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/libcalls.c

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

128 lines
5.3 KiB
C
Raw Normal View History

Fix FileCheck --check-prefix lines. Various tests had sprung up over the years which had --check-prefix=ABC on the RUN line, but "CHECK-ABC:" later on. This happened to work before, but was strictly incorrect. FileCheck is getting stricter soon though. Patch by Ron Ofir. llvm-svn: 188174
2013-08-12 20:51:05 +08:00
// RUN: %clang_cc1 -fmath-errno -emit-llvm -o - %s -triple i386-unknown-unknown | FileCheck -check-prefix CHECK-YES %s
// RUN: %clang_cc1 -emit-llvm -o - %s -triple i386-unknown-unknown | FileCheck -check-prefix CHECK-NO %s
Restore the sqrt -> llvm.sqrt mapping in fast-math mode This restores the sqrt -> llvm.sqrt mapping, but only in fast-math mode (specifically, when the UnsafeFPMath or NoNaNsFPMath CodeGen options are enabled). The @llvm.sqrt* intrinsics have slightly different semantics from the libm call, specifically, they are undefined when given a non-zero negative number (the libm calls will always return NaN for any negative number). This mapping was removed in r100613, and replaced with a TODO, but at that time the fast-math flags were not yet implemented. Now that we have these, restoring this mapping is important because it will enable autovectorization of sqrt calls in loops (at least in fast-math mode). llvm-svn: 190646
2013-09-13 07:57:55 +08:00
// RUN: %clang_cc1 -menable-unsafe-fp-math -emit-llvm -o - %s -triple i386-unknown-unknown | FileCheck -check-prefix CHECK-FAST %s
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +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
// CHECK-YES-LABEL: define void @test_sqrt
// CHECK-NO-LABEL: define void @test_sqrt
// CHECK-FAST-LABEL: define void @test_sqrt
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
void test_sqrt(float a0, double a1, long double a2) {
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: call float @sqrtf
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set The LLVM sqrt intrinsic definition changed with: D28797 ...so we don't have to use any relaxed FP settings other than errno handling. This patch sidesteps a question raised in PR27435: https://bugs.llvm.org/show_bug.cgi?id=27435 Is a programmer using __builtin_sqrt() invoking the compiler's intrinsic definition of sqrt or the mathlib definition of sqrt? But we have an answer now: the builtin should match the behavior of the libm function including errno handling. Differential Revision: https://reviews.llvm.org/D39204 llvm-svn: 317031
2017-11-01 04:19:39 +08:00
// CHECK-NO: call float @llvm.sqrt.f32(float
// CHECK-FAST: call float @llvm.sqrt.f32(float
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
float l0 = sqrtf(a0);
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: call double @sqrt
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set The LLVM sqrt intrinsic definition changed with: D28797 ...so we don't have to use any relaxed FP settings other than errno handling. This patch sidesteps a question raised in PR27435: https://bugs.llvm.org/show_bug.cgi?id=27435 Is a programmer using __builtin_sqrt() invoking the compiler's intrinsic definition of sqrt or the mathlib definition of sqrt? But we have an answer now: the builtin should match the behavior of the libm function including errno handling. Differential Revision: https://reviews.llvm.org/D39204 llvm-svn: 317031
2017-11-01 04:19:39 +08:00
// CHECK-NO: call double @llvm.sqrt.f64(double
// CHECK-FAST: call double @llvm.sqrt.f64(double
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
double l1 = sqrt(a1);
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: call x86_fp80 @sqrtl
