Prepare for release

5 files changed, 3082 insertions, 2563 deletions

diff --git a/Changelog b/Changelog
index ac806380f..7e6b5d19c 100644
--- a/Changelog
+++ b/Changelog
@@ -1,3 +1,31 @@
+0.9.22	13 November 2003
+
+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
+    long as you take care to avoid any configuraton changes that will
+    break things.  Enabling or disabling things like soft-float,
+    locale, wide wchar support, or changing cpu type are all good
+    examples of things that will break binary compatibility.
+
+Release highlights:
+    o Fixed several silly configuration problems that were
+	present in the 0.9.22 release.
+    o Fixed compilation problem with soft-float support on
+	several architectures.
+    o Lots of cleanup work on the powepc shared lib loader
+	thanks to Joakim Tjernlund.
+    o Updated the debian packaging for use in a standalone uClibc
+	system, rather than being a subordinate library under a
+	glibc based system.
+
+ -Erik
+
+
+
+
+
 0.9.22	8 November 2003
 
 See Changelog.full for the complete list of who did what. 
diff --git a/Changelog.full b/Changelog.full
index b7bcbbd2f..ba8457c8a 100644
--- a/Changelog.full
+++ b/Changelog.full
@@ -1,2737 +1,469 @@
 ---------------------
-PatchSet 2680 
-Date: 2003/09/11 11:24:28
+PatchSet 2849 
+Date: 2003/11/08 07:42:34
 Author: andersen
 Branch: HEAD
 Tag: (none) 
 Log:
-Stefan Allius writes:
-
-1. Under Solaris the test command with the 'string' argument don't
-   work well. So I added the '-n' argument, which might work on all
-   platforms.
-
-2. The sed under Solaris don't support the '-s' argument, which is a
-   GNU extension. I changed it to '-e' which works fine for me.
-
-Members: 
-	extra/scripts/fix_includes.sh:1.4->1.5 
-	extra/scripts/get-needed-libgcc-objects.sh:1.20->1.21 
-
----------------------
-PatchSet 2681 
-Date: 2003/09/11 11:26:16
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Stefan Allius writes:
-
-Hello,
-
-my patch changed the format of the ldso debug output to the same
-format as on the i386 systems.
-
-By Stefan
-
-Members: 
-	ldso/ldso/sh/elfinterp.c:1.6->1.7 
-
----------------------
-PatchSet 2682 
-Date: 2003/09/11 11:47:23
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Simplify an expression.  Use static const on strings to avoid
-a gcc 2.95 compiler problem on powerpc.
-
-Members: 
-	ldso/ldso/readelflib1.c:1.45->1.46 
-
----------------------
-PatchSet 2683 
-Date: 2003/09/11 12:12:55
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Peter Kjellerstedt writes:
-
-The attached patch allows $(CC) to contain spaces
-(something which we use).
-
-Members: 
-	libc/Makefile:1.27->1.28 
-
----------------------
-PatchSet 2684 
-Date: 2003/09/11 12:25:20
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix several stupid bugs I let slip into the release
-
-Members: 
-	Rules.mak:1.139->1.140 
-
----------------------
-PatchSet 2685 
-Date: 2003/09/11 17:57:15
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch by Atsushi Nemoto <anemo@mba.ocn.ne.jp> to do arch-required
-mapping of signal strings (alpha, mips, hppa, sparc).
-
-Members: 
-	libc/string/wstring.c:1.17->1.18 
-
----------------------
-PatchSet 2686 
-Date: 2003/09/13 17:09:42
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix a problem reported by Atsushi Nemoto <anemo@mba.ocn.ne.jp>
-for environments where long and long long are the same size.
-
-Members: 
-	libc/stdio/scanf.c:1.30->1.31 
-
----------------------
-PatchSet 2687 
-Date: 2003/09/16 23:22:30
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-This should enabled building of uClibc locale support when using uClibc itself.
-Sorry I didn't test this before the release.
-
-Please remember that the locale data generation tools are not very robust,
-so doing something like disabling 8-bit codeset support is likely to break
-things.  As it stands, UTF-8 support is required, but I'm not sure I test
-for that.
-
-Also, you will notice a difference in the locale data generated by uClibc
-verses glibc.  That's because the bg_BG locale specifies use of grouping
-in LC_NUMERIC, but supplies no grouping char.  The uClibc locale code
-tests for and works around this (at the moment) by disabling grouping.
-But the result is slightly different data which ripples throughout the
-rest of the tables.
-
-Members: 
-	extra/locale/gen_wc8bit.c:1.4->1.5 
-	extra/locale/gen_wctype.c:1.3->1.4 
-	libc/sysdeps/linux/common/bits/uClibc_locale.h:1.9->1.10 
-
----------------------
-PatchSet 2688 
-Date: 2003/09/17 11:37:37
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix fcntl64 when UCLIBC_HAS_LFS is enabled, but __NR_fcntl64
-is not defined.
-
-Members: 
-	libc/sysdeps/linux/common/syscalls.c:1.114->1.115 
-
----------------------
-PatchSet 2689 
-Date: 2003/09/17 11:39:08
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from Philip Nye fixing mmu-less
-
-Members: 
-	extra/scripts/fix_includes.sh:1.5->1.6 
-
----------------------
-PatchSet 2690 
-Date: 2003/09/17 12:50:18
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-Make sure static heaps are aligned correctly.
-
-Members: 
-	libc/stdlib/malloc/heap.h:1.12->1.13 
-
----------------------
-PatchSet 2691 
-Date: 2003/09/17 15:53:37
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Make sure we use the local copy of elf.h, since some system
-versions are insufficient....
-
-Members: 
-	ldso/util/Makefile:1.33->1.34 
-
----------------------
-PatchSet 2692 
-Date: 2003/09/19 11:01:08
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-Add dependencies on header files.
-
-Members: 
-	libc/stdlib/malloc/Makefile:1.28->1.29 
-
----------------------
-PatchSet 2693 
-Date: 2003/09/19 11:10:33
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-Make sure we don't allocate too little space for static free-areas
-because of our fiddling with alignment (because doing so is VERY BAD).
-
-Members: 
-	libc/stdlib/malloc/heap.h:1.13->1.14 
-
----------------------
-PatchSet 2694 
-Date: 2003/09/19 13:06:05
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Added assembler version of startup code. Fix Makefile so it uses the new
-code.
-
-Members: 
-	libc/sysdeps/linux/cris/Makefile:1.6->1.7 
-	libc/sysdeps/linux/cris/crt0.S:INITIAL->1.1 
-
----------------------
-PatchSet 2695 
-Date: 2003/09/19 13:07:22
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Re-add cris to "define list" when fixing auxvt.
-
-Members: 
-	ldso/ldso/ldso.c:1.72->1.73 
-
----------------------
-PatchSet 2696 
-Date: 2003/09/19 13:11:14
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Initialize strings as pointers instead of arrays.
-
-Members: 
-	ldso/ldso/readelflib1.c:1.46->1.47 
-
----------------------
-PatchSet 2697 
-Date: 2003/09/19 13:11:43
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Corrected LD_BOOT macro
-
-Members: 
-	ldso/ldso/cris/boot1_arch.h:1.2->1.3 
-
----------------------
-PatchSet 2698 
-Date: 2003/09/19 13:12:29
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-* Partial re-write to match other archs and remove duplicate code.
-* Assign insead of add when doing relocations.
-
-Members: 
-	ldso/ldso/cris/elfinterp.c:1.4->1.5 
-
----------------------
-PatchSet 2699 
-Date: 2003/09/21 20:41:53
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix "subst -g,," problem for SAFECFLAGS.
-
-Members: 
-	libc/sysdeps/linux/arm/Makefile:1.43->1.44 
-	libc/sysdeps/linux/common/Makefile:1.85->1.86 
-	libc/sysdeps/linux/i386/Makefile:1.56->1.57 
-	libc/sysdeps/linux/powerpc/Makefile:1.20->1.21 
-	libc/sysdeps/linux/sh/Makefile:1.21->1.22 
-
----------------------
-PatchSet 2700 
-Date: 2003/09/22 03:57:18
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Ugh... EOF handling by scanf was completely broken.  :-(  Regretably,
-I got my mind fixed in one mode and didn't comply with the standards.
-Things should be fixed now, but comparision testing is difficult when
-glibc's scanf is broken and they stubbornly refuse to even acknowledge
-that it is... even when confronted by specific examples from the C99
-standards and from an official C standard defect report.
-
-Members: 
-	libc/stdio/scanf.c:1.31->1.32 
-
----------------------
-PatchSet 2701 
-Date: 2003/09/22 04:07:51
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Modify _stdio_READ to conform with C99, as stdio input behavior upon
-   encountering EOF changed with Defect Report #141.  In the current
-   standard, the stream's EOF indicator is "sticky".  Once it is set,
-   all further input from the stream should fail until the application
-   explicitly clears the EOF indicator (clearerr(), file positioning),
-   even if more data becomes available.
-Fixed a bug in fgets.  Wasn't checking for read errors.
-Minor thread locking optimizations to avoid some unnecessary locking.
-Remove the explicit calls to __builtin_* funcs, as we really need to
-   implement a more general solution.
-
-Members: 
-	libc/stdio/stdio.c:1.71->1.72 
-
----------------------
-PatchSet 2702 
-Date: 2003/09/22 04:29:37
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Document some glibc bugs that Ulrich Drepper refuses to acknowledge.  :-(
-
-Members: 
-	docs/Glibc_vs_uClibc_Differences.txt:1.11->1.12 
-
----------------------
-PatchSet 2703 
-Date: 2003/09/22 05:01:11
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-More fiddling with static free-areas: make sure both their size _and_
-their alignment are correct.
-
-Members: 
-	libc/stdlib/malloc/heap.h:1.14->1.15 
-
----------------------
-PatchSet 2704 
-Date: 2003/09/22 06:43:44
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-One more scanf difference.
-
-Members: 
-	docs/Glibc_vs_uClibc_Differences.txt:1.12->1.13 
-
----------------------
-PatchSet 2705 
-Date: 2003/09/23 23:49:24
-Author: mjn3
-Branch: HEAD
-Tag: (none) 
-Log:
-Quick fix to make sure vfprintf always checks stream orientation when
-wchar support is enabled.
-
-Members: 
-	libc/stdio/printf.c:1.48->1.49 
-
----------------------
-PatchSet 2706 
-Date: 2003/09/24 09:45:36
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Hm. Keep using crt0.c for now. My fever-hack of crt0.S didn't do the
-expected thing. A so called "D'oh!".
-
-Members: 
-	libc/sysdeps/linux/cris/Makefile:1.7->1.8 
-
----------------------
-PatchSet 2707 
-Date: 2003/09/26 02:33:57
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-Revert to using uClibc-specific c-symbol-prefix stuff
-(__C_SYMBOL_PREFIX__).
-
-Members: 
-	include/sys/cdefs.h:1.8->1.9 
-
----------------------
-PatchSet 2708 
-Date: 2003/09/29 22:46:00
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix SUPPORT_LD_DEBUG_EARLY on powerpc
-
-Members: 
-	ldso/include/ld_string.h:1.1->1.2 
-
----------------------
-PatchSet 2709 
-Date: 2003/09/30 08:06:02
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-oops.
-
-Members: 
-	extra/scripts/fix_includes.sh:1.6->1.7 
-
----------------------
-PatchSet 2710 
-Date: 2003/09/30 09:45:28
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Mention that I have updated the dev systems.
-
-Members: 
-	docs/uclibc.org/index.html:1.69->1.70 
-
----------------------
-PatchSet 2711 
-Date: 2003/09/30 09:54:18
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-grammar fixes
-
-Members: 
-	docs/uclibc.org/index.html:1.70->1.71 
-
----------------------
-PatchSet 2712 
-Date: 2003/09/30 10:32:48
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-typo
-
-Members: 
-	docs/uclibc.org/index.html:1.71->1.72 
-
----------------------
-PatchSet 2713 
-Date: 2003/09/30 12:51:11
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Don't forget addend on relocations + minor other fixes.
-
-Members: 
-	ldso/ldso/cris/elfinterp.c:1.5->1.6 
-
----------------------
-PatchSet 2714 
-Date: 2003/10/01 12:57:45
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Updated cris configuration.
-
-Members: 
-	extra/Configs/Config.cris.default:1.2->1.3 
-
----------------------
-PatchSet 2715 
-Date: 2003/10/03 02:17:36
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Quick implementation of sigwait.  Doesn't handle thread cancelation
-yet, but should work...
-
-Members: 
-	libc/signal/Makefile:1.20->1.21 
-	libc/signal/sigwait.c:INITIAL->1.1 
-
----------------------
-PatchSet 2716 
-Date: 2003/10/04 18:31:22
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix a stupid bug causing 'ldd' output to not contain the full
-path to the target library
-
-Members: 
-	ldso/ldso/ldso.c:1.73->1.74 
-
----------------------
-PatchSet 2717 
-Date: 2003/10/06 08:46:14
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-s/root_fs-/root_fs_/g
+Joakim Tjernlund writes:
 
-Members: 
-	docs/uclibc.org/FAQ.html:1.16->1.17 
-	docs/uclibc.org/index.html:1.72->1.73 
-	docs/uclibc.org/old-news.html:1.13->1.14 
-
----------------------
-PatchSet 2718 
-Date: 2003/10/08 19:15:41
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from George Thanos adding support for the "e1" architecture.
-
-Members: 
-	extra/Configs/Config.e1:INITIAL->1.1 
-	extra/Configs/Config.e1.default:INITIAL->1.1 
-	libc/sysdeps/linux/e1/Makefile:INITIAL->1.1 
-	libc/sysdeps/linux/e1/crt0.S:INITIAL->1.1 
-	libc/sysdeps/linux/e1/crt1.c:INITIAL->1.1 
-	libc/sysdeps/linux/e1/longjmp.c:INITIAL->1.1 
-	libc/sysdeps/linux/e1/setjmp.c:INITIAL->1.1 
-	libc/sysdeps/linux/e1/syscalls.c:INITIAL->1.1 
-	libc/sysdeps/linux/e1/vfork.c:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/endian.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/fcntl.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/fenv.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/fenvinline.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/kernel_stat.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/kernel_types.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/mman.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/proto.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/setjmp.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/syscalls.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/unistd.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/bits/wordsize.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/sys/procfs.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/sys/reg.h:INITIAL->1.1 
-	libc/sysdeps/linux/e1/sys/ucontext.h:INITIAL->1.1 
-
----------------------
-PatchSet 2719 
-Date: 2003/10/08 19:27:19
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Apply a trivial patch that doesn't change anything, but which is alledged to
-workaround a toolchain specifi bug for the e1.
-
-Members: 
-	libc/misc/time/time.c:1.14->1.15 
-
----------------------
-PatchSet 2720 
-Date: 2003/10/08 20:30:30
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from Atsushi Nemoto (with some additions):
-
-Current uClibc contains only one fpu_control.h and it is i386 version.
-This is a patch to use platform specific fpu_control.h.  All new files
-come from glibc 2.3.2.  This patch is against 0.9.21 but also can be
-applied to CVS as is.
-
-Members: 
-	libc/sysdeps/linux/alpha/Makefile:1.5->1.6 
-	libc/sysdeps/linux/alpha/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/arm/Makefile:1.44->1.45 
-	libc/sysdeps/linux/arm/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/common/Makefile:1.86->1.87 
-	libc/sysdeps/linux/common/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/i386/Makefile:1.57->1.58 
-	libc/sysdeps/linux/i386/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/m68k/Makefile:1.26->1.27 
-	libc/sysdeps/linux/m68k/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/mips/Makefile:1.29->1.30 
-	libc/sysdeps/linux/mips/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/powerpc/Makefile:1.21->1.22 
-	libc/sysdeps/linux/powerpc/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/sh/Makefile:1.22->1.23 
-	libc/sysdeps/linux/sh/fpu_control.h:INITIAL->1.1 
-	libc/sysdeps/linux/sparc/Makefile:1.13->1.14 
-	libc/sysdeps/linux/sparc/fpu_control.h:INITIAL->1.1 
-
----------------------
-PatchSet 2721 
-Date: 2003/10/08 20:35:11
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Atsushi Nemoto writes:
-
-I found inappropriate data types are used in some places in networking
-codes.
-
-* tcp_seq is 32bit (u_long -> u_int32_t)
-* in_addt_t should be used for internet address (unsigned long -> in_addr_t)
-* socklen_t should be used for accept()
-
-This is a patch against uclibc-0.9.21 (can be applied for current
-CVS).  64bit platforms (sizeof(int)!=sizeof(long)) will need this.  I
-believe this patch does not harm any 32bit platforms.
-
-Members: 
-	include/netinet/ip_tcp.h:1.3->1.4 
-	libc/inet/addr.c:1.13->1.14 
-	libc/inet/inet_net.c:1.3->1.4 
-	libc/inet/rpc/rcmd.c:1.8->1.9 
-	libc/inet/rpc/rexec.c:1.1->1.2 
+    > Very interesting.  Do you have any suggestions for how
+    > we could fix our powerpc shared library loader
 
----------------------
-PatchSet 2722 
-Date: 2003/10/09 07:05:35
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-When someone would dlopen a shared lib that was already opened,
-we would try to re-fixup the lib's relocations with rather
-horrible results.  So fix that by checking the the dlopened lib
-has already had its init functions called, which will never be
-the case for newly loaded libs, and skip the rest in that case.
+    Removing those instr. comes with a very big performance
+    penalty. To flush the dcache you will have read up to 8KB
+    dummy data and to invalidate the icache you will have to
+    execute up to 16KB nops. I don't know of any other way from
+    user space.
 
-also apply a few minor fixups
+    hmm, actually I think it will work reliable to perform a
+    store to the same page(s) as the dcbst/icbi will act on. That
+    way you will make the DTLB Error happen(if any) prior to the
+    dcbst/icbi. The worst thing that can happen then is a regular
+    DTLB Miss and that works for dcbst/icbi.
 
-Members: 
-	ldso/libdl/dlib.c:1.30->1.31 
+    You will have to lookout for if dcbst/icbi crosses a page
+    boundary. Then you will have to perform a store to both
+    pages.
 
----------------------
-PatchSet 2723 
-Date: 2003/10/09 07:07:24
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-move the test for loading the wrong C library into the
-_dl_check_if_named_library_is_loaded function
+     Jocke
 
-Members: 
-	ldso/ldso/readelflib1.c:1.47->1.48 
+# And again later writes:
 
----------------------
-PatchSet 2724 
-Date: 2003/10/09 10:02:05
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from Rob McMullen:
+    Hi again
 
-Here's a patch...  Since they aren't SUSv3 functions, I don't know if
-they'll ever get officially added, but it helps with BSD porting and
-allows quite a few Gentoo ebuilds to compile without changing anything.
+    I think I know what the problem is. The
+    PPC_DCBST;PPC_SYNC;PPC_ICBI;PPC_ISYNC sequence is executed
+    even if no modification has been done i some cases:
 
-Rob
+     _dl_linux_resolver(), the last else has no store for insns[0].
+			   these is a insns[1] = OPCODE_B(delta - 4) that
+			   does not have a PPC_DCBST.
 
-Members: 
-	libc/misc/error/error.c:1.2->1.3 
+     _dl_do_lazy_reloc(), for R_PPC_NONE there is no store.
+			  for R_PPC_JMP_SLOT there is a
+			  insns[1] = OPCODE_B(delta)that does not
+			  have a PPC_DCBST.
 
----------------------
-PatchSet 2725 
-Date: 2003/10/10 08:34:26
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Implement getgrgid_r and getgrnam_r.  Rework group handling code to be fully
-reentrant, since there was still a shared static value.  indent stuff,
+     _dl_do_reloc(), for R_PPC_COPY there is no store.
+		     for R_PPC_JMP_SLOT there is a
+		     reloc_addr[1] = OPCODE_B(delta) that does not
+		     have a PPC_DCBST.
 
