From abd383584b16dd0fb1bbf40e4ece65ebe7b839ec Mon Sep 17 00:00:00 2001 From: Joseph Myers Date: Fri, 10 Sep 2021 20:56:22 +0000 Subject: Add narrowing square root functions This patch adds the narrowing square root functions from TS 18661-1 / TS 18661-3 / C2X to glibc's libm: fsqrt, fsqrtl, dsqrtl, f32sqrtf64, f32sqrtf32x, f32xsqrtf64 for all configurations; f32sqrtf64x, f32sqrtf128, f64sqrtf64x, f64sqrtf128, f32xsqrtf64x, f32xsqrtf128, f64xsqrtf128 for configurations with _Float64x and _Float128; __f32sqrtieee128 and __f64sqrtieee128 aliases in the powerpc64le case (for calls to fsqrtl and dsqrtl when long double is IEEE binary128). Corresponding tgmath.h macro support is also added. The changes are mostly similar to those for the other narrowing functions previously added, so the description of those generally applies to this patch as well. However, the not-actually-narrowing cases (where the two types involved in the function have the same floating-point format) are aliased to sqrt, sqrtl or sqrtf128 rather than needing a separately built not-actually-narrowing function such as was needed for add / sub / mul / div. Thus, there is no __nldbl_dsqrtl name for ldbl-opt because no such name was needed (whereas the other functions needed such a name since the only other name for that entry point was e.g. f32xaddf64, not reserved by TS 18661-1); the headers are made to arrange for sqrt to be called in that case instead. The DIAG_* calls in sysdeps/ieee754/soft-fp/s_dsqrtl.c are because they were observed to be needed in GCC 7 testing of riscv32-linux-gnu-rv32imac-ilp32. The other sysdeps/ieee754/soft-fp/ files added didn't need such DIAG_* in any configuration I tested with build-many-glibcs.py, but if they do turn out to be needed in more files with some other configuration / GCC version, they can always be added there. I reused the same test inputs in auto-libm-test-in as for non-narrowing sqrt rather than adding extra or separate inputs for narrowing sqrt. The tests in libm-test-narrow-sqrt.inc also follow those for non-narrowing sqrt. Tested as followed: natively with the full glibc testsuite for x86_64 (GCC 11, 7, 6) and x86 (GCC 11); with build-many-glibcs.py with GCC 11, 7 and 6; cross testing of math/ tests for powerpc64le, powerpc32 hard float, mips64 (all three ABIs, both hard and soft float). The different GCC versions are to cover the different cases in tgmath.h and tgmath.h tests properly (GCC 6 has _Float* only as typedefs in glibc headers, GCC 7 has proper _Float* support, GCC 8 adds __builtin_tgmath). --- sysdeps/ieee754/float128/float128_private.h | 5 +++++ sysdeps/ieee754/float128/s_f32sqrtf128.c | 6 ++++++ sysdeps/ieee754/float128/s_f64sqrtf128.c | 10 ++++++++++ sysdeps/ieee754/float128/s_f64xsqrtf128.c | 2 ++ 4 files changed, 23 insertions(+) create mode 100644 sysdeps/ieee754/float128/s_f32sqrtf128.c create mode 100644 sysdeps/ieee754/float128/s_f64sqrtf128.c create mode 100644 sysdeps/ieee754/float128/s_f64xsqrtf128.c (limited to 'sysdeps/ieee754/float128') diff --git a/sysdeps/ieee754/float128/float128_private.h b/sysdeps/ieee754/float128/float128_private.h index b6b6d3d5fd..13d1d63baf 100644 --- a/sysdeps/ieee754/float128/float128_private.h +++ b/sysdeps/ieee754/float128/float128_private.h @@ -138,6 +138,9 @@ #define libm_alias_float_ldouble(func) libm_alias_float32_float128 (func) #undef libm_alias_double_ldouble #define libm_alias_double_ldouble(func) libm_alias_float64_float128 (func) +#undef libm_alias_ldouble_narrow +#define libm_alias_ldouble_narrow(from, to) \ + libm_alias_float128_narrow (from, to) #include #undef USE_NEARBYINTL_BUILTIN @@ -284,6 +287,8 @@ #define __ddivl __f64divf128 #define __fmull __f32mulf128 #define __dmull __f64mulf128 +#define __fsqrtl __f32sqrtf128 +#define __dsqrtl __f64sqrtf128 #define __fsubl __f32subf128 #define __dsubl __f64subf128 diff --git a/sysdeps/ieee754/float128/s_f32sqrtf128.c b/sysdeps/ieee754/float128/s_f32sqrtf128.c new file mode 100644 index 0000000000..74e91abd4d --- /dev/null +++ b/sysdeps/ieee754/float128/s_f32sqrtf128.c @@ -0,0 +1,6 @@ +#define f32sqrtf64x __hide_f32sqrtf64x +#define f32sqrtf128 __hide_f32sqrtf128 +#include +#undef f32sqrtf64x +#undef f32sqrtf128 +#include "../ldbl-128/s_fsqrtl.c" diff --git a/sysdeps/ieee754/float128/s_f64sqrtf128.c b/sysdeps/ieee754/float128/s_f64sqrtf128.c new file mode 100644 index 0000000000..8373879d42 --- /dev/null +++ b/sysdeps/ieee754/float128/s_f64sqrtf128.c @@ -0,0 +1,10 @@ +#define f32xsqrtf64x __hide_f32xsqrtf64x +#define f32xsqrtf128 __hide_f32xsqrtf128 +#define f64sqrtf64x __hide_f64sqrtf64x +#define f64sqrtf128 __hide_f64sqrtf128 +#include +#undef f32xsqrtf64x +#undef f32xsqrtf128 +#undef f64sqrtf64x +#undef f64sqrtf128 +#include "../ldbl-128/s_dsqrtl.c" diff --git a/sysdeps/ieee754/float128/s_f64xsqrtf128.c b/sysdeps/ieee754/float128/s_f64xsqrtf128.c new file mode 100644 index 0000000000..383f92a94e --- /dev/null +++ b/sysdeps/ieee754/float128/s_f64xsqrtf128.c @@ -0,0 +1,2 @@ +#include +#include "../ldbl-128/s_f64xsqrtf128.c" -- cgit 1.4.1