Update for release

7 files changed, 1315 insertions, 907 deletions

diff --git a/Changelog b/Changelog
index 72f336406..6ce013ace 100644
--- a/Changelog
+++ b/Changelog
@@ -1,12 +1,48 @@
+0.9.25	15 December 2003
+
+See Changelog.full for the complete list of who did what.
+
+Note:
+    This release remains binary compatible with uClibc 0.9.21-24 as long as
+    you take care to avoid any configuraton changes that will break things.
+    We _were_ planning to break binary compatibilty in this release, but
+    decided to hold those changes so we could push out a bugfix release.
+
+    We _will_ break binary compatibilty in the upcoming 0.9.26 release to
+    implement a few things we have been postponing.  That should hopefully be
+    the last ABI change before we freeze the ABI for the upcoming 1.0.x
+    stable uClibc series.
+
+Release highlights:
+    o Fixed a ton of problems found using the LTP and NIST test suites.
+	The few remaining test failures are obscure corner cases, such as
+	a few functions that fail (correctly) with incorrect errno values.
+    o Fixed a longstanding pthreads bug -- amoung other things, you no
+	longer need to explicitly add -lpthread for perl to work.
+    o Added some sh optimized string functions
+    o Added sh64 shared library support
+    o The default malloc implementation has been replaced, and will
+	now actually free memory when told to do so.
+    o Both popen() and exec*() now behave themselves per SuSv3
+    o Better 2.6.x kernel header support
+    o Added support for Position Independent Executables (PIE) on x86
+    o Lots of other minor cleanups
+
+ -Erik
+
+
+
+
+
 0.9.24	15 December 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
     This release remains binary compatible with uClibc 0.9.21-23 as
     long as you take care to avoid any configuraton changes that will
-    break things.  
-    
+    break things.
+
     We currently plan to break binary compatibilty in the upcoming 0.9.25 to
     implement a few things we have been postponing, which will hopefully be
     the last change....
@@ -24,7 +60,7 @@ Release highlights:
 
 0.9.23	13 November 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
     This release is binary compatible with uClibc 0.9.21 and 0.9.22 as
@@ -52,7 +88,7 @@ Release highlights:
 
 0.9.22	8 November 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
     This release is binary compatible with the 0.9.21 release
@@ -82,7 +118,7 @@ Release highlights:
 
 0.9.21	9 September 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
     This release is not binary compatible with earlier releases.
@@ -113,7 +149,7 @@ Release highlights:
 
 0.9.20	30 June 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
     This release remains binary compatible with 0.9.18 and 0.9.19.
@@ -141,7 +177,7 @@ Release highlights:
 
 0.9.19	3 March 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
     This release remains binary compatible with 0.9.18 (except
@@ -190,7 +226,7 @@ Release highlights:
 
 0.9.18	12 February 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
 
@@ -241,7 +277,7 @@ Release highlights:
 
 0.9.17	25 January 2003
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Note:
 
@@ -331,7 +367,7 @@ Release highlights:
 
 0.9.16	8 November 2002
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 
 Release highlights:
     o   CRIS architecture and shared library support from Tobias Anderberg 
@@ -356,7 +392,7 @@ Release highlights:
 
 This is a minor bugfix release.
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 Release highlights:
     o   Eliminated the HAS_LONG_LONG option.  gcc always supports 
 	long long, and the option never excluded all long long anyways.
@@ -380,7 +416,7 @@ Release highlights:
 
 This is a minor bugfix release.
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 Release highlights:
     o   Fix a compile error when RPC and Pthread support
 	were both enabled.
@@ -406,7 +442,7 @@ Security Fixes
     o   There was an integer overflow bug in the xdr_array 
 	RPC code, per http://online.securityfocus.com/bid/5356
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 Release highlights:
     o   Add full shared library support for Hitachi SuperH (sh)
 	thanks to Stefan Allius and Edie C. Dost 
@@ -433,7 +469,7 @@ Release highlights:
 