-Members: 
-	libc/pwd_grp/Makefile:1.21->1.22 
-	libc/pwd_grp/__getgrent.c:1.9->1.10(DEAD) 
-	libc/pwd_grp/__getgrent_r.c:INITIAL->1.1 
-	libc/pwd_grp/__getpwent_r.c:1.3->1.4 
-	libc/pwd_grp/__getspent_r.c:1.5->1.6 
-	libc/pwd_grp/__sgetspent_r.c:1.3->1.4 
-	libc/pwd_grp/config.h:1.6->1.7 
-	libc/pwd_grp/fgetgrent.c:1.7->1.8 
-	libc/pwd_grp/fgetpwent.c:1.9->1.10 
-	libc/pwd_grp/fgetspent.c:1.5->1.6 
-	libc/pwd_grp/getgrgid.c:1.8->1.9 
-	libc/pwd_grp/getgrnam.c:1.8->1.9 
-	libc/pwd_grp/getpw.c:1.4->1.5 
-	libc/pwd_grp/getpwnam.c:1.10->1.11 
-	libc/pwd_grp/getpwuid.c:1.9->1.10 
-	libc/pwd_grp/getspnam.c:1.5->1.6 
-	libc/pwd_grp/getspuid.c:1.5->1.6 
-	libc/pwd_grp/grent.c:1.9->1.10 
-	libc/pwd_grp/initgroups.c:1.11->1.12 
-	libc/pwd_grp/lckpwdf.c:1.3->1.4 
-	libc/pwd_grp/putpwent.c:1.5->1.6 
-	libc/pwd_grp/putspent.c:1.2->1.3 
-	libc/pwd_grp/pwent.c:1.12->1.13 
-	libc/pwd_grp/sgetspent.c:1.5->1.6 
-	libc/pwd_grp/spent.c:1.6->1.7 
-
----------------------
-PatchSet 2726 
-Date: 2003/10/10 09:05:08
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Remove some stuff that is now done
+     _dl_init_got(), I THINK that the
+	    PPC_DCBST(plt);
+	    PPC_DCBST(plt+4);
+	    PPC_DCBST(plt+8);
+	    PPC_SYNC;
+	    PPC_ICBI(plt);
+	    PPC_ICBI(plt+4);
+	    PPC_ICBI(plt+8);
+	    PPC_ISYNC;
+	    is off a bit. The address range does not match the sum
+	    of the plt[] and tramp[] address range.
 
-Members: 
-	TODO:1.43->1.44 
-	docs/threads.txt:1.4->1.5 
+       Jocke
 
----------------------
-PatchSet 2727 
-Date: 2003/10/10 10:21:54
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-We do not provide gmtime(), so disable the prototype to
-prevent confusing autoconf
 
-Members: 
-	include/time.h:1.9->1.10 
-
----------------------
-PatchSet 2728 
-Date: 2003/10/11 09:03:48
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-These are wanted by net-tools
-
-Members: 
-	include/netax25/ax25.h:INITIAL->1.1 
-	include/neteconet/ec.h:INITIAL->1.1 
-
----------------------
-PatchSet 2729 
-Date: 2003/10/11 19:14:32
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-I forgot to remove this
-
-Members: 
-	include/fpu_control.h:1.1->1.2(DEAD) 
-
----------------------
-PatchSet 2730 
-Date: 2003/10/12 05:19:38
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Sigh.  It seems that csh wants glob_pattern_p(), we need to
-export it,
-
-Members: 
-	libc/misc/glob/glob.c:1.5->1.6 
-
----------------------
-PatchSet 2731 
-Date: 2003/10/12 05:41:11
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Don't make two copies
-
-Members: 
-	libc/misc/glob/glob.c:1.6->1.7 
-	libc/misc/glob/glob64.c:1.3->1.4 
-
----------------------
-PatchSet 2732 
-Date: 2003/10/12 09:49:31
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Always build and install ldconfig
-
-Members: 
-	Makefile:1.217->1.218 
-	ldso/include/ld_elf.h:1.2->1.3 
-	ldso/util/Makefile:1.34->1.35 
-	ldso/util/ldconfig.c:1.10->1.11 
-	ldso/util/readsoname.c:1.2->1.3 
-
----------------------
-PatchSet 2733 
-Date: 2003/10/13 09:23:34
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Fix an ldd bug, more ldconfig cleanup
-
-Members: 
-	ldso/util/ldconfig.c:1.11->1.12 
-	ldso/util/ldd.c:1.33->1.34 
-
----------------------
-PatchSet 2734 
-Date: 2003/10/13 09:34:51
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from Peter Kjellerstedt:
-
-The attached patch performs a clean up of
-extra/gcc-uClibc/Makefile:
-
-* Better dependencies to only rebuild what is necessary.
-* Use $< and $@ where appropriate.
-* Suppress warning messages from which about
-  commands that cannot be found.
-
-//Peter
-
-Members: 
-	extra/gcc-uClibc/Makefile:1.58->1.59 
-
----------------------
-PatchSet 2735 
-Date: 2003/10/13 11:01:10
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-sigh.  the warn funcs seem to be broken.  Use local versions
-till I can straighten that out.
-
-Members: 
-	ldso/util/ldconfig.c:1.12->1.13 
-
----------------------
-PatchSet 2736 
-Date: 2003/10/14 07:23:00
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Enable full debug support for ld-uClibc and libdl
+# And then later added the comment:
 
