Update.

1999-12-17  Ulrich Drepper  <drepper@cygnus.com>

	* string/bits/string2.h (__strtok_r_1c): Help gcc optimizing string
	access.

	* locale/programs/ld-collate.c: Implement handling of absolute
	ellipsis.  Parsing of file and constructing the internal data
	structures should now be complete.
	(collate_finish): Start adding support to generate the data
	structures which are written out to the file.

	* intl/dcgettext.c: Rewrite to handle caching of previous results here
	instead of in the dcgettext macro.
	* intl/libintl.h (dcgettext): Don't define for systems using this
	glibc or systems with tsearch.

	* sysdeps/generic/mathdef.h: Protect definitions for math.h
	against double inclusion.
	* sysdeps/alpha/fpu/bits/mathdef.h: Likewise.
	* sysdeps/i386/fpu/bits/mathdef.h: Likewise.
	* sysdeps/m68k/fpu/bits/mathdef.h: Likewise.
	* sysdeps/powerpc/fpu/bits/mathdef.h: Likewise.

	* sysdeps/i386/fpu/libm-test-ulps: Add more deltas (are mobile PIIs
	that different?).

1999-12-17  Andreas Jaeger  <aj@suse.de>

	* rt/aio.h (struct aiocb64): Add member __next_prio to sync the
	struct with aiocb.

	* rt/Makefile (tests): Added tst-aio64.
	Added dependency rules for tst-aio64.

	* rt/tst-aio64.c: New file, copied from tst-aio.c and changed for
	64bit tests.

1999-12-15  Thorsten Kukuk  <kukuk@suse.de>

	* sysdeps/unix/sysv/linux/alpha/oldgetrlimit64.c: Removed.
	* sysdeps/unix/sysv/linux/alpha/oldsetrlimit64.c: Removed.
	* sysdeps/unix/sysv/linux/bits/resource.h: Change RLIM_INFINITY back to
	old value (signed long).
	* sysdeps/unix/sysv/linux/i386/bits/resource.h: New, with unsigned
	long RLIM_INFINITY.
	* sysdeps/unix/sysv/linux/getrlimit.c: Moved from here to ...
	* sysdeps/unix/sysv/linux/i386/getrlimit.c: ... here.
	* sysdeps/unix/sysv/linux/getrlimit64.c: Moved from here to ...
	* sysdeps/unix/sysv/linux/i386/getrlimit64.c: ... here.
	* sysdeps/unix/sysv/linux/oldgetrlimit64.c: Moved from here to ...
	* sysdeps/unix/sysv/linux/i386/oldgetrlimit64.c: ... here.
	* sysdeps/unix/sysv/linux/oldsetrlimit64.c: Moved from here to ...
	* sysdeps/unix/sysv/linux/i386/oldsetrlimit64.c: ... here.
	* sysdeps/unix/sysv/linux/setrlimit.c: Moved from here to ...
	* sysdeps/unix/sysv/linux/i386/setrlimit.c: ... here.
	* sysdeps/unix/sysv/linux/setrlimit64.c: Moved from here to ...
	* sysdeps/unix/sysv/linux/i386/setrlimit64.c: ... here.
	* sysdeps/unix/sysv/linux/sparc/bits/resource.h: New.
	* sysdeps/unix/sysv/linux/sparc/sparc64/oldgetrlimit64.c: Removed.
	* sysdeps/unix/sysv/linux/sparc/sparc64/oldsetrlimit64.c: Removed.

1999-12-17  Andreas Jaeger  <aj@suse.de>

	* elf/ldconfig.c: Add new option -l to manualy link shared
	libraries.
	(options): Added option.
	(parse_opt): Set option.
	(main): Handle option.
	(manual_link): New function.

1999-12-17  Thorsten Kukuk  <kukuk@suse.de>

	* string/bits/string2.h: Fix patch from 1999-12-07.

1999-12-16  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/generic/strsep.c: If delim string has only one character
	don't run over end of string.

	* locale/programs/ld-collate.c (insert_weights): Also update next
	pointer of last cursor element.
	(insert_value): Return nonzero value if nothing got inserted.
	(handle_ellipsis): Don't do anything if to-value cannot be inserted.

1999-12-10  Jakub Jelinek  <jakub@redhat.com>

	* stdlib/longlong.h (__sparc_v9__): Use %rDIGIT instead of %DIGIT
	where appropriate.

1999-12-10  Jakub Jelinek  <jakub@redhat.com>

	* sysdeps/unix/sysv/linux/sparc/sparc64/sigaction.c (__sigaction):
	Copy sa_flags into kernel sigaction structure.

1999-12-14  Andreas Jaeger  <aj@suse.de>

	* string/tester.c (test_strsep): More tests for access beyond
	the final NUL.  The first two tests come from PR libc/1486 by
	martinea@iro.umontreal.ca.

1999-12-14  Thorsten Kukuk  <kukuk@suse.de>

	* nis/ypclnt.c: Correct handling of cached client handles.
	(__xdr_ypresp_all): Call callback function for errors, too,
	like Solaris does.
	* nis/nss_compat/compat-grp.c: Make sure errno is always set correct.
	* nis/nss_compat/compat-initgroups.c: Likewise.
	* nis/nss_compat/compat-spwd.c: Likewise.
	* nis/nss_nis/nis-alias.c: Likewise.
	* nis/nss_nis/nis-ethers.c: Likewise.
	* nis/nss_nis/nis-grp.c: Likewise.
	* nis/nss_nis/nis-hosts.c: Likewise.
	* nis/nss_nis/nis-netgrp.c: Likewise.
	* nis/nss_nis/nis-publickey.c: Likewise.
	* nis/nss_nis/nis-service.c: Likewise. Also use services.byservicename
	Map if available, optimize query if name/port and protocol is known.

2 files changed, 52 insertions, 15 deletions

diff --git a/sysdeps/i386/fpu/bits/mathdef.h b/sysdeps/i386/fpu/bits/mathdef.h
index ba5129895d..9436877a77 100644
--- a/sysdeps/i386/fpu/bits/mathdef.h
+++ b/sysdeps/i386/fpu/bits/mathdef.h
@@ -20,7 +20,9 @@
 # error "Never use <bits/mathdef.h> directly; include <math.h> instead"
 #endif
 
-#if defined __USE_ISOC99 && defined _MATH_H
+#if defined __USE_ISOC99 && defined _MATH_H && !defined _MATH_H_MATHDEF
+# define _MATH_H_MATHDEF	1
+
 /* The ix87 FPUs evaluate all values in the 80 bit floating-point format
    which is also available for the user as `long double'.  Therefore we
    define:  */
diff --git a/sysdeps/i386/fpu/libm-test-ulps b/sysdeps/i386/fpu/libm-test-ulps
index 443117da63..195bcca168 100644
--- a/sysdeps/i386/fpu/libm-test-ulps
+++ b/sysdeps/i386/fpu/libm-test-ulps
@@ -488,10 +488,14 @@ Test "Real part of: ctanh (-2 - 3 i) == -0.9653858790221331242 + 0.0098843750383
 ildouble: 2
 ldouble: 2
 Test "Imaginary part of: ctanh (-2 - 3 i) == -0.9653858790221331242 + 0.0098843750383224937 i":
+float: 1
+ifloat: 1
 ildouble: 23
 ldouble: 23
 Test "Real part of: ctanh (0 + pi/4 i) == 0.0 + 1.0 i":
 Test "Imaginary part of: ctanh (0 + pi/4 i) == 0.0 + 1.0 i":
+float: 1
+ifloat: 1
 double: 0.5
 idouble: 0.5
 Test "Real part of: ctanh (0.7 + 1.2 i) == 1.3472197399061191630 + 0.4778641038326365540 i":
@@ -636,6 +640,9 @@ idouble: 1
 Test "j0 (8.0) == 0.17165080713755390609":
 float: 1
 ifloat: 1
+Test "j0 (10.0) == -0.24593576445134833520":
+double: 2
+idouble: 2
 
