glibc/manual
Wilco Dijkstra c3d466cba1 Remove slow paths from pow
Remove the slow paths from pow.  Like several other double precision math
functions, pow is exactly rounded.  This is not required from math functions
and causes major overheads as it requires multiple fallbacks using higher
precision arithmetic if a result is close to 0.5ULP.  Ridiculous slowdowns
of up to 100000x have been reported when the highest precision path triggers.

All GLIBC math tests pass on AArch64 and x64 (with ULP of pow set to 1).
The worst case error is ~0.506ULP.  A simple test over a few hundred million
values shows pow is 10% faster on average.  This fixes BZ #13932.

	[BZ #13932]
	* sysdeps/ieee754/dbl-64/uexp.h (err_1): Remove.
	* benchtests/pow-inputs: Update comment for slow path cases.
	* manual/probes.texi (slowpow_p10): Delete removed probe.
	(slowpow_p10): Likewise.
	* math/Makefile: Remove halfulp.c and slowpow.c.
	* sysdeps/aarch64/libm-test-ulps: Set ULP of pow to 1.
	* sysdeps/generic/math_private.h (__exp1): Remove error argument.
	(__halfulp): Remove.
	(__slowpow): Remove.
	* sysdeps/i386/fpu/halfulp.c: Delete file.
	* sysdeps/i386/fpu/slowpow.c: Likewise.
	* sysdeps/ia64/fpu/halfulp.c: Likewise.
	* sysdeps/ia64/fpu/slowpow.c: Likewise.
	* sysdeps/ieee754/dbl-64/e_exp.c (__exp1): Remove error argument,
	improve comments and add error analysis.
	* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Add error analysis.
	(power1): Remove function:
	(log1): Remove error argument, add error analysis.
	(my_log2): Remove function.
	* sysdeps/ieee754/dbl-64/halfulp.c: Delete file.
	* sysdeps/ieee754/dbl-64/slowpow.c: Likewise.
	* sysdeps/m68k/m680x0/fpu/halfulp.c: Likewise.
	* sysdeps/m68k/m680x0/fpu/slowpow.c: Likewise.
	* sysdeps/powerpc/power4/fpu/Makefile: Remove CPPFLAGS-slowpow.c.
	* sysdeps/x86_64/fpu/libm-test-ulps: Set ULP of pow to 1.
	* sysdeps/x86_64/fpu/multiarch/Makefile: Remove slowpow-fma.c,
	slowpow-fma4.c, halfulp-fma.c, halfulp-fma4.c.
	* sysdeps/x86_64/fpu/multiarch/e_pow-fma.c (__slowpow): Remove define.
	* sysdeps/x86_64/fpu/multiarch/e_pow-fma4.c (__slowpow): Likewise.
	* sysdeps/x86_64/fpu/multiarch/halfulp-fma.c: Delete file.
	* sysdeps/x86_64/fpu/multiarch/halfulp-fma4.c: Likewise.
	* sysdeps/x86_64/fpu/multiarch/slowpow-fma.c: Likewise.
	* sysdeps/x86_64/fpu/multiarch/slowpow-fma4.c: Likewise.
2018-02-12 10:47:09 +00:00
..
examples Update copyright dates with scripts/update-copyrights. 2018-01-01 00:32:25 +00:00
argp.texi manual: Complete @standards in argp.texi. 2017-06-16 01:19:30 -07:00
arith.texi Add narrowing add functions. 2018-02-10 02:08:43 +00:00
charset.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
check-safety.sh Update copyright dates with scripts/update-copyrights. 2018-01-01 00:32:25 +00:00
conf.texi manual/conf.texi: add a missing underscore in front of SC_SSIZE_MAX [BZ #22588] 2017-12-12 00:11:29 +01:00
contrib.texi Update contributions in the manual 2018-02-01 13:07:39 +00:00
creature.texi manual: Fix Texinfo warnings about improper node names. 2018-02-09 04:40:48 -08:00
crypt.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
ctype.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
debug.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
dir
errno.texi manual: clarify errno value on success [BZ #22615] 2017-12-29 14:44:57 +01:00
fdl-1.3.texi
filesys.texi manual: Document the linkat function 2017-11-04 00:28:37 +01:00
freemanuals.texi
getopt.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
header.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
install-plain.texi
install.texi Fix typo in the previous commit 2018-01-31 23:13:00 +00:00
intro.texi manual: fix typo in the introduction 2016-05-19 23:22:59 -04:00
io.texi
ipc.texi manual/ipc.texi: Fix AC-safety notes. 2014-04-08 17:12:15 -04:00
job.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
lang.texi manual: Rewrite the section on widths of integer types. 2017-08-10 20:28:28 -07:00
lgpl-2.1.texi
libc-texinfo.sh Remove add-ons mechanism. 2017-10-05 15:58:13 +00:00
libc.texinfo Update copyright dates not handled by scripts/update-copyrights. 2018-01-01 00:41:16 +00:00
libcbook.texi
libdl.texi * manual/libdl.texi: New. 2014-01-31 23:23:59 -02:00
libm-err-tab.pl Update copyright dates with scripts/update-copyrights. 2018-01-01 00:32:25 +00:00
llio.texi preadv2/pwritev2: Handle offset == -1 [BZ #22753] 2018-02-02 10:46:26 +01:00
locale.texi manual: Touch up documentation for strftime and ALTMON_* constants. 2018-01-22 19:36:51 -08:00
macros.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
maint.texi Remove add-ons mechanism. 2017-10-05 15:58:13 +00:00
Makefile Update copyright dates with scripts/update-copyrights. 2018-01-01 00:32:25 +00:00
math.texi Skeleton documentation for the RISC-V port 2018-01-29 10:22:26 -08:00
memory.texi Linux: Implement interfaces for memory protection keys 2017-12-05 15:20:35 +01:00
message.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
nss.texi Remove compat from DEFAULT_CONFIG lookup strings 2017-09-12 10:21:48 -07:00
nsswitch.texi
pattern.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
pipe.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
platform.texi Add documentation for __riscv_flush_icache 2018-01-29 10:25:23 -08:00
probes.texi Remove slow paths from pow 2018-02-12 10:47:09 +00:00
process.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
README.pretty-printers Remove obsolete notes at top level of source tree. 2017-09-01 08:04:22 -04:00
README.tunables Remove obsolete notes at top level of source tree. 2017-09-01 08:04:22 -04:00
resource.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
search.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
setjmp.texi manual: Document getcontext uc_stack value on Linux [BZ #759] 2017-08-08 16:16:43 -03:00
signal.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
socket.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
startup.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
stdio-fp.c
stdio.texi manual: Fix Texinfo warnings about improper node names. 2018-02-09 04:40:48 -08:00
string.texi manual: Fix Texinfo warnings about improper node names. 2018-02-09 04:40:48 -08:00
summary.pl Update copyright dates with scripts/update-copyrights. 2018-01-01 00:32:25 +00:00
sysinfo.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
syslog.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
terminal.texi manual: Update to mention ENODEV for ttyname and ttyname_r 2017-11-15 20:46:45 +01:00
texinfo.tex Update miscellaneous files from upstream sources. 2018-01-01 18:05:06 +00:00
texis.awk
threads.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
time.texi manual: Touch up documentation for strftime and ALTMON_* constants. 2018-01-22 19:36:51 -08:00
tsort.awk Update copyright dates with scripts/update-copyrights. 2018-01-01 00:32:25 +00:00
tunables.texi manual: Fix spelling of "Auxiliary." 2018-01-23 11:40:44 -08:00
users.texi manual: Replace summary.awk with summary.pl. 2017-06-15 21:26:20 -07:00
xtract-typefun.awk Make shebang interpreter directives consistent 2016-01-07 04:03:21 -05:00

			TUNABLE FRAMEWORK
			=================

Tunables is a feature in the GNU C Library that allows application authors and
distribution maintainers to alter the runtime library behaviour to match their
workload.