 0.9.12	20 June 2002
 
-See Changelog.full for the complete list of who did what. 
+See Changelog.full for the complete list of who did what.
 Release highlights:
     o   Add full shared library support for mips (big and little
 	endian), thanks to a lot of hard work from Steven J. Hill
diff --git a/Changelog.full b/Changelog.full
index 0ec024ca8..fe74bea4c 100644
--- a/Changelog.full
+++ b/Changelog.full
@@ -1,928 +1,812 @@
 ---------------------
-PatchSet 2873 
-Date: 2003/11/13 12:45:50
+PatchSet 2938 
+Date: 2003/12/16 06:45:48
 Author: andersen
 Branch: HEAD
-Tag: uClibc_0_9_23 
-Log:
-Prepare for release
-
-Members: 
-	Changelog:1.21->1.22 
-	Changelog.full:1.15->1.16 
-	Changelog.full.last:INITIAL->1.1 
-	docs/uclibc.org/FAQ.html:1.19->1.20 
-	docs/uclibc.org/index.html:1.75->1.76 
-
----------------------
-PatchSet 2874 
-Date: 2003/11/16 04:50:49
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix a build problem when using 2.2.x kernel headers.
-
-Members: 
-	libc/sysdeps/linux/common/syscalls.c:1.120->1.121 
-
----------------------
-PatchSet 2875 
-Date: 2003/11/16 15:58:33
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Don't install shadow.h if shadow password support is disabled.
-
-Members: 
-	Makefile:1.237->1.238 
-
----------------------
-PatchSet 2876 
-Date: 2003/11/16 16:26:34
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from Jeffrey Baitis <baitisj@evolution.com> for mips arch tuning.
-
-Members: 
-	Rules.mak:1.155->1.156 
-	extra/Configs/Config.mips:1.38->1.39 
-
----------------------
-PatchSet 2877 
-Date: 2003/11/16 21:10:07
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-I forgot to update the thread locking in the last dst fix.
-
-Members: 
-	libc/misc/time/time.c:1.16->1.17 
-
----------------------
-PatchSet 2878 
-Date: 2003/11/17 01:24:07
-Author: mjn3
-Branch: HEAD
 Tag: (none) 
 Log:
-If __NR_ftruncate64 isn't defined, do the best we can.
+Update the website
 
 Members: 
-	libc/sysdeps/linux/common/ftruncate64.c:1.9->1.10 
+	docs/uclibc.org/news.html:1.2->1.3 
+	docs/uclibc.org/products.html:1.1->1.2 
 
 ---------------------
-PatchSet 2879 
-Date: 2003/11/18 19:57:05
-Author: mjn3
+PatchSet 2939 
+Date: 2003/12/16 06:47:27
+Author: andersen
 Branch: HEAD
-Tag: (none) 
+Tag: uClibc_0_9_24 
 Log:
-Kill ARCH_HAS_NO_MMU and use HAS_FPU only.  Wether floating point ops
-are implemented in hardware or via kernel emulation doesn't matter to
-the libc code.
+Update for release (no really this time ;-)
 
 Members: 
-	extra/Configs/Config.alpha:1.14->1.15 
-	extra/Configs/Config.arm:1.39->1.40 
-	extra/Configs/Config.cris:1.9->1.10 
-	extra/Configs/Config.e1:1.5->1.6 
-	extra/Configs/Config.h8300:1.17->1.18 
-	extra/Configs/Config.i386:1.36->1.37 
-	extra/Configs/Config.i960:1.13->1.14 
-	extra/Configs/Config.in.arch:1.11->1.12 
-	extra/Configs/Config.m68k:1.34->1.35 
-	extra/Configs/Config.microblaze:1.4->1.5 
-	extra/Configs/Config.sh:1.40->1.41 
-	extra/Configs/Config.v850:1.6->1.7 
+	Changelog:1.23->1.24 
+	Changelog.full:1.17->1.18 
 
 ---------------------
-PatchSet 2880 
-Date: 2003/11/19 07:21:10
-Author: mjn3
+PatchSet 2940 
+Date: 2003/12/16 07:11:11
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Fix the return value for fputs when passed an empty string.
-Indirectly detected by gmp-4.1.2 self-tests and reported by
-"Peter S. Mazinger" <ps.m@gmx.net>.
+fix grammer such it is possible to understand what is written...
 
 Members: 
-	libc/stdio/stdio.c:1.72->1.73 
+	docs/uclibc.org/news.html:1.3->1.4 
 
 ---------------------
-PatchSet 2881 
-Date: 2003/11/20 14:35:06
+PatchSet 2941 
+Date: 2003/12/16 08:00:28
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Fixup dependancies to ensure arch specific string functions
-will always win
+Add LEAF Bering-uClibc
 
 Members: 
-	libc/string/Makefile:1.56->1.57 
+	docs/uclibc.org/products.html:1.2->1.3 
 
 ---------------------
-PatchSet 2882 
-Date: 2003/11/20 14:40:50
+PatchSet 2942 
+Date: 2003/12/17 04:23:42
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Make certain that the arch specific stuff is always compiled last
+Add in the axis network cameras
 
 Members: 
-	libc/sysdeps/linux/Makefile:1.19->1.20 
+	docs/uclibc.org/products.html:1.3->1.4 
 
 ---------------------
-PatchSet 2883 
-Date: 2003/11/20 14:42:42
+PatchSet 2943 
+Date: 2003/12/17 06:41:20
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Make certain that arch specific stuff is compiled last
+When I switched from using stack allocated space for printf, I missed a case
+where a sizeof(foo) was changed to the sizeof a pointer.  This caused
+_dl_printf to complain a lot when debug is enabled (which itself revealed a bug
+since it should have exited on buffer overflow), and let me to find another
+bug, where memory failures would try to recursively call _dl_printf....
+What a mess.
 
 Members: 
-	libm/Makefile:1.38->1.39 
+	ldso/ldso/readelflib1.c:1.52->1.53 
 
 ---------------------
-PatchSet 2884 
-Date: 2003/11/20 15:05:27
+PatchSet 2944 
+Date: 2003/12/17 07:57:31
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Lets try that one again...
+Patch from Paul Mundt <lethal@linux-sh.org>:
+
+The subject says it all.. optimized memset/memcpy/strcpy, lifted from SuperH's
+glibc tree.
 
 Members: 
-	libc/sysdeps/linux/Makefile:1.20->1.21 
+	libc/string/Makefile:1.58->1.59 
+	libc/string/sh64/Makefile:INITIAL->1.1 
+	libc/string/sh64/memcpy.S:INITIAL->1.1 
+	libc/string/sh64/memset.S:INITIAL->1.1 
+	libc/string/sh64/strcpy.S:INITIAL->1.1 
 
 ---------------------
-PatchSet 2885 
-Date: 2003/11/20 15:20:22
+PatchSet 2945 
+Date: 2003/12/17 08:05:40
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Add an initial set of arm optimized string functions.  These
-do seem to make noticable speed improvement...
- -Erik
+Patch from Paul Mundt <lethal@linux-sh.org>:
 
-Members: 
-	libc/string/Makefile:1.57->1.58 
-	libc/string/arm/Makefile:INITIAL->1.1 
-	libc/string/arm/_memcpy.S:INITIAL->1.1 
-	libc/string/arm/bcopy.S:INITIAL->1.1 
-	libc/string/arm/bzero.S:INITIAL->1.1 
-	libc/string/arm/memcmp.S:INITIAL->1.1 
-	libc/string/arm/memcpy.S:INITIAL->1.1 
-	libc/string/arm/memmove.S:INITIAL->1.1 
-	libc/string/arm/memset.S:INITIAL->1.1 
-	libc/string/arm/strcmp.S:INITIAL->1.1 
-	libc/string/arm/strlen.S:INITIAL->1.1 
-	libc/string/arm/strncmp.S:INITIAL->1.1 
+For sh64 we need implicit access to the symtab, primarily to get at the
+->st_other value. This presently isn't possible, as PERFORM_BOOTSTRAP_RELOC()
+is invoked as such:
 
----------------------
-PatchSet 2886 
-Date: 2003/11/20 15:28:24
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Don't depend on BSD naming
+        PERFORM_BOOTSTRAP_RELOC(rpnt, reloc_addr, symbol_addr, load_addr);
 
-Members: 
-	libc/string/arm/_memcpy.S:1.1->1.2 
+while we can easily get the symtab_index value from rpnt->r_info, this still
+doesn't buy us easy access to the actual table. As such, I've modified
+PERFORM_BOOTSTRAP_RELOC() to take an additional SYMTAB argument. Most
+architectures aren't going to care about this, but unfortunately we don't
+have any other options for sh64.
 
----------------------
-PatchSet 2887 
-Date: 2003/11/20 17:13:53
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Always include locale.h, which was included by libintl.h only when
-optimizing.
+The following patch fixes up the API for what we need for sh64, and updates
+the other architectures appropriately.
 
 Members: 
-	libc/misc/gnu/obstack.c:1.3->1.4 
+	ldso/ldso/ldso.c:1.75->1.76 
+	ldso/ldso/arm/ld_sysdep.h:1.6->1.7 
+	ldso/ldso/cris/ld_sysdep.h:1.2->1.3 
+	ldso/ldso/i386/ld_sysdep.h:1.8->1.9 
+	ldso/ldso/m68k/ld_sysdep.h:1.3->1.4 
+	ldso/ldso/mips/ld_sysdep.h:1.6->1.7 
+	ldso/ldso/powerpc/ld_sysdep.h:1.7->1.8 
+	ldso/ldso/sh/ld_sysdep.h:1.7->1.8 
+	ldso/ldso/sparc/ld_sysdep.h:1.4->1.5 
 
 ---------------------
-PatchSet 2888 
-Date: 2003/11/20 20:05:53
+PatchSet 2946 
+Date: 2003/12/17 08:07:14
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Add missing weak alias for bcmp
+Patch from Paul Mundt <lethal@linux-sh.org>:
 
-Members: 
-	libc/string/arm/memcmp.S:1.1->1.2 
+Here's the patch for the ldso bits for sh64. This is still in need of a bunch
+of debugging, testing, etc. and is really only being submitted for general
+completeness. This assumes that the previous patches I've submitted have
+already been applied.
 
