From 3e08ff544b86834cd24795de159f16b8c65c7b8f Mon Sep 17 00:00:00 2001 From: Szabolcs Nagy Date: Wed, 13 Jun 2018 17:48:52 +0100 Subject: Add new log2 implementation Similar algorithm is used as in log: log2(2^k x) = k + log2(c) + log2(x/c) where the last term is approximated by a polynomial of x/c - 1, the first order coefficient is about 1/ln2 in this case. There is separate code path when fma instruction is not available for computing x/c - 1 precisely, for which the table size is doubled. The worst case error is 0.547 ULP (0.55 without fma), the read only global data size is 1168 bytes (2192 without fma) on aarch64. The non-nearest rounding error is less than 1 ULP. Improvements on Cortex-A72 compared to current glibc master: log2 thruput: 2.00x in [0.01 11.1] log2 latency: 2.04x in [0.01 11.1] log2 thruput: 2.17x in [0.999 1.001] log2 latency: 2.88x in [0.999 1.001] Tested on aarch64-linux-gnu (defined __FP_FAST_FMA) arm-linux-gnueabihf (!defined __FP_FAST_FMA) x86_64-linux-gnu (!defined __FP_FAST_FMA) powerpc64le-linxu-gnu (defined __FP_FAST_FMA) targets. * NEWS: Mention log2 improvements. * math/Makefile (type-double-routines): Add e_log2_data. * sysdeps/i386/fpu/e_log2_data.c: New file. * sysdeps/ia64/fpu/e_log2_data.c: New file. * sysdeps/ieee754/dbl-64/e_log2.c: Rewrite. * sysdeps/ieee754/dbl-64/e_log2_data.c: New file. * sysdeps/ieee754/dbl-64/math_config.h (__log2_data): Add. * sysdeps/ieee754/dbl-64/wordsize-64/e_log2.c: Remove. * sysdeps/m68k/m680x0/fpu/e_log2_data.c: New file. --- sysdeps/m68k/m680x0/fpu/e_log2_data.c | 1 + 1 file changed, 1 insertion(+) create mode 100644 sysdeps/m68k/m680x0/fpu/e_log2_data.c (limited to 'sysdeps/m68k/m680x0/fpu') diff --git a/sysdeps/m68k/m680x0/fpu/e_log2_data.c b/sysdeps/m68k/m680x0/fpu/e_log2_data.c new file mode 100644 index 0000000000..1cc8931700 --- /dev/null +++ b/sysdeps/m68k/m680x0/fpu/e_log2_data.c @@ -0,0 +1 @@ +/* Not needed. */ -- cgit 1.4.1