
This patch fixes a crash when trying to emit a constant compound literal. For C++ Clang evaluates either casts or binary operations at translation time, but doesn't pass on the InConstantContext information that was inferred when parsing the statement. Because of this, strict FP evaluation (-ftrapping-math) which shouldn't be in effect yet, then causes checkFloatingpointResult to return false, which in tryEmitGlobalCompoundLiteral will trigger an assert that the compound literal wasn't constant. The discussion here around 'manifestly constant evaluated contexts' was very helpful to me when trying to understand what LLVM's position is on what evaluation context should be in effect, together with the explanatory text in that patch itself: https://reviews.llvm.org/D87528 Reviewed By: rjmccall, DavidSpickett Differential Revision: https://reviews.llvm.org/D131555
193 lines
6.2 KiB
C
193 lines
6.2 KiB
C
// setting strict FP behaviour in the run line below tests that the compiler
|
|
// does the right thing for global compound literals (compoundliteral test)
|
|
// RUN: %clang_cc1 -no-opaque-pointers -triple i386-pc-linux-gnu -ffreestanding -Wno-pointer-to-int-cast -Wno-int-conversion -ffp-exception-behavior=strict -emit-llvm -o - %s | FileCheck %s
|
|
|
|
#include <stdint.h>
|
|
|
|
// Brace-enclosed string array initializers
|
|
char a[] = { "asdf" };
|
|
// CHECK: @a ={{.*}} global [5 x i8] c"asdf\00"
|
|
|
|
char a2[2][5] = { "asdf" };
|
|
// CHECK: @a2 ={{.*}} global [2 x [5 x i8]] {{\[}}[5 x i8] c"asdf\00", [5 x i8] zeroinitializer]
|
|
|
|
// Double-implicit-conversions of array/functions (not legal C, but
|
|
// clang accepts it for gcc compat).
|
|
intptr_t b = a;
|
|
int c(void);
|
|
void *d = c;
|
|
intptr_t e = c;
|
|
|
|
int f, *g = __extension__ &f, *h = (1 != 1) ? &f : &f;
|
|
|
|
union s2 {
|
|
struct {
|
|
struct { } *f0;
|
|
} f0;
|
|
};
|
|
|
|
int g0 = (int)(&(((union s2 *) 0)->f0.f0) - 0);
|
|
|
|
// CHECK: @g1x ={{.*}} global { double, double } { double 1.000000e+00{{[0]*}}, double 0.000000e+00{{[0]*}} }
|
|
_Complex double g1x = 1.0f;
|
|
// CHECK: @g1y ={{.*}} global { double, double } { double 0.000000e+00{{[0]*}}, double 1.000000e+00{{[0]*}} }
|
|
_Complex double g1y = 1.0fi;
|
|
// CHECK: @g1 ={{.*}} global { i8, i8 } { i8 1, i8 10 }
|
|
_Complex char g1 = (char) 1 + (char) 10 * 1i;
|
|
// CHECK: @g2 ={{.*}} global { i32, i32 } { i32 1, i32 10 }
|
|
_Complex int g2 = 1 + 10i;
|
|
// CHECK: @g3 ={{.*}} global { float, float } { float 1.000000e+00{{[0]*}}, float 1.000000e+0{{[0]*}}1 }
|
|
_Complex float g3 = 1.0 + 10.0i;
|
|
// CHECK: @g4 ={{.*}} global { double, double } { double 1.000000e+00{{[0]*}}, double 1.000000e+0{{[0]*}}1 }
|
|
_Complex double g4 = 1.0 + 10.0i;
|
|
// CHECK: @g5 ={{.*}} global { i32, i32 } zeroinitializer
|
|
_Complex int g5 = (2 + 3i) == (5 + 7i);
|
|
// CHECK: @g6 ={{.*}} global { double, double } { double -1.100000e+0{{[0]*}}1, double 2.900000e+0{{[0]*}}1 }
|
|
_Complex double g6 = (2.0 + 3.0i) * (5.0 + 7.0i);
|
|
// CHECK: @g7 ={{.*}} global i32 1
|
|
int g7 = (2 + 3i) * (5 + 7i) == (-11 + 29i);
|
|
// CHECK: @g8 ={{.*}} global i32 1
|
|
int g8 = (2.0 + 3.0i) * (5.0 + 7.0i) == (-11.0 + 29.0i);
|
|
// CHECK: @g9 ={{.*}} global i32 0
|
|
int g9 = (2 + 3i) * (5 + 7i) != (-11 + 29i);
|
|
// CHECK: @g10 ={{.*}} global i32 0
|
|
int g10 = (2.0 + 3.0i) * (5.0 + 7.0i) != (-11.0 + 29.0i);
|
|
|
|
// PR5108
|
|
// CHECK: @gv1 ={{.*}} global %struct.anon <{ i32 0, i8 7 }>, align 1
|
|
struct {
|
|
unsigned long a;
|
|
unsigned long b:3;
|
|
} __attribute__((__packed__)) gv1 = { .a = 0x0, .b = 7, };
|
|
|
|
// PR5118
|
|
// CHECK: @gv2 ={{.*}} global %struct.anon.0 <{ i8 1, i8* null }>, align 1
|
|
struct {
|
|
unsigned char a;
|
|
char *b;
|
|
} __attribute__((__packed__)) gv2 = { 1, (void*)0 };
|
|
|
|
// Global references
|
|
// CHECK: @g11.l0 = internal global i32 ptrtoint (i32 ()* @g11 to i32)
|
|
long g11(void) {
|
|