----------------------
-PatchSet 2889 
-Date: 2003/11/21 07:47:43
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-prevent some needless warnings when building w/2.6.x kernel headers
+I plan on playing with this and buildroot some more later, as I'd definitely
+like to see buildroot images for sh64.
 
 Members: 
-	include/features.h:1.55->1.56 
+	extra/Configs/Config.sh:1.45->1.46 
+	ldso/ldso/sh64/boot1_arch.h:INITIAL->1.1 
+	ldso/ldso/sh64/elfinterp.c:INITIAL->1.1 
+	ldso/ldso/sh64/ld_syscalls.h:INITIAL->1.1 
+	ldso/ldso/sh64/ld_sysdep.h:INITIAL->1.1 
+	ldso/ldso/sh64/resolve.S:INITIAL->1.1 
 
 ---------------------
-PatchSet 2890 
-Date: 2003/11/22 04:42:22
-Author: mjn3
+PatchSet 2947 
+Date: 2003/12/17 18:34:14
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Make sure the utils build gets the setting for CROSS.
+Use PAGE_SIZE rather than hard coding 4096
 
 Members: 
-	Makefile:1.238->1.239 
+	ldso/include/ldso.h:1.1->1.2 
+	ldso/ldso/ldso.c:1.76->1.77 
+	ldso/ldso/readelflib1.c:1.53->1.54 
+	ldso/ldso/arm/elfinterp.c:1.16->1.17 
 
 ---------------------
-PatchSet 2891 
-Date: 2003/11/22 06:51:17
-Author: mjn3
+PatchSet 2948 
+Date: 2003/12/19 11:45:21
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Fix arm detection for libfloat.
+dev systems updated
 
 Members: 
-	utils/Makefile:1.3->1.4 
+	docs/uclibc.org/FAQ.html:1.23->1.24 
+	docs/uclibc.org/news.html:1.4->1.5 
 
 ---------------------
-PatchSet 2892 
-Date: 2003/11/22 18:11:47
+PatchSet 2949 
+Date: 2003/12/22 08:53:21
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Oops.  Only enable debug when debugging...
+Update the config used when building a .deb
 
 Members: 
-	ldso/ldso/Makefile:1.49->1.50 
+	debian/changelog:1.17->1.18 
+	debian/config:1.2->1.3 
 
 ---------------------
-PatchSet 2893 
-Date: 2003/11/22 18:32:59
+PatchSet 2950 
+Date: 2003/12/22 08:56:37
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Only enable debugging when SUPPORT_LD_DEBUG is enabled, not DEBUGGING
+Use 'sed -i -e' rather than 'sed -ie' since we don't want
+to leave backup files with "e" appended scattered all over
+the place.
 
 Members: 
-	ldso/ldso/Makefile:1.50->1.51 
+	debian/rules:1.27->1.28 
 
 ---------------------
-PatchSet 2894 
-Date: 2003/11/22 21:21:39
+PatchSet 2951 
+Date: 2003/12/22 09:17:19
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Remove SAFECFLAGS.  It was a workaround for failures in old awk script
-used to generate the crti.S and crtn.S files.  Since we don't use that
-anymore, keeping the workaround makes no sense.
-
-Furthermore, in most cases, SAFECFLAGS was not picking up all the
-needed flags, causing crti.o and crtn.o to not be built PIC.
-Which is very bad.  Removing SAFECFLAGS and using CFLAGS fixes
-that as well.
+That was awfully stupid.
 
 Members: 
-	libc/sysdeps/linux/alpha/Makefile:1.10->1.11 
-	libc/sysdeps/linux/arm/Makefile:1.50->1.51 
-	libc/sysdeps/linux/cris/Makefile:1.11->1.12 
-	libc/sysdeps/linux/e1/Makefile:1.5->1.6 
-	libc/sysdeps/linux/h8300/Makefile:1.10->1.11 
-	libc/sysdeps/linux/i386/Makefile:1.63->1.64 
-	libc/sysdeps/linux/i960/Makefile:1.7->1.8 
-	libc/sysdeps/linux/m68k/Makefile:1.31->1.32 
-	libc/sysdeps/linux/microblaze/Makefile:1.4->1.5 
-	libc/sysdeps/linux/mips/Makefile:1.34->1.35 
-	libc/sysdeps/linux/powerpc/Makefile:1.27->1.28 
-	libc/sysdeps/linux/sh/Makefile:1.28->1.29 
-	libc/sysdeps/linux/sh64/Makefile:1.4->1.5 
-	libc/sysdeps/linux/sparc/Makefile:1.18->1.19 
-	libc/sysdeps/linux/v850/Makefile:1.17->1.18 
+	debian/rules:1.28->1.29 
 
 ---------------------
-PatchSet 2895 
-Date: 2003/11/23 18:45:12
-Author: mjn3
+PatchSet 2952 
+Date: 2003/12/22 10:47:04
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Protect against quoting changes.
+In unistd.h, getpgrp() is redirected to __getpgid(), but
+we did not have a __getpgid().   Fix that.
 
