpermi.d $xr0, $xr0, 75 is enough?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, a single xvpermi.d is enough for this case.

But now, when we legalize vector_shuffle for lasx, we firstly use canonicalizeShuffleVectorByLane() to convert the source vector to avoid cross-lane access. Thus the xvpermi.d $xr0, $xr0, 78 will be generated. Then the converted vector will be used to match lasx shuffle instructions' patterns, and xvshuf maches.

To avoid this, we may should modify the order of the current processing logic. And the pattern for xvpermi.d should also be implemented.

; CHECK-NEXT: ret
%shuffle = shufflevector <4 x double> %a, <4 x double> poison, <4 x i32> <i32 3, i32 2, i32 0, i32 1>
ret <4 x double> %shuffle
}