-; RUN: llvm-as %s -o /dev/null

-

-; This testcase failed due to a bad assertion in SymbolTable.cpp, removed in

-; the 1.20 revision. Basically the symbol table assumed that if there was an

-; abstract type in the symbol table, [in this case for the entry %foo of type

-; void(opaque)* ], that there should have also been named types by now. This

-; was obviously not the case here, and this is valid. Assertion disabled.

-

-%bb = type i32

-

-declare void @foo(i32)

-; RUN: llvm-as %s -o /dev/null

-

-; This testcase used to fail due to a lack of this diff in Value.cpp:

-; diff -r1.16 Value.cpp

-; 11c11

-; < #include "llvm/Type.h"

-; ---

-; > #include "llvm/DerivedTypes.h"

-; 74c74,76

-; < assert(Ty.get() == (const Type*)OldTy &&"Can't refine anything but my type!");

-; ---

-; > assert(Ty.get() == OldTy &&"Can't refine anything but my type!");

-; > if (OldTy == NewTy && !OldTy->isAbstract())

-; > Ty.removeUserFromConcrete();

-;

-; This was causing an assertion failure, due to the "foo" Method object never

-; releasing it's reference to the opaque %bb value.

-;

-

-%bb = type i32

-%exception_descriptor = type i32

-

-declare void @foo(i32)

-; RUN: llvm-as %s -o /dev/null

-

-%Hosp = type { i32, i32, i32, { \2*, { i32, i32, i32, { [4 x \3], \2, \5, %Hosp, i32, i32 }* }*, \2* }, { \2*, { i32, i32, i32, { [4 x \3], \2, \5, %Hosp, i32, i32 }* }*, \2* }, { \2*, { i32, i32, i32, { [4 x \3], \2, \5, %Hosp, i32, i32 }* }*, \2* }, { \2*, { i32, i32, i32, { [4 x \3], \2, \5, %Hosp, i32, i32 }* }*, \2* } }

-; RUN: llvm-as %s -o /dev/null

-

- type { { \2 *, \4 ** },

- { \2 *, \4 ** }

- }

-

-; RUN: llvm-as %s -o /dev/null

-

-%t = type { { \2*, \2 },

- { \2*, \2 }

- }

-; RUN: llvm-as %s -o /dev/null

-

- %Hosp = type { { \2*, { \2, %Hosp }* }, { \2*, { \2, %Hosp }* } }

-; RUN: llvm-as %s -o /dev/null

-

-%T = type i32 *

-

-define %T @test() {

- ret %T null

-}

-; This file takes about 48 __MINUTES__ to assemble using as. This is WAY too

-; long. The type resolution code needs to be sped up a lot.

-; RUN: llvm-as %s -o /dev/null

- %ALL_INTERSECTIONS_METHOD = type i32 (%OBJECT*, %RAY*, %ISTACK*)*

- %BBOX = type { %BBOX_VECT, %BBOX_VECT }

- %BBOX_TREE = type { i16, i16, %BBOX, %BBOX_TREE** }

- %BBOX_VECT = type [3 x float]

- %BLEND_MAP = type { i16, i16, i16, i32, %BLEND_MAP_ENTRY* }

- %BLEND_MAP_ENTRY = type { float, i8, { %COLOUR, %PIGMENT*, %TNORMAL*, %TEXTURE*, %UV_VECT } }

- %CAMERA = type { %VECTOR, %VECTOR, %VECTOR, %VECTOR, %VECTOR, %VECTOR, double, double, i32, double, double, i32, double, %TNORMAL* }

- %COLOUR = type [5 x float]

- %COPY_METHOD = type i8* (%OBJECT*)*

- %COUNTER = type { i32, i32 }

- %DENSITY_FILE = type { i32, %DENSITY_FILE_DATA* }

- %DENSITY_FILE_DATA = type { i32, i8*, i32, i32, i32, i8*** }

- %DESTROY_METHOD = type void (%OBJECT*)*

- %FILE = type { i32, i8*, i8*, i8, i8, i32, i32, i32 }

- %FILE_HANDLE = type { i8*, i32, i32, i32, i32, i8*, %FILE*, i32, i32 (%FILE_HANDLE*, i8*, i32*, i32*, i32, i32)*, void (%FILE_HANDLE*, %COLOUR*, i32)*, i32 (%FILE_HANDLE*, %COLOUR*, i32*)*, void (%IMAGE*, i8*)*, void (%FILE_HANDLE*)* }

- %FINISH = type { float, float, float, float, float, float, float, float, float, float, float, float, float, float, float, %BBOX_VECT, %BBOX_VECT }

- %FOG = type { i32, double, double, double, %COLOUR, %VECTOR, %TURB*, float, %FOG* }

- %FRAME = type { %CAMERA*, i32, i32, i32, %LIGHT_SOURCE*, %OBJECT*, double, double, %COLOUR, %COLOUR, %COLOUR, %IMEDIA*, %FOG*, %RAINBOW*, %SKYSPHERE* }

- %FRAMESEQ = type { i32, double, i32, i32, double, i32, i32, double, i32, double, i32, double, i32, i32 }

- %IMAGE = type { i32, i32, i32, i32, i32, i16, i16, %VECTOR, float, float, i32, i32, i16, %IMAGE_COLOUR*, { %IMAGE_LINE*, i8** } }

- %IMAGE_COLOUR = type { i16, i16, i16, i16, i16 }

- %IMAGE_LINE = type { i8*, i8*, i8*, i8* }

- %IMEDIA = type { i32, i32, i32, i32, i32, double, double, i32, i32, i32, i32, %COLOUR, %COLOUR, %COLOUR, %COLOUR, double, double, double, double*, %PIGMENT*, %IMEDIA* }

- %INSIDE_METHOD = type i32 (double*, %OBJECT*)*

- %INTERIOR = type { i32, i32, float, float, float, float, float, %IMEDIA* }

- %INTERSECTION = type { double, %VECTOR, %VECTOR, %OBJECT*, i32, i32, double, double, i8* }

- %INVERT_METHOD = type void (%OBJECT*)*

- %ISTACK = type { %ISTACK*, %INTERSECTION*, i32 }

- %LIGHT_SOURCE = type { %METHODS*, i32, %OBJECT*, %TEXTURE*, %INTERIOR*, %OBJECT*, %OBJECT*, %BBOX, i32, %OBJECT*, %COLOUR, %VECTOR, %VECTOR, %VECTOR, %VECTOR, %VECTOR, double, double, double, double, double, %LIGHT_SOURCE*, i8, i8, i8, i8, i32, i32, i32, i32, i32, %COLOUR**, %OBJECT*, [6 x %PROJECT_TREE_NODE*] }

- %MATRIX = type [4 x %VECTOR_4D]

- %METHODS = type { %ALL_INTERSECTIONS_METHOD, %INSIDE_METHOD, %NORMAL_METHOD, %COPY_METHOD, %ROTATE_METHOD, %ROTATE_METHOD, %ROTATE_METHOD, %TRANSFORM_METHOD, %DESTROY_METHOD, %DESTROY_METHOD }

- %NORMAL_METHOD = type void (double*, %OBJECT*, %INTERSECTION*)*

- %OBJECT = type { %METHODS*, i32, %OBJECT*, %TEXTURE*, %INTERIOR*, %OBJECT*, %OBJECT*, %BBOX, i32 }

- %Opts = type { i32, i32, i8, i8, i8, i32, [150 x i8], [150 x i8], [150 x i8], [150 x i8], [150 x i8], double, double, i32, i32, double, double, i32, [25 x i8*], i32, i32, i32, double, double, i32, i32, double, double, double, i32, i32, i32, i32, i32, %FRAMESEQ, double, i32, double, double, double, double, double, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, [150 x i8], %SHELLDATA*, [150 x i8], i32, i32 }

- %PIGMENT = type { i16, i16, i16, i32, float, float, float, %WARP*, %TPATTERN*, %BLEND_MAP*, { %DENSITY_FILE*, %IMAGE*, %VECTOR, float, i16, i16, i16, { float, %VECTOR }, %complex.float }, %COLOUR }

- %PRIORITY_QUEUE = type { i32, i32, %QELEM* }

- %PROJECT = type { i32, i32, i32, i32 }

- %PROJECT_QUEUE = type { i32, i32, %PROJECT_TREE_NODE** }

- %PROJECT_TREE_NODE = type { i16, %BBOX_TREE*, %PROJECT, i16, %PROJECT_TREE_NODE** }

- %QELEM = type { double, %BBOX_TREE* }

- %RAINBOW = type { double, double, double, double, double, double, double, %VECTOR, %VECTOR, %VECTOR, %PIGMENT*, %RAINBOW* }

- %RAY = type { %VECTOR, %VECTOR, i32, [100 x %INTERIOR*] }

- %RAYINFO = type { %VECTOR, %VECTOR, %VECTORI, %VECTORI }

- %RGB = type [3 x float]

- %ROTATE_METHOD = type void (%OBJECT*, double*, %TRANSFORM*)*

- %SCALE_METHOD = type void (%OBJECT*, double*, %TRANSFORM*)*

- %SHELLDATA = type { i32, i32, [250 x i8] }

- %SKYSPHERE = type { i32, %PIGMENT**, %TRANSFORM* }

- %SNGL_VECT = type [3 x float]

- %TEXTURE = type { i16, i16, i16, i32, float, float, float, %WARP*, %TPATTERN*, %BLEND_MAP*, { %DENSITY_FILE*, %IMAGE*, %VECTOR, float, i16, i16, i16, { float, %VECTOR }, %complex.float }, %TEXTURE*, %PIGMENT*, %TNORMAL*, %FINISH*, %TEXTURE*, i32 }

