Linux 2.4.25-rc2

Linux 2.4.25-rc2

[Marcelo Tosatti]

Ola,

Here goes -rc2, with small number of fixes and corrections.

Most notably acpi4asus and toshiba ACPI updates.

Detailed changelog follows.

Summary of changes from v2.4.25-rc1 to v2.4.25-rc2
============================================

<brad:wasp.net.au>:
  o backport 2.6.x yenta detection fix

<davem:nuts.davemloft.net>:
  o [IPV4]: Use same sysctl number for IGMP version forcing as 2.6.x
  o [SPARC64]: Fix exception remaining length calcs in VIS copy routines

<john:neggie.net>:
  o toshiba_acpi 0.17 from John Belmonte

<khali:linux-fr.org>:
  o Small i2c maintainer correction

<len.brown:intel.com>:
  o [ACPI] proposed fix for AML parameter passing from Bob Moore http://bugzilla.kernel.org/show_bug.cgi?id=1766
  o [ACPI] proposed fix for AML parameter passing from Bob Moore http://bugzilla.kernel.org/show_bug.cgi?id=1766
  o [ACPI] fix IA64 build warning from Martin Hicks

<macro:ds2.pg.dga.pl>:
  o [NET]: Fix comment typo in net/socket.c

<marcelo:logos.cnet>:
  o Changed EXTRAVERSION to -rc2

<shaggy:kleikamp.dyn.webahead.ibm.com>:
  o JFS: rename should update mtime on source and target directories
  o JFS: Threads should exit with complete_and_exit

<sziwan:hell.org.pl>:
  o acpi4asus update from Karol 'sziwan' Kozimor

Adrian Bunk:
  o Fix amd7930_fn.c compilation with CONFIG_HOTPLUG=n

Alan Cox:
  o sstfb oops fix

Andrew Morton:
  o Improper handling of %c in vsscanf

Geert Uytterhoeven:
  o Fix fs/inode.c warning if !HIGHMEM

Keith Owens:
  o [XFS] No need to have xfs in mod-subdirs in the fs/Makefile anymore

Paul Mackerras:
  o [PPC64] Two small fixes for hvc_console (the hypervisor virtual console)

smb_ops_unix warning (was: Re: Linux 2.4.25-rc2)

[Geert Uytterhoeven]

On Wed, 11 Feb 2004, Marcelo Tosatti wrote:
> Here goes -rc2, with small number of fixes and corrections.

This patch adds a missing include (fortunately it was included in some other
include, but its' not clean for files that check CONFIG_* flags) and kills a
compiler warning (introduced in -rc1, IIRC) if CONFIG_SMB_UNIX is not set.

Alternatively, perhaps the code can be reshuffled a bit to avoid too many
ifdefs?

--- linux-2.4.25-rc2/fs/smbfs/proc.c	2004-01-30 09:17:07.000000000 +0100
+++ linux-m68k-2.4.25-rc2/fs/smbfs/proc.c	2004-02-12 13:46:14.000000000 +0100
@@ -7,6 +7,7 @@
  *  Please add a note about your changes to smbfs in the ChangeLog file.
  */

+#include <linux/config.h>
 #include <linux/types.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
@@ -49,7 +50,9 @@
 static struct smb_ops smb_ops_os2;
 static struct smb_ops smb_ops_win95;
 static struct smb_ops smb_ops_winNT;
+#ifdef CONFIG_SMB_UNIX
 static struct smb_ops smb_ops_unix;
+#endif

 static void
 smb_init_dirent(struct smb_sb_info *server, struct smb_fattr *fattr);
@@ -2636,6 +2639,7 @@
 	return result;
 }

+#ifdef CONFIG_SMB_UNIX
 /*
  * Note: called with the server locked.
  */
@@ -2680,6 +2684,7 @@
 out:
 	return result;
 }
+#endif /* CONFIG_SMB_UNIX */

 static int
 smb_proc_getattr_null(struct smb_sb_info *server, struct dentry *dir,
@@ -3332,6 +3337,7 @@
 	.truncate       = smb_proc_trunc64,
 };

+#ifdef CONFIG_SMB_UNIX
 /* Samba w/ unix extensions. Others? */
 static struct smb_ops smb_ops_unix =
 {
@@ -3341,6 +3347,7 @@
 	.getattr        = smb_proc_getattr_unix,
 	.truncate       = smb_proc_trunc64,
 };
+#endif /* CONFIG_SMB_UNIX */

 /* Place holder until real ops are in place */
 static struct smb_ops smb_ops_null =

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: smb_ops_unix warning (was: Re: Linux 2.4.25-rc2)

[Urban Widmark]

On Thu, 12 Feb 2004, Geert Uytterhoeven wrote:

> On Wed, 11 Feb 2004, Marcelo Tosatti wrote:
> > Here goes -rc2, with small number of fixes and corrections.
> 
> This patch adds a missing include (fortunately it was included in some other
> include, but its' not clean for files that check CONFIG_* flags) and kills a
> compiler warning (introduced in -rc1, IIRC) if CONFIG_SMB_UNIX is not set.

Ah, thanks.


> Alternatively, perhaps the code can be reshuffled a bit to avoid too many
> ifdefs?

Code below builds cleanly for me either way, but I have not tested it yet.
Will do that later.

The config could also be removed, but I like that 2.4 users can keep the
previous 2.4 behaviour if they want to.

/Urban


diff -ur linux-2.4.25-rc2-orig/fs/smbfs/proc.c linux-2.4.25-rc2-smbfs/fs/smbfs/proc.c
--- linux-2.4.25-rc2-orig/fs/smbfs/proc.c	2004-02-12 17:52:37.000000000 +0100
+++ linux-2.4.25-rc2-smbfs/fs/smbfs/proc.c	2004-02-12 17:54:41.000000000 +0100
@@ -7,6 +7,7 @@
  *  Please add a note about your changes to smbfs in the ChangeLog file.
  */
 
+#include <linux/config.h>
 #include <linux/types.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
@@ -915,7 +916,9 @@
 		SB_of(server)->s_maxbytes = ~0ULL >> 1;
 		VERBOSE("LFS enabled\n");
 	}
-#ifdef CONFIG_SMB_UNIX
+#ifndef CONFIG_SMB_UNIX
+	server->opt.capabilities &= ~SMB_CAP_UNIX;
+#endif
 	if (server->opt.capabilities & SMB_CAP_UNIX) {
 		struct inode *inode;
 		VERBOSE("Using UNIX CIFS extensions\n");
@@ -924,9 +927,6 @@
 		if (inode)
 			inode->i_op = &smb_dir_inode_operations_unix;
 	}
-#else
-	server->opt.capabilities &= ~SMB_CAP_UNIX;
-#endif
 
 	VERBOSE("protocol=%d, max_xmit=%d, pid=%d capabilities=0x%x\n",
 		server->opt.protocol, server->opt.max_xmit, server->conn_pid,

Re: smb_ops_unix warning (was: Re: Linux 2.4.25-rc2)

[Geert Uytterhoeven]

On Thu, 12 Feb 2004, Urban Widmark wrote:
> On Thu, 12 Feb 2004, Geert Uytterhoeven wrote:
> > On Wed, 11 Feb 2004, Marcelo Tosatti wrote:
> > > Here goes -rc2, with small number of fixes and corrections.
> >
> > This patch adds a missing include (fortunately it was included in some other
> > include, but its' not clean for files that check CONFIG_* flags) and kills a
> > compiler warning (introduced in -rc1, IIRC) if CONFIG_SMB_UNIX is not set.
>
> Ah, thanks.
>
>
> > Alternatively, perhaps the code can be reshuffled a bit to avoid too many
> > ifdefs?
>
> Code below builds cleanly for me either way, but I have not tested it yet.
> Will do that later.
>
> The config could also be removed, but I like that 2.4 users can keep the
> previous 2.4 behaviour if they want to.

I can confirm it has the same result: no more compiler warnings.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds