patching grub-0.92 with lvm-diff fails (ref. Bug report #977)

[Erich Waelde]

Hi all,

I have Linux running on scsi+md+raid1+lvm+ext3 and ide+lvm+reiserfs.
The /boot partition resides outside of lvm, I hacked mkinitrd and
lvmcreate_initrd to create initrd images with all modules and
executables needed. So far, so nice.

As others, too, I would like to have /boot reside in lvm as well.
The order of lvols on disk (/boot, swap, / --- all contiguous)
would be just fine with me. However, I prefer grub over lilo.

When searching I found a patch by J"org Walter on the grub home page
  http://savannah.gnu.org/projects/grub/
listed as bug report # 977
  http://savannah.gnu.org/bugs/?func=detailbug&bug_id=977&group_id=68

md5sum:
0f36630c321312f8e5369af8ca89ba76  grub-lvm.diff
50f2786e7bf96b1115ee82b9c101a816  grub-0.92.tar.gz


Unfortunately the posting does not mention the exact version
of grub it was applied to. However, the diff mentions "grub-0.92"
in the file names, e.g.:
  diff -rubBP -x ltmain.sh -x ...
--- /home/tmp/portage/grub-0.92/work/grub-0.92/ChangeLog Thu Aug  8
20:54:29 2002
+++ grub-0.92/ChangeLog Sun Aug 11 14:17:55 2002

Unfortunately the patch does not apply cleanly:

  ls -l
  -rw-------    1 ew       users      877112 Oct 27 09:26
grub-0.92.tar.gz
  -rw-------    1 ew       users       39578 Oct 27 09:23 grub-lvm.diff
  tar xzf grub-0.92.tar.gz
  ...
  patch  -p0 -bl -V t --verbose < grub-lvm.diff | tee patch.log 
  ...

Did I do something stupid, like trying to patch the wrong version ?

A closer look at the *.rej files indeed suggests that patch does
not find what it expects (see ./grub-0.92/lib/device.c).


Does someone know what the problem is? Is a solution available? Or can
I help other than trying to hack the source code ?

Any help is appreciated,

Erich



please find appended:
1. patch.log
2. the rejected context diffs
./grub-0.92/configure.in.rej
./grub-0.92/stage2/Makefile.am.rej
./grub-0.92/lib/device.c.rej

---snip-8<------------------------------------------------------------
patch.log:

Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/ChangeLog grub-0.92/ChangeLog
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/ChangeLog       Thu Aug  8
20:54:29 2002
|+++ grub-0.92/ChangeLog        Sun Aug 11 14:17:55 2002
--------------------------
Patching file grub-0.92/ChangeLog using Plan A...
Hunk #1 succeeded at 1.
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/config.h.in
grub-0.92/config.h.in
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/config.h.in     Tue Apr 16
00:09:16 2002
|+++ grub-0.92/config.h.in      Sun Aug 11 15:07:27 2002
--------------------------
Patching file grub-0.92/config.h.in using Plan A...
Hunk #1 succeeded at 47.
Hunk #2 succeeded at 86.
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/configure.in
grub-0.92/configure.in
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/configure.in    Thu Aug  8
20:54:29 2002
|+++ grub-0.92/configure.in     Sun Aug 11 15:16:21 2002
--------------------------
Patching file grub-0.92/configure.in using Plan A...
Hunk #1 FAILED at 110.
Hunk #2 succeeded at 185 (offset -1 lines).
1 out of 2 hunks FAILED -- saving rejects to file
grub-0.92/configure.in.rej
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/docs/grub.texi
grub-0.92/docs/grub.texi
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/docs/grub.texi  Thu Aug 
8 20:54:29 2002
|+++ grub-0.92/docs/grub.texi   Sun Aug 11 14:17:50 2002
--------------------------
Patching file grub-0.92/docs/grub.texi using Plan A...
Hunk #1 succeeded at 406.
Hunk #2 succeeded at 1576.
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/lib/device.c
grub-0.92/lib/device.c
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/lib/device.c    Thu Aug  8
20:54:29 2002
|+++ grub-0.92/lib/device.c     Sun Aug 11 14:45:26 2002
--------------------------
Patching file grub-0.92/lib/device.c using Plan A...
Hunk #1 succeeded at 648.
Hunk #2 FAILED at 680.
1 out of 2 hunks FAILED -- saving rejects to file
grub-0.92/lib/device.c.rej
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/stage2/Makefile.am
grub-0.92/stage2/Makefile.am
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/stage2/Makefile.am      Thu
Aug  8 20:54:29 2002
|+++ grub-0.92/stage2/Makefile.am       Sun Aug 11 14:17:54 2002
--------------------------
Patching file grub-0.92/stage2/Makefile.am using Plan A...
Hunk #1 succeeded at 17.
Hunk #2 succeeded at 85 with fuzz 2.
Hunk #3 succeeded at 123 with fuzz 1.
Hunk #4 succeeded at 132 with fuzz 1.
Hunk #5 FAILED at 141.
Hunk #6 succeeded at 159 with fuzz 1.
Hunk #7 succeeded at 168 with fuzz 1.
Hunk #8 succeeded at 177 with fuzz 1.
Hunk #9 succeeded at 186 with fuzz 1.
1 out of 9 hunks FAILED -- saving rejects to file
grub-0.92/stage2/Makefile.am.rej
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/stage2/builtins.c
grub-0.92/stage2/builtins.c
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/stage2/builtins.c       Thu
Aug  8 20:54:29 2002
|+++ grub-0.92/stage2/builtins.c        Sun Aug 11 14:47:46 2002
--------------------------
Patching file grub-0.92/stage2/builtins.c using Plan A...
Hunk #1 succeeded at 1225 (offset -115 lines).
Hunk #2 succeeded at 3109 (offset -148 lines).
Hunk #3 succeeded at 3750 (offset -326 lines).
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/stage2/char_io.c
grub-0.92/stage2/char_io.c
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/stage2/char_io.c        Thu
Aug  8 20:54:29 2002
|+++ grub-0.92/stage2/char_io.c Sun Aug 11 14:44:23 2002
--------------------------
Patching file grub-0.92/stage2/char_io.c using Plan A...
Hunk #1 succeeded at 1029.
Hunk #2 succeeded at 1046.
Hunk #3 succeeded at 1403.
Hunk #4 succeeded at 1415.
Hunk #5 succeeded at 1506.
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/stage2/disk_io.c
grub-0.92/stage2/disk_io.c
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/stage2/disk_io.c        Thu
Aug  8 20:54:29 2002
|+++ grub-0.92/stage2/disk_io.c Sun Aug 11 15:18:53 2002
--------------------------
Patching file grub-0.92/stage2/disk_io.c using Plan A...
Hunk #1 succeeded at 34.
Hunk #2 succeeded at 126.
Hunk #3 succeeded at 299.
Hunk #4 succeeded at 368.
Hunk #5 succeeded at 688.
Hunk #6 succeeded at 753.
Hunk #7 succeeded at 919.
Hunk #8 succeeded at 959.
Hunk #9 succeeded at 1103.
Hunk #10 succeeded at 1153.
Hunk #11 succeeded at 1217.
Hunk #12 succeeded at 1326.
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/stage2/linuxlvm.c
grub-0.92/stage2/linuxlvm.c
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/stage2/linuxlvm.c       Thu
Jan  1 01:00:00 1970
|+++ grub-0.92/stage2/linuxlvm.c        Sun Aug 11 14:56:34 2002
--------------------------
Patching file grub-0.92/stage2/linuxlvm.c using Plan A...
Hunk #1 succeeded at 1.
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|diff -rubBP -x ltmain.sh -x aclocal.m4 -x Makefile.in -x configure
/home/tmp/portage/grub-0.92/work/grub-0.92/stage2/shared.h
grub-0.92/stage2/shared.h
|--- /home/tmp/portage/grub-0.92/work/grub-0.92/stage2/shared.h Thu Aug 
8 20:54:29 2002
|+++ grub-0.92/stage2/shared.h  Sun Aug 11 15:20:21 2002
--------------------------
Patching file grub-0.92/stage2/shared.h using Plan A...
Hunk #1 succeeded at 837 (offset -69 lines).
Hunk #2 succeeded at 925 (offset -69 lines).
done
---snip-8<------------------------------------------------------------
./grub-0.92/configure.in.rej:

***************
*** 110,116 ****
  ])
  
  # Force no alignment to save space.
- CPPFLAGS="$CPPFLAGS -falign-jumps=1 -falign-loops=1
-falign-functions=1"
  
  if test "x$undef_flag" = xyes; then
    CPPFLAGS="$CPPFLAGS -Wundef"
--- 110,116 ----
  ])
  
  # Force no alignment to save space.
+ CPPFLAGS="$CPPFLAGS -malign-jumps=0 -malign-loops=0
-malign-functions=0"
  
  if test "x$undef_flag" = xyes; then
    CPPFLAGS="$CPPFLAGS -Wundef"

---snip-8<------------------------------------------------------------
./grub-0.92/stage2/Makefile.am.rej:

***************
*** 141,156 ****
  
  # For ffs_stage1_5 target.
  ffs_stage1_5_exec_SOURCES = start.S asm.S common.c char_io.c disk_io.c
\
-       stage1_5.c fsys_ffs.c bios.c
  ffs_stage1_5_exec_CFLAGS = $(STAGE1_5_COMPILE) -DFSYS_FFS=1 \
-       -DNO_BLOCK_FILES=1
  ffs_stage1_5_exec_CCASFLAGS = $(STAGE1_5_COMPILE) -DFSYS_FFS=1 \
        -DNO_BLOCK_FILES=1
  ffs_stage1_5_exec_LDFLAGS = $(STAGE1_5_LINK)
  
  # For minix_stage1_5 target.
  minix_stage1_5_exec_SOURCES = start.S asm.S common.c char_io.c
disk_io.c \
-       stage1_5.c fsys_minix.c bios.c
  minix_stage1_5_exec_CFLAGS = $(STAGE1_5_COMPILE) -DFSYS_MINIX=1 \
        -DNO_BLOCK_FILES=1
  minix_stage1_5_exec_CCASFLAGS = $(STAGE1_5_COMPILE) -DFSYS_MINIX=1 \
--- 141,156 ----
  
  # For ffs_stage1_5 target.
  ffs_stage1_5_exec_SOURCES = start.S asm.S common.c char_io.c disk_io.c
\
+       stage1_5.c fsys_ffs.c bios.c linuxlvm.c
  ffs_stage1_5_exec_CFLAGS = $(STAGE1_5_COMPILE) -DFSYS_FFS=1 \
+       -DNO_BLOCK_FILES=1 -DNO_LVM_IN_STAGE1_5=1
  ffs_stage1_5_exec_CCASFLAGS = $(STAGE1_5_COMPILE) -DFSYS_FFS=1 \
        -DNO_BLOCK_FILES=1
  ffs_stage1_5_exec_LDFLAGS = $(STAGE1_5_LINK)
  
  # For minix_stage1_5 target.
  minix_stage1_5_exec_SOURCES = start.S asm.S common.c char_io.c
disk_io.c \
+       stage1_5.c fsys_minix.c bios.c linuxlvm.c
  minix_stage1_5_exec_CFLAGS = $(STAGE1_5_COMPILE) -DFSYS_MINIX=1 \
        -DNO_BLOCK_FILES=1
  minix_stage1_5_exec_CCASFLAGS = $(STAGE1_5_COMPILE) -DFSYS_MINIX=1 \

---snip-8<------------------------------------------------------------
./grub-0.92/lib/device.c.rej:

***************
*** 671,676 ****
            strncmp(dev, "/dev/cciss/", 11) == 0 ||
            strncmp(dev, "/dev/rd/", 8) == 0) ? "p" : "",
           ((partition >> 16) & 0xFF) + 1);
    
    /* Open the partition.  */
    fd = open (dev, O_RDWR);
--- 680,686 ----
            strncmp(dev, "/dev/cciss/", 11) == 0 ||
            strncmp(dev, "/dev/rd/", 8) == 0) ? "p" : "",
           ((partition >> 16) & 0xFF) + 1);
+     }
    
    /* Open the partition.  */
    fd = open (dev, O_RDWR);