Skip to contentSkip to conversation

Compile fix for 2.6.0-test5-mm4 in net/atm/proc.c

PATCH In review Remi Colinet 3 participants 4 messages Sep 22, 2003 – Sep 23, 2003

Subsystems (2)
Networking ATM
lore
Share & tools
lore

Remi Colinet

author
#0 lore raw

PATCH 1/1 Compile fix for 2.6.0-test5-mm4 in net/atm/proc.c

Hi,

net/atm/proc.c 2003-09-22 21:44:24.000000000 +0200 renamed from net/atm/proc.c 2003-09-22 21:42:09.000000000 +0200 +2 -0
@@ -192,7 +192,9 @@
               goto out_kfree;
 
       state->family = family;
+#if defined(CONFIG_ATM_CLIP) || defined(CONFIG_ATM_CLIP_MODULE)
       state->clip_info = try_atm_clip_ops();
+#endif
 
       seq = file->private_data;
       seq->private = state;

This patch fixes the following compile error :

 CC      arch/i386/lib/usercopy.o
 AR      arch/i386/lib/lib.a
 GEN     .version
 CHK     include/linux/compile.h
 UPD     include/linux/compile.h
 CC      init/version.o
 LD      init/built-in.o
 LD      .tmp_vmlinