Check in Request for WIND00398740 : autofs_5.0.7 do_compile failed with the msg "#error Failed to locate modprobe

Submitted by yanjun.zhu on Jan. 17, 2013, 2:34 a.m.

Details

Message ID 50F76342.80505@windriver.com
State Accepted, archived
Headers show

Commit Message

yanjun.zhu Jan. 17, 2013, 2:34 a.m.
Hi, JoeM

GK && TK is OK.

Please check it this patch. The patch is in the attachment. Please check it.

Thanks a lot.
Zhu Yanjun

Patch hide | download patch | download mbox

From 4fff087eb52462e0817a8d22d6187f9dcd32f924 Mon Sep 17 00:00:00 2001
From: yanjun.zhu <yanjun.zhu@windriver.com>
Date: Tue, 15 Jan 2013 12:45:44 +0800
Subject: [PATCH 1/1] autofs:do not check for modprobe

CQID: WIND00398740

Reference:http://www.spinics.net/lists/autofs/msg00139.html

Loading autofs module is #ifdef'ed in the source, so
there is no need to check for /proc (which is only used
to load module) or modprobe.  Both modprobe and /proc
are always in the fixed location so there's no need to
check for these to start with.

Signed-off-by: yanjun.zhu <yanjun.zhu@windriver.com>
---
 .../autofs-5.0.7-do-not-check-for-modprobe.patch   |   66 ++++++++++++++++++++
 recipes-daemons/autofs/autofs_5.0.7.bb             |    1 +
 2 files changed, 67 insertions(+), 0 deletions(-)
 create mode 100644 recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch

diff --git a/recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch b/recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch
new file mode 100644
index 0000000..9f89b3c
--- /dev/null
+++ b/recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch
@@ -0,0 +1,66 @@ 
+diff -urpN a/configure.in b/configure.in
+--- a/configure.in	2013-01-15 11:30:22.000000000 +0800
++++ b/configure.in	2013-01-15 11:31:45.000000000 +0800
+@@ -34,11 +34,6 @@ AC_MSG_CHECKING([for binaries in])
+ AC_MSG_RESULT([$searchpath])
+ 
+ #
+-# Make sure we have "/proc"
+-#
+-AF_LINUX_PROCFS()
+-
+-#
+ # Location of init.d directory?
+ #
+ AF_INIT_D()
+@@ -142,7 +137,6 @@ AF_PATH_INCLUDE(UMOUNT, umount, /bin/umo
+ AF_PATH_INCLUDE(E2FSCK, fsck.ext2 e2fsck, , $searchpath)
+ AF_PATH_INCLUDE(E3FSCK, fsck.ext3 e3fsck, , $searchpath)
+ AF_PATH_INCLUDE(E4FSCK, fsck.ext4 e4fsck, , $searchpath)
+-AF_PATH_INCLUDE(MODPROBE, modprobe, , $searchpath)
+ 
+ AF_CHECK_PROG(LEX, flex lex, , $searchpath)
+ AF_CHECK_PROG(YACC, bison, , $searchpath)
+diff -urpN a/daemon/module.c b/daemon/module.c
+--- a/daemon/module.c	2013-01-15 11:30:49.000000000 +0800
++++ b/daemon/module.c	2013-01-15 11:32:00.000000000 +0800
+@@ -18,6 +18,8 @@
+ #include <stdlib.h>
+ #include "automount.h"
+ 
++#if 0
++/* see comment in daemon/automount.c around load_autofs4_module() call */
+ int load_autofs4_module(void)
+ {
+ 	FILE *fp;
+@@ -52,6 +54,7 @@ int load_autofs4_module(void)
+ 
+ 	return 1;
+ }
++#endif
+ 
+ struct lookup_mod *open_lookup(const char *name, const char *err_prefix,
+ 			       const char *mapfmt, int argc, const char *const *argv)
+diff -urpN a/include/automount.h b/include/automount.h
+--- a/include/automount.h	2013-01-15 11:31:10.000000000 +0800
++++ b/include/automount.h	2013-01-15 11:32:06.000000000 +0800
+@@ -50,16 +50,11 @@
+ #error Failed to locate umount(8)!
+ #endif
+ 
+-#ifndef HAVE_MODPROBE
+-#error Failed to locate modprobe(8)!
+-#endif
+-
+-#ifndef HAVE_LINUX_PROCFS
+-#error Failed to verify existence of procfs filesystem!
+-#endif
+-
++#if 0
++/* see comment in daemon/automount.c around load_autofs4_module() call */
+ #define FS_MODULE_NAME  "autofs4"
+ int load_autofs4_module(void);
++#endif
+ 
+ /* The -s (sloppy) option to mount is good, if we have it... */
+ 
diff --git a/recipes-daemons/autofs/autofs_5.0.7.bb b/recipes-daemons/autofs/autofs_5.0.7.bb
index ad737e3..bf679b4 100644
--- a/recipes-daemons/autofs/autofs_5.0.7.bb
+++ b/recipes-daemons/autofs/autofs_5.0.7.bb
@@ -30,6 +30,7 @@  SRC_URI = "${KERNELORG_MIRROR}/linux/daemons/autofs/v5/autofs-${PV}.tar.bz2 \
            file://cross.patch \
            file://libtirpc.patch \
            file://libtirpc-name-clash-backout.patch \
+           file://autofs-5.0.7-do-not-check-for-modprobe.patch \
           "
 