-Members: 
-	ldso/ldso/Makefile:1.43->1.44 
-	ldso/libdl/Makefile:1.28->1.29 
-
----------------------
-PatchSet 2737 
-Date: 2003/10/14 12:52:27
-Author: davidm
-Branch: HEAD
-Tag: (none) 
-Log:
-
-Microblaze port from the uClinux-dist,
-contributed by John Williams <jwilliams@itee.uq.edu.au>
-
-Members: 
-	extra/Configs/Config.microblaze:INITIAL->1.1 
-	extra/Configs/Config.microblaze.default:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/Makefile:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/__longjmp.S:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/_mmap.c:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/clinkage.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/clone.c:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/crt0.S:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/longjmp.S:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/setjmp.S:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/syscall.c:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/vfork.S:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/byteswap.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/endian.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/fcntl.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/kernel_stat.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/kernel_types.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/mman.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/poll.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/select.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/setjmp.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/syscalls.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/bits/wordsize.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/sys/ptrace.h:INITIAL->1.1 
-	libc/sysdeps/linux/microblaze/sys/ucontext.h:INITIAL->1.1 
-
----------------------
-PatchSet 2738 
-Date: 2003/10/15 09:55:22
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Doh!  I was returning with an incorrect pointer type when dlopening
-already loaded libs, which unsurprisingly would cause dlsym() to
-not work at all...
- -Erik
-
-Members: 
-	ldso/libdl/dlib.c:1.31->1.32 
-
----------------------
-PatchSet 2739 
-Date: 2003/10/15 10:04:40
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Patch from Tony J. White:
-
-I've created a patch for adding dn_expand() to uClibc 0.9.21.
-
-dn_expand() is used by at least ipsec-tools and also openldap I think.
-
-Members: 
-	libc/inet/Makefile:1.39->1.40 
-	libc/inet/resolv.c:1.47->1.48 
-
----------------------
-PatchSet 2740 
-Date: 2003/10/15 10:24:35
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Brett Nash writes:
-
-Hello,
-        Attached is a patch of some changes I made to the gcc wrapper to
-get it to compile XFree CVS.
-
-Basically it supports the use of '-' on the command line to read stdin
-for gcc, as well as setting the file type back to none before adding the
-last too .o files if the file type was set on the command line.
-
-It is applied against uClibc-0.9.20
-
-        Regards,
-        nash
-
-Members: 
-	extra/gcc-uClibc/gcc-uClibc.c:1.62->1.63 
-
----------------------
-PatchSet 2741 
-Date: 2003/10/15 14:03:03
-Author: tobiasa
-Branch: HEAD
-Tag: (none) 
-Log:
-Updated cris configuration; added support for %m.
-
-Members: 
-	extra/Configs/Config.cris.default:1.3->1.4 
-