 Members: 
-	utils/Makefile:1.4->1.5 
+	libc/sysdeps/linux/common/syscalls.c:1.121->1.122 
 
 ---------------------
-PatchSet 2896 
-Date: 2003/11/23 18:46:00
-Author: mjn3
+PatchSet 2953 
+Date: 2003/12/22 11:19:49
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Hopefully this is correct for cris.
+implement the worthless cuserid() function we claim to support.
+This isn't in SuSv3, but is expected by at least some apps such
+as emacs...
 
 Members: 
-	utils/ldd.c:1.2->1.3 
+	libc/unistd/getlogin.c:1.4->1.5 
 
 ---------------------
-PatchSet 2897 
-Date: 2003/11/23 18:47:56
-Author: mjn3
+PatchSet 2954 
+Date: 2003/12/22 11:22:59
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Make sure we use the right CC setting when building utils.
-Not doing so was breaking the buildroot softfloat arm build.
+oops
 
 Members: 
-	Makefile:1.239->1.240 
+	libc/unistd/getlogin.c:1.5->1.6 
 
 ---------------------
-PatchSet 2898 
-Date: 2003/11/23 20:24:30
+PatchSet 2955 
+Date: 2003/12/27 23:30:30
 Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Ok.. this fixes the buildroot soft-float arm build.
-Also try to make sure the build breaks if we want soft float but
-don't know how to request it.
-
-Members: 
-	Rules.mak:1.156->1.157 
-	utils/Makefile:1.5->1.6 
-
----------------------
-PatchSet 2899 
-Date: 2003/11/24 04:23:31
+Fix a long-standing bug with pthreads.  A couple of linuxthreads files
+were including libc-lock.h which had a bunch of weak pragmas.  Also,
+uClibc supplied a number of no-op weak thread functions even though
+many weren't needed.  This combined result was that sometimes the
+functional versions of thread functions in pthread would not override
+the weaks in libc.
+
+While fixing this, I also prepended double-underscore to all necessary
+weak thread funcs in uClibc, and removed all unused weaks.
+
+I did a test build, but haven't tested this since these changes are
+a backport from my working tree.  I did test the changes there and
+no longer need to explicitly add -lpthread in the perl build for
+perl to pass its thread self tests.
+
+Members: 
+	include/pthread.h:1.3->1.4 
+	libc/inet/getnetent.c:1.4->1.5 
+	libc/inet/getproto.c:1.5->1.6 
+	libc/inet/getservice.c:1.5->1.6 
+	libc/inet/resolv.c:1.49->1.50 
+	libc/inet/rpc/create_xid.c:1.2->1.3 
+	libc/misc/dirent/closedir.c:1.5->1.6 
+	libc/misc/dirent/opendir.c:1.6->1.7 
+	libc/misc/dirent/readdir.c:1.8->1.9 
+	libc/misc/dirent/readdir64.c:1.9->1.10 
+	libc/misc/dirent/readdir64_r.c:1.3->1.4 
+	libc/misc/dirent/readdir_r.c:1.3->1.4 
+	libc/misc/dirent/rewinddir.c:1.5->1.6 
+	libc/misc/dirent/seekdir.c:1.5->1.6 
+	libc/misc/mntent/mntent.c:1.5->1.6 
+	libc/misc/pthread/weaks.c:1.3->1.4 
+	libc/misc/syslog/syslog.c:1.11->1.12 
+	libc/misc/time/time.c:1.18->1.19 
+	libc/misc/utmp/utent.c:1.9->1.10 
+	libc/misc/wchar/wstdio.c:1.4->1.5 
+	libc/pwd_grp/lckpwdf.c:1.4->1.5 
+	libc/pwd_grp/pwd_grp.c:1.4->1.5 
+	libc/stdio/stdio.c:1.73->1.74 
+	libc/stdlib/abort.c:1.13->1.14 
+	libc/stdlib/atexit.c:1.24->1.25 
+	libc/stdlib/random.c:1.4->1.5 
+	libc/stdlib/setenv.c:1.13->1.14 
+	libc/stdlib/malloc/heap.h:1.15->1.16 
+	libc/stdlib/malloc/malloc.h:1.17->1.18 
+	libc/stdlib/malloc-930716/malloc.c:1.13->1.14 
+	libc/stdlib/malloc-930716/memalign.c:1.3->1.4 
+	libc/stdlib/malloc-930716/realloc.c:1.4->1.5 
+	libc/sysdeps/linux/common/bits/uClibc_pthread.h:INITIAL->1.1 
+	libc/sysdeps/linux/common/bits/uClibc_stdio.h:1.14->1.15 
+	libpthread/linuxthreads/lockfile.c:1.2->1.3 
+	libpthread/linuxthreads/mutex.c:1.6->1.7 
+	libpthread/linuxthreads/specific.c:1.5->1.6 
+
+---------------------
+PatchSet 2956 
+Date: 2003/12/27 23:36:22
 Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Don't bother checking, since we want an error if -msoft-float doesn't work.
+Handle the app_fini stuff in exit without requiring atexit().
+This avoids pulling in all the malloc/free code for a simple true/false app.
 