The tunable framework allows modules within glibc to register variables that
may be tweaked through an environment variable.  It aims to enforce a strict
namespace rule to bring consistency to naming of these tunable environment
variables across the project.  This document is a guide for glibc developers to
add tunables to the framework.

ADDING A NEW TUNABLE
--------------------

The TOP_NAMESPACE macro is defined by default as 'glibc'.  If distributions
intend to add their own tunables, they should do so in a different top
namespace by overriding the TOP_NAMESPACE macro for that tunable.  Downstream
implementations are discouraged from using the 'glibc' top namespace for
tunables they don't already have consensus to push upstream.

There are three steps to adding a tunable:

1. Add a tunable to the list and fully specify its properties:

For each tunable you want to add, make an entry in elf/dl-tunables.list.  The
format of the file is as follows:

TOP_NAMESPACE {
  NAMESPACE1 {
    TUNABLE1 {
      # tunable attributes, one per line
    }
    # A tunable with default attributes, i.e. string variable.
    TUNABLE2
    TUNABLE3 {
      # its attributes
    }
  }
  NAMESPACE2 {
    ...
  }
}

The list of allowed attributes are:

- type:			Data type.  Defaults to STRING.  Allowed types are:
			INT_32, UINT_64, SIZE_T and STRING.  Numeric types may
			be in octal or hexadecimal format too.

- minval:		Optional minimum acceptable value.  For a string type
			this is the minimum length of the value.

- maxval:		Optional maximum acceptable value.  For a string type
			this is the maximum length of the value.

- default:		Specify an optional default value for the tunable.

- env_alias:		An alias environment variable

- security_level:	Specify security level of the tunable.  Valid values:

			SXID_ERASE: (default) Don't read for AT_SECURE binaries and
				    removed so that child processes can't read it.
			SXID_IGNORE: Don't read for AT_SECURE binaries, but retained for
				     non-AT_SECURE subprocesses.
			NONE: Read all the time.

2. Use TUNABLE_GET/TUNABLE_SET to get and set tunables.

3. OPTIONAL: If tunables in a namespace are being used multiple times within a
   specific module, set the TUNABLE_NAMESPACE macro to reduce the amount of
   typing.

GETTING AND SETTING TUNABLES
----------------------------

When the TUNABLE_NAMESPACE macro is defined, one may get tunables in that
module using the TUNABLE_GET macro as follows:

  val = TUNABLE_GET (check, int32_t, TUNABLE_CALLBACK (check_callback))

where 'check' is the tunable name, 'int32_t' is the C type of the tunable and
'check_callback' is the function to call if the tunable got initialized to a
non-default value.  The macro returns the value as type 'int32_t'.

The callback function should be defined as follows:

  void
  TUNABLE_CALLBACK (check_callback) (int32_t *valp)
  {
  ...
  }

where it can expect the tunable value to be passed in VALP.

Tunables in the module can be updated using:

  TUNABLE_SET (check, int32_t, val)

where 'check' is the tunable name, 'int32_t' is the C type of the tunable and
'val' is a value of same type.

To get and set tunables in a different namespace from that module, use the full
form of the macros as follows:

  val = TUNABLE_GET_FULL (glibc, tune, hwcap_mask, uint64_t, NULL)

  TUNABLE_SET_FULL (glibc, tune, hwcap_mask, uint64_t, val)

where 'glibc' is the top namespace, 'tune' is the tunable namespace and the
remaining arguments are the same as the short form macros.

When TUNABLE_NAMESPACE is not defined in a module, TUNABLE_GET is equivalent to
TUNABLE_GET_FULL, so you will need to provide full namespace information for
both macros.  Likewise for TUNABLE_SET and TUNABLE_SET_FULL.

** IMPORTANT NOTE **

The tunable list is set as read-only after the dynamic linker relocates itself,
so setting tunable values must be limited only to tunables within the dynamic
linker, that too before relocation.

FUTURE WORK
-----------

The framework currently only allows a one-time initialization of variables
through environment variables and in some cases, modification of variables via
an API call.  A future goals for this project include:

- Setting system-wide and user-wide defaults for tunables through some
  mechanism like a configuration file.

- Allow tweaking of some tunables at runtime