 # j1
 Test "j1 (10.0) == 0.043472746168861436670":
@@ -654,6 +661,8 @@ ifloat: 1
 Test "jn (0, 10.0) == -0.24593576445134833520":
 float: 1
 ifloat: 1
+double: 2
+idouble: 2
 Test "jn (0, 2.0) == 0.22389077914123566805":
 float: 1
 ifloat: 1
@@ -673,33 +682,46 @@ idouble: 1
 Test "jn (1, 8.0) == 0.23463634685391462438":
 float: 1
 ifloat: 1
+Test "jn (10, -1.0) == 0.26306151236874532070e-9":
+float: 1
+ifloat: 1
 Test "jn (10, 0.1) == 0.26905328954342155795e-19":
 double: 4
 float: 2
 idouble: 4
 ifloat: 2
 Test "jn (10, 0.7) == 0.75175911502153953928e-11":
-double: 3
+double: 4
 float: 1
-idouble: 3
+idouble: 4
 ifloat: 1
-Test "jn (10, 2.0) == 0.25153862827167367096e-6":
+Test "jn (10, 1.0) == 0.26306151236874532070e-9":
 float: 1
 ifloat: 1
-double: 1
-idouble: 1
-Test "jn (10, 10.0) == 0.20748610663335885770":
+Test "jn (10, 2.0) == 0.25153862827167367096e-6":
 float: 1
 ifloat: 1
+double: 2
+idouble: 2
+Test "jn (10, 10.0) == 0.20748610663335885770":
+float: 2
+ifloat: 2
+double: 4
+idouble: 4
 Test "jn (3, 0.1) == 0.000020820315754756261429":
 double: 1
 idouble: 1
 Test "jn (3, 0.7) == 0.0069296548267508408077":
 double: 2
 idouble: 2
+Test "jn (3, 2.0) == 0.12894324947440205110":
+double: 1
+idouble: 1
 Test "jn (3, 10.0) == 0.058379379305186812343":
 float: 1
 ifloat: 1
+double: 3
+idouble: 3
 
 # lgamma
 Test "lgamma (-0.5) == log(2*sqrt(pi))":
@@ -847,6 +869,8 @@ float: 1
 idouble: 1
 ifloat: 1
 Test "y0 (0.7) == -0.19066492933739506743":
+float: 1
+ifloat: 1
 double: 2
 idouble: 2
 Test "y0 (1.0) == 0.088256964215676957983":
@@ -857,6 +881,8 @@ ifloat: 1
 Test "y0 (1.5) == 0.38244892379775884396":
 float: 1
 ifloat: 1
+double: 1
+idouble: 1
 Test "y0 (2.0) == 0.51037567264974511960":
 double: 1
 idouble: 1
@@ -891,9 +917,9 @@ idouble: 3
 ifloat: 2
 Test "y1 (2.0) == -0.10703243154093754689":
 double: 1
-float: 1
+float: 2
 idouble: 1
-ifloat: 1
+ifloat: 2
 Test "y1 (8.0) == -0.15806046173124749426":
 float: 2
 ifloat: 2
@@ -905,6 +931,8 @@ float: 1
 idouble: 1
 ifloat: 1
 Test "yn (0, 0.7) == -0.19066492933739506743":
+float: 1
+ifloat: 1
 double: 2
 idouble: 2
 Test "yn (0, 1.0) == 0.088256964215676957983":
@@ -915,6 +943,8 @@ ifloat: 1
 Test "yn (0, 1.5) == 0.38244892379775884396":
 float: 1
 ifloat: 1
+double: 1
+idouble: 1
 Test "yn (0, 2.0) == 0.51037567264974511960":
 double: 1
 idouble: 1
@@ -947,9 +977,9 @@ idouble: 3
 ifloat: 2
 Test "yn (1, 2.0) == -0.10703243154093754689":
 double: 1
-float: 1
+float: 2
 idouble: 1
-ifloat: 1
+ifloat: 2
 Test "yn (1, 8.0) == -0.15806046173124749426":
 float: 2
 ifloat: 2
@@ -974,8 +1004,8 @@ ifloat: 1
 Test "yn (10, 2.0) == -129184.54220803928264":
 float: 1
 ifloat: 1
-double: 1
-idouble: 1
+double: 2
+idouble: 2
 Test "yn (3, 0.1) == -5099.3323786129048894":
 double: 1
 float: 1
@@ -984,6 +1014,9 @@ ifloat: 1
 Test "yn (3, 0.7) == -15.819479052819633505":
 double: 2
 idouble: 2
+Test "yn (3, 2.0) == -1.1277837768404277861":
+double: 1
+idouble: 1
 Test "yn (3, 10.0) == -0.25136265718383732978":
 double: 1
 float: 1
@@ -1263,6 +1296,8 @@ ildouble: 286
 ldouble: 286
 
 Function: Imaginary part of "ctanh":
+float: 1
+ifloat: 1
 double: 1
 idouble: 1
 ildouble: 3074
@@ -1313,8 +1348,8 @@ ldouble: 560
 Function: "j0":
 float: 1
 ifloat: 1
-double: 1
-idouble: 1
+double: 2
+idouble: 2
 
 Function: "j1":
 double: 1