 Members: 
-	utils/Makefile:1.6->1.7 
+	libc/misc/internals/__uClibc_main.c:1.28->1.29 
+	libc/stdlib/atexit.c:1.25->1.26 
 
 ---------------------
-PatchSet 2900 
-Date: 2003/11/24 04:24:31
-Author: mjn3
+PatchSet 2957 
+Date: 2003/12/28 00:25:31
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Add -msoft-float to CPU_CFLAGS instead of CFLAGS, since ldso and libdl
-Makefiles ignore CFLAGS.
+remove unused file
 
 Members: 
-	Rules.mak:1.157->1.158 
+	libpthread/linuxthreads/Makefile:1.12->1.13 
+	libpthread/linuxthreads/weaks.c:1.3->1.4(DEAD) 
 
 ---------------------
-PatchSet 2901 
-Date: 2003/11/28 18:28:39
-Author: mjn3
+PatchSet 2958 
+Date: 2003/12/28 07:41:24
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Add sysmips(), based on a patch by Atsushi Nemoto <anemo@mba.ocn.ne.jp>.
+Fix whitespace
 
 Members: 
-	libc/sysdeps/linux/mips/Makefile:1.35->1.36 
-	libc/sysdeps/linux/mips/sysmips.c:INITIAL->1.1 
-	libc/sysdeps/linux/mips/sys/sysmips.h:1.1->1.2 
+	libc/inet/getservice.c:1.6->1.7 
 
 ---------------------
-PatchSet 2902 
-Date: 2003/11/28 19:39:51
-Author: mjn3
+PatchSet 2959 
+Date: 2003/12/28 08:25:16
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Another missing mips function, needed by the boehm garbage collector.
+Fix a really dumb bug introduced in version 1.4 of this file (a patch for ipv6
+support) which could cause things like EOF and read errors while reading
+/etc/services to always return a TRY_AGAIN.  The perl test suite would alloc a
+larger buffer and try again until all memory was exhausted.  When we get a read
+error, or EOF, it means we didn't get what we wanted, and so we should return
+an error.  Doing so fixes the failing perl 5.8.2 test.
+ -Erik
 
 Members: 
-	libc/sysdeps/linux/mips/Makefile:1.36->1.37 
-	libc/sysdeps/linux/mips/_test_and_set.c:INITIAL->1.1 
+	libc/inet/getservice.c:1.7->1.8 
 
 ---------------------
-PatchSet 2903 
-Date: 2003/11/29 02:17:24
-Author: mjn3
+PatchSet 2960 
+Date: 2003/12/30 01:41:14
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Patch by "Bradley D. LaRonde" <brad@ltc.com> for boehm garbage collector.
+Update and restore malloc-simple.  Slow as molasses, but trivially
+simple and releases memory immediately when asked to do so.
+ -Erik
 
 Members: 
-	libc/sysdeps/linux/mips/crt0.S:1.10->1.11 
+	libc/stdlib/malloc-simple/Makefile:1.11->1.12 
+	libc/stdlib/malloc-simple/alloc.c:1.11->1.12 
 
 ---------------------
-PatchSet 2904 
-Date: 2003/11/29 20:48:18
-Author: mjn3
+PatchSet 2961 
+Date: 2003/12/30 10:40:47
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Minor update... Config no longer used.
+Rework malloc.  The new default implementation is based on dlmalloc from Doug
+Lea.  It is about 2x faster than the old malloc-930716, and behave itself much
+better -- it will properly release memory back to the system, and it uses a
+combination of brk() for small allocations and mmap() for larger allocations.
+ -Erik
 
