3 files changed, 192 insertions, 0 deletions
diff --git a/test/CodeGen/X86/vec_shuffle-28.ll b/test/CodeGen/X86/vec_shuffle-28.ll
new file mode 100644
index 0000000..e73b824
--- /dev/null
+++ b/test/CodeGen/X86/vec_shuffle-28.ll
@@ -0,0 +1,33 @@
+; RUN: llvm-as < %s | llc -march=x86 -mattr=sse41 -o %t -f
+; RUN: grep punpcklwd %t | count 1
+; RUN: grep pextrw %t | count 8
+; RUN: grep pinsrw %t | count 8
+
+
+; Pack various elements via shuffles.
+define <8 x i16> @shuf1(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp7 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 1, i32 8, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp7
+}
+
+
+define <8 x i16> @shuf2(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp8 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 undef, i32 undef, i32 7, i32 2, i32 8, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp8
+}
+
+
+define <8 x i16> @shuf3(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp9 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 0, i32 1, i32 undef, i32 undef, i32 3, i32 11, i32 undef , i32 undef >
+	ret <8 x i16> %tmp9
+}
+
+
+define <8 x i16> @shuf4(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp9 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 8, i32 9, i32 undef, i32 undef, i32 11, i32 3, i32 undef , i32 undef >
+	ret <8 x i16> %tmp9
+}
diff --git a/test/CodeGen/X86/vec_splat-3.ll b/test/CodeGen/X86/vec_splat-3.ll
new file mode 100644
index 0000000..1f1a214
--- /dev/null
+++ b/test/CodeGen/X86/vec_splat-3.ll
@@ -0,0 +1,55 @@
+; RUN: llvm-as < %s | llc -march=x86 -mattr=sse41 -o %t -f
+; RUN: grep punpcklwd %t | count 4
+; RUN: grep punpckhwd %t | count 4
+; RUN: grep "pshufd" %t | count 8
+
+; Splat test for v8i16
+; Should generate with pshufd with masks $0, $85, $170, $255 (each mask is used twice)
+define <8 x i16> @shuf_8i16_0(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 0, i32 undef, i32 undef, i32 0, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+define <8 x i16> @shuf_8i16_1(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 1, i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+define <8 x i16> @shuf_8i16_2(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 2, i32 undef, i32 undef, i32 2, i32 undef, i32 2, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+define <8 x i16> @shuf_8i16_3(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 3, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+define <8 x i16> @shuf_8i16_4(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 4, i32 undef, i32 undef, i32 undef, i32 4, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+define <8 x i16> @shuf_8i16_5(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 5, i32 undef, i32 undef, i32 5, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+define <8 x i16> @shuf_8i16_6(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 6, i32 6, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
+
+
+define <8 x i16> @shuf_8i16_7(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 7, i32 undef, i32 undef, i32 7, i32 undef, i32 undef, i32 undef , i32 undef >
+	ret <8 x i16> %tmp6
+}
diff --git a/test/CodeGen/X86/vec_splat-4.ll b/test/CodeGen/X86/vec_splat-4.ll
new file mode 100644
index 0000000..220e1cd
--- /dev/null
+++ b/test/CodeGen/X86/vec_splat-4.ll
@@ -0,0 +1,104 @@
+; RUN: llvm-as < %s | llc -march=x86 -mattr=sse41 -o %t -f
+; RUN: grep punpcklbw %t | count 16
+; RUN: grep punpckhbw %t | count 16
+; RUN: grep "pshufd" %t | count 16
+
+; Should generate with pshufd with masks $0, $85, $170, $255 (each mask is used 4 times)
+
+; Splat test for v16i8
+define <16 x i8 > @shuf_16i8_0(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 0, i32 undef, i32 undef, i32 0, i32 undef, i32 0, i32 0 , i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_1(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 1, i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef, i32 undef, i32 undef, i32 undef  >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_2(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 2, i32 undef, i32 undef, i32 2, i32 undef, i32 2, i32 2 , i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_3(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 3, i32 undef, i32 undef, i32 3, i32 undef, i32 3, i32 3 , i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3 >
+	ret <16 x i8 > %tmp6
+}
+
+
+define <16 x i8 > @shuf_16i8_4(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 4, i32 undef, i32 undef, i32 undef, i32 4, i32 undef, i32 undef , i32 undef, i32 undef, i32 undef, i32 undef , i32 undef, i32 undef, i32 undef, i32 undef , i32 undef  >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_5(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 5, i32 undef, i32 undef, i32 5, i32 undef, i32 5, i32 5 , i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_6(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 6, i32 undef, i32 undef, i32 6, i32 undef, i32 6, i32 6 , i32 6, i32 6, i32 6, i32 6, i32 6, i32 6, i32 6, i32 6, i32 6 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_7(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 7, i32 undef, i32 undef, i32 7, i32 undef, i32 undef, i32 undef , i32 undef, i32 undef, i32 undef, i32 undef , i32 undef , i32 undef, i32 undef, i32 undef , i32 undef  >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_8(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 8, i32 undef, i32 undef, i32 8, i32 undef, i32 8, i32 8 , i32 8, i32 8, i32 8, i32 8, i32 8, i32 8, i32 8, i32 8, i32 8 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_9(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 9, i32 undef, i32 undef, i32 9, i32 undef, i32 9, i32 9 , i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9, i32 9 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_10(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 10, i32 undef, i32 undef, i32 10, i32 undef, i32 10, i32 10 , i32 10, i32 10, i32 10, i32 10, i32 10, i32 10, i32 10, i32 10, i32 10 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_11(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 11, i32 undef, i32 undef, i32 11, i32 undef, i32 11, i32 11 , i32 11, i32 11, i32 11, i32 11, i32 11, i32 11, i32 11, i32 11, i32 11 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_12(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 12, i32 undef, i32 undef, i32 12, i32 undef, i32 12, i32 12 , i32 12, i32 12, i32 12, i32 12, i32 12, i32 12, i32 12, i32 12, i32 12 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_13(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 13, i32 undef, i32 undef, i32 13, i32 undef, i32 13, i32 13 , i32 13, i32 13, i32 13, i32 13, i32 13, i32 13, i32 13, i32 13, i32 13 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_14(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 14, i32 undef, i32 undef, i32 14, i32 undef, i32 14, i32 14 , i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14 >
+	ret <16 x i8 > %tmp6
+}
+
+define <16 x i8 > @shuf_16i8_15(<16 x i8 > %T0, <16 x i8 > %T1) nounwind readnone {
+entry:
+	%tmp6 = shufflevector <16 x i8 > %T0, <16 x i8 > %T1, <16 x i32> < i32 15, i32 undef, i32 undef, i32 15, i32 undef, i32 15, i32 15 , i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15 >
+	ret <16 x i8 > %tmp6
+}