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set The LLVM sqrt intrinsic definition changed with: D28797 ...so we don't have to use any relaxed FP settings other than errno handling. This patch sidesteps a question raised in PR27435: https://bugs.llvm.org/show_bug.cgi?id=27435 Is a programmer using __builtin_sqrt() invoking the compiler's intrinsic definition of sqrt or the mathlib definition of sqrt? But we have an answer now: the builtin should match the behavior of the libm function including errno handling. Differential Revision: https://reviews.llvm.org/D39204 llvm-svn: 317031
2017-11-01 04:19:39 +08:00
// CHECK-NO: call x86_fp80 @llvm.sqrt.f80(x86_fp80
// CHECK-FAST: call x86_fp80 @llvm.sqrt.f80(x86_fp80
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
long double l2 = sqrtl(a2);
}
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: declare float @sqrtf(float)
// CHECK-YES: declare double @sqrt(double)
// CHECK-YES: declare x86_fp80 @sqrtl(x86_fp80)
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set The LLVM sqrt intrinsic definition changed with: D28797 ...so we don't have to use any relaxed FP settings other than errno handling. This patch sidesteps a question raised in PR27435: https://bugs.llvm.org/show_bug.cgi?id=27435 Is a programmer using __builtin_sqrt() invoking the compiler's intrinsic definition of sqrt or the mathlib definition of sqrt? But we have an answer now: the builtin should match the behavior of the libm function including errno handling. Differential Revision: https://reviews.llvm.org/D39204 llvm-svn: 317031
2017-11-01 04:19:39 +08:00
// CHECK-NO: declare float @llvm.sqrt.f32(float)
// CHECK-NO: declare double @llvm.sqrt.f64(double)
// CHECK-NO: declare x86_fp80 @llvm.sqrt.f80(x86_fp80)
Restore the sqrt -> llvm.sqrt mapping in fast-math mode This restores the sqrt -> llvm.sqrt mapping, but only in fast-math mode (specifically, when the UnsafeFPMath or NoNaNsFPMath CodeGen options are enabled). The @llvm.sqrt* intrinsics have slightly different semantics from the libm call, specifically, they are undefined when given a non-zero negative number (the libm calls will always return NaN for any negative number). This mapping was removed in r100613, and replaced with a TODO, but at that time the fast-math flags were not yet implemented. Now that we have these, restoring this mapping is important because it will enable autovectorization of sqrt calls in loops (at least in fast-math mode). llvm-svn: 190646
2013-09-13 07:57:55 +08:00
// CHECK-FAST: declare float @llvm.sqrt.f32(float)
// CHECK-FAST: declare double @llvm.sqrt.f64(double)
// CHECK-FAST: declare x86_fp80 @llvm.sqrt.f80(x86_fp80)
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +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
// CHECK-YES-LABEL: define void @test_pow
// CHECK-NO-LABEL: define void @test_pow
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
void test_pow(float a0, double a1, long double a2) {
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: call float @powf
// CHECK-NO: call float @llvm.pow.f32
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
float l0 = powf(a0, a0);
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: call double @pow
// CHECK-NO: call double @llvm.pow.f64
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
double l1 = pow(a1, a1);
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: call x86_fp80 @powl
// CHECK-NO: call x86_fp80 @llvm.pow.f80
Support IRgen of sqrt -> llvm.sqrt, pow -> llvm.pow. - Define pow[lf]?, sqrt[lf]? as builtins. - Add -fmath-errno option which binds to LangOptions.MathErrno - Add new builtin flag Builtin::Context::isConstWithoutErrno for functions which can be marked as const if errno isn't respected for math functions. Sema automatically marks these functions as const when they are defined, if MathErrno=0. - IRgen uses const attribute on sqrt and pow library functions to decide if it can use the llvm intrinsic. llvm-svn: 64689
2009-02-17 06:43:43 +08:00
long double l2 = powl(a2, a2);
}
@llvm.sqrt isn't really close enough to C's sqrt to justify emitting calls to the intrinsic, even when math-errno is off. Fixes rdar://problem/7828230 by falling back on the library function. llvm-svn: 100613
2010-04-07 16:20:20 +08:00
// CHECK-YES: declare float @powf(float, float)
// CHECK-YES: declare double @pow(double, double)
// CHECK-YES: declare x86_fp80 @powl(x86_fp80, x86_fp80)
Update clang test to cover for new treatment of intrinsics as readnone. llvm-svn: 203056
2014-03-06 09:37:10 +08:00
// CHECK-NO: declare float @llvm.pow.f32(float, float) [[NUW_RNI:#[0-9]+]]
// CHECK-NO: declare double @llvm.pow.f64(double, double) [[NUW_RNI]]
// CHECK-NO: declare x86_fp80 @llvm.pow.f80(x86_fp80, x86_fp80) [[NUW_RNI]]
Add codegen support for the fma/fmal/fmaf builtins. llvm-svn: 134743
2011-07-09 05:39:34 +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
// CHECK-YES-LABEL: define void @test_fma
// CHECK-NO-LABEL: define void @test_fma
Add codegen support for the fma/fmal/fmaf builtins. llvm-svn: 134743
2011-07-09 05:39:34 +08:00
void test_fma(float a0, double a1, long double a2) {
[CodeGen] fix const-ness of cbrt and fma cbrt() is always constant because it can't overflow or underflow. Therefore, it can't set errno. fma() is not always constant because it can overflow or underflow. Therefore, it can set errno. But we know that it never sets errno on GNU / MSVC, so make it constant in those environments. Differential Revision: https://reviews.llvm.org/D39641 llvm-svn: 318093
2017-11-14 06:11:49 +08:00
// CHECK-YES: call float @fmaf
Add codegen support for the fma/fmal/fmaf builtins. llvm-svn: 134743
2011-07-09 05:39:34 +08:00
// CHECK-NO: call float @llvm.fma.f32
float l0 = fmaf(a0, a0, a0);
[CodeGen] fix const-ness of cbrt and fma cbrt() is always constant because it can't overflow or underflow. Therefore, it can't set errno. fma() is not always constant because it can overflow or underflow. Therefore, it can set errno. But we know that it never sets errno on GNU / MSVC, so make it constant in those environments. Differential Revision: https://reviews.llvm.org/D39641 llvm-svn: 318093
2017-11-14 06:11:49 +08:00
// CHECK-YES: call double @fma
Add codegen support for the fma/fmal/fmaf builtins. llvm-svn: 134743
2011-07-09 05:39:34 +08:00
// CHECK-NO: call double @llvm.fma.f64
double l1 = fma(a1, a1, a1);
[CodeGen] fix const-ness of cbrt and fma cbrt() is always constant because it can't overflow or underflow. Therefore, it can't set errno. fma() is not always constant because it can overflow or underflow. Therefore, it can set errno. But we know that it never sets errno on GNU / MSVC, so make it constant in those environments. Differential Revision: https://reviews.llvm.org/D39641 llvm-svn: 318093
2017-11-14 06:11:49 +08:00
// CHECK-YES: call x86_fp80 @fmal
Add codegen support for the fma/fmal/fmaf builtins. llvm-svn: 134743
2011-07-09 05:39:34 +08:00
// CHECK-NO: call x86_fp80 @llvm.fma.f80
long double l2 = fmal(a2, a2, a2);
}
[CodeGen] fix const-ness of cbrt and fma cbrt() is always constant because it can't overflow or underflow. Therefore, it can't set errno. fma() is not always constant because it can overflow or underflow. Therefore, it can set errno. But we know that it never sets errno on GNU / MSVC, so make it constant in those environments. Differential Revision: https://reviews.llvm.org/D39641 llvm-svn: 318093
2017-11-14 06:11:49 +08:00
// CHECK-YES: declare float @fmaf(float, float, float)
// CHECK-YES: declare double @fma(double, double, double)
// CHECK-YES: declare x86_fp80 @fmal(x86_fp80, x86_fp80, x86_fp80)
Fix testcases to not rely upon target-* attributes. llvm-svn: 176135
2013-02-27 07:08:48 +08:00
// CHECK-NO: declare float @llvm.fma.f32(float, float, float) [[NUW_RN2:#[0-9]+]]
// CHECK-NO: declare double @llvm.fma.f64(double, double, double) [[NUW_RN2]]
// CHECK-NO: declare x86_fp80 @llvm.fma.f80(x86_fp80, x86_fp80, x86_fp80) [[NUW_RN2]]