- %TNORMAL = type { i16, i16, i16, i32, float, float, float, %WARP*, %TPATTERN*, %BLEND_MAP*, { %DENSITY_FILE*, %IMAGE*, %VECTOR, float, i16, i16, i16, { float, %VECTOR }, %complex.float }, float }

- %TPATTERN = type { i16, i16, i16, i32, float, float, float, %WARP*, %TPATTERN*, %BLEND_MAP*, { %DENSITY_FILE*, %IMAGE*, %VECTOR, float, i16, i16, i16, { float, %VECTOR }, %complex.float } }

- %TRANSFORM = type { %MATRIX, %MATRIX }

- %TRANSFORM_METHOD = type void (%OBJECT*, %TRANSFORM*)*

- %TRANSLATE_METHOD = type void (%OBJECT*, double*, %TRANSFORM*)*

- %TURB = type { i16, %WARP*, %VECTOR, i32, float, float }

- %UV_VECT = type [2 x double]

- %VECTOR = type [3 x double]

- %VECTORI = type [3 x i32]

- %VECTOR_4D = type [4 x double]

- %WARP = type { i16, %WARP* }

- %__FILE = type { i32, i8*, i8*, i8, i8, i32, i32, i32 }

- %_h_val = type { [2 x i32], double }

- %complex.float = type { float, float }

-; Test to make sure that the 'internal' tag is not lost!

-;

-; RUN: llvm-as < %s | llvm-dis | grep internal

-

-declare void @foo()

-

-define internal void @foo() {

- ret void

-}

-; Make sure we don't get an assertion failure, even though this is a parse

-; error

-; RUN: not llvm-as %s -o /dev/null |& grep {'@foo' defined with}

-

-%ty = type void (i32)

-

-declare %ty* @foo()

-

-define void @test() {

- call %ty* @foo( ) ; <%ty*>:0 [#uses=0]

- ret void

-}

-

-%ty = type void (i32)

-

-declare %ty* @foo()

- call %ty* ()* @foo( ) ; <%ty*>:1 [#uses=0]

-; RUN: llvm-as %s -o /dev/null

-

-declare i32 @"ArrayRef"([100 x i32] * %Array)

-

-define i32 @"ArrayRef"([100 x i32] * %Array) {

- ret i32 0

-}

-%T = type i32

-; RUN: not llvm-as < %s >/dev/null |& grep {constant expression type mismatch}

-; RUN: not llvm-as < %s > /dev/null |& grep {constant expression type mismatch}

-; RUN: llvm-as %s -o /dev/null

-

-%MidFnTy = type void (%MidFnTy*)

-; RUN: llvm-as %s -o /dev/null

-

-%T = type i32

-@X = global i32* null ; <i32**> [#uses=0]

-@Y = global i32* null ; <i32**> [#uses=0]

-

-; RUN: not llvm-as %s -o /dev/null |& grep {use of undefined type named 'struct.D_Scope'}

-; END.

-

-@d_reduction_0_dparser_gram = global {

- i32 (i8*, i8**, i32, i32, {

- %struct.Grammar*, void (\4, %struct.d_loc_t*, i8**)*, %struct.D_Scope*,

- void (\4)*, { i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }* (\4, i32, { i32, %struct.d_loc_t, i8*, i8*,

- %struct.D_Scope*, void (\9, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }**)*,

- void ({ i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*,

- %struct.Grammar*, %struct.ParseNode_User }*)*,

- %struct.d_loc_t, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32,

- i32 }*)*,

- i32 (i8*, i8**, i32, i32, { %struct.Grammar*,

- void (\4, %struct.d_loc_t*, i8**)*, %struct.D_Scope*, void (\4)*, {

- i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }* (\4, i32, { i32, %struct.d_loc_t, i8*, i8*,

- %struct.D_Scope*, void (\9, %struct.d_loc_t*, i8**)*,

- %struct.Grammar*, %struct.ParseNode_User }**)*,

- void ({ i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }*)*, %struct.d_loc_t, i32, i32, i32, i32,

- i32, i32, i32, i32, i32, i32, i32, i32 }*)** }

-

- { i32 (i8*, i8**, i32, i32, {

- %struct.Grammar*, void (\4, %struct.d_loc_t*, i8**)*,

- %struct.D_Scope*, void (\4)*, {

- i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User

- }* (\4, i32, { i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\9, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }**)*,

- void ({ i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }*)*,

- %struct.d_loc_t, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32,

- i32, i32 }*)* null,

- i32 (i8*, i8**, i32, i32, {

- %struct.Grammar*, void (\4, %struct.d_loc_t*, i8**)*,

- %struct.D_Scope*, void (\4)*, { i32, %struct.d_loc_t, i8*, i8*,

- %struct.D_Scope*, void (\8, %struct.d_loc_t*, i8**)*,

- %struct.Grammar*, %struct.ParseNode_User }* (\4, i32, { i32,

- %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\9, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }**)*,

- void ({ i32, %struct.d_loc_t, i8*, i8*, %struct.D_Scope*,

- void (\8, %struct.d_loc_t*, i8**)*, %struct.Grammar*,

- %struct.ParseNode_User }*)*, %struct.d_loc_t, i32, i32, i32,

- i32, i32, i32, i32, i32, i32, i32, i32, i32 }*)** null

- }

-; RUN: not llvm-as %s -o /dev/null |& grep {invalid type for null constant}

-; RUN: llvm-as < %s | llvm-dis | llvm-as

-

-; Ensure that the asm writer emits types before both operands of the

-; store, even though they can be the same.

