r1285 - trunk/glibc-2.3-head/patches/fixes

2006-03-02 Thread Petr Salinger
Author: ps-guest
Date: 2006-03-02 08:31:54 + (Thu, 02 Mar 2006)
New Revision: 1285

Modified:
   trunk/glibc-2.3-head/patches/fixes/thread_signals.patch
Log:
add status


Modified: trunk/glibc-2.3-head/patches/fixes/thread_signals.patch
===
--- trunk/glibc-2.3-head/patches/fixes/thread_signals.patch 2006-03-01 
19:31:19 UTC (rev 1284)
+++ trunk/glibc-2.3-head/patches/fixes/thread_signals.patch 2006-03-02 
08:31:54 UTC (rev 1285)
@@ -1,3 +1,7 @@
+
+Status: submitted upstream
+   see http://sourceware.org/bugzilla/show_bug.cgi?id=2383
+
 diff -urN linuxthreads/pthread.c linuxthreads/pthread.c
 --- linuxthreads/pthread.c 2006-01-17 13:28:26.0 +0100
 +++ linuxthreads/pthread.c 2006-01-17 13:26:42.0 +0100


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1286 - trunk/glibc-2.3-head/sysdeps/kfreebsd

2006-03-02 Thread Aurelien Jarno
Author: aurel32
Date: 2006-03-02 09:24:02 + (Thu, 02 Mar 2006)
New Revision: 1286

Modified:
   trunk/glibc-2.3-head/sysdeps/kfreebsd/configure
   trunk/glibc-2.3-head/sysdeps/kfreebsd/configure.in
Log:
Put the libraries in /lib64 and (/usr)/lib64 on x86_64, this is where the 
upstream 
put 64-bit version of libraries for architectures that can do biarch.

Consequences:
- the amd64 glibc package on i386 will go directly in /lib64 and (/usr)/lib64
- the path has to be fixed for the native amd64 port. This is already the case 
in the 
  Debian package (glibc 2.3.6-3).



Modified: trunk/glibc-2.3-head/sysdeps/kfreebsd/configure
===
--- trunk/glibc-2.3-head/sysdeps/kfreebsd/configure 2006-03-02 08:31:54 UTC 
(rev 1285)
+++ trunk/glibc-2.3-head/sysdeps/kfreebsd/configure 2006-03-02 09:24:02 UTC 
(rev 1286)
@@ -201,7 +201,7 @@
 /usr | /usr/)
   # 64-bit libraries on bi-arch platforms go in /lib64 instead of /lib
   case $machine in
-  sparc/sparc64 | powerpc/powerpc64 | s390/s390-64 | \
+  sparc/sparc64 | x86_64 | powerpc/powerpc64 | s390/s390-64 | \
   mips/mips64/n64/* )
 libc_cv_slibdir="/lib64"
 if test "$libdir" = '${exec_prefix}/lib'; then

Modified: trunk/glibc-2.3-head/sysdeps/kfreebsd/configure.in
===
--- trunk/glibc-2.3-head/sysdeps/kfreebsd/configure.in  2006-03-02 08:31:54 UTC 
(rev 1285)
+++ trunk/glibc-2.3-head/sysdeps/kfreebsd/configure.in  2006-03-02 09:24:02 UTC 
(rev 1286)
@@ -133,7 +133,7 @@
 /usr | /usr/)
   # 64-bit libraries on bi-arch platforms go in /lib64 instead of /lib
   case $machine in
-  sparc/sparc64 | powerpc/powerpc64 | s390/s390-64 | \
+  sparc/sparc64 | x86_64 | powerpc/powerpc64 | s390/s390-64 | \
   mips/mips64/n64/* )
 libc_cv_slibdir="/lib64"
 if test "$libdir" = '${exec_prefix}/lib'; then


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1287 - in trunk/kfreebsd-6/debian: . patches

2006-03-02 Thread Robert Millan
Author: rmh
Date: 2006-03-02 10:39:47 + (Thu, 02 Mar 2006)
New Revision: 1287

Modified:
   trunk/kfreebsd-6/debian/changelog
   trunk/kfreebsd-6/debian/patches/005_binutils.diff
Log:
- patches/binutils.diff: Add binutils fixes for amd64 code.

Modified: trunk/kfreebsd-6/debian/changelog
===
--- trunk/kfreebsd-6/debian/changelog   2006-03-02 09:24:02 UTC (rev 1286)
+++ trunk/kfreebsd-6/debian/changelog   2006-03-02 10:39:47 UTC (rev 1287)
@@ -4,6 +4,7 @@
   * New upstream version pulled from RELENG_6 CVSup.
 - ReiserFS support.  (Closes: #335019)
   * Disabled the non-compiling hwpmc driver.
+
   [ Robert Millan ]
   * Merge 5.x branch changes (from rev 654 to rev 1058)
   * Add src/usr.sbin/config into the source and build/use it dynamicaly.
@@ -40,10 +41,12 @@
 - arch/amd64/em64t-p4.config: New.  P4 config.
 - patches/907_cpu_class.diff: Set hw.machine to "x86_64".  This improves
 compatibility with old config.sub scripts.
+  * patches/binutils.diff: Add binutils fixes for amd64 code.
+
   [ Aurelien Jarno ]
   * Fix a remote denial of service in NFS server (FreeBSD-SA-06:10.nfs / 
CVE-2006-0900).
 
- -- Robert Millan <[EMAIL PROTECTED]>  Tue, 28 Feb 2006 12:42:18 +0100
+ -- Robert Millan <[EMAIL PROTECTED]>  Thu,  2 Mar 2006 11:38:52 +0100
 
 kfreebsd-5 (5.3-6) unstable; urgency=low
 

Modified: trunk/kfreebsd-6/debian/patches/005_binutils.diff
===
--- trunk/kfreebsd-6/debian/patches/005_binutils.diff   2006-03-02 09:24:02 UTC 
(rev 1286)
+++ trunk/kfreebsd-6/debian/patches/005_binutils.diff   2006-03-02 10:39:47 UTC 
(rev 1287)
@@ -110,3 +110,338 @@
ret
  
