Update.cvs/libc-ud-970810

1997-08-10 18:48  Ulrich Drepper  <drepper@cygnus.com>

	* Makeconfig: Define build-programs to no if not set and $(config-os)
	is none.
	* Makerules (+depfiles): Don't add $(others) if not build-programs.
	* Rules (others): Don't depend on $(install-bin), $(install-rootbin)
	and $(install-sbin) if not build-programs.
	* Makefile (subdirs): Remove login.
	* sysdeps/mach/hurd/Subdirs: Add login.
	* sysdeps/unix/Subdirs: Add login.

	* sysdeps/generic/sys/mman.h: Test feature macros before defining
	non-POSIX things.  Add MAP_FAILED.
	* sysdeps/unix/bsd/osf/sys/mman.h: Likewise.
	* sysdeps/unix/bsd/sun/sunos4/sys/mman.h: Likewise.
	* sysdeps/unix/bsd/ultrix4/sys/mman.h: Likewise.
	* sysdeps/unix/sysv/irix4/sys/mman.h: Likewise.
	* sysdeps/unix/sysv/linux/sys/mman.h: Rewrite to not use kernel header
	but bits/mman.h.  Test feature macros before definitions non-POSIX
	functions.
	* sysdeps/unix/sysv/linux/alpha/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/i386/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/m68k/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/mips/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/powerpc/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/sparc/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/sparc64/bits/mman.h: New file.
	* sysdeps/unix/sysv/linux/alpha/Dist: Add bits/mman.h.
	* sysdeps/unix/sysv/linux/i386/Dist: Likewise.
	* sysdeps/unix/sysv/linux/m68k/Dist: Likewise.
	* sysdeps/unix/sysv/linux/mips/Dist: Likewise.
	* sysdeps/unix/sysv/linux/powerpc/Dist: Likewise.
	* sysdeps/unix/sysv/linux/sparc/Dist: Likewise.
	* sysdeps/unix/sysv/linux/sparc64/Dist: New file.  Likewise.

	* catgets/open_catalog.c (__open_catalog): Compare result of mmap
	with MAP_FAILED and not -1.
	* elf/dl-load.c (_dl_map_object_from_fd): Likewise.
	* elf/dl-minimal.c (malloc): Likewise.
	* elf/dl-misc.c (_dl_sysdep_read_whole_file): Likewise.
	* locale/loadlocale.c (_nl_load_locale): Likewise.
	* locale/programs/localedef.c: Likewise.
	* malloc/malloc.c (mmap_chunk): Likewise.
	(new_heap): Likewise.
	* string/stratcliff.c: Likewise.
	* string/tester.c: Add tests for stpcpy.

	* gmon/gmon.c (__monstartup): Use calloc instead of malloc+memset.

	* gmon/sys/gmon.h: Add prototype for write_profiling.

	* elf/dl-profile.c: Add decsriptions and implement reading of
	existing data in profiling file.
	* gmon/sys/gmon_out.h (GMON_SHOBJ_VERSION): New macro for shared
	lib profiling data.

	* sysdeps/stub/bits/libc-lock.h (__libc_once_define): Add new first
	parameter for class.

	* inet/inet_ntoa.c: Make once control variable static.
	Correctly handle case where key cannot be retrieved.
	* inet/Makefile (tests): Add tst-ntoa.
	* inet/tst-ntoa.c: New file.

	* manual/time.texi: Document localtime_r, gmtime_r, asctime_r, and
	ctime_r.
	Mention that tzname values must not be changed.

	* manual/users.texi: Document UTMP/WTMP handling functions.
	Mainly written by Mark Kettenis.

	* math/libm-test.c (cbrt_test): Add test for number with aboslute
	value < 1.0.

	* nss/nss_files/files-XXX.c (internal_setent): If opening of file
	failed with EGAIN return NSS_STATUS_TRYAGAIN.
	* nss/nss_files/files-alias.c (internal_setent): Likewise.

	* sysdeps/generic/bits/utmp.h: Pretty print.

	* sysdeps/i386/i586/strcpy.S: New file.  Much faster implementation.
	* sysdeps/i386/i586/stpcpy.S: New file.
	* sysdeps/i386/i686/Implies: Don't fall back on i586 versions since
	these are less efficient than i486/i386 versions.

	* sysdeps/libm-i387/e_rem_pio2.c: Add empty file to prevent the
	version from libm-ieee754 be compiled since the later is not needed.
	* sysdeps/libm-i387/e_rem_pio2f.c: Likewise.
	* sysdeps/libm-i387/e_rem_pio2l.c: Likewise.
	* sysdeps/libm-i387/k_rem_pio2.c: Likewise.
	* sysdeps/libm-i387/k_rem_pio2f.c: Likewise.
	* sysdeps/libm-i387/k_rem_pio2l.c: Likewise.
	* sysdeps/m68k/fpu/e_rem_pio2.c: Likewise.
	* sysdeps/m68k/fpu/e_rem_pio2f.c: Likewise.
	* sysdeps/m68k/fpu/e_rem_pio2l.c: Likewise.
	* sysdeps/m68k/fpu/k_rem_pio2.c: Likewise.
	* sysdeps/m68k/fpu/k_rem_pio2f.c: Likewise.
	* sysdeps/m68k/fpu/k_rem_pio2l.c: Likewise.

	* sysdeps/libm-i387/s_cbrt.S: Fix several bugs.
	* sysdeps/libm-i387/s_cbrtf.S: Likewise.
	* sysdeps/libm-i387/s_cbrtl.S: Likewise.

	* sysdeps/unix/sysv/linux/if_index.c (if_indextoname): Correct
	little bug.

	* sysdeps/unix/sysv/linux/bits/socket.h (struct msghdr): Make field
	msg_iovlen of type int.

	* time/tzfile.c: Change return type of __tzstring to char *.
	* time/tzset.c: Likewise.  Change definition of __tzstring.

	* time/tzset.c: Interpret no DST information in user provided time
	zone specification as it is meant: no DST.
	Patch by Paul Eggert <eggert@twinsun.com>.

1997-07-20 03:01  Geoff Keating  <geoffk@ozemail.com.au>

	* sysdeps/elf/dl-load.c (add_name_to_object): New procedure.
	(_dl_map_object_from_fd): Use add_name_to_object.
	(_dl_map_object): If an object's SONAME is used to resolve a
	dependency, add it to the list of the object's names.

	* inet/getnameinfo.c: Make `domain' non-const.

	* sysdeps/unix/sysv/linux/powerpc/kernel_termios.c: Clean up.

	* math/test-fenv.c (feenv_nomask_test): Don't execute if
	FE_NOMASK_ENV is not defined, or if fesetenv(FE_NOMASK_ENV)
	sets errno to ENOSYS.

	* sysdeps/powerpc/dl-machine.h: Print proper error message on
	unknown reloc type (rather than using assert).

	* sysdeps/unix/sysv/linux/powerpc/profil-counter.h: New file.
	* sysdeps/unix/sysv/linux/powerpc/profil.c: Deleted.

1997-07-16 12:47  Geoff Keating  <geoffk@ozemail.com.au>

	* sysdeps/powerpc/bits/fenv.h (feraiseexcept): New optimising macro.
	(feclearexcept): New optimising macro.
	(FE_NOMASK_ENV): Change to procedure so it can enable exceptions.
	(FE_ENABLED_ENV): Add.
	(FE_NONIEEE_ENV): Add.
	* sysdeps/powerpc/bits/fenv_const.c: Add __fe_enabled_env,
	__fe_nonieee_env; delete __fe_nomask_env.
	* sysdeps/powerpc/bits/fe_nomask.c: New file (stub warning until it
	gets implemented in the kernel).

	* sysdeps/powerpc/fraiseenv.c: Deal with chips that don't have
	FE_INVALID_SOFTWARE implemented. Use macros for bit names for clarity.
	* sysdeps/powerpc/fsetexcptflag.c: Likewise.

	* io/ftw.c: Don't compare holes in structures.
	* sysdeps/unix/sysv/linux/sys/sysmacros.h: Cast the result of the
	macros to `int', because otherwise it might be `long long' which
	the calling code is probably not expecting.

	* sysdeps/libm-ieee754/s_lround.c [NO_LONG_DOUBLE]: Fix a few bugs,
	document the existence of some more.

	* sysdeps/powerpc/s_llrint.c: New file.
	* sysdeps/powerpc/s_lrint.c: New file.
	* sysdeps/powerpc/s_llround.c: New file.
	* sysdeps/powerpc/s_lround.c: New file.

	* sysdeps/powerpc/s_sqrt.c: New file.
	* sysdeps/powerpc/s_sqrtf.c: New file.
	* sysdeps/powerpc/w_sqrt.s: New empty file.
	* sysdeps/powerpc/w_sqrtf.s: New empty file.
	* sysdeps/powerpc/t_sqrt.c: New file.
	* sysdeps/powerpc/test-arithf.c: New file.
	* sysdeps/powerpc/Makefile [subdir=math]: Add t_sqrt to support
	routines.  Add test-arithf to test programs.

	* sysdeps/powerpc/bits/mathdef.h: Add FP_ILOGB0, FP_ILOGBNAN.

	* sysdeps/powerpc/strcmp.s: Simplify drastically. Now much neater,
	and possibly faster (or possibly slower, depending on input).

1997-06-08 22:55  Geoff Keating  <geoffk@ozemail.com.au>

	* sysdeps/powerpc/fenvbits.h: Correct FE_DFL_ENV and FE_NOMASK_ENV
	macros.
	* sysdeps/powerpc/s_rint.c: New file.

1997-05-22 08:47  Geoff Keating  <geoffk@ozemail.com.au>

	* sysdeps/powerpc/Makefile [subdir=math]: Add q_* routines.
	* sysdeps/powerpc/Dist: Add quad_float.h.
	* sysdeps/powerpc/q_dtoq.c: New file.
	* sysdeps/powerpc/q_itoq.c: New file.
	* sysdeps/powerpc/q_lltoq.c: New file.
	* sysdeps/powerpc/q_neg.c: New file.
	* sysdeps/powerpc/q_qtoi.c: New file.
	* sysdeps/powerpc/q_qtoll.c: New file.
	* sysdeps/powerpc/q_qtos.c: New file.
	* sysdeps/powerpc/q_qtou.c: New file.
	* sysdeps/powerpc/q_qtoull.c: New file.
	* sysdeps/powerpc/q_stoq.c: New file.
	* sysdeps/powerpc/q_ulltoq.c: New file.
	* sysdeps/powerpc/q_utoq.c: New file.
	* sysdeps/powerpc/quad_float.h: New file.
	* sysdeps/powerpc/test-arith.c: New file.

	* sysdeps/powerpc/fpu_control.h: Fix _FPU_GETCW.
	* sysdeps/powerpc/fegetround.c: Use mcrfs to be faster and not
	require a stack frame.
	* sysdeps/powerpc/bits/fenv.h: Include inline macro for fegetround.

1997-05-18 05:55  Geoff Keating  <geoffk@ozemail.com.au>

	* sysdeps/powerpc/fenv_libc.h (fegetenv_register,
	fesetenv_register): Add 'volatile'.
	(set_fpscr_bit, reset_fpscr_bit): New macros, FPSCR_*
	constants to use with them.
	* sysdeps/powerpc/s_copysign.S: New file.
	* sysdeps/powerpc/s_copysignf.s: New file.
	* sysdeps/powerpc/s_fabs.S: New file.
	* sysdeps/powerpc/s_fabsf.s: New file.
	* sysdeps/powerpc/s_isnan.c: New file.
	* sysdeps/powerpc/s_isnanf.s: New file.
	* sysdeps/powerpc/s_rintf.c: New file.

	* sysdeps/powerpc/fenvbits.h: Make FE_INVALID the summary
	bit in the FPSCR, not the enable bit.
	* sysdeps/powerpc/fraiseexcpt.c: Consequent change to the above.
	* sysdeps/powerpc/fclrexcpt.c: Correct.
	* sysdeps/powerpc/fsetexcptflag.c: Correct.
	* sysdeps/powerpc/ftestexcpt.c: Is now much simpler.
	* sysdeps/powerpc/fgetexcptflg.c: Simplify.

	* sysdeps/powerpc/strlen.s: Schedule better, save 3 clocks :-).

	* sysdeps/powerpc/dl-machine.h (elf_machine_rela): Add comment
	explaining some reentrancy issues with lazy PLT entries.

1997-08-09 13:04  Mark Kettenis  <kettenis@phys.uva.nl>

	* login/logout.c (logout): utmpname returns -1 on error.

	* login/libutil.map: Remove updwtmp.

	* login/getutline.c: Rename getutline to __getutline and make
	getutline a weak alias.  Make getutxline a weak alias for
	__getutline.
	* login/getutid.c: Rename getutid to __getutid and make getutid a
	weak alias.  Make getutxid a weak alias for __getutid.
	* libc.map: Add getutxid, getutxline.

	* login/utmpname.c (__utmpname): Reset backend right after backend
	endutent call.
	* login/utmp_file.c: Reordered functions.  Remove unecessary
	header files.
	(getutent_r_file, pututline_file): Do not call setutent_file.  At this
	point the file is guaranteed to be open (assert!).
	(getutid_r_file, getutline_r_file): Remove check for validity of file
	descriptor.  At this point the file is guaranteed to be open.
	(getutent_r_file, internal_getut_r, getutline_r_file, pututline_file,
	updwtmp_file): Do not wait when unlocking file.
	* login/utmp_daemon.c: General cleanup and a bit of reordering.
	(getutent_r_daemon, pututline_daemon): Do not call setutent_daemon.
	At this point the socket is guaranteed to be open (assert!).
	(getutid_r_daemon, getutline_r_daemon): Do not check if daemon_sock is
	valid.  At this point the socket is guaranteed to be open (assert!).
	* login/getutline_r.c: Remove unnecessary header files.
	(__getutline_r): Do not call backend setutent.
	* login/getutid_r.c: Remove unnecessary header files.
	(__getutid_r): Do not call backend setutent.
	* login/getutent_r.c: Remove unneccesary header files.
	(__libc_utmp_unknown_functions): Added getutid_r_unknown,
	getutline_r_unknown.
	(setutent_unknown): Only set file backend if setutent for the file
	backend was successful.
	(getutent_r_unknown, pututline_unknown): Call setutent_unknown instead
	of __setutent.  Report failure if no backend was selected.
	(getutid_r_unknown): New function.
	(getutline_r_unknown): New function.
	(__endutent): Reset backend.  This makes sure all backends are checked
	on the next setutent call.

1997-08-08 20:20  Thorsten Kukuk  <kukuk@vt.uni-paderborn.de>

	* nis_cache.c: Replace dummy functions.

	* libc.map: Add xdr_sizeof symbol.

	* sunrpc/Makefile: Add xdr_sizeof to routines.
	* sunrpc/rpc/xdr.h: Add xdr_sizeof prototype.
	* sunrpc/xdr_sizeof.c: New, from tirpc 2.3.

1997-08-08  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/bits/libc-lock.h (__libc_once): Define correctly.

1997-08-07  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/hurd/profil.c (fetch_samples): Put parens in the
	right place.

1997-08-06  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/hurd/profil.c (fetch_samples): Do arithmetic on
	PC's in long long to avoid overflow.

1997-08-07  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/bits/libc-lock.h (__libc_once, __libc_once_define):
	New macros.

1997-08-06  Andreas Jaeger  <aj@arthur.rhein-neckar.de>

	* nis/Makefile (headers): Remove bits/nislib.h.

1997-08-06 14:54  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/mach/hurd/Subdirs: Add login.

1997-08-06 14:23  Klaus Espenlaub  <kespenla@hydra.informatik.uni-ulm.de>

	* db/hash/hash.c (init_hash): Don't use stat() if it doesn't provide
	the preferred block size.

	* login/programs/database.c (store_state_entry): Don't compile if
	there is no ut_type field.
	(store_state_entry, store_process_entry): Use the ut_tv field
	for timestamps if supported.

	* login/programs/utmpdump.c (print_entry): Always use ut_tv field.

	* login/programs/xtmp.c: Fix numerous xtmp/utmp typos.  Use the ut_tv
	field for timestamps if supported.

	* login/programs/xtmp.h: Fix xtmp/utmp typo.

	* sysdeps/posix/defs.c (stdstream): Change (PTR) to (void *).

	* sysdeps/stub/connect.c (connect): Change to __connect, add alias.
	* sysdeps/stub/send.c (send): Likewise.
	* sysdeps/stub/s_exp2f.c: Emit correct stub_warning().
	* sysdeps/stub/statfs.c: Move stub_warning() to the usual place.
	* sysdeps/stub/init-first.c: Add definition of __libc_pid.