-

-%RecTy = type %RecTy*

-

-define void @foo() {

- %A = malloc %RecTy ; <%RecTy> [#uses=1]

- %B = malloc %RecTy ; <%RecTy> [#uses=1]

- store %RecTy %B, %RecTy %A

- ret void

-}

-

-; RUN: llvm-as < %s | llvm-dis | grep {tail call void.*sret null}

-

-declare void @foo({ }* sret , ...)

-

-define void @bar() {

- tail call void ({ }* sret , ...)* @foo( { }* null sret , i32 0 )

- ret void

-}

-; Test that upgrading zext/sext attributes to zeroext and signext

-; works correctly.

-; PR1553

-; RUN: llvm-as < %s > /dev/null

-

-define i32 @bar() {

- %t = call i8 @foo( i8 10 sext ) zext

- %x = zext i8 %t to i32

- ret i32 %x

-}

-

-declare i8 @foo(i8 signext ) zeroext

-; RUN: llvm-as < %s

-

-@FP = weak global i8 (...) signext * null

-; RUN: opt < %s -verify -S | llvm-as -disable-output

-

-define {i32, i8} @foo(i32 %p) {

- ret i32 1, i8 2

-}

-

-define i8 @f2(i32 %p) {

- %c = call {i32, i8} @foo(i32 %p)

- %d = getresult {i32, i8} %c, 1

- %e = add i8 %d, 1

- ret i8 %e

-}

-

-define i32 @f3(i32 %p) {

- %c = invoke {i32, i8} @foo(i32 %p)

- to label %L unwind label %L2

- L:

- %d = getresult {i32, i8} %c, 0

- ret i32 %d

- L2:

- ret i32 0

-}

-; RUN: llvm-as < %s | llvm-dis | llvm-as | llvm-dis

-; PR2733

-

-%t1 = type i32

-%t2 = type { %t1 }

-@i1 = constant %t2 { %t1 15 }

-type i32

- bitcast %0* null to i32*

-; RUN: llvm-as < %s | llvm-dis | not grep {i32 @llvm\\.ct}

-; RUN: llvm-as < %s | llvm-dis | \

-; RUN: not grep {llvm\\.part\\.set\\.i\[0-9\]*\\.i\[0-9\]*\\.i\[0-9\]*}

-; RUN: llvm-as < %s | llvm-dis | \

-; RUN: not grep {llvm\\.part\\.select\\.i\[0-9\]*\\.i\[0-9\]*}

-; RUN: llvm-as < %s | llvm-dis | \

-; RUN: not grep {llvm\\.bswap\\.i\[0-9\]*\\.i\[0-9\]*}

-; RUN: llvm-as < %s | llvm-dis | \

-; RUN: not grep {llvm\\.x86\\.sse2\\.loadu}

-; RUN: llvm-as < %s | llvm-dis | \

-; RUN: grep {llvm\\.x86\\.mmx\\.ps} | grep {x86_mmx} | count 16

-declare i32 @llvm.ctpop.i28(i28 %val)

-declare i32 @llvm.cttz.i29(i29 %val)

-declare i32 @llvm.ctlz.i30(i30 %val)

-

-define i32 @test_ct(i32 %A) {

- %c1 = call i32 @llvm.ctpop.i28(i28 1234)

- %c2 = call i32 @llvm.cttz.i29(i29 2345)

- %c3 = call i32 @llvm.ctlz.i30(i30 3456)

- %r1 = add i32 %c1, %c2

- %r2 = add i32 %r1, %c3

- ret i32 %r2

-}

-

-declare i32 @llvm.part.set.i32.i32.i32(i32 %x, i32 %rep, i32 %hi, i32 %lo)

-declare i16 @llvm.part.set.i16.i16.i16(i16 %x, i16 %rep, i32 %hi, i32 %lo)

-define i32 @test_part_set(i32 %A, i16 %B) {

- %a = call i32 @llvm.part.set.i32.i32.i32(i32 %A, i32 27, i32 8, i32 0)

- %b = call i16 @llvm.part.set.i16.i16.i16(i16 %B, i16 27, i32 8, i32 0)

- %c = zext i16 %b to i32

- %d = add i32 %a, %c

- ret i32 %d

-}

-

-declare i32 @llvm.part.select.i32.i32(i32 %x, i32 %hi, i32 %lo)

-declare i16 @llvm.part.select.i16.i16(i16 %x, i32 %hi, i32 %lo)