Add atan, atan2, exp, and log to the builtin math library functions. With -fno-math-errno (the default for Darwin) or -ffast-math these library function can be marked readnone enabling more opportunities for CSE and other optimizations. rdar://11251464 llvm-svn: 155498
2012-04-25 06:40:01 +08:00
// Just checking to make sure these library functions are marked readnone
void test_builtins(double d, float f, long double ld) {
Fix broken CHECK lines llvm-svn: 201477
2014-02-16 15:29:41 +08:00
// CHECK-NO: @test_builtins
// CHECK-YES: @test_builtins
Add atan, atan2, exp, and log to the builtin math library functions. With -fno-math-errno (the default for Darwin) or -ffast-math these library function can be marked readnone enabling more opportunities for CSE and other optimizations. rdar://11251464 llvm-svn: 155498
2012-04-25 06:40:01 +08:00
double atan_ = atan(d);
long double atanl_ = atanl(ld);
float atanf_ = atanf(f);
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set The LLVM sqrt intrinsic definition changed with: D28797 ...so we don't have to use any relaxed FP settings other than errno handling. This patch sidesteps a question raised in PR27435: https://bugs.llvm.org/show_bug.cgi?id=27435 Is a programmer using __builtin_sqrt() invoking the compiler's intrinsic definition of sqrt or the mathlib definition of sqrt? But we have an answer now: the builtin should match the behavior of the libm function including errno handling. Differential Revision: https://reviews.llvm.org/D39204 llvm-svn: 317031
2017-11-01 04:19:39 +08:00
// CHECK-NO: declare double @atan(double) [[NUW_RN:#[0-9]+]]
Reapply r176133 with testcase fixes. llvm-svn: 176145
2013-02-27 08:06:04 +08:00
// CHECK-NO: declare x86_fp80 @atanl(x86_fp80) [[NUW_RN]]
// CHECK-NO: declare float @atanf(float) [[NUW_RN]]
Fix testcases to not rely upon target-* attributes. llvm-svn: 176135
2013-02-27 07:08:48 +08:00
// CHECK-YES-NOT: declare double @atan(double) [[NUW_RN]]
// CHECK-YES-NOT: declare x86_fp80 @atanl(x86_fp80) [[NUW_RN]]
// CHECK-YES-NOT: declare float @atanf(float) [[NUW_RN]]
Add atan, atan2, exp, and log to the builtin math library functions. With -fno-math-errno (the default for Darwin) or -ffast-math these library function can be marked readnone enabling more opportunities for CSE and other optimizations. rdar://11251464 llvm-svn: 155498
2012-04-25 06:40:01 +08:00
double atan2_ = atan2(d, 2);
long double atan2l_ = atan2l(ld, ld);
float atan2f_ = atan2f(f, f);
Reapply r176133 with testcase fixes. llvm-svn: 176145
2013-02-27 08:06:04 +08:00
// CHECK-NO: declare double @atan2(double, double) [[NUW_RN]]
// CHECK-NO: declare x86_fp80 @atan2l(x86_fp80, x86_fp80) [[NUW_RN]]
// CHECK-NO: declare float @atan2f(float, float) [[NUW_RN]]
Fix testcases to not rely upon target-* attributes. llvm-svn: 176135
2013-02-27 07:08:48 +08:00
// CHECK-YES-NOT: declare double @atan2(double, double) [[NUW_RN]]
// CHECK-YES-NOT: declare x86_fp80 @atan2l(x86_fp80, x86_fp80) [[NUW_RN]]
// CHECK-YES-NOT: declare float @atan2f(float, float) [[NUW_RN]]
Add atan, atan2, exp, and log to the builtin math library functions. With -fno-math-errno (the default for Darwin) or -ffast-math these library function can be marked readnone enabling more opportunities for CSE and other optimizations. rdar://11251464 llvm-svn: 155498
2012-04-25 06:40:01 +08:00
double exp_ = exp(d);
long double expl_ = expl(ld);
float expf_ = expf(f);
[CodeGen] convert math libcalls/builtins to equivalent LLVM intrinsics There are 20 LLVM math intrinsics that correspond to mathlib calls according to the LangRef: http://llvm.org/docs/LangRef.html#standard-c-library-intrinsics We were only converting 3 mathlib calls (sqrt, fma, pow) and 12 builtin calls (ceil, copysign, fabs, floor, fma, fmax, fmin, nearbyint, pow, rint, round, trunc) to their intrinsic-equivalents. This patch pulls the transforms together and handles all 20 cases. The switch is guarded by a check for const-ness to make sure we're not doing the transform if errno could possibly be set by the libcall or builtin. Differential Revision: https://reviews.llvm.org/D40044 llvm-svn: 319593
