Compile error with Paul's pmac-devel rsync tree

Compile error with Paul's pmac-devel rsync tree

[Michel Lanners]

[-- Attachment #1: Type: TEXT/plain, Size: 1058 bytes --]

Hi all,

Thought I let you all know so you don't have to search for yourself.
Patch is below...

- There is a problem in compiling ext2 fs support, in fs/ext2/dir.c.
  This looks like part of a patch submitted to linux-kernel sometime in
  June. I haven't checked whether this is wrong in kernel.org kernels as
  well...

- The same goes for mark_buffer_dirty_inode() in fs/ext2/truncate.c,
  which seems to come out of the same patch...

- arch/ppc/kernel/chrp_setup.c has a conditional in the wrong place. Can
  somebody correct this? Paul? It only appears as an error if sysrq key
  is not enabled, so that's probably why nobody noticed...

Now I'll check whether it actually works ;-)

Happy compiling!

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


[-- Attachment #2: rsync.patch --]
[-- Type: TEXT/plain, Size: 1202 bytes --]

--- linux-2.4.paul/fs/ext2/dir.c	Mon Jun 12 02:10:39 2000
+++ linux/fs/ext2/dir.c	Wed Aug  9 20:21:23 2000
@@ -26,7 +26,7 @@
 	read:		generic_read_dir,
 	readdir:	ext2_readdir,
 	ioctl:		ext2_ioctl,
-	fsync:		ext2_fsync_file,
+	fsync:		ext2_sync_file,
 };

 int ext2_check_dir_entry (const char * function, struct inode * dir,
--- linux-2.4.paul/arch/ppc/kernel/chrp_setup.c	Tue Aug  8 08:07:45 2000
+++ linux/arch/ppc/kernel/chrp_setup.c	Wed Aug  9 20:40:04 2000
@@ -477,8 +477,8 @@
 #ifdef CONFIG_MAGIC_SYSRQ
 		ppc_md.ppc_kbd_sysrq_xlate	 = mackbd_sysrq_xlate;
 		SYSRQ_KEY = 0x69;
-	}
 #endif /* CONFIG_MAGIC_SYSRQ */
+	}
 #endif /* CONFIG_VT && CONFIG_ADB_KEYBOARD */
 }

--- linux-2.4.paul/fs/ext2/truncate.c	Mon Jun 12 02:10:39 2000
+++ linux/fs/ext2/truncate.c	Wed Aug  9 20:47:42 2000
@@ -211,7 +211,7 @@
 			inode->i_ino, tmp);
 		*p = 0;
 		if (dind_bh)
-			mark_buffer_dirty_inode(dind_bh, 1, inode);
+			mark_buffer_dirty(dind_bh, 1);
 		else
 			mark_inode_dirty(inode);
 		return 0;
@@ -279,7 +279,7 @@
 			inode->i_ino, tmp);
 		*p = 0;
 		if (tind_bh)
-			mark_buffer_dirty_inode(tind_bh, 1, inode);
+			mark_buffer_dirty(tind_bh, 1);
 		else
 			mark_inode_dirty(inode);
 		return 0;

Re: Compile error with Paul's pmac-devel rsync tree

[Martin Costabel]

Michel Lanners wrote:
>
> Hi all,
>
> Thought I let you all know so you don't have to search for yourself.
> Patch is below...
>
> - There is a problem in compiling ext2 fs support, in fs/ext2/dir.c.
>   This looks like part of a patch submitted to linux-kernel sometime in
>   June. I haven't checked whether this is wrong in kernel.org kernels as
>   well...

I think this has been corrected already earlier today. Just rsync again.
It's funny to look at this in bitkeeper: it has been changed back and
forth several times.

> - The same goes for mark_buffer_dirty_inode() in fs/ext2/truncate.c,
>   which seems to come out of the same patch...

Also fixed as of 8:30 GMT today.

I had sometimes problems loading modules with this version, even with
modutils-2.3.14. I am not quite sure what was the cause, it just didn't
work ("module de4x5 not found"). Sometimes it just worked (I am running
it right now).

Otherwise standard bugs persist:
- No mounting of HFS CDs
- No mounting of loop devices
- No strace

Another bug of Paul's devel tree in comparison with the bitkeeper tree:
There is something wrong with the passing of environment variables. I
have been seeing this at least since 2.4.0-test1-ac18, but I still
haven't quite understood what is going on.

--
Martin

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Compile error with Paul's pmac-devel rsync tree

[Michel Lanners]

Hi all,

On   9 Aug, this message from Martin Costabel echoed through cyberspace:
[errors snip'ed]
> I think this has been corrected already earlier today. Just rsync again.
> It's funny to look at this in bitkeeper: it has been changed back and
> forth several times.
>
>> - The same goes for mark_buffer_dirty_inode() in fs/ext2/truncate.c,
>>   which seems to come out of the same patch...
>
> Also fixed as of 8:30 GMT today.

Hmm... I had rsync'ed just minutes before sending my mail, so I guess it
has been changing back and forth in Paul's rsync tree as well ;-)

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/