maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/CodeGen/mips64-padding-arg.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

72 lines
2.9 KiB
C
Raw Normal View History

Remove compile time PreserveName in favor of a runtime cc1 -discard-value-names option Summary: This flag is enabled by default in the driver when NDEBUG is set. It is forwarded on the LLVMContext to discard all value names (but GlobalValue) for performance purpose. This an improved version of D18024 Reviewers: echristo, chandlerc Subscribers: cfe-commits Differential Revision: http://reviews.llvm.org/D18127 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 263394
2016-03-14 05:05:23 +08:00
// RUN: %clang_cc1 -triple mipsel-unknown-linux -O3 -S -o - -emit-llvm %s | FileCheck %s -check-prefix=O32
// RUN: %clang_cc1 -triple mips64el-unknown-linux -O3 -S -target-abi n64 -o - -emit-llvm %s | FileCheck %s -check-prefix=N64
// RUN: %clang_cc1 -triple mipsel-unknown-linux -target-feature "+fp64" -O3 -S -o - -emit-llvm %s | FileCheck %s -check-prefix=O32
Add field PaddingType to ABIArgInfo which specifies the type of padding that is inserted before the real argument. Padding is needed to ensure the backend reads from or writes to the correct argument slots when the original alignment of a byval structure is unavailable due to flattening. llvm-svn: 147699
2012-01-07 08:25:33 +08:00
typedef struct {
double d;
long double ld;
} S0;
Insert padding before unaligned long double arguments. llvm-svn: 147791
2012-01-10 03:31:25 +08:00
// Insert padding to ensure arguments of type S0 are aligned to 16-byte boundaries.
Add field PaddingType to ABIArgInfo which specifies the type of padding that is inserted before the real argument. Padding is needed to ensure the backend reads from or writes to the correct argument slots when the original alignment of a byval structure is unavailable due to flattening. llvm-svn: 147699
2012-01-07 08:25:33 +08:00
Revert "[CodeGenModule] Assume dso_local for -fpic -fno-semantic-interposition" This reverts commit 789a46f2d742e11edaade28cb59a0f4d2a1d770e. Accidentally committed.
2020-02-04 02:09:39 +08:00
// N64-LABEL: define void @foo1(i32 signext %a0, i64 %0, double inreg %a1.coerce0, i64 inreg %a1.coerce1, i64 inreg %a1.coerce2, i64 inreg %a1.coerce3, double inreg %a2.coerce0, i64 inreg %a2.coerce1, i64 inreg %a2.coerce2, i64 inreg %a2.coerce3, i32 signext %b, i64 %1, double inreg %a3.coerce0, i64 inreg %a3.coerce1, i64 inreg %a3.coerce2, i64 inreg %a3.coerce3)
[mips] Mark aggregate arguments passed in registers with the inreg attribute Summary: This allows us to easily identify them in the backend which in turn allows us to handle them correctly for big-endian targets (where they must be shifted into the upper bits of the register). Depends on D5961 Reviewers: atanasyan Reviewed By: atanasyan Subscribers: cfe-commits, theraven Differential Revision: http://reviews.llvm.org/D5962 llvm-svn: 220566
2014-10-24 23:30:16 +08:00
// N64: tail call void @foo2(i32 signext 1, i32 signext 2, i32 signext %a0, i64 undef, double inreg %a1.coerce0, i64 inreg %a1.coerce1, i64 inreg %a1.coerce2, i64 inreg %a1.coerce3, double inreg %a2.coerce0, i64 inreg %a2.coerce1, i64 inreg %a2.coerce2, i64 inreg %a2.coerce3, i32 signext 3, i64 undef, double inreg %a3.coerce0, i64 inreg %a3.coerce1, i64 inreg %a3.coerce2, i64 inreg %a3.coerce3)
// N64: declare void @foo2(i32 signext, i32 signext, i32 signext, i64, double inreg, i64 inreg, i64 inreg, i64 inreg, double inreg, i64 inreg, i64 inreg, i64 inreg, i32 signext, i64, double inreg, i64 inreg, i64 inreg, i64 inreg)
Add field PaddingType to ABIArgInfo which specifies the type of padding that is inserted before the real argument. Padding is needed to ensure the backend reads from or writes to the correct argument slots when the original alignment of a byval structure is unavailable due to flattening. llvm-svn: 147699
2012-01-07 08:25:33 +08:00
extern void foo2(int, int, int, S0, S0, int, S0);
void foo1(int a0, S0 a1, S0 a2, int b, S0 a3) {
foo2(1, 2, a0, a1, a2, 3, a3);
}
Insert padding before unaligned long double arguments. llvm-svn: 147791
2012-01-10 03:31:25 +08:00
// Insert padding before long double argument.