2017-12-02 07:15:52 +08:00
// CHECK-NO: declare double @llvm.exp.f64(double) [[NUW_RNI]]
// CHECK-NO: declare x86_fp80 @llvm.exp.f80(x86_fp80) [[NUW_RNI]]
// CHECK-NO: declare float @llvm.exp.f32(float) [[NUW_RNI]]
Fix testcases to not rely upon target-* attributes. llvm-svn: 176135
2013-02-27 07:08:48 +08:00
// CHECK-YES-NOT: declare double @exp(double) [[NUW_RN]]
// CHECK-YES-NOT: declare x86_fp80 @expl(x86_fp80) [[NUW_RN]]
// CHECK-YES-NOT: declare float @expf(float) [[NUW_RN]]
Add atan, atan2, exp, and log to the builtin math library functions. With -fno-math-errno (the default for Darwin) or -ffast-math these library function can be marked readnone enabling more opportunities for CSE and other optimizations. rdar://11251464 llvm-svn: 155498
2012-04-25 06:40:01 +08:00
double log_ = log(d);
long double logl_ = logl(ld);
float logf_ = logf(f);
[CodeGen] convert math libcalls/builtins to equivalent LLVM intrinsics There are 20 LLVM math intrinsics that correspond to mathlib calls according to the LangRef: http://llvm.org/docs/LangRef.html#standard-c-library-intrinsics We were only converting 3 mathlib calls (sqrt, fma, pow) and 12 builtin calls (ceil, copysign, fabs, floor, fma, fmax, fmin, nearbyint, pow, rint, round, trunc) to their intrinsic-equivalents. This patch pulls the transforms together and handles all 20 cases. The switch is guarded by a check for const-ness to make sure we're not doing the transform if errno could possibly be set by the libcall or builtin. Differential Revision: https://reviews.llvm.org/D40044 llvm-svn: 319593
2017-12-02 07:15:52 +08:00
// CHECK-NO: declare double @llvm.log.f64(double) [[NUW_RNI]]
// CHECK-NO: declare x86_fp80 @llvm.log.f80(x86_fp80) [[NUW_RNI]]
// CHECK-NO: declare float @llvm.log.f32(float) [[NUW_RNI]]
Fix testcases to not rely upon target-* attributes. llvm-svn: 176135
2013-02-27 07:08:48 +08:00
// CHECK-YES-NOT: declare double @log(double) [[NUW_RN]]
// CHECK-YES-NOT: declare x86_fp80 @logl(x86_fp80) [[NUW_RN]]
// CHECK-YES-NOT: declare float @logf(float) [[NUW_RN]]
Add atan, atan2, exp, and log to the builtin math library functions. With -fno-math-errno (the default for Darwin) or -ffast-math these library function can be marked readnone enabling more opportunities for CSE and other optimizations. rdar://11251464 llvm-svn: 155498
2012-04-25 06:40:01 +08:00
}
Modify the tests to use attribute group references instead of listing the function attributes. llvm-svn: 175606
2013-02-20 15:22:19 +08:00
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set The LLVM sqrt intrinsic definition changed with: D28797 ...so we don't have to use any relaxed FP settings other than errno handling. This patch sidesteps a question raised in PR27435: https://bugs.llvm.org/show_bug.cgi?id=27435 Is a programmer using __builtin_sqrt() invoking the compiler's intrinsic definition of sqrt or the mathlib definition of sqrt? But we have an answer now: the builtin should match the behavior of the libm function including errno handling. Differential Revision: https://reviews.llvm.org/D39204 llvm-svn: 317031
2017-11-01 04:19:39 +08:00
// CHECK-NO-DAG: attributes [[NUW_RN]] = { nounwind readnone{{.*}} }
[FunctionAttrs] Annotate "willreturn" for intrinsics Summary: In D62801, new function attribute `willreturn` was introduced. In short, a function with `willreturn` is guaranteed to come back to the call site(more precise definition is in LangRef). In this patch, willreturn is annotated for LLVM intrinsics. Reviewers: jdoerfert Reviewed By: jdoerfert Subscribers: jvesely, nhaehnle, sstefan1, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D64904 llvm-svn: 367184
2019-07-28 14:09:56 +08:00
// CHECK-NO-DAG: attributes [[NUW_RNI]] = { nounwind readnone speculatable willreturn }