  ENTRY(x86_gettid)
+--- sys/amd64/amd64/cpu_switch.S.old   2006-03-01 20:12:04.0 +0100
 sys/amd64/amd64/cpu_switch.S   2006-03-01 20:04:59.0 +0100
+@@ -107,10 +107,10 @@
+   jz  1f  /* no, skip over */
+ 
+   /* Save segment selector numbers */
+-  movl%ds,PCB_DS(%r8)
+-  movl%es,PCB_ES(%r8)
+-  movl%fs,PCB_FS(%r8)
+-  movl%gs,PCB_GS(%r8)
++  movw%ds,PCB_DS(%r8)
++  movw%es,PCB_ES(%r8)
++  movw%fs,PCB_FS(%r8)
++  movw%gs,PCB_GS(%r8)
+   jmp 2f
+ 1:
+ 
+@@ -191,14 +191,14 @@
+   jz  1f  /* no, skip over */
+ 
+   /* Restore segment selector numbers */
+-  movlPCB_DS(%r8),%ds
+-  movlPCB_ES(%r8),%es
+-  movlPCB_FS(%r8),%fs
++  movwPCB_DS(%r8),%ds
++  movwPCB_ES(%r8),%es
++  movwPCB_FS(%r8),%fs
+ 
+   /* Restore userland %gs while preserving kernel gsbase */
+   movl$MSR_GSBASE,%ecx
+   rdmsr
+-  movlPCB_GS(%r8),%gs
++  movwPCB_GS(%r8),%gs
+   wrmsr
+   jmp 2f
+ 1:
+--- sys/amd64/ia32/ia32_signal.c~  2006-03-01 22:23:22.0 +0100
 sys/amd64/ia32/ia32_signal.c   2006-03-01 22:44:47.0 +0100
+@@ -197,8 +197,8 @@
+   sf.sf_uc.uc_mcontext.mc_onstack = (oonstack) ? 1 : 0;
+   sf.sf_uc.uc_mcontext.mc_gs = rgs();
+   sf.sf_uc.uc_mcontext.mc_fs = rfs();
+-  __asm __volatile("movl %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
+-  __asm __volatile("movl %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
++  __asm __volatile("mov %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
++  __asm __volatile("mov %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
+   sf.sf_uc.uc_mcontext.mc_edi = regs->tf_rdi;
+   sf.sf_uc.uc_mcontext.mc_esi = regs->tf_rsi;
+   sf.sf_uc.uc_mcontext.mc_ebp = regs->tf_rbp;
+@@ -309,8 +309,8 @@
+   sf.sf_uc.uc_mcontext.mc_onstack = (oonstack) ? 1 : 0;
+   sf.sf_uc.uc_mcontext.mc_gs = rgs();
+   sf.sf_uc.uc_mcontext.mc_fs = rfs();
+-  __asm __volatile("movl %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
+-  __asm __volatile("movl %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
++  __asm __volatile("mov %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
++  __asm __volatile("mov %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
+   sf.sf_uc.uc_mcontext.mc_edi = regs->tf_rdi;
+   sf.sf_uc.uc_mcontext.mc_esi = regs->tf_rsi;
+   sf.sf_uc.uc_mcontext.mc_ebp = regs->tf_rbp;
+--- sys/amd64/include/pcpu.h~  2006-03-01 23:10:56.0 +0100
 sys/amd64/include/pcpu.h   2006-03-01 23:40:19.0 +0100
+@@ -80,7 +80,7 @@
+ #define   __PCPU_PTR(name) __extension__ ({   
\
+   __pcpu_type(name) *__p; \
+   \
+-  __asm __volatile("movq %%gs:%1,%0; addq %2,%0"  \
++  __asm __volatile("mov %%gs:%1,%0; addq %2,%0"   \
+   : "=r" (__p)\
+   : "m" (*(struct pcpu *)(__pcpu_offset(pc_prvspace))), 

r1288 - trunk/kfreebsd-common/debian

2006-03-02 Thread Robert Millan
Author: rmh
Date: 2006-03-02 10:43:07 + (Thu, 02 Mar 2006)
New Revision: 1288

Modified:
   trunk/kfreebsd-common/debian/changelog
   trunk/kfreebsd-common/debian/init.d
Log:
init.d: Fix /dev/da[0-9]* too.

Modified: trunk/kfreebsd-common/debian/changelog
===
--- trunk/kfreebsd-common/debian/changelog  2006-03-02 10:39:47 UTC (rev 
1287)
+++ trunk/kfreebsd-common/debian/changelog  2006-03-02 10:43:07 UTC (rev 
1288)
@@ -1,3 +1,9 @@
+kfreebsd-common (0.1.1) UNRELEASED; urgency=low
+
+  * init.d: Fix /dev/da[0-9]* too.
+
+ -- Robert Millan <[EMAIL PROTECTED]>  Thu,  2 Mar 2006 11:42:58 +0100
+
 kfreebsd-common (0.1.0) unreleased; urgency=low
 
   * Rename to kfreebsd-common.

Modified: trunk/kfreebsd-common/debian/init.d
===
--- trunk/kfreebsd-common/debian/init.d 2006-03-02 10:39:47 UTC (rev 1287)
+++ trunk/kfreebsd-common/debian/init.d 2006-03-02 10:43:07 UTC (rev 1288)
@@ -40,7 +40,7 @@
 chgrp floppy $i
   fi
 done
-for i in /dev/ad[0-9]* ; do
+for i in /dev/ad[0-9]* /dev/da[0-9]* ; do
   if test -e $i ; then
 chgrp disk $i
   fi


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1289 - trunk/kfreebsd-6/debian

2006-03-02 Thread Robert Millan
Author: rmh
Date: 2006-03-02 11:39:33 + (Thu, 02 Mar 2006)
New Revision: 1289

Modified:
   trunk/kfreebsd-6/debian/changelog
   trunk/kfreebsd-6/debian/rules
Log:
rules: Replace all "dh_* -a" with "dh_* -s".  -a should NEVER be used.  See 
debhelper(7).

Modified: trunk/kfreebsd-6/debian/changelog
===
--- trunk/kfreebsd-6/debian/changelog   2006-03-02 10:43:07 UTC (rev 1288)
+++ trunk/kfreebsd-6/debian/changelog   2006-03-02 11:39:33 UTC (rev 1289)
@@ -41,12 +41,14 @@
 - arch/amd64/em64t-p4.config: New.  P4 config.
 - patches/907_cpu_class.diff: Set hw.machine to "x86_64".  This improves
 compatibility with old config.sub scripts.
-  * patches/binutils.diff: Add binutils fixes for amd64 code.
+- patches/binutils.diff: Add binutils fixes for amd64 code.
+- rules: Replace all "dh_* -a" with "dh_* -s".  -a should NEVER be
+used.  See debhelper(7).
 
   [ Aurelien Jarno ]
   * Fix a remote denial of service in NFS server (FreeBSD-SA-06:10.nfs / 
CVE-2006-0900).
 
- -- Robert Millan <[EMAIL PROTECTED]>  Thu,  2 Mar 2006 11:38:52 +0100
+ -- Robert Millan <[EMAIL PROTECTED]>  Thu,  2 Mar 2006 12:39:08 +0100
 
 kfreebsd-5 (5.3-6) unstable; urgency=low
 

Modified: trunk/kfreebsd-6/debian/rules
===
--- trunk/kfreebsd-6/debian/rules   2006-03-02 10:43:07 UTC (rev 1288)
+++ trunk/kfreebsd-6/debian/rules   2006-03-02 11:39:33 UTC (rev 1289)
@@ -208,18 +208,18 @@
 binary-arch: build-arch install-arch
dh_testdir
dh_testroot
-   dh_installdocs -a
-   dh_installchangelogs -a
-   dh_link -a
-   dh_strip -a -X/boot
-   dh_compress -a
-   dh_fixperms -a
-   dh_makeshlibs -a
-   dh_installdeb -a
-   dh_shlibdeps -a -X/boot
-   dh_gencontrol -a
-   dh_md5sums -a
-   dh_builddeb -a
+   dh_installdocs -s
+   dh_installchangelogs -s
+   dh_link -s
+   dh_strip -s -X/boot
+   dh_compress -s
+   dh_fixperms -s
+   dh_makeshlibs -s
+   dh_installdeb -s
+   dh_shlibdeps -s -X/boot
+   dh_gencontrol -s
+   dh_md5sums -s
+   dh_builddeb -s
 else
 binary-arch:
# We have nothing to do


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1290 - trunk/glibc-2.3-head/sysdeps/kfreebsd

2006-03-02 Thread Petr Salinger
Author: ps-guest
Date: 2006-03-02 12:30:54 + (Thu, 02 Mar 2006)
New Revision: 1290

Modified:
   trunk/glibc-2.3-head/sysdeps/kfreebsd/open.c
Log:
forward fix for glibc 2.4:  __futimes() is hidden -> not available within 
libpthread



Modified: trunk/glibc-2.3-head/sysdeps/kfreebsd/open.c
===
--- trunk/glibc-2.3-head/sysdeps/kfreebsd/open.c2006-03-02 11:39:33 UTC 
(rev 1289)
+++ trunk/glibc-2.3-head/sysdeps/kfreebsd/open.c2006-03-02 12:30:54 UTC 
(rev 1290)
@@ -25,7 +25,7 @@
 #include 
 #include 
 
-extern int __futimes (int fd, const struct timeval tvp[2]);
+extern int futimes (int fd, const struct timeval tvp[2]);
 
 int
 __libc_open (const char *file, int oflag, ...)
@@ -67,7 +67,7 @@
  tv[0].tv_sec = statbuf.st_atime;
  tv[0].tv_usec = 0;
 
- __futimes (fd, tv);
+ futimes (fd, tv);
}
}
   __set_errno (saved_errno);


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1291 - trunk/kfreebsd-6/debian

2006-03-02 Thread Petr Salinger
Author: ps-guest
Date: 2006-03-02 12:34:40 + (Thu, 02 Mar 2006)
New Revision: 1291

Modified:
   trunk/kfreebsd-6/debian/changelog
Log:
revert of r1274 (r1281) also in changelog


Modified: trunk/kfreebsd-6/debian/changelog
===
--- trunk/kfreebsd-6/debian/changelog   2006-03-02 12:30:54 UTC (rev 1290)
+++ trunk/kfreebsd-6/debian/changelog   2006-03-02 12:34:40 UTC (rev 1291)
@@ -39,8 +39,6 @@
 - arch/amd64/amd64-generic.config: New.  Generic x86_64 config.
 - arch/amd64/amd64-k8.config: New.  K8 config.
 - arch/amd64/em64t-p4.config: New.  P4 config.
-- patches/907_cpu_class.diff: Set hw.machine to "x86_64".  This improves
-compatibility with old config.sub scripts.
 - patches/binutils.diff: Add binutils fixes for amd64 code.
 - rules: Replace all "dh_* -a" with "dh_* -s".  -a should NEVER be
 used.  See debhelper(7).


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1292 - trunk/glibc-2.3-head

2006-03-02 Thread Petr Salinger
Author: ps-guest
Date: 2006-03-02 12:57:48 + (Thu, 02 Mar 2006)
New Revision: 1292

Modified:
   trunk/glibc-2.3-head/NOTES-2.4
Log:
* update glibc 2.4 status



Modified: trunk/glibc-2.3-head/NOTES-2.4
===
--- trunk/glibc-2.3-head/NOTES-2.4  2006-03-02 12:34:40 UTC (rev 1291)
+++ trunk/glibc-2.3-head/NOTES-2.4  2006-03-02 12:57:48 UTC (rev 1292)
@@ -11,13 +11,14 @@
 test build:
 ---
 
-* get glibc-20060123.tar.bz2 + glibc-ports-20060123.tar.bz2
+* get main glibc module and linuxthreads module from snapshot or directly from 
CVS
 
-* move linuxthreads & linuxthreads_db from ports into top dir
+* move linuxthreads & linuxthreads_db from module linuxthreads into top dir
 
 * try to apply patches
failures in configure/configure.in ignore
failure in nscd/mem.c fix manually
+   of course, patches in subdir upstream are already in CVS
 
 * define MSG_NOSIGNAL
enable definition in kfreebsd/bits/socket.h
@@ -29,7 +30,16 @@
unconditionally assumes mremap and uses MREMAP_FIXED
it also unconditionally uses __thread :-(
 
-* there is no longer sysdeps/generic, files are moved
+* io/ftw.c uses unconditionally O_DIRECTORY :-((
+   it have to be rewritten correctly 
+   add "enum { O_DIRECTORY = 0 };" into bits/fcnt.h
+   this way it will have value, but will not be #define'd
+   it is used unconditionally also in testsuite :-((
+
+* again in io/ftw.c, here is needed openat64_not_cancel_3()
+   temporary hack - use openat64() instead
+
+* many files are moved from sysdeps/generic,
we #include<> some of them, use new location

 * new sysdeps files (based on linux)
@@ -44,30 +54,18 @@
i386/sysdep.h - drop definition of SETUP_PIC_REG(reg)
dl-osinfo.h - add function _dl_setup_stack_chk_guard()
 
-../src/configure --enable-add-ons=linuxthreads --with-tls --with-__thread 
i486-kfreebsd
+../src/configure --enable-add-ons=linuxthreads --with-tls --with-__thread 
i486-kfreebsd --prefix=...
 
 "make check" results:
 -
+(as of 2006-03-02)
 
-make[2]: *** [/build/g24/build/libio/tst_wprintf2.out] Error 139
 make[2]: *** [/build/g24/build/dirent/tst-fdopendir.o] Error 1
 make[2]: *** [/build/g24/build/posix/tst-waitid.out] Error 1
 
-make[2]: *** [/build/g24/build/io/tst-openat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-unlinkat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-fstatat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-futimesat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-renameat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-fchownat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-fchmodat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-faccessat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-symlinkat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-linkat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-readlinkat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-mkdirat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-mknodat.o] Error 1
-make[2]: *** [/build/g24/build/io/tst-mkfifoat.o] Error 1
-
+make[2]: *** [/build/g24/build/io/bug-ftw2.out] Error 1
+make[2]: *** [/build/g24/build/io/bug-ftw4.out] Error 1
+make[2]: *** [/build/g24/build/io/tst-readlinkat.out] Error 1
 make[2]: *** [/build/g24/build/misc/tst-pselect.out] Error 1
 
 make[2]: *** [/build/g24/build/rt/tst-timer2.o] Error 1


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1293 - trunk/glibc-2.3-head

2006-03-02 Thread Petr Salinger
Author: ps-guest
Date: 2006-03-02 13:17:04 + (Thu, 02 Mar 2006)
New Revision: 1293

Added:
   trunk/glibc-2.3-head/glibc24-extra-sysdeps.diff
Log:
* add changes to our sysdeps for glibc 2.4 snapshots as .diff


Added: trunk/glibc-2.3-head/glibc24-extra-sysdeps.diff
===
--- trunk/glibc-2.3-head/glibc24-extra-sysdeps.diff 2006-03-02 12:57:48 UTC 
(rev 1292)
+++ trunk/glibc-2.3-head/glibc24-extra-sysdeps.diff 2006-03-02 13:17:04 UTC 
(rev 1293)
@@ -0,0 +1,138 @@
+
+ patch to kfreebsd sysdeps for building glibc 2.4 snapshot
+ see also NOTES-2.4
+
+diff -urN linuxthreads/kfreebsd/allocrtsig.c linuxthreads/kfreebsd/allocrtsig.c
+--- linuxthreads/kfreebsd/allocrtsig.c 2005-12-19 08:43:29.0 +0100
 linuxthreads/kfreebsd/allocrtsig.c 2006-03-02 10:34:13.0 +0100
+@@ -1,4 +1,4 @@
+-#include 
++#include 
+ strong_alias (__libc_current_sigrtmin, __libc_current_sigrtmin_private);
+ strong_alias (__libc_current_sigrtmax, __libc_current_sigrtmax_private);
+ strong_alias (__libc_allocate_rtsig, __libc_allocate_rtsig_private);
+diff -urN -x .svn sysdeps/kfreebsd/kernel-posix-cpu-timers.h 
sysdeps/kfreebsd/kernel-posix-cpu-timers.h
+--- sysdeps/kfreebsd/kernel-posix-cpu-timers.h 1970-01-01 01:00:00.0 
+0100
 sysdeps/kfreebsd/kernel-posix-cpu-timers.h 2006-03-02 14:04:19.0 
+0100
+@@ -0,0 +1 @@
++/ * placeholder */
+diff -urN -x .svn sysdeps/kfreebsd/bits/fcntl.h sysdeps/kfreebsd/bits/fcntl.h
+--- sysdeps/kfreebsd/bits/fcntl.h  2005-12-16 16:29:14.0 +0100
 sysdeps/kfreebsd/bits/fcntl.h  2006-03-02 11:24:01.0 +0100
+@@ -43,6 +43,8 @@
+ #define O_NOFOLLOW0x0100  /* Don't follow symlinks.  */
+ #endif
+ 
++enum { O_DIRECTORY = 0 };
++
+ /* File status flags for `open' and `fcntl'.  */
+ #define   O_APPEND0x0008  /* Writes append to the file.  */
+ #define   O_NONBLOCK  0x0004  /* Non-blocking I/O.  */
+diff -urN -x .svn sysdeps/kfreebsd/bits/socket.h sysdeps/kfreebsd/bits/socket.h
+--- sysdeps/kfreebsd/bits/socket.h 2006-01-23 17:15:53.0 +0100
 sysdeps/kfreebsd/bits/socket.h 2006-03-02 10:26:38.0 +0100
+@@ -191,7 +191,7 @@
+ #define MSG_EOF MSG_EOF
+ MSG_COMPAT= 0x8000 /* Used in sendit().  */
+ #define MSG_COMPAT MSG_COMPAT
+-#if 0 /* available since FreeBSD 6.0 */
++#if 1 /* available since FreeBSD 6.0 */
+ ,MSG_NOSIGNAL = 0x2 /* do not generate SIGPIPE on EOF */
+ #define MSG_NOSIGNAL MSG_NOSIGNAL 
+ #endif
+diff -urN -x .svn sysdeps/kfreebsd/check_fds.c sysdeps/kfreebsd/check_fds.c
+--- sysdeps/kfreebsd/check_fds.c   1970-01-01 01:00:00.0 +0100
 sysdeps/kfreebsd/check_fds.c   2006-03-02 10:30:14.0 +0100
+@@ -0,0 +1 @@
++void __libc_check_standard_fds (void) {;}
+diff -urN -x .svn sysdeps/kfreebsd/check_pf.c sysdeps/kfreebsd/check_pf.c
+--- sysdeps/kfreebsd/check_pf.c1970-01-01 01:00:00.0 +0100
 sysdeps/kfreebsd/check_pf.c2006-03-02 10:39:53.0 +0100
+@@ -0,0 +1 @@
++#include 
+diff -urN -x .svn sysdeps/kfreebsd/dl-osinfo.h sysdeps/kfreebsd/dl-osinfo.h
+--- sysdeps/kfreebsd/dl-osinfo.h   2006-02-14 18:12:55.0 +0100
 sysdeps/kfreebsd/dl-osinfo.h   2006-03-02 11:24:32.0 +0100
+@@ -97,3 +97,24 @@
+   } \
+   }   
  \
+   } while (0)
++
++static inline uintptr_t __attribute__ ((always_inline))
++_dl_setup_stack_chk_guard (void)
++{
++  uintptr_t ret;
++#ifdef ENABLE_STACKGUARD_RANDOMIZE 
++  int fd = __open ("/dev/urandom", O_RDONLY);
++  if (fd >= 0)
++{
++  ssize_t reslen = __read (fd, &ret, sizeof (ret));
++  __close (fd);
++  if (reslen == (ssize_t) sizeof (ret))
++return ret;
++}
++#endif
++  ret = 0;
++  unsigned char *p = (unsigned char *) &ret;
++  p[sizeof (ret) - 1] = 255;
++  p[sizeof (ret) - 2] = '\n';
++  return ret;
++}
+diff -urN -x .svn sysdeps/kfreebsd/glob.c sysdeps/kfreebsd/glob.c
+--- sysdeps/kfreebsd/glob.c2006-01-23 11:02:39.0 +0100
 sysdeps/kfreebsd/glob.c2006-03-02 10:37:07.0 +0100
+@@ -1,3 +1,3 @@
+ /* 'glob64' is different from 'glob', because
+'struct stat64' != 'struct stat'.  */
+-#include 
++#include 
+diff -urN -x .svn sysdeps/kfreebsd/i386/sysdep.h sysdeps/kfreebsd/i386/sysdep.h
+--- sysdeps/kfreebsd/i386/sysdep.h 2006-01-19 14:02:55.0 +0100
 sysdeps/kfreebsd/i386/sysdep.h 2006-03-02 10:31:06.0 +0100
+@@ -81,27 +81,6 @@
+ # define SYSCALL_ERROR_HANDLER/* Nothing here; code in sysdep.S is 
used.  */
+ #else
+ 
+-# ifndef HAVE_HIDDEN
+-#  define SETUP_PIC_REG(reg) \
+-  call 1f;  \
+-  .subsection 1;\
+-1:movl (%esp), %e##reg; 

Re: r1287 - in trunk/kfreebsd-6/debian: . patches

2006-03-02 Thread Robert Millan
On Thu, Mar 02, 2006 at 10:39:48AM +, Robert Millan wrote:
> Author: rmh
> Date: 2006-03-02 10:39:47 + (Thu, 02 Mar 2006)
> New Revision: 1287
> 
> Modified:
>trunk/kfreebsd-6/debian/changelog
>trunk/kfreebsd-6/debian/patches/005_binutils.diff
> Log:
> - patches/binutils.diff: Add binutils fixes for amd64 code.

I would appreciate if someone reviewed this before we send it upstream.  I'm not
sure everything is 1:1 correct.