//
Revert "[CodeGenModule] Assume dso_local for -fpic -fno-semantic-interposition" This reverts commit 789a46f2d742e11edaade28cb59a0f4d2a1d770e. Accidentally committed.
2020-02-04 02:09:39 +08:00
// N64-LABEL: define void @foo3(i32 signext %a0, i64 %0, fp128 %a1)
[mips] Promote all integral/enumeration types to the GPR width Summary: Ensure all integral/enumeration types are appropriately annotated with signext/zeroext. In particular, i32 now has these attributes when using the N32/N64 ABI. This paves the way for accurately representing the way the N32/N64 ABI's promotes integer arguments to i64. Reviewers: atanasyan Reviewed By: atanasyan Subscribers: cfe-commits, theraven Differential Revision: http://reviews.llvm.org/D5961 llvm-svn: 220563
2014-10-24 22:42:42 +08:00
// N64: tail call void @foo4(i32 signext 1, i32 signext 2, i32 signext %a0, i64 undef, fp128 %a1)
// N64: declare void @foo4(i32 signext, i32 signext, i32 signext, i64, fp128)
Insert padding before unaligned long double arguments. llvm-svn: 147791
2012-01-10 03:31:25 +08:00
extern void foo4(int, int, int, long double);
void foo3(int a0, long double a1) {
foo4(1, 2, a0, a1);
}
Take into account the pointer to an aggregate that is passed as a hidden argument when Offset is initialized. llvm-svn: 147986
2012-01-12 09:10:09 +08:00
// Insert padding after hidden argument.
//
[CodeGen] Add an alignment attribute to all sret parameters This fixes a miscompile when the parameter is actually underaligned. rdar://58316406 Differential revision: https://reviews.llvm.org/D74183
2020-03-25 00:36:19 +08:00
// N64-LABEL: define void @foo5(%struct.S0* noalias sret align 16 %agg.result, i64 %0, fp128 %a0)
// N64: call void @foo6(%struct.S0* sret align 16 %agg.result, i32 signext 1, i32 signext 2, i64 undef, fp128 %a0)
// N64: declare void @foo6(%struct.S0* sret align 16, i32 signext, i32 signext, i64, fp128)
Take into account the pointer to an aggregate that is passed as a hidden argument when Offset is initialized. llvm-svn: 147986
2012-01-12 09:10:09 +08:00
extern S0 foo6(int, int, long double);
S0 foo5(long double a0) {
Fix test case committed in r147986. llvm-svn: 147993
2012-01-12 09:27:24 +08:00
return foo6(1, 2, a0);
Take into account the pointer to an aggregate that is passed as a hidden argument when Offset is initialized. llvm-svn: 147986
2012-01-12 09:10:09 +08:00
}
[mips] Do not emit i32 padding if target ABI is O32. This was causing backend to pass floating point arguments to be passed in integer registers. llvm-svn: 173375
2013-01-25 05:47:33 +08:00
// Do not insert padding if ABI is O32.
//
Revert "[CodeGenModule] Assume dso_local for -fpic -fno-semantic-interposition" This reverts commit 789a46f2d742e11edaade28cb59a0f4d2a1d770e. Accidentally committed.
2020-02-04 02:09:39 +08:00
// O32-LABEL: define void @foo7(float %a0, double %a1)
[mips] Do not emit i32 padding if target ABI is O32. This was causing backend to pass floating point arguments to be passed in integer registers. llvm-svn: 173375
2013-01-25 05:47:33 +08:00
// O32: declare void @foo8(float, double)
extern void foo8(float, double);
void foo7(float a0, double a1) {
foo8(a0 + 1.0f, a1 + 2.0);
}
Revert "[CodeGenModule] Assume dso_local for -fpic -fno-semantic-interposition" This reverts commit 789a46f2d742e11edaade28cb59a0f4d2a1d770e. Accidentally committed.
2020-02-04 02:09:39 +08:00
// O32-LABEL: define void @foo9()
[mips] Promote all integral/enumeration types to the GPR width Summary: Ensure all integral/enumeration types are appropriately annotated with signext/zeroext. In particular, i32 now has these attributes when using the N32/N64 ABI. This paves the way for accurately representing the way the N32/N64 ABI's promotes integer arguments to i64. Reviewers: atanasyan Reviewed By: atanasyan Subscribers: cfe-commits, theraven Differential Revision: http://reviews.llvm.org/D5961 llvm-svn: 220563
2014-10-24 22:42:42 +08:00
// O32: declare void @foo10(i32 signext, i32
[mips] Align the stack to 16-bytes for -mfp64. llvm-svn: 193640
2013-10-30 03:00:35 +08:00
typedef struct __attribute__((aligned(16))) {
int a;
} S16;
S16 s16;
void foo10(int, S16);
void foo9(void) {
foo10(1, s16);
}