1997-08-05 13:28  Philip Blundell  <pb@spring.nexus.co.uk>

	* sysdeps/standalone/arm/bits/errno.h: Add EISDIR, EOPNOTSUPP;
	tidy up formatting.

	* Makefile (subdirs): Remove `login'.
	* sysdeps/unix/Subdirs: New file; build `login' subdirectory
	for Unix systems.

1997-08-05  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* sysdeps/generic/bits/utmpx.h: New file.

	* sysdeps/mach/hurd/Dist: Add some files.
	* sysdeps/mips/Dist: Likewise.
	* sysdeps/mips/mips64/Dist: Likewise.
	* sysdeps/sparc/Dist: Likewise.
	* sysdeps/unix/sysv/linux/mips/Dist: Likewise.
	* sysdeps/unix/sysv/linux/sparc/Dist: Likewise.
	* sysdeps/mips/mipsel/Dist: New file.
	* sysdeps/sparc64/elf/Dist: New file.
	* sysdeps/unix/sysv/linux/sparc64/Dist: New file.

1997-08-05  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* libc.map: Add missing symbols.

1997-08-05  Andreas Jaeger  <aj@arthur.rhein-neckar.de>

	* manual/socket.texi: Correct typos.

	* manual/math.texi: Correct typos.

	* manual/time.texi (Formatting Date and Time): Likewise.

1997-08-04 13:06  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* gmon/gmon.c (write_gmon): New function; guts from _mcleanup.
	(_mcleanup): Use write_gmon.
	(write_profiling): This function can be safely called at any time
	to write a current histogram without interfering with ongoing
	profiling.

	* sysdeps/mach/hurd/profil.c (fetch_samples): Initialize NSAMPLES.

1997-08-01 17:53  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/hurd/profil.c (fetch_samples): Sample buffer need
	not be vm_deallocated; it's a stack buffer.
	(profil_vm_deallocate): Delete prototype.
	(#include <../mach/RPC_vm_deallocate_rpc.c>): Drop this
	inclusion.
	* sysdeps/mach/hurd/Makefile
	($(common-objpfx)hurd/../mach/RPC_vm_deallocate_rpc.c): Delete
	this rule.

	* sysdeps/mach/hurd/profil.c (fetch_samples): New function, guts
	from profile_waiter.
	(profile_waiter): Use fetch_samples.
	(profil): When turning off profiling, fetch the last bunch of
	samples before disabling PC sampling.
	(fetch_samples): Add prototype.

1997-07-30 12:53  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/hurd/Makefile: Give normal rules for the mach RPC
	source included by profil.c instead of trying to use
	before-compile grot.

1997-07-23 15:04  Thomas Bushnell, n/BSG  <thomas@gnu.ai.mit.edu>

	* sysdeps/mach/hurd/profil.c (profile_waiter): Do vm_deallocate
	after releasing lock.  Use special RPC stubs and avoid assert,
	thus making this function entirely independent of the threadvar
	mechanism and of cthreads.
	(lock): Convert to be a spin lock instead of a mutex, so that the
	waiter thread need not use cthreads.  Use a fork prepare hook
	instead of _hurd_fork_locks to make sure we are clean before
	forking.
	(fork_profil_child): Renamed from fork_profil.
	(profil_reply_port): New variable.
	(profil_vm_deallocate, profil_task_get_sampled_pcs): Special RPC
	stubs made by special hacks.
	* sysdeps/mach/hurd/Makefile (before-compile): Add the mach RPC
	source files that profil.c needs if we are in the gmon directory.

	* mach/setup-thread.c (__mach_setup_thread): Delete avoidance of a
	cthreads bug that hasn't existed for two years.

	* stdio-common/printf_fp.c (__printf_fp): Correct rounding of numbers
	* locale/programs/locale.c (print_escaped): New function.
	* time/australasia: Likewise.
	(open_database, synchronize_database, initialize_database):
	* nss/nss_files/files-netgrp.c (_nss_netgroup_parseline): Don't
	* Makerules (load-map-file): Currectly handle missing map file.
	cannot rely on M_SQRT2 being defined.
	(log1p): Use __M_SQRT2 not M_SQRT2.

1 files changed, 28 insertions, 28 deletions

diff --git a/manual/math.texi b/manual/math.texi
index d4449bb..71b3651 100644
--- a/manual/math.texi
+++ b/manual/math.texi
@@ -232,7 +232,7 @@ infinite.
 @item
 Squre root if the operand is less then zero.
 @item
-Conversion of an internal floating-point number to an integer or toa
+Conversion of an internal floating-point number to an integer or to a
 decimal string when overflow, infinity, or NaN precludes a faithful
 representation in that format and this cannot otherwise be signaled.
 @item
@@ -247,7 +247,7 @@ If the exception does not cause a trap handler to be called the result
 of the operation is taken as a quiet NaN.
 
 @item Division by Zero
-This exception is raised of the devisor is zero and the dividend is a
+This exception is raised if the devisor is zero and the dividend is a
 finite nonzero number.  If no trap occurs the result is either
 @iftex
 @tex
@@ -270,8 +270,8 @@ $-\infty$
 signs of the operands.
 
 @item Overflow
-This exception is signalled whenever if the result cannot be represented
-as a finite value in the destination precision's format.  If no trap
+This exception is signalled whenever the result cannot be represented
+as a finite value in the precision format of the destination.  If no trap
 occurs the result depends on the sign of the intermediate result and the
 current rounding mode (@w{IEEE 754}, @w{section 7.3}):
 @enumerate
@@ -334,7 +334,7 @@ $\infty$.
 @end enumerate
 
 @item Underflow
-The underflow exception is created when a intermediate result is too
+The underflow exception is created when an intermediate result is too
 small for the operation or if the operations result rounded to the
 destination precision causes a loss of accuracy by approximating the
 result by denormalized numbers.
@@ -342,17 +342,17 @@ result by denormalized numbers.
 When no trap is installed for the underflow exception, underflow shall
 be signaled (via the underflow flag) only when both tininess and loss of
 accuracy have been detected.  If no trap handler is installed the
-operation continues witht he inprecise small value or zero if the
+operation continues with an inprecise small value or zero if the
 destination precision cannot hold the small exact result.
 
 @item Inexact
-This exception is signaled if the rounded result is not exact (such as
+This exception is signalled if the rounded result is not exact (such as
 computing the square root of two) or the result overflows without an
 overflow trap.
 @end table
 
 To control whether an exception causes a trap to occur all @w{IEEE 754}
-conformant floating-point implementation (either hardware or software)
+conformant floating-point implementations (either hardware or software)
 have a control word.  By setting specific bits for each exception in
 this control word the programmer can decide whether a trap is wanted or
 not.
@@ -378,42 +378,42 @@ actual implementation can be very different, depending on the underlying
 hardware or software.
 
 To address the single exception the @file{fenv.h} headers defines a
-number macros:
+number of macros:
 
 @vtable @code
 @comment fenv.h
 @comment ISO
 @item FE_INEXACT
-Represent the inexact exception iff the FPU supports this exception.
+Represents the inexact exception iff the FPU supports this exception.
 @comment fenv.h
 @comment ISO
 @item FE_DIVBYZERO
-Represent the divide by zero exception iff the FPU supports this exception.
+Represents the divide by zero exception iff the FPU supports this exception.
 @comment fenv.h
 @comment ISO
 @item FE_UNDERFLOW
-Represent the underflow exception iff the FPU supports this exception.
+Represents the underflow exception iff the FPU supports this exception.
 @comment fenv.h
 @comment ISO
 @item FE_OVERFLOW
-Represent the overflow exception iff the FPU supports this exception.
+Represents the overflow exception iff the FPU supports this exception.
 @comment fenv.h
 @comment ISO
 @item FE_INVALID
-Represent the invalid exception iff the FPU supports this exception.
+Represents the invalid exception iff the FPU supports this exception.
 @end vtable
 
 The macro @code{FE_ALL_EXCEPT} is the bitwise OR of all exception macros
 which are supported by the FP implementation.
 
-Each of the supported exception flag can either be set or unset.  The
+Each of the supported exception flags can either be set or unset.  The
 @w{ISO C 9X} standard defines functions to set, unset and test the
 status of the flags.
 
 @comment fenv.h
 @comment ISO
 @deftypefun void feclearexcept (int @var{excepts})
-This functions clears all of the supported exception flags denoted by
+This function clears all of the supported exception flags denoted by
 @var{excepts} in the status word.
 @end deftypefun
 
@@ -430,7 +430,7 @@ exception flags indicated by the parameter @var{excepts}.
 @end deftypefun
 
 @noindent
-To restore the previously saved values one can use this functions:
+To restore the previously saved values one can use this function:
 
 @comment fenv.h
 @comment ISO
@@ -486,7 +486,7 @@ control word for the exceptions and the rounding mode can be safed.
 
 The file @file{fenv.h} defines the type @code{fenv_t}.  The layout of a
 variable of this type is implementation defined but the variable is able
-to contain the complete status informations.  To fill a variable of this
+to contain the complete status information.  To fill a variable of this
 type one can use this function:
 
 @comment fenv.h
@@ -497,7 +497,7 @@ Store the current floating-point environment in the object pointed to by
 @end deftypefun
 
 @noindent
-Another possibility which is useful is several situations is
+Another possibility which is useful in several situations is
 
 @comment fenv.h
 @comment ISO
@@ -505,7 +505,7 @@ Another possibility which is useful is several situations is
 Store the current floating-point environment in the object pointed to by
 @var{envp}.  Afterwards, all exception flags are cleared and if
 available a mode is installed which continues on all exception and does
-not cause a trap to occur.  In ths case a nonzero value is returned.
+not cause a trap to occur.  In this case a nonzero value is returned.
 
 If the floating-point implementation does not support such a non-stop
 mode, the return value is zero.
@@ -605,14 +605,14 @@ $-\infty$.
 Round toward zero.
 @end vtable
 
-At any time one of the four rounding modes above is selected.  To get
+At any time one of the above four rounding modes is selected.  To get
 information about the currently selected mode one can use this function:
 
 @comment fenv.h
 @comment ISO
 @deftypefun int fegetround (void)
 Return the currently selected rounding mode, represented by one of the
-values of the defined rouding mode macros.
+values of the defined rounding mode macros.
 @end deftypefun
 
 @noindent
@@ -625,18 +625,18 @@ Change the currently selected rounding mode to the mode described by the
 parameter @var{round}.  If @var{round} does not correspond to one of the
 supported rounding modes nothing is changed.
 
-The function return a nonzero value iff the requested rounding mode can
+The function returns a nonzero value iff the requested rounding mode can
 be established.  Otherwise zero is returned.
 @end deftypefun
 
-Changing the rounding mode can be necessary for various reasons.  But
+Changing the rounding mode might be necessary for various reasons.  But
 changing the mode only to round a given number normally is no good idea.
 The standard defines a set of functions which can be used to round an
 argument according to some rules and for all of the rounding modes there
 is a corresponding function.
 
 If a large set of number has to be rounded it might be good to change
-the rounding mode and do not use the function the library provides.  So
+the rounding mode and to not use the function the library provides.  So
 the perhaps necessary switching of the rounding mode in the library
 function can be avoided.  But since not all rounding modes are
 guaranteed to exist on any platform this possible implementation cannot
@@ -684,7 +684,7 @@ The value is the reziprocal of the square root of the value of the number pi.
 
 ALl values are defined as @code{long double} values unless the compiler
 does not support this type or @code{__STDC__} is not defined (both is
-unlikey).  Historically the numbers were @code{double} values and some
+unlikely).  Historically the numbers were @code{double} values and some
 old code still relies on this so you might want to add explizit casts if
 the extra precision of the @code{long double} value is not needed.  One
 critical case are functions with a variable number of arguments, such as
@@ -705,7 +705,7 @@ this value is not added to @file{math.h}.  Every program should use
 @section Floating-Point Comparison Functions
 @cindex unordered comparison
 
-The @w{IEEE 754} standards defines s'a set of functions which allows to
+The @w{IEEE 754} standards defines a set of functions which allows to
 compare even those numbers which normally would cause an exception to be
 raised since they are unordered.  E.g., the expression
 
@@ -774,7 +774,7 @@ All the macros are defined in a way to ensure that both arguments are
 evaluated exactly once and so they can be used exactly like the builtin
 operators.
 
-On several platform these macros are mapped on very efficient functions
+On several platform these macros are mapped to very efficient functions
 the processor understands.  But on machines missing these functions, the
 macros above might be rather slow.  So it is best to use the builtin
 operators unless it is necessary to use unordered comparisons.