> Modified: trunk/kfreebsd-6/debian/changelog
> ===
> --- trunk/kfreebsd-6/debian/changelog 2006-03-02 09:24:02 UTC (rev 1286)
> +++ trunk/kfreebsd-6/debian/changelog 2006-03-02 10:39:47 UTC (rev 1287)
> @@ -4,6 +4,7 @@
>* New upstream version pulled from RELENG_6 CVSup.
>  - ReiserFS support.  (Closes: #335019)
>* Disabled the non-compiling hwpmc driver.
> +
>[ Robert Millan ]
>* Merge 5.x branch changes (from rev 654 to rev 1058)
>* Add src/usr.sbin/config into the source and build/use it dynamicaly.
> @@ -40,10 +41,12 @@
>  - arch/amd64/em64t-p4.config: New.  P4 config.
>  - patches/907_cpu_class.diff: Set hw.machine to "x86_64".  This improves
>  compatibility with old config.sub scripts.
> +  * patches/binutils.diff: Add binutils fixes for amd64 code.
> +
>[ Aurelien Jarno ]
>* Fix a remote denial of service in NFS server (FreeBSD-SA-06:10.nfs / 
> CVE-2006-0900).
>  
> - -- Robert Millan <[EMAIL PROTECTED]>  Tue, 28 Feb 2006 12:42:18 +0100
> + -- Robert Millan <[EMAIL PROTECTED]>  Thu,  2 Mar 2006 11:38:52 +0100
>  
>  kfreebsd-5 (5.3-6) unstable; urgency=low
>  
> 
> Modified: trunk/kfreebsd-6/debian/patches/005_binutils.diff
> ===
> --- trunk/kfreebsd-6/debian/patches/005_binutils.diff 2006-03-02 09:24:02 UTC 
> (rev 1286)
> +++ trunk/kfreebsd-6/debian/patches/005_binutils.diff 2006-03-02 10:39:47 UTC 
> (rev 1287)
> @@ -110,3 +110,338 @@
>   ret
>   
>   ENTRY(x86_gettid)
> +--- sys/amd64/amd64/cpu_switch.S.old 2006-03-01 20:12:04.0 +0100
>  sys/amd64/amd64/cpu_switch.S 2006-03-01 20:04:59.0 +0100
> +@@ -107,10 +107,10 @@
> + jz  1f  /* no, skip over */
> + 
> + /* Save segment selector numbers */
> +-movl%ds,PCB_DS(%r8)
> +-movl%es,PCB_ES(%r8)
> +-movl%fs,PCB_FS(%r8)
> +-movl%gs,PCB_GS(%r8)
> ++movw%ds,PCB_DS(%r8)
> ++movw%es,PCB_ES(%r8)
> ++movw%fs,PCB_FS(%r8)
> ++movw%gs,PCB_GS(%r8)
> + jmp 2f
> + 1:
> + 
> +@@ -191,14 +191,14 @@
> + jz  1f  /* no, skip over */
> + 
> + /* Restore segment selector numbers */
> +-movlPCB_DS(%r8),%ds
> +-movlPCB_ES(%r8),%es
> +-movlPCB_FS(%r8),%fs
> ++movwPCB_DS(%r8),%ds
> ++movwPCB_ES(%r8),%es
> ++movwPCB_FS(%r8),%fs
> + 
> + /* Restore userland %gs while preserving kernel gsbase */
> + movl$MSR_GSBASE,%ecx
> + rdmsr
> +-movlPCB_GS(%r8),%gs
> ++movwPCB_GS(%r8),%gs
> + wrmsr
> + jmp 2f
> + 1:
> +--- sys/amd64/ia32/ia32_signal.c~2006-03-01 22:23:22.0 +0100
>  sys/amd64/ia32/ia32_signal.c 2006-03-01 22:44:47.0 +0100
> +@@ -197,8 +197,8 @@
> + sf.sf_uc.uc_mcontext.mc_onstack = (oonstack) ? 1 : 0;
> + sf.sf_uc.uc_mcontext.mc_gs = rgs();
> + sf.sf_uc.uc_mcontext.mc_fs = rfs();
> +-__asm __volatile("movl %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
> +-__asm __volatile("movl %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
> ++__asm __volatile("mov %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
> ++__asm __volatile("mov %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
> + sf.sf_uc.uc_mcontext.mc_edi = regs->tf_rdi;
> + sf.sf_uc.uc_mcontext.mc_esi = regs->tf_rsi;
> + sf.sf_uc.uc_mcontext.mc_ebp = regs->tf_rbp;
> +@@ -309,8 +309,8 @@
> + sf.sf_uc.uc_mcontext.mc_onstack = (oonstack) ? 1 : 0;
> + sf.sf_uc.uc_mcontext.mc_gs = rgs();
> + sf.sf_uc.uc_mcontext.mc_fs = rfs();
> +-__asm __volatile("movl %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
> +-__asm __volatile("movl %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
> ++__asm __volatile("mov %%es,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_es));
> ++__asm __volatile("mov %%ds,%0" : "=rm" (sf.sf_uc.uc_mcontext.mc_ds));
> + sf.sf_uc.uc_mcontext.mc_edi = regs->tf_rdi;
> + sf.sf_uc.uc_mcontext.mc_esi = regs->tf_rsi;
> + sf.sf_uc.uc_mcontext.mc_ebp = regs->tf_rbp;
> +--- sys/amd64/include/pcpu.h~2006-03-01 23:10:56.0 +0100
>  sys/amd64/include/pcpu.h 2006-03-01 23:40:19.0 +0100
> +@@ -80,7 +80,7 @@
> + #define __PCPU_PTR(name) __extension__ ({   
> \
> + __pcpu_type(name) *__p; \
> + \

kfreebsd-amd64 base system ready

2006-03-02 Thread Robert Millan

Hi!

For those who want to play with it, there's a Debian GNU/kFreeBSD system for
amd64 now:

  http://io.debian.net/~rmh/amd64/

Its stage is very preliminar and only useful for developers.  Instructions:

  - Install FreeBSD 6.0 for amd64, and setup network.
  - Extract /boot/kernel from kfreebsd-image-6.0-1-amd64-generic, and use it to
  replace the stock kernel from FreeBSD.
  - Get base.tgz and extract it somewhere.
  - Mount /dev (devfs) and /proc (linprocfs).
  - chroot and run /native-install.
  - Get /etc/passwd and /etc/group from a sane Debian system and copy them over
(/etc/shadow is NOT needed).
  - Done.  You should be able to install things via apt-get now.

This is due to Petr Salinger who did all the actual work of porting Glibc and
debugging it.  Thanks a lot!

-- 
Robert Millan


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: r1287 - in trunk/kfreebsd-6/debian: . patches

2006-03-02 Thread Petr Salinger

Hi Robert,

many thanks for bootstraping kfreebsd-amd64 port.

I would appreciate if someone reviewed this before we send it upstream. 
I'm not sure everything is 1:1 correct.


Unfortunately no. Please, drop all patches to sys/amd64/include/pcpu.h.
I am afraid, rebuild of kernel should probably follow.

Instructions like "movq %%gs:%1,%0" should retain as they are.
It is move from memory with offset %1 with segment override from ds to gs
to somewhere (%0).

Only moves "mov %[fg]s, something" and "mov something, %[fg]s" have to be fixed.

Once again, many thanks for your hard work with bootstraping.

Petr


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: r1287 - in trunk/kfreebsd-6/debian: . patches

2006-03-02 Thread Robert Millan
On Thu, Mar 02, 2006 at 06:21:25PM +0100, Petr Salinger wrote:
> Hi Robert,
> 
> many thanks for bootstraping kfreebsd-amd64 port.
> 
> >I would appreciate if someone reviewed this before we send it upstream. 
> >I'm not sure everything is 1:1 correct.
> 
> Unfortunately no. Please, drop all patches to sys/amd64/include/pcpu.h.
> I am afraid, rebuild of kernel should probably follow.
> 
> Instructions like "movq %%gs:%1,%0" should retain as they are.
> It is move from memory with offset %1 with segment override from ds to gs
> to somewhere (%0).
> 
> Only moves "mov %[fg]s, something" and "mov something, %[fg]s" have to be 
> fixed.

Ok, done.  Let me know if I missed something.

> Once again, many thanks for your hard work with bootstraping.