 Members: 
-	.cvsignore:1.8->1.9 
+	debian/config:1.3->1.4 
+	extra/Configs/Config.in:1.41->1.42 
+	libc/stdlib/Makefile:1.54->1.55 
+	libc/stdlib/calloc.c:1.1->1.2(DEAD) 
+	libc/stdlib/malloc/Makefile:1.30->1.31 
+	libc/stdlib/malloc/calloc.c:1.3->1.4 
+	libc/stdlib/malloc-930716/Makefile:1.11->1.12(DEAD) 
+	libc/stdlib/malloc-930716/README:1.1->1.2(DEAD) 
+	libc/stdlib/malloc-930716/malloc.c:1.14->1.15(DEAD) 
+	libc/stdlib/malloc-930716/malloc.h:1.3->1.4(DEAD) 
+	libc/stdlib/malloc-930716/memalign.c:1.4->1.5(DEAD) 
+	libc/stdlib/malloc-930716/realloc.c:1.5->1.6(DEAD) 
+	libc/stdlib/malloc-simple/alloc.c:1.12->1.13 
+	libc/stdlib/malloc-standard/Makefile:INITIAL->1.1 
+	libc/stdlib/malloc-standard/calloc.c:INITIAL->1.1 
+	libc/stdlib/malloc-standard/free.c:INITIAL->1.1 
+	libc/stdlib/malloc-standard/mallinfo.c:INITIAL->1.1 
+	libc/stdlib/malloc-standard/malloc.c:INITIAL->1.1 
+	libc/stdlib/malloc-standard/malloc.h:INITIAL->1.1 
+	libc/stdlib/malloc-standard/mallopt.c:INITIAL->1.1 
+	libc/stdlib/malloc-standard/memalign.c:INITIAL->1.1 
+	libc/stdlib/malloc-standard/realloc.c:INITIAL->1.1 
 
 ---------------------
-PatchSet 2905 
-Date: 2003/11/29 21:23:42
-Author: mjn3
+PatchSet 2962 
+Date: 2003/12/30 10:45:45
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Minor typo fix.
+Make sleep behave itself properly inthe presence of SIGCHLD
 
 Members: 
-	extra/scripts/fix_includes.sh:1.8->1.9 
+	libc/unistd/sleep.c:1.6->1.7 
+	libc/unistd/usleep.c:1.1->1.2 
 
 ---------------------
-PatchSet 2906 
-Date: 2003/12/02 04:58:41
-Author: mjn3
+PatchSet 2963 
+Date: 2003/12/30 10:57:16
+Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Put in a dummy reference to main so busybox will link.
+Fix the manditory typo in my last minute untested fix.
+ -Erik
 
 Members: 
-	libc/sysdeps/linux/cris/crt0.c:1.4->1.5 
+	libc/stdlib/malloc-standard/calloc.c:1.1->1.2 
 
 ---------------------
-PatchSet 2907 
-Date: 2003/12/02 17:58:24
+PatchSet 2964 
+Date: 2003/12/31 11:50:08
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Avoid a potentially undefined operation with a veriable++ in a macro
+Add extra checks for sigprocmask and rt_sigprocmask syscalls.
+The rt_sigprocmask syscall has broken error handling in 2.4.x
+kernels, while the sigprocmask syscall appears to get things
+right.  Regardless we should be extra careful, and add these
+checks.
 
 Members: 
-	libc/inet/rpc/bindresvport.c:1.8->1.9 
+	libc/sysdeps/linux/common/syscalls.c:1.122->1.123 
 
 ---------------------
-PatchSet 2908 
-Date: 2003/12/02 17:59:06
+PatchSet 2965 
+Date: 2003/12/31 12:13:27
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Add function versions of ntohl and friends
+Fir errno return when the file/path is NULL or points to an empty string.
+ -Erik
 
 Members: 
-	libc/inet/Makefile:1.41->1.42 
-	libc/inet/ntohl.c:INITIAL->1.1 
+	libc/sysdeps/linux/common/syscalls.c:1.123->1.124 
 
 ---------------------
-PatchSet 2909 
-Date: 2003/12/02 19:06:31
+PatchSet 2966 
+Date: 2004/01/02 07:00:29
 Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Add a syscall() implementation using a hacked version of the syscall6 macro.
-Untested, but syscall() is needed by busybox for pivot_root at least.
+Kept the same approach, but rewrote the code for the most part.
+Fixed some minor issues plus (as I recall) one SUSv3 errno case.
 
 Members: 
-	libc/sysdeps/linux/sh/Makefile:1.29->1.30 
-	libc/sysdeps/linux/sh/syscall.c:INITIAL->1.1 
+	libc/termios/ttyname.c:1.11->1.12 
 
 ---------------------
-PatchSet 2910 
-Date: 2003/12/02 20:37:08
+PatchSet 2967 
+Date: 2004/01/02 07:01:13
 Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Include Rules.mk for consistency.  Trying not to was simply too painful.
+Minor fix from upstream glibc.
 
