aboutsummaryrefslogtreecommitdiff
path: root/manual
diff options
context:
space:
mode:
authorPatrick McGehearty <patrick.mcgehearty@oracle.com>2017-12-19 17:25:14 +0000
committerJoseph Myers <joseph@codesourcery.com>2017-12-19 17:27:31 +0000
commit6fd0a3c6a887a91b1554730c977657a7e65334cc (patch)
tree6a0b347f2435cdd23fb55f3f39f22e7a0fb83096 /manual
parent3bb1ef58b989012f8199b82af6ec136da2f9fda3 (diff)
downloadglibc-6fd0a3c6a887a91b1554730c977657a7e65334cc.tar
glibc-6fd0a3c6a887a91b1554730c977657a7e65334cc.tar.gz
glibc-6fd0a3c6a887a91b1554730c977657a7e65334cc.tar.bz2
glibc-6fd0a3c6a887a91b1554730c977657a7e65334cc.zip
Improve __ieee754_exp() performance by greater than 5x on sparc/x86.
These changes will be active for all platforms that don't provide their own exp() routines. They will also be active for ieee754 versions of ccos, ccosh, cosh, csin, csinh, sinh, exp10, gamma, and erf. Typical performance gains is typically around 5x when measured on Sparc s7 for common values between exp(1) and exp(40). Using the glibc perf tests on sparc, sparc (nsec) x86 (nsec) old new old new max 17629 395 5173 144 min 399 54 15 13 mean 5317 200 1349 23 The extreme max times for the old (ieee754) exp are due to the multiprecision computation in the old algorithm when the true value is very near 0.5 ulp away from an value representable in double precision. The new algorithm does not take special measures for those cases. The current glibc exp perf tests overrepresent those values. Informal testing suggests approximately one in 200 cases might invoke the high cost computation. The performance advantage of the new algorithm for other values is still large but not as large as indicated by the chart above. Glibc correctness tests for exp() and expf() were run. Within the test suite 3 input values were found to cause 1 bit differences (ulp) when "FE_TONEAREST" rounding mode is set. No differences in exp() were seen for the tested values for the other rounding modes. Typical example: exp(-0x1.760cd2p+0) (-1.46113312244415283203125) new code: 2.31973271630014299393707e-01 0x1.db14cd799387ap-3 old code: 2.31973271630014271638132e-01 0x1.db14cd7993879p-3 exp = 2.31973271630014285508337 (high precision) Old delta: off by 0.49 ulp New delta: off by 0.51 ulp In addition, because ieee754_exp() is used by other routines, cexp() showed test results with very small imaginary input values where the imaginary portion of the result was off by 3 ulp when in upward rounding mode, but not in the other rounding modes. For x86, tgamma showed a few values where the ulp increased to 6 (max ulp for tgamma is 5). Sparc tgamma did not show these failures. I presume the tgamma differences are due to compiler optimization differences within the gamma function.The gamma function is known to be difficult to compute accurately. * sysdeps/ieee754/dbl-64/e_exp.c: Include <math-svid-compat.h> and <errno.h>. Include "eexp.tbl". (half): New constant. (one): Likewise. (__ieee754_exp): Rewrite. (__slowexp): Remove prototype. * sysdeps/ieee754/dbl-64/eexp.tbl: New file. * sysdeps/ieee754/dbl-64/slowexp.c: Remove file. * sysdeps/i386/fpu/slowexp.c: Likewise. * sysdeps/ia64/fpu/slowexp.c: Likewise. * sysdeps/m68k/m680x0/fpu/slowexp.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-avx.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-fma.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-fma4.c: Likewise. * sysdeps/generic/math_private.h (__slowexp): Remove prototype. * sysdeps/ieee754/dbl-64/e_pow.c: Remove mention of slowexp.c in comment. * sysdeps/powerpc/power4/fpu/Makefile [$(subdir) = math] (CPPFLAGS-slowexp.c): Remove variable. * sysdeps/x86_64/fpu/multiarch/Makefile (libm-sysdep_routines): Remove slowexp-fma, slowexp-fma4 and slowexp-avx. (CFLAGS-slowexp-fma.c): Remove variable. (CFLAGS-slowexp-fma4.c): Likewise. (CFLAGS-slowexp-avx.c): Likewise. * sysdeps/x86_64/fpu/multiarch/e_exp-avx.c (__slowexp): Do not define as macro. * sysdeps/x86_64/fpu/multiarch/e_exp-fma.c (__slowexp): Likewise. * sysdeps/x86_64/fpu/multiarch/e_exp-fma4.c (__slowexp): Likewise. * math/Makefile (type-double-routines): Remove slowexp. * manual/probes.texi (slowexp_p6): Remove. (slowexp_p32): Likewise.
Diffstat (limited to 'manual')
-rw-r--r--manual/probes.texi14
1 files changed, 0 insertions, 14 deletions
diff --git a/manual/probes.texi b/manual/probes.texi
index 8ab67562d7..f8ae64be33 100644
--- a/manual/probes.texi
+++ b/manual/probes.texi
@@ -258,20 +258,6 @@ Unless explicitly mentioned otherwise, a precision of 1 implies 24 bits of
precision in the mantissa of the multiple precision number. Hence, a precision
level of 32 implies 768 bits of precision in the mantissa.
-@deftp Probe slowexp_p6 (double @var{$arg1}, double @var{$arg2})
-This probe is triggered when the @code{exp} function is called with an
-input that results in multiple precision computation with precision
-6. Argument @var{$arg1} is the input value and @var{$arg2} is the
-computed output.
-@end deftp
-
-@deftp Probe slowexp_p32 (double @var{$arg1}, double @var{$arg2})
-This probe is triggered when the @code{exp} function is called with an
-input that results in multiple precision computation with precision
-32. Argument @var{$arg1} is the input value and @var{$arg2} is the
-computed output.
-@end deftp
-
@deftp Probe slowpow_p10 (double @var{$arg1}, double @var{$arg2}, double @var{$arg3}, double @var{$arg4})
This probe is triggered when the @code{pow} function is called with
inputs that result in multiple precision computation with precision