No problem, I just put the cpu.  The real merit is yours for doing the though
stuff of porting Glibc and debugging.  I couldn't have handled that myself!

-- 
Robert Millan


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1294 - trunk/kfreebsd-6/debian/patches

2006-03-02 Thread Robert Millan
Author: rmh
Date: 2006-03-02 18:23:01 + (Thu, 02 Mar 2006)
New Revision: 1294

Modified:
   trunk/kfreebsd-6/debian/patches/005_binutils.diff
Log:
Remove wrong hunks as pointed by Petr.

Modified: trunk/kfreebsd-6/debian/patches/005_binutils.diff
===
--- trunk/kfreebsd-6/debian/patches/005_binutils.diff   2006-03-02 13:17:04 UTC 
(rev 1293)
+++ trunk/kfreebsd-6/debian/patches/005_binutils.diff   2006-03-02 18:23:01 UTC 
(rev 1294)
@@ -170,86 +170,6 @@
sf.sf_uc.uc_mcontext.mc_edi = regs->tf_rdi;
sf.sf_uc.uc_mcontext.mc_esi = regs->tf_rsi;
sf.sf_uc.uc_mcontext.mc_ebp = regs->tf_rbp;
 sys/amd64/include/pcpu.h~  2006-03-01 23:10:56.0 +0100
-+++ sys/amd64/include/pcpu.h   2006-03-01 23:40:19.0 +0100
-@@ -80,7 +80,7 @@
- #define   __PCPU_PTR(name) __extension__ ({   
\
-   __pcpu_type(name) *__p; \
-   \
--  __asm __volatile("movq %%gs:%1,%0; addq %2,%0"  \
-+  __asm __volatile("mov %%gs:%1,%0; addq %2,%0"   \
-   : "=r" (__p)\
-   : "m" (*(struct pcpu *)(__pcpu_offset(pc_prvspace))),   \
- "i" (__pcpu_offset(name)));   \
-@@ -96,25 +96,25 @@
-   \
-   if (sizeof(__result) == 1) {\
-   u_char __b; \
--  __asm __volatile("movb %%gs:%1,%0"  \
-+  __asm __volatile("mov %%gs:%1,%0"   \
-   : "=r" (__b)\
-   : "m" (*(u_char *)(__pcpu_offset(name;  \
-   __result = *(__pcpu_type(name) *)&__b;  \
-   } else if (sizeof(__result) == 2) { \
-   u_short __w;\
--  __asm __volatile("movw %%gs:%1,%0"  \
-+  __asm __volatile("mov %%gs:%1,%0"   \
-   : "=r" (__w)\
-   : "m" (*(u_short *)(__pcpu_offset(name; \
-   __result = *(__pcpu_type(name) *)&__w;  \
-   } else if (sizeof(__result) == 4) { \
-   u_int __i;  \
--  __asm __volatile("movl %%gs:%1,%0"  \
-+  __asm __volatile("mov %%gs:%1,%0"   \
-   : "=r" (__i)\
-   : "m" (*(u_int *)(__pcpu_offset(name;   \
-   __result = *(__pcpu_type(name) *)&__i;  \
-   } else if (sizeof(__result) == 8) { \
-   u_long __l; \
--  __asm __volatile("movq %%gs:%1,%0"  \
-+  __asm __volatile("mov %%gs:%1,%0"   \
-   : "=r" (__l)\
-   : "m" (*(u_long *)(__pcpu_offset(name;  \
-   __result = *(__pcpu_type(name) *)&__l;  \
-@@ -134,25 +134,25 @@
-   if (sizeof(__val) == 1) {   \
-   u_char __b; \
-   __b = *(u_char *)&__val;\
--  __asm __volatile("movb %1,%%gs:%0"  \
-+  __asm __volatile("mov %1,%%gs:%0"   \
-   : "=m" (*(u_char *)(__pcpu_offset(name)))   \
-   : "r" (__b));   \
-   } else if (sizeof(__val) == 2) {\
-   u_short __w;\
-   __w = *(u_short *)&__val;   \
--  __asm __volatile("movw %1,%%gs:%0"  \
-+  __asm __volatile("mov %1,%%gs:%0"   \
-   : "=m" (*(u_short *)(__pcpu_offset(name)))  \
-   : "r" (__w));   \
-   } else if (sizeof(__val) == 4) {\
-   u_int __i;  \
-   __i = *(u_int *)&__val; \
--  __asm __volatile("movl %1,%%gs:%0"  \
-+  __asm __volatile("mov %1,

kfreebsd-amd64

2006-03-02 Thread Robert Millan

Hi!

Please could you consider the attached patch?  (for packages module)

It adds "kfreebsd-amd64" and a few more GNU/kFreeBSD mirrors to the list.

Thanks

-- 
Robert Millan


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: kfreebsd-amd64

2006-03-02 Thread Robert Millan

Oops, forgot the patch.

Here is it.

-- 
Robert Millan
Index: cgi-bin/download.pl
===
RCS file: /cvs/webwml/packages/cgi-bin/download.pl,v
retrieving revision 1.20
diff -u -r1.20 download.pl
--- cgi-bin/download.pl 17 Jan 2006 21:20:26 -  1.20
+++ cgi-bin/download.pl 2 Mar 2006 19:16:14 -
@@ -111,7 +111,12 @@
 "kfreebsd-gnu.debian.net/debian",
 "ftp.easynet.be/ftp/gnuab/debian",
"ftp.de.debian.org/debian-kfreebsd",
+   "ftp.sh.cvut.cz/MIRRORS/debian-non-linux",
+   "ftp.eila.jussieu.fr/pub/Debian-gnuab",
   );
+my @kfreebsd_asian_sites = (
+   "ftp.citkit.ru/pub/HURD/debian",
+   );
 my @nonus_north_american_sites = (
 #  "ftp.ca.debian.org/debian-non-US",
"debian.yorku.ca/debian/non-US",
@@ -171,7 +176,8 @@
 s390=> 'IBM S/390',
"hurd-i386" => 'Hurd (i386)',
amd64   => 'AMD64',
-   "kfreebsd-i386" => 'GNU/kFreeBSD (i386)'
+   "kfreebsd-i386" => 'GNU/kFreeBSD (i386)',
+   "kfreebsd-amd64" => 'GNU/kFreeBSD (amd64)',
 );
 
 my $urlbase = "http://www.debian.org";;
@@ -241,13 +247,13 @@
   current information.
 END
 
-} elsif ($arch eq 'kfreebsd-i386') {
+} elsif (($arch eq 'kfreebsd-i386') || ($arch eq 'kfreebsd-amd64')) {
 
 print_links( "North America", $file, @kfreebsd_north_american_sites );
 print_links( "Europe", $file, @kfreebsd_european_sites );
 #print_links( "Australia and New Zealand", $file,
 #   @nonus_australian_sites );
-#print_links( "Asia", $file, @amd64_asian_sites );
+print_links( "Asia", $file, @kfreebsd_asian_sites );
 #print_links( "South America", $file, @nonus_south_american_sites );
 
print <

r1295 - trunk/glibc-2.3-head

2006-03-02 Thread Robert Millan
Author: rmh
Date: 2006-03-02 20:07:10 + (Thu, 02 Mar 2006)
New Revision: 1295

Modified:
   trunk/glibc-2.3-head/AMD64
Log:
Update testsuite failure list (got much better, i'm not sure why..)