-define i32 @test_part_select(i32 %A, i16 %B) {

- %a = call i32 @llvm.part.select.i32.i32(i32 %A, i32 8, i32 0)

- %b = call i16 @llvm.part.select.i16.i16(i16 %B, i32 8, i32 0)

- %c = zext i16 %b to i32

- %d = add i32 %a, %c

- ret i32 %d

-}

-

-declare i32 @llvm.bswap.i32.i32(i32 %x)

-declare i16 @llvm.bswap.i16.i16(i16 %x)

-define i32 @test_bswap(i32 %A, i16 %B) {

- %a = call i32 @llvm.bswap.i32.i32(i32 %A)

- %b = call i16 @llvm.bswap.i16.i16(i16 %B)

- %c = zext i16 %b to i32

- %d = add i32 %a, %c

- ret i32 %d

-}

-

-declare <4 x i16> @llvm.x86.mmx.psra.w(<4 x i16>, <2 x i32>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psll.w(<4 x i16>, <2 x i32>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psrl.w(<4 x i16>, <2 x i32>) nounwind readnone

-define void @sh16(<4 x i16> %A, <2 x i32> %B) {

- %r1 = call <4 x i16> @llvm.x86.mmx.psra.w( <4 x i16> %A, <2 x i32> %B ) ; <<4 x i16>> [#uses=0]

- %r2 = call <4 x i16> @llvm.x86.mmx.psll.w( <4 x i16> %A, <2 x i32> %B ) ; <<4 x i16>> [#uses=0]

- %r3 = call <4 x i16> @llvm.x86.mmx.psrl.w( <4 x i16> %A, <2 x i32> %B ) ; <<4 x i16>> [#uses=0]

- ret void

-}

-

-declare <2 x i32> @llvm.x86.mmx.psra.d(<2 x i32>, <2 x i32>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psll.d(<2 x i32>, <2 x i32>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psrl.d(<2 x i32>, <2 x i32>) nounwind readnone

-define void @sh32(<2 x i32> %A, <2 x i32> %B) {

- %r1 = call <2 x i32> @llvm.x86.mmx.psra.d( <2 x i32> %A, <2 x i32> %B ) ; <<2 x i32>> [#uses=0]

- %r2 = call <2 x i32> @llvm.x86.mmx.psll.d( <2 x i32> %A, <2 x i32> %B ) ; <<2 x i32>> [#uses=0]

- %r3 = call <2 x i32> @llvm.x86.mmx.psrl.d( <2 x i32> %A, <2 x i32> %B ) ; <<2 x i32>> [#uses=0]

- ret void

-}

-

-declare <1 x i64> @llvm.x86.mmx.psll.q(<1 x i64>, <2 x i32>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.psrl.q(<1 x i64>, <2 x i32>) nounwind readnone

-define void @sh64(<1 x i64> %A, <2 x i32> %B) {

- %r1 = call <1 x i64> @llvm.x86.mmx.psll.q( <1 x i64> %A, <2 x i32> %B ) ; <<1 x i64>> [#uses=0]

- %r2 = call <1 x i64> @llvm.x86.mmx.psrl.q( <1 x i64> %A, <2 x i32> %B ) ; <<1 x i64>> [#uses=0]

- ret void

-}

-; Tests to make sure MMX intrinsics are automatically upgraded.

-; RUN: llvm-as < %s | llvm-dis -o %t

-; RUN: grep {llvm\\.x86\\.mmx} %t | not grep {\\\<1 x i64\\\>}

-; RUN: grep {llvm\\.x86\\.mmx} %t | not grep {\\\<2 x i32\\\>}

-; RUN: grep {llvm\\.x86\\.mmx} %t | not grep {\\\<4 x i16\\\>}

-; RUN: grep {llvm\\.x86\\.mmx} %t | not grep {\\\<8 x i8\\\>}

-; RUN: grep {llvm\\.x86\\.sse\\.pshuf\\.w} %t | not grep i32

-

-; Addition

-declare <8 x i8> @llvm.x86.mmx.padd.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.padd.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.padd.d(<2 x i32>, <2 x i32>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.padd.q(<1 x i64>, <1 x i64>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.padds.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.padds.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.paddus.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.paddus.w(<4 x i16>, <4 x i16>) nounwind readnone

-define void @add(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D,

- <2 x i32> %E, <2 x i32> %F, <1 x i64> %G, <1 x i64> %H) {

- %r1 = call <8 x i8> @llvm.x86.mmx.padd.b(<8 x i8> %A, <8 x i8> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.padd.w(<4 x i16> %C, <4 x i16> %D)

- %r3 = call <2 x i32> @llvm.x86.mmx.padd.d(<2 x i32> %E, <2 x i32> %F)

- %r4 = call <1 x i64> @llvm.x86.mmx.padd.q(<1 x i64> %G, <1 x i64> %H)