 Members: 
-	utils/Makefile:1.7->1.8 
+	libc/sysdeps/linux/common/bits/posix1_lim.h:1.1->1.2 
 
 ---------------------
-PatchSet 2911 
-Date: 2003/12/02 23:16:59
-Author: andersen
+PatchSet 2968 
+Date: 2004/01/02 07:11:35
+Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Disable pthreads for sh64 for the time being (since it is not
-currently supported)
+Rewrite popen for SUSv3 compliance.
+  Added a list of popen()'d to store pids and use waitpid() in pclose().
+  Loop on waitpid() failure due to EINTR as required.
+  Close parent's popen()'d FILEs in the {v}fork()'d child.
+  Fix failure exit code for failed execve().
 
 Members: 
-	extra/Configs/Config.sh:1.41->1.42 
+	libc/stdio/popen.c:1.8->1.9 
 
 ---------------------
-PatchSet 2912 
-Date: 2003/12/03 02:35:36
-Author: andersen
+PatchSet 2969 
+Date: 2004/01/02 08:44:58
+Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-force enable FORCE_SHAREABLE_TEXT_SEGMENTS for sh4
+Redo the exec functions to comply with SUSv3.
 
 Members: 
-	extra/Configs/Config.sh:1.42->1.43 
+	libc/unistd/Makefile:1.35->1.36 
+	libc/unistd/exec.c:INITIAL->1.1 
+	libc/unistd/execl.c:1.4->1.5(DEAD) 
+	libc/unistd/execle.c:1.2->1.3(DEAD) 
+	libc/unistd/execlp.c:1.5->1.6(DEAD) 
+	libc/unistd/execv.c:1.2->1.3(DEAD) 
+	libc/unistd/execvep.c:1.2->1.3(DEAD) 
+	libc/unistd/execvp.c:1.3->1.4(DEAD) 
 
 ---------------------
-PatchSet 2913 
-Date: 2003/12/03 20:12:45
-Author: andersen
+PatchSet 2970 
+Date: 2004/01/02 08:47:22
+Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Rob McMullen writes:
-There is code in ldconfig that checks to see if a shared library is named
-either ``lib*'' or ``ld-*'' before ldconfig -l will work on it. gawk uses this
-feature of ldconfig to make /lib/rcscripts/filefuncs.so, so the build fails.
+The checking Erik added seems to be unnecessary.  Without it, I am not
+seeing any LTP failures.
 
 Members: 
-	utils/ldconfig.c:1.1->1.2 
+	libc/sysdeps/linux/common/syscalls.c:1.124->1.125 
 
 ---------------------
-PatchSet 2914 
-Date: 2003/12/03 20:36:49
-Author: andersen
+PatchSet 2971 
+Date: 2004/01/02 09:20:33
+Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-remove some garbage
+Oops... temporarily support the old stdio code.
 
 Members: 
-	libc/sysdeps/linux/mips/vfork.c:1.1->1.2 
+	libc/stdio/popen.c:1.9->1.10 
 
 ---------------------
-PatchSet 2915 
-Date: 2003/12/03 21:35:01
-Author: andersen
+PatchSet 2972 
+Date: 2004/01/02 09:21:33
+Author: mjn3
 Branch: HEAD
 Tag: (none) 
 Log:
-Eliminate some static buffers based on a patch from Tobias Anderberg
+Fix __freadable and __fwritable... were using '~' instead of '!'. (ugh)
+Fix (hopefully) a potential problem with failed freopen() calls.  The
+  fix isn't tested since I've been working on the replacement stdio
+  core code which will go in after the next release.
 
 Members: 
-	ldso/ldso/readelflib1.c:1.50->1.51 
+	libc/stdio/stdio.c:1.74->1.75 
 
 ---------------------
-PatchSet 2916 
-Date: 2003/12/03 23:03:59
+PatchSet 2973 
+Date: 2004/01/02 10:52:38
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Paul Mundt <lethal@linux-sh.org> writes:
+rohde at soap dot adsl dot dk writes:
 
-Nothing overly interesting here, this renames Hitachi/Mitsubishi to Renesas
-for the relevant platforms (in this case, h8, sh, and m32r). The same changes
-have already been going on in gcc/binutils/gdb/glibc/etc.
+The macro to do some floating point checks in libc/sysdeps/linux/powerpc/setjmp.S is incorrect.
+The following should fix it.
 
 Members: 
-	extra/Configs/Config.sh:1.43->1.44 
-	include/elf.h:1.9->1.10 
-	utils/readelf.c:1.1->1.2 
+	libc/sys