Modified: trunk/glibc-2.3-head/AMD64
===
--- trunk/glibc-2.3-head/AMD64  2006-03-02 18:23:01 UTC (rev 1294)
+++ trunk/glibc-2.3-head/AMD64  2006-03-02 20:07:10 UTC (rev 1295)
@@ -2,43 +2,18 @@
 
 * glibc testsuite failures:
 
-[catgets/de/libc.cat] Error 1
-[catgets/test-gencat.out] Error 1
-[catgets/test1.cat] Error 1
-[dlfcn/tststatic.out] Error 139
-[dlfcn/tststatic2.out] Error 139
-[libio/bug-ftell.out] Error 1
-[libio/bug-ungetwc2.out] Error 1
-[libio/tst-swscanf.out] Error 1
-[libio/tst-ungetwc2.out] Error 1
-[libio/tst-widetext.out] Error 1
-[linuxthreads/tst-clock1.out] Error 1
-[posix/annexc.out] Error 138 (ignored)
-[posix/bug-regex17.out] Error 1
-[posix/bug-regex18.out] Error 1
-[posix/bug-regex19.out] Error 1
-[posix/bug-regex20.out] Error 1
-[posix/bug-regex22.out] Error 1
-[posix/bug-regex5.out] Error 1
-[posix/bug-regex6.out] Error 1
-[posix/tst-fnmatch.out] Error 1
-[posix/tst-regex.out] Error 1
-[posix/tst-regexloc.out] Error 1
-[posix/tst-rxspencer.out] Error 1
-[posix/tst-waitid.out] Error 1 (already known, see NOTES)
-[stdio-common/bug14.out] Error 1
-[stdio-common/tst-swprintf.out] Error 1
-[stdlib/testmb2.out] Error 1
-[stdlib/tst-strtod.out] Error 1
-[string/tst-strxfrm.out] Error 1
-[time/tst-ftime_l.out] Error 1
-[time/tst-strptime.out] Error 1
-[wcsmbs/tst-btowc.out] Error 1
-[wcsmbs/tst-mbrtowc.out] Error 1
+[localedata/sort-test.out] Error 1
+[stdlib/tst-setcontext.out] Error 255
+[posix/tst-rxspencer.out] Error 139
+[posix/tst-waitid.out] Error 1
+[posix/annexc.out] Error 1 (ignored)
+[rt/tst-timer2.o] Error 1
+[rt/tst-aio.out] Error 1
+[rt/tst-aio64.out] Error 1
+[rt/tst-mqueue5.o] Error 1
+[rt/tst-timer4.o] Error 1
+[rt/tst-timer5.o] Error 1
 
-In linuxthreads/tst-context, I had to interrupt the tests, because it looked
-stalled.  Might be a deadlock, I don't know.. (will give it more time later)
-
 * decide ld.so location
It should be unique between all (at least Debian) ports,
namely not the same as on kfreebsd-i386 (/lib/ld.so.1).


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1296 - trunk/glibc-2.3-head

2006-03-02 Thread Petr Salinger
Author: ps-guest
Date: 2006-03-02 20:55:43 + (Thu, 02 Mar 2006)
New Revision: 1296

Modified:
   trunk/glibc-2.3-head/AMD64
   trunk/glibc-2.3-head/NOTES
Log:
  classify testsuite failures



Modified: trunk/glibc-2.3-head/AMD64
===
--- trunk/glibc-2.3-head/AMD64  2006-03-02 20:07:10 UTC (rev 1295)
+++ trunk/glibc-2.3-head/AMD64  2006-03-02 20:55:43 UTC (rev 1296)
@@ -2,8 +2,13 @@
 
 * glibc testsuite failures:
 
+kfreebsd-amd64 only:
+
+[stdlib/tst-setcontext.out] Error 255
+
+fails also on kfreebsd-i386:
+
 [localedata/sort-test.out] Error 1
-[stdlib/tst-setcontext.out] Error 255
 [posix/tst-rxspencer.out] Error 139
 [posix/tst-waitid.out] Error 1
 [posix/annexc.out] Error 1 (ignored)

Modified: trunk/glibc-2.3-head/NOTES
===
--- trunk/glibc-2.3-head/NOTES  2006-03-02 20:07:10 UTC (rev 1295)
+++ trunk/glibc-2.3-head/NOTES  2006-03-02 20:55:43 UTC (rev 1296)
@@ -83,6 +83,10 @@
i486-kfreebsd-gnu (with -march=i486 -mtune=i686)
i686-kfreebsd-gnu (with -march=i686 -mtune=i686)
 
+  failure within Debian package 2.3.6-3, but no with pristine glibc 2.3.6
+make[3]: *** 
[/home/aurel32/tmp/glibc/glibc-2.3.6/build-tree/kfreebsd-i386-libc/localedata/sort-test.out]
 Erreur 1
+make[3]: *** 
[/home/aurel32/tmp/glibc/glibc-2.3.6/build-tree/kfreebsd-i386-libc/posix/tst-rxspencer.out]
 Erreur 139
+
   examine why, maybe due to non-rt behaviour of signals:
 make[2]: *** [/build/glibc-2.3-head/build/rt/tst-aio.out] Error 1
 make[2]: *** [/build/glibc-2.3-head/build/rt/tst-aio64.out] Error 1


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



r1297 - trunk/web/patches/upstream-only

2006-03-02 Thread Robert Millan
Author: rmh
Date: 2006-03-02 23:14:15 + (Thu, 02 Mar 2006)
New Revision: 1297

Added:
   trunk/web/patches/upstream-only/gcc.diff
Log:
Add (untested) patch for gcc trunk.