- %r5 = call <8 x i8> @llvm.x86.mmx.padds.b(<8 x i8> %A, <8 x i8> %B)

- %r6 = call <4 x i16> @llvm.x86.mmx.padds.w(<4 x i16> %C, <4 x i16> %D)

- %r7 = call <8 x i8> @llvm.x86.mmx.paddus.b(<8 x i8> %A, <8 x i8> %B)

- %r8 = call <4 x i16> @llvm.x86.mmx.paddus.w(<4 x i16> %C, <4 x i16> %D)

- ret void

-}

-

-; Subtraction

-declare <8 x i8> @llvm.x86.mmx.psub.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psub.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psub.d(<2 x i32>, <2 x i32>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.psub.q(<1 x i64>, <1 x i64>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.psubs.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psubs.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.psubus.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psubus.w(<4 x i16>, <4 x i16>) nounwind readnone

-define void @sub(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D,

- <2 x i32> %E, <2 x i32> %F, <1 x i64> %G, <1 x i64> %H) {

- %r1 = call <8 x i8> @llvm.x86.mmx.psub.b(<8 x i8> %A, <8 x i8> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.psub.w(<4 x i16> %C, <4 x i16> %D)

- %r3 = call <2 x i32> @llvm.x86.mmx.psub.d(<2 x i32> %E, <2 x i32> %F)

- %r4 = call <1 x i64> @llvm.x86.mmx.psub.q(<1 x i64> %G, <1 x i64> %H)

- %r5 = call <8 x i8> @llvm.x86.mmx.psubs.b(<8 x i8> %A, <8 x i8> %B)

- %r6 = call <4 x i16> @llvm.x86.mmx.psubs.w(<4 x i16> %C, <4 x i16> %D)

- %r7 = call <8 x i8> @llvm.x86.mmx.psubus.b(<8 x i8> %A, <8 x i8> %B)

- %r8 = call <4 x i16> @llvm.x86.mmx.psubus.w(<4 x i16> %C, <4 x i16> %D)

- ret void

-}

-

-; Multiplication

-declare <4 x i16> @llvm.x86.mmx.pmulh.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pmull.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pmulhu.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pmulu.dq(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.pmadd.wd(<4 x i16>, <4 x i16>) nounwind readnone

-define void @mul(<4 x i16> %A, <4 x i16> %B) {

- %r1 = call <4 x i16> @llvm.x86.mmx.pmulh.w(<4 x i16> %A, <4 x i16> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.pmull.w(<4 x i16> %A, <4 x i16> %B)

- %r3 = call <4 x i16> @llvm.x86.mmx.pmulhu.w(<4 x i16> %A, <4 x i16> %B)

- %r4 = call <4 x i16> @llvm.x86.mmx.pmulu.dq(<4 x i16> %A, <4 x i16> %B)

- %r5 = call <2 x i32> @llvm.x86.mmx.pmadd.wd(<4 x i16> %A, <4 x i16> %B)

- ret void

-}

-

-; Bitwise operations

-declare <1 x i64> @llvm.x86.mmx.pand(<1 x i64>, <1 x i64>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.pandn(<1 x i64>, <1 x i64>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.por(<1 x i64>, <1 x i64>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.pxor(<1 x i64>, <1 x i64>) nounwind readnone

-define void @bit(<1 x i64> %A, <1 x i64> %B) {

- %r1 = call <1 x i64> @llvm.x86.mmx.pand(<1 x i64> %A, <1 x i64> %B)

- %r2 = call <1 x i64> @llvm.x86.mmx.pandn(<1 x i64> %A, <1 x i64> %B)

- %r3 = call <1 x i64> @llvm.x86.mmx.por(<1 x i64> %A, <1 x i64> %B)

- %r4 = call <1 x i64> @llvm.x86.mmx.pxor(<1 x i64> %A, <1 x i64> %B)

- ret void

-}

-

-; Averages

-declare <8 x i8> @llvm.x86.mmx.pavg.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pavg.w(<4 x i16>, <4 x i16>) nounwind readnone

-define void @avg(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D) {

- %r1 = call <8 x i8> @llvm.x86.mmx.pavg.b(<8 x i8> %A, <8 x i8> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.pavg.w(<4 x i16> %C, <4 x i16> %D)

- ret void

-}

-

-; Maximum

-declare <8 x i8> @llvm.x86.mmx.pmaxu.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pmaxs.w(<4 x i16>, <4 x i16>) nounwind readnone

-define void @max(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D) {

- %r1 = call <8 x i8> @llvm.x86.mmx.pmaxu.b(<8 x i8> %A, <8 x i8> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.pmaxs.w(<4 x i16> %C, <4 x i16> %D)

- ret void

-}

-

-; Minimum

-declare <8 x i8> @llvm.x86.mmx.pminu.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pmins.w(<4 x i16>, <4 x i16>) nounwind readnone

-define void @min(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D) {

- %r1 = call <8 x i8> @llvm.x86.mmx.pminu.b(<8 x i8> %A, <8 x i8> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.pmins.w(<4 x i16> %C, <4 x i16> %D)

- ret void

-}

-

-; Packed sum of absolute differences

-declare <4 x i16> @llvm.x86.mmx.psad.bw(<8 x i8>, <8 x i8>) nounwind readnone

-define void @psad(<8 x i8> %A, <8 x i8> %B) {

- %r1 = call <4 x i16> @llvm.x86.mmx.psad.bw(<8 x i8> %A, <8 x i8> %B)

- ret void

-}

-

-; Shift left

-declare <4 x i16> @llvm.x86.mmx.psll.w(<4 x i16>, <1 x i64>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psll.d(<2 x i32>, <1 x i64>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.psll.q(<1 x i64>, <1 x i64>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pslli.w(<4 x i16>, i32) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.pslli.d(<2 x i32>, i32) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.pslli.q(<1 x i64>, i32) nounwind readnone

-define void @shl(<4 x i16> %A, <2 x i32> %B, <1 x i64> %C, i32 %D) {

- %r1 = call <4 x i16> @llvm.x86.mmx.psll.w(<4 x i16> %A, <1 x i64> %C)

- %r2 = call <2 x i32> @llvm.x86.mmx.psll.d(<2 x i32> %B, <1 x i64> %C)

- %r3 = call <1 x i64> @llvm.x86.mmx.psll.q(<1 x i64> %C, <1 x i64> %C)

- %r4 = call <4 x i16> @llvm.x86.mmx.pslli.w(<4 x i16> %A, i32 %D)

- %r5 = call <2 x i32> @llvm.x86.mmx.pslli.d(<2 x i32> %B, i32 %D)

- %r6 = call <1 x i64> @llvm.x86.mmx.pslli.q(<1 x i64> %C, i32 %D)

- ret void

-}

-

-; Shift right logical

-declare <4 x i16> @llvm.x86.mmx.psrl.w(<4 x i16>, <1 x i64>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psrl.d(<2 x i32>, <1 x i64>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.psrl.q(<1 x i64>, <1 x i64>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psrli.w(<4 x i16>, i32) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psrli.d(<2 x i32>, i32) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.psrli.q(<1 x i64>, i32) nounwind readnone

-define void @shr(<4 x i16> %A, <2 x i32> %B, <1 x i64> %C, i32 %D) {

- %r1 = call <4 x i16> @llvm.x86.mmx.psrl.w(<4 x i16> %A, <1 x i64> %C)

- %r2 = call <2 x i32> @llvm.x86.mmx.psrl.d(<2 x i32> %B, <1 x i64> %C)

- %r3 = call <1 x i64> @llvm.x86.mmx.psrl.q(<1 x i64> %C, <1 x i64> %C)

- %r4 = call <4 x i16> @llvm.x86.mmx.psrli.w(<4 x i16> %A, i32 %D)

- %r5 = call <2 x i32> @llvm.x86.mmx.psrli.d(<2 x i32> %B, i32 %D)

- %r6 = call <1 x i64> @llvm.x86.mmx.psrli.q(<1 x i64> %C, i32 %D)

- ret void

-}

-

-; Shift right arithmetic

-declare <4 x i16> @llvm.x86.mmx.psra.w(<4 x i16>, <1 x i64>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psra.d(<2 x i32>, <1 x i64>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.psrai.w(<4 x i16>, i32) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.psrai.d(<2 x i32>, i32) nounwind readnone

-define void @sra(<4 x i16> %A, <2 x i32> %B, <1 x i64> %C, i32 %D) {

- %r1 = call <4 x i16> @llvm.x86.mmx.psra.w(<4 x i16> %A, <1 x i64> %C)

- %r2 = call <2 x i32> @llvm.x86.mmx.psra.d(<2 x i32> %B, <1 x i64> %C)

- %r3 = call <4 x i16> @llvm.x86.mmx.psrai.w(<4 x i16> %A, i32 %D)

- %r4 = call <2 x i32> @llvm.x86.mmx.psrai.d(<2 x i32> %B, i32 %D)

- ret void

-}

-

-; Pack/Unpack ops

-declare <8 x i8> @llvm.x86.mmx.packsswb(<4 x i16>, <4 x i16>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.packssdw(<2 x i32>, <2 x i32>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.packuswb(<4 x i16>, <4 x i16>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.punpckhbw(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.punpckhwd(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.punpckhdq(<2 x i32>, <2 x i32>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.punpcklbw(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.punpcklwd(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.punpckldq(<2 x i32>, <2 x i32>) nounwind readnone

-define void @pack_unpack(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D,

- <2 x i32> %E, <2 x i32> %F) {

- %r1 = call <8 x i8> @llvm.x86.mmx.packsswb(<4 x i16> %C, <4 x i16> %D)

- %r2 = call <4 x i16> @llvm.x86.mmx.packssdw(<2 x i32> %E, <2 x i32> %F)

- %r3 = call <8 x i8> @llvm.x86.mmx.packuswb(<4 x i16> %C, <4 x i16> %D)

- %r4 = call <8 x i8> @llvm.x86.mmx.punpckhbw(<8 x i8> %A, <8 x i8> %B)

- %r5 = call <4 x i16> @llvm.x86.mmx.punpckhwd(<4 x i16> %C, <4 x i16> %D)

- %r6 = call <2 x i32> @llvm.x86.mmx.punpckhdq(<2 x i32> %E, <2 x i32> %F)

- %r7 = call <8 x i8> @llvm.x86.mmx.punpcklbw(<8 x i8> %A, <8 x i8> %B)

- %r8 = call <4 x i16> @llvm.x86.mmx.punpcklwd(<4 x i16> %C, <4 x i16> %D)

- %r9 = call <2 x i32> @llvm.x86.mmx.punpckldq(<2 x i32> %E, <2 x i32> %F)

- ret void

-}

-

-; Integer comparison ops

-declare <8 x i8> @llvm.x86.mmx.pcmpeq.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pcmpeq.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.pcmpeq.d(<2 x i32>, <2 x i32>) nounwind readnone

-declare <8 x i8> @llvm.x86.mmx.pcmpgt.b(<8 x i8>, <8 x i8>) nounwind readnone

-declare <4 x i16> @llvm.x86.mmx.pcmpgt.w(<4 x i16>, <4 x i16>) nounwind readnone

-declare <2 x i32> @llvm.x86.mmx.pcmpgt.d(<2 x i32>, <2 x i32>) nounwind readnone

-define void @cmp(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D,

- <2 x i32> %E, <2 x i32> %F) {

- %r1 = call <8 x i8> @llvm.x86.mmx.pcmpeq.b(<8 x i8> %A, <8 x i8> %B)

- %r2 = call <4 x i16> @llvm.x86.mmx.pcmpeq.w(<4 x i16> %C, <4 x i16> %D)

- %r3 = call <2 x i32> @llvm.x86.mmx.pcmpeq.d(<2 x i32> %E, <2 x i32> %F)

- %r4 = call <8 x i8> @llvm.x86.mmx.pcmpgt.b(<8 x i8> %A, <8 x i8> %B)

- %r5 = call <4 x i16> @llvm.x86.mmx.pcmpgt.w(<4 x i16> %C, <4 x i16> %D)

- %r6 = call <2 x i32> @llvm.x86.mmx.pcmpgt.d(<2 x i32> %E, <2 x i32> %F)

- ret void

-}

-

-; Miscellaneous

-declare void @llvm.x86.mmx.maskmovq(<8 x i8>, <8 x i8>, i32*) nounwind readnone

-declare i32 @llvm.x86.mmx.pmovmskb(<8 x i8>) nounwind readnone

-declare void @llvm.x86.mmx.movnt.dq(i32*, <1 x i64>) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.palignr.b(<1 x i64>, <1 x i64>, i8) nounwind readnone

-declare i32 @llvm.x86.mmx.pextr.w(<1 x i64>, i32) nounwind readnone

-declare <1 x i64> @llvm.x86.mmx.pinsr.w(<1 x i64>, i32, i32) nounwind readnone

-declare <4 x i16> @llvm.x86.ssse3.pshuf.w(<4 x i16>, i32) nounwind readnone

-define void @misc(<8 x i8> %A, <8 x i8> %B, <4 x i16> %C, <4 x i16> %D,

- <2 x i32> %E, <2 x i32> %F, <1 x i64> %G, <1 x i64> %H,

- i32* %I, i8 %J, i16 %K, i32 %L) {

- call void @llvm.x86.mmx.maskmovq(<8 x i8> %A, <8 x i8> %B, i32* %I)

- %r1 = call i32 @llvm.x86.mmx.pmovmskb(<8 x i8> %A)

- call void @llvm.x86.mmx.movnt.dq(i32* %I, <1 x i64> %G)

- %r2 = call <1 x i64> @llvm.x86.mmx.palignr.b(<1 x i64> %G, <1 x i64> %H, i8 %J)

- %r3 = call i32 @llvm.x86.mmx.pextr.w(<1 x i64> %G, i32 37)

- %r4 = call <1 x i64> @llvm.x86.mmx.pinsr.w(<1 x i64> %G, i32 37, i32 927)

- %r5 = call <4 x i16> @llvm.x86.ssse3.pshuf.w(<4 x i16> %C, i32 37)

- ret void

-}

-@y = global i32* getelementptr ({i32, i32}* @x, i16 42, i32 0)

-; CHECK: @y = global i32* getelementptr (%0* @x, i16 42, i32 0)

-; CHECK: %B = getelementptr %0* %t, i92 %n, i32 0

-; Test to make sure that the 'private' tag is not lost!

-;

-; RUN: llvm-as < %s | llvm-dis | grep private

-

-declare void @foo()

-

-define private void @foo() {

- ret void

-}