 SRC_URI[md5sum] = "bc46838dece83c02d800ff144ed9f431"
-- 
1.7.1


Comments

Martin Jansa Jan. 17, 2013, 12:50 p.m.
On Thu, Jan 17, 2013 at 10:34:42AM +0800, yzhu1 wrote:
> Hi, JoeM
> 
> GK && TK is OK.
> 
> Please check it this patch. The patch is in the attachment. Please check it.
> 
> Thanks a lot.
> Zhu Yanjun

It looks like missing meta-networking prefix in patch and subject tag.

Cheers,

> From 4fff087eb52462e0817a8d22d6187f9dcd32f924 Mon Sep 17 00:00:00 2001
> From: yanjun.zhu <yanjun.zhu@windriver.com>
> Date: Tue, 15 Jan 2013 12:45:44 +0800
> Subject: [PATCH 1/1] autofs:do not check for modprobe
> 
> CQID: WIND00398740
> 
> Reference:http://www.spinics.net/lists/autofs/msg00139.html
> 
> Loading autofs module is #ifdef'ed in the source, so
> there is no need to check for /proc (which is only used
> to load module) or modprobe.  Both modprobe and /proc
> are always in the fixed location so there's no need to
> check for these to start with.
> 
> Signed-off-by: yanjun.zhu <yanjun.zhu@windriver.com>
> ---
>  .../autofs-5.0.7-do-not-check-for-modprobe.patch   |   66 ++++++++++++++++++++
>  recipes-daemons/autofs/autofs_5.0.7.bb             |    1 +
>  2 files changed, 67 insertions(+), 0 deletions(-)
>  create mode 100644 recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch
> 
> diff --git a/recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch b/recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch
> new file mode 100644
> index 0000000..9f89b3c
> --- /dev/null
> +++ b/recipes-daemons/autofs/autofs-5.0.7/autofs-5.0.7-do-not-check-for-modprobe.patch
> @@ -0,0 +1,66 @@
> +diff -urpN a/configure.in b/configure.in
> +--- a/configure.in	2013-01-15 11:30:22.000000000 +0800
> ++++ b/configure.in	2013-01-15 11:31:45.000000000 +0800
> +@@ -34,11 +34,6 @@ AC_MSG_CHECKING([for binaries in])
> + AC_MSG_RESULT([$searchpath])
> + 
> + #
> +-# Make sure we have "/proc"
> +-#
> +-AF_LINUX_PROCFS()
> +-
> +-#
> + # Location of init.d directory?
> + #
> + AF_INIT_D()
> +@@ -142,7 +137,6 @@ AF_PATH_INCLUDE(UMOUNT, umount, /bin/umo
> + AF_PATH_INCLUDE(E2FSCK, fsck.ext2 e2fsck, , $searchpath)
> + AF_PATH_INCLUDE(E3FSCK, fsck.ext3 e3fsck, , $searchpath)
> + AF_PATH_INCLUDE(E4FSCK, fsck.ext4 e4fsck, , $searchpath)
> +-AF_PATH_INCLUDE(MODPROBE, modprobe, , $searchpath)
> + 
> + AF_CHECK_PROG(LEX, flex lex, , $searchpath)
> + AF_CHECK_PROG(YACC, bison, , $searchpath)
> +diff -urpN a/daemon/module.c b/daemon/module.c
> +--- a/daemon/module.c	2013-01-15 11:30:49.000000000 +0800
> ++++ b/daemon/module.c	2013-01-15 11:32:00.000000000 +0800
> +@@ -18,6 +18,8 @@
> + #include <stdlib.h>
> + #include "automount.h"
> + 
> ++#if 0
> ++/* see comment in daemon/automount.c around load_autofs4_module() call */
> + int load_autofs4_module(void)
> + {
> + 	FILE *fp;
> +@@ -52,6 +54,7 @@ int load_autofs4_module(void)
> + 
> + 	return 1;
> + }
> ++#endif
> + 
> + struct lookup_mod *open_lookup(const char *name, const char *err_prefix,
> + 			       const char *mapfmt, int argc, const char *const *argv)
> +diff -urpN a/include/automount.h b/include/automount.h
> +--- a/include/automount.h	2013-01-15 11:31:10.000000000 +0800
> ++++ b/include/automount.h	2013-01-15 11:32:06.000000000 +0800
> +@@ -50,16 +50,11 @@
> + #error Failed to locate umount(8)!
> + #endif
> + 
> +-#ifndef HAVE_MODPROBE
> +-#error Failed to locate modprobe(8)!
> +-#endif
> +-
> +-#ifndef HAVE_LINUX_PROCFS
> +-#error Failed to verify existence of procfs filesystem!
> +-#endif
> +-
> ++#if 0
> ++/* see comment in daemon/automount.c around load_autofs4_module() call */
> + #define FS_MODULE_NAME  "autofs4"
> + int load_autofs4_module(void);
> ++#endif
> + 
> + /* The -s (sloppy) option to mount is good, if we have it... */
> + 
> diff --git a/recipes-daemons/autofs/autofs_5.0.7.bb b/recipes-daemons/autofs/autofs_5.0.7.bb
> index ad737e3..bf679b4 100644
> --- a/recipes-daemons/autofs/autofs_5.0.7.bb
> +++ b/recipes-daemons/autofs/autofs_5.0.7.bb
> @@ -30,6 +30,7 @@ SRC_URI = "${KERNELORG_MIRROR}/linux/daemons/autofs/v5/autofs-${PV}.tar.bz2 \
>             file://cross.patch \
>             file://libtirpc.patch \
>             file://libtirpc-name-clash-backout.patch \
> +           file://autofs-5.0.7-do-not-check-for-modprobe.patch \
>            "
>  
>  SRC_URI[md5sum] = "bc46838dece83c02d800ff144ed9f431"
> -- 
> 1.7.1
> 

> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
Joe MacDonald Jan. 17, 2013, 2:04 p.m.
[Check in Request for WIND00398740 : autofs_5.0.7 do_compile failed with the msg "#error Failed to locate modprobe] On 13.01.17 (Thu 10:34) yzhu1 wrote:

> Hi, JoeM
> 
> GK && TK is OK.
> 
> Please check it this patch. The patch is in the attachment. Please check it.
> 
> Thanks a lot.
> Zhu Yanjun

Hi Yanjun,

There's a few hiccups with this one.

   - the patch appears to be against meta-networking, not
     meta-openembedded.  I can work with that, but nothing should be
     committed to meta-networking, that's purely my own side-project and
     everything going to it needs to come through commits in
     meta-openembedded.


   - the commit log includes a reference to an internal bug tracking
     ID, that won't mean anything to anyone outside of Wind River (or to
     most inside :-)).

   - you copied an internal WR mailing list on this, which I think ends
     up with bounces to anyone who replies and isn't inside WR.

   - you should try to ensure your patches follow the OE guidelines, in
     particular with an upstream-status line, not all patches in the
     tree have them, but it's nice if we can stay on top of them.

   - the submission didn't follow the standard guidelines for patches.

There's a recipe in the meta-networking layer (it's basically the same
as all the other meta-openembedded layers) that describes how to easily
send patches to the list in a way that ensures the machinery recognizes
it and that anyone can work with it.  It's here:

   meta-openembedded/
      meta-networking/
         MAINTAINERS

Something like this will help next time around:

   git send-email -1 \
        --to openembedded-devel@lists.openembedded.org \
        --subject-prefix=meta-networking][PATCH

Since this is a straight back-port of an upstream patch, I don't expect
it to be a big deal, so I've updated your commit log and the patch
header with the information from the link you quoted above.  Mainly just
FYI for the next time.

Thanks.