Added: trunk/web/patches/upstream-only/gcc.diff
===
--- trunk/web/patches/upstream-only/gcc.diff2006-03-02 20:55:43 UTC (rev 
1296)
+++ trunk/web/patches/upstream-only/gcc.diff2006-03-02 23:14:15 UTC (rev 
1297)
@@ -0,0 +1,156 @@
+
+Author: rmh
+Status: untested, unsent
+
+diff -ur gcc.old/config/i386/kfreebsd-gnu.h gcc/config/i386/kfreebsd-gnu.h
+--- gcc.old/config/i386/kfreebsd-gnu.h 2006-03-02 21:22:05.0 +0100
 gcc/config/i386/kfreebsd-gnu.h 2006-03-03 00:08:05.0 +0100
+@@ -1,5 +1,5 @@
+-/* Definitions for Intel 386 running kFreeBSD-based GNU systems with ELF 
format
+-   Copyright (C) 2004
++/* Definitions for Intel 386 (and AMD x86-64) running kFreeBSD-based GNU 
systems with ELF format
++   Copyright (C) 2004, 2006
+Free Software Foundation, Inc.
+Contributed by Robert Millan.
+ 
+@@ -20,7 +20,28 @@
+ the Free Software Foundation, 51 Franklin Street, Fifth Floor,
+ Boston, MA 02110-1301, USA.  */
+ 
++#ifdef GLIBC_DYNAMIC_LINKER32
++#undef GLIBC_DYNAMIC_LINKER32
++#define GLIBC_DYNAMIC_LINKER32 "/lib/ld.so.1"
++#endif
++
++#ifdef GLIBC_DYNAMIC_LINKER64
++#undef GLIBC_DYNAMIC_LINKER64
++#define GLIBC_DYNAMIC_LINKER64 "/lib/ld-kfreebsd-x86-64.so.1"
++#endif
++
+ #undef LINK_EMULATION
+ #define LINK_EMULATION "elf_i386_fbsd"
++
++#ifdef LINK_EMULATION32
++#undef LINK_EMULATION32
++#define LINK_EMULATION32 LINK_EMULATION
++#endif
++
++#ifdef LINK_EMULATION64
++#undef LINK_EMULATION64
++#define LINK_EMULATION64 "elf_x86_64_fbsd"
++#endif
++
+ #undef REG_NAME
+ #define REG_NAME(reg) sc_ ## reg
+diff -ur gcc.old/config/i386/knetbsd-gnu.h gcc/config/i386/knetbsd-gnu.h
+--- gcc.old/config/i386/knetbsd-gnu.h  2006-03-02 21:22:05.0 +0100
 gcc/config/i386/knetbsd-gnu.h  2006-03-03 00:08:52.0 +0100
+@@ -1,5 +1,5 @@
+-/* Definitions for Intel 386 running kNetBSD-based GNU systems with ELF format
+-   Copyright (C) 2004
++/* Definitions for Intel 386 (and AMD x86-64) running kNetBSD-based GNU 
systems with ELF format
++   Copyright (C) 2004, 2006
+Free Software Foundation, Inc.
+Contributed by Robert Millan.
+ 
+@@ -20,5 +20,15 @@
+ the Free Software Foundation, 51 Franklin Street, Fifth Floor,
+ Boston, MA 02110-1301, USA.  */
+ 
++#ifdef GLIBC_DYNAMIC_LINKER32
++#undef GLIBC_DYNAMIC_LINKER32
++#define GLIBC_DYNAMIC_LINKER32 "/lib/ld.so.1"
++#endif
++
++#ifdef GLIBC_DYNAMIC_LINKER64
++#undef GLIBC_DYNAMIC_LINKER64
++#define GLIBC_DYNAMIC_LINKER64 "/lib/ld-knetbsd-x86-64.so.1"
++#endif
++
+ #undef REG_NAME
+ #define REG_NAME(reg) sc_ ## reg
+diff -ur gcc.old/config/i386/linux.h gcc/config/i386/linux.h
+--- gcc.old/config/i386/linux.h2006-03-02 21:22:05.0 +0100
 gcc/config/i386/linux.h2006-03-03 00:04:02.0 +0100
+@@ -99,6 +99,7 @@
+ 
+ /* If ELF is the default format, we should not use /lib/elf.  */
+ 
++/* These macros may be overridden in k*bsd-gnu.h and i386/k*bsd-gnu.h. */
+ #define LINK_EMULATION "elf_i386"
+ #define GLIBC_DYNAMIC_LINKER "/lib/ld-linux.so.2"
+ 
+diff -ur gcc.old/config/i386/linux64.h gcc/config/i386/linux64.h
+--- gcc.old/config/i386/linux64.h  2006-03-02 21:22:05.0 +0100
 gcc/config/i386/linux64.h  2006-03-03 00:03:42.0 +0100
+@@ -49,11 +49,14 @@
+When the -shared link option is used a final link is not being
+done.  */
+ 
++/* These macros may be overridden in k*bsd-gnu.h and i386/k*bsd-gnu.h. */
+ #define GLIBC_DYNAMIC_LINKER32 "/lib/ld-linux.so.2"
+ #define GLIBC_DYNAMIC_LINKER64 "/lib64/ld-linux-x86-64.so.2"
++#define LINK_EMULATION32  "elf_i386"
++#define LINK_EMULATION64  "elf_x86_64"
+ 
+ #undefLINK_SPEC
+-#define LINK_SPEC "%{!m32:-m elf_x86_64} %{m32:-m elf_i386} \
++#define LINK_SPEC "%{!m32:-m " LINK_EMULATION64 "} %{m32:-m " 
LINK_EMULATION32 "} \
+   %{shared:-shared} \
+   %{!shared: \
+ %{!static: \
+diff -ur gcc.old/config/kfreebsd-gnu.h gcc/config/kfreebsd-gnu.h
+--- gcc.old/config/kfreebsd-gnu.h  2006-03-02 21:23:05.0 +0100
 gcc/config/kfreebsd-gnu.h  2006-03-03 00:47:03.0 +0100
+@@ -1,5 +1,5 @@
+ /* Definitions for kFreeBSD-based GNU systems with ELF format
+-   Copyright (C) 2004
++   Copyright (C) 2004, 2006
+Free Software Foundation, Inc.
+Contributed by Robert Millan.
+ 
+@@ -32,5 +32,7 @@
+ } \
+   while (0)
+ 
+-#undef DYNAMIC_LINKER
+-#define DYNAMIC_LINKER "/lib/ld.so.1"
++#ifdef GLIBC_DYNAMIC_LINKER
++#undef GLIBC_DYNAMIC_LINKER
++#define GLIBC_DYNAMIC_LINKER "/lib/ld.so.1"
++#endif
+diff -ur gcc.old/config/knetbsd-gnu.h gcc/config/knetbsd-gnu.h
+--- gcc.old/config/knetbsd-gnu.h   2006-03-02 21:23:05.0 +0100
 gcc/config/knetbsd-gnu.h   2006-03-03 00:47:09.0 +0100
+@@ -1,5 +1,5 @@

Re: kfreebsd-amd64

2006-03-02 Thread Frank Lichtenheld
On Thu, Mar 02, 2006 at 08:46:49PM +0100, Robert Millan wrote:
> 
> Oops, forgot the patch.
> 
> Here is it.

Just FYI: That patch alone will not add kfreebsd-amd64 support to
packages.debian.org since you need some fixes to the cron job, too.

No need to provide a complete patch though. I will fix that myself.

Gruesse,
-- 
Frank Lichtenheld <[EMAIL PROTECTED]>
www: http://www.djpig.de/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



D-I Etch Beta2 - Status update (3)

2006-03-02 Thread Frans Pop
I've made a complete mess of CD images for Beta2 so far as the result of a 
wrong assumption. This means, as some installation reports and comments 
have shown, that CD images linked from [1] have been broken since last 
Friday.

The good news is that there are now good Beta2 netinst and businesscard CD 
test images available from [1]. Note that these are not the regular 
etch_d-i or sid_d-i. The correct link for Beta2 images is currently:
http://cdimage.debian.org/cdimage/beta2-test-build/

Initial tests using these images for i386 and sparc64 have shown no 
problems. Tests for other arches are very much appreciated [2].

This also means that there are no good full CD images available yet. 
However, some issues with full CD images have been identified [3] and are 
being worked on.


I've also uploaded the (hopefully final) 20060302 build of 
debian-installer today which includes the new 2.6.15-7 kernels. It has 
been accepted for i386 and so should now start building for all arches.

We will probably have a small further delay in the Beta2 release, but 
because no blocking issues have been identified as yet, it looks as if we 
can keep it minimal. I'm somewhat reluctant to post hard dates ATM.

I'm very sorry for any confusion. Please blame it on my inexperience as 
release manager for d-i and the complexity of the whole process.

Cheers,
FJP

[1] http://www.debian.org/devel/debian-installer/
[2] SVN: installer/doc/devel/release-checklist
[3] http://lists.debian.org/debian-boot/2006/03/msg00050.html


pgp6UtCZYLFdT.pgp
Description: PGP signature