----------------------
-PatchSet 2742 
-Date: 2003/10/16 05:08:02
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-More detail on the use of MALLOC_DEBUG
-
-Members: 
-	extra/Configs/Config.in:1.32->1.33 
-
----------------------
-PatchSet 2743 
-Date: 2003/10/16 11:12:45
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-Don't set errno if for zero SIZE returns.
-Code formatting cleanup.
-
-Members: 
-	libc/stdlib/malloc/malloc.c:1.35->1.36 
-
----------------------
-PatchSet 2744 
-Date: 2003/10/16 11:13:53
-Author: miles
-Branch: HEAD
-Tag: (none) 
-Log:
-Code formatting cleanup.
-
-Members: 
-	libc/stdlib/malloc/realloc.c:1.19->1.20 
-
----------------------
-PatchSet 2745 
-Date: 2003/10/16 15:21:26
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Do not set errno when they asked for size 0
-
-Members: 
-	libc/stdlib/malloc-930716/malloc.c:1.11->1.12 
-
----------------------
-PatchSet 2746 
-Date: 2003/10/17 08:52:54
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Point people to the kernel.org mirrors for downloading the dev
-systems to take some load off my server
-
-Members: 
-	docs/uclibc.org/FAQ.html:1.17->1.18 
-	docs/uclibc.org/index.html:1.73->1.74 
-	docs/uclibc.org/old-news.html:1.14->1.15 
-
----------------------
-PatchSet 2747 
-Date: 2003/10/17 10:05:20
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Doh!!!
-
-Members: 
-	libc/stdlib/malloc-930716/malloc.c:1.12->1.13 
-
----------------------
-PatchSet 2748 
-Date: 2003/10/18 10:14:57
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Implement putgrent to make Peter S. Mazing happy
-
-Members: 
-	libc/pwd_grp/Makefile:1.22->1.23 
-	libc/pwd_grp/putgrent.c:INITIAL->1.1 
-
----------------------
-PatchSet 2749 
-Date: 2003/10/18 11:07:28
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Peter Kjellerstedt writes:
-
-install.patch:
-* Define $(INSTALL) as install in Rules.mak.
-* Change all occurrences of install into $(INSTALL).
-* Change all occurrences of mkdir -p into $(INSTALL) -d.
-  install -d is already used in a number of places so
-  this should not be an additional compatibility problem.
-
-Members: 
-	Makefile:1.218->1.219 
-	Rules.mak:1.140->1.141 
-	extra/gcc-uClibc/Makefile:1.59->1.60 
-	ldso/ldso/Makefile:1.44->1.45 
-	ldso/libdl/Makefile:1.29->1.30 
-	libc/Makefile:1.28->1.29 
-	libc/sysdeps/linux/common/Makefile:1.87->1.88 
-	libcrypt/Makefile:1.21->1.22 
-	libintl/Makefile:1.1->1.2 
-	libm/Makefile:1.34->1.35 
-	libnsl/Makefile:1.2->1.3 
-	libpthread/Makefile:1.22->1.23 
-	libresolv/Makefile:1.12->1.13 
-	libutil/Makefile:1.18->1.19 
-
----------------------
-PatchSet 2750 
-Date: 2003/10/18 11:18:19
-Author: andersen
-Branch: HEAD
-Tag: (none) 
-Log:
-Peter Kjellerstedt writes:
-
-rm.patch:
-* Define $(RM) as rm -f in Rules.mak and test/Rules.mak
-  (this is the same definition as gmake uses by default).
-* Change all occurrences of rm and rm -f into $(RM).
-
-Members: 
-	Makefile:1.219->1.220 
-	Rules.mak:1.141->1.142 
-	extra/Makefile:1.4->1.5 
-	extra/config/Makefile:1.8->1.9 
-	extra/gcc-uClibc/Makefile:1.60->1.61 
-	extra/locale/Makefile:1.13->1.14 
-	ldso/Makefile:1.23->1.24 
-	ldso/ldso/Makefile:1.45->1.46 
-	ldso/libdl/Makefile:1.30->1.31 
-	ldso/man/Makefile:1.2->1.3 
-	ldso/util/Makefile:1.35->1.36 
-	libc/Makefile:1.29->1.30 
-	libc/inet/Makefile:1.40->1.41 
-	libc/inet/rpc/Makefile:1.22->1.23