static long l0 = (long) g11;
|
|
return l0;
|
|
}
|
|
|
|
// CHECK: @g12 ={{.*}} global i32 ptrtoint (i8* @g12_tmp to i32)
|
|
static char g12_tmp;
|
|
long g12 = (long) &g12_tmp;
|
|
|
|
// CHECK: @g13 ={{.*}} global [1 x %struct.g13_s0] [%struct.g13_s0 { i32 ptrtoint (i8* @g12_tmp to i32) }]
|
|
struct g13_s0 {
|
|
long a;
|
|
};
|
|
struct g13_s0 g13[] = {
|
|
{ (long) &g12_tmp }
|
|
};
|
|
|
|
// CHECK: @g14 ={{.*}} global i8* inttoptr (i32 100 to i8*)
|
|
void *g14 = (void*) 100;
|
|
|
|
// CHECK: @g15 ={{.*}} global i32 -1
|
|
int g15 = (int) (char) ((void*) 0 + 255);
|
|
|
|
// CHECK: @g16 ={{.*}} global i64 4294967295
|
|
long long g16 = (long long) ((void*) 0xFFFFFFFF);
|
|
|
|
// CHECK: @g17 ={{.*}} global i32* @g15
|
|
int *g17 = (int *) ((long) &g15);
|
|
|
|
// CHECK: @g18.p = internal global [1 x i32*] [i32* @g19]
|
|
void g18(void) {
|
|
extern int g19;
|
|
static int *p[] = { &g19 };
|
|
}
|
|
|
|
// CHECK: @g20.l0 = internal global %struct.g20_s1 { %struct.g20_s0* null, %struct.g20_s0** getelementptr inbounds (%struct.g20_s1, %struct.g20_s1* @g20.l0, i32 0, i32 0) }
|
|
struct g20_s0;
|
|
struct g20_s1 {
|
|
struct g20_s0 *f0, **f1;
|
|
};
|
|
void *g20(void) {
|
|
static struct g20_s1 l0 = { ((void*) 0), &l0.f0 };
|
|
return l0.f1;
|
|
}
|
|
|
|
// PR4108
|
|
struct g21 {int g21;};
|
|
const struct g21 g21 = (struct g21){1};
|
|
|
|
// PR5474
|
|
struct g22 {int x;} __attribute((packed));
|
|
struct g23 {char a; short b; char c; struct g22 d;};
|
|
struct g23 g24 = {1,2,3,4};
|
|
|
|
// CHECK: @g25.g26 = internal global i8* getelementptr inbounds ([4 x i8], [4 x i8]* @[[FUNC:.*]], i32 0, i32 0)
|
|
// CHECK: @[[FUNC]] = private unnamed_addr constant [4 x i8] c"g25\00"
|
|
int g25(void) {
|
|
static const char *g26 = __func__;
|
|
return *g26;
|
|
}
|
|
|
|
// CHECK: @g27.x = internal global i8* bitcast (i8** @g27.x to i8*), align 4
|
|
void g27(void) { // PR8073
|
|
static void *x = &x;
|
|
}
|
|
|
|
void g28(void) {
|
|
typedef long long v1i64 __attribute((vector_size(8)));
|
|
typedef short v12i16 __attribute((vector_size(24)));
|
|
typedef long double v2f80 __attribute((vector_size(24)));
|
|
// CHECK: @g28.a = internal global <1 x i64> <i64 10>
|
|
// CHECK: @g28.b = internal global <12 x i16> <i16 0, i16 0, i16 0, i16 -32768, i16 16383, i16 0, i16 0, i16 0, i16 0, i16 -32768, i16 16384, i16 0>
|
|
// CHECK: @g28.c = internal global <2 x x86_fp80> <x86_fp80 0xK3FFF8000000000000000, x86_fp80 0xK40008000000000000000>, align 32
|
|
static v1i64 a = (v1i64)10LL;
|
|
static v12i16 b = (v12i16)(v2f80){1,2};
|
|
static v2f80 c = (v2f80)(v12i16){0,0,0,-32768,16383,0,0,0,0,-32768,16384,0};
|
|
}
|
|
|
|
// PR13643
|
|
void g29(void) {
|
|
typedef char DCC_PASSWD[2];
|
|
typedef struct
|
|
{
|
|
DCC_PASSWD passwd;
|
|
} DCC_SRVR_NM;
|
|
// CHECK: @g29.a = internal global %struct.DCC_SRVR_NM { [2 x i8] c"@\00" }, align 1
|
|
// CHECK: @g29.b = internal global [1 x i32] [i32 ptrtoint ([5 x i8]* @.str.1 to i32)], align 4
|
|
// CHECK: @g29.c = internal global [1 x i32] [i32 97], align 4
|
|
static DCC_SRVR_NM a = { {"@"} };
|
|
static int b[1] = { "asdf" };
|
|
static int c[1] = { L"a" };
|
|
}
|
|
|
|
// PR21300
|
|
void g30(void) {
|
|
#pragma pack(1)
|
|
static struct {
|
|
int : 1;
|
|
int x;
|
|
} a = {};
|
|
// CHECK: @g30.a = internal global %struct.anon.1 <{ i8 undef, i32 0 }>, align 1
|
|
#pragma pack()
|
|
}
|
|
|
|
void g31(void) {
|
|
#pragma pack(4)
|
|
static struct {
|
|
short a;
|
|
long x;
|
|
short z;
|
|
} a = {23122, -12312731, -312};
|
|
#pragma pack()
|
|
// CHECK: @g31.a = internal global %struct.anon.2 { i16 23122, i32 -12312731, i16 -312 }, align 4
|
|
}
|
|
|
|
// Clang should evaluate this in constant context, so floating point mode should
|
|
// have no effect.
|
|
// CHECK: @.compoundliteral = internal global [1 x float] [float 0x3FB99999A0000000], align 4
|
|
struct { const float *floats; } compoundliteral = {
|
|
(float[1]) { 0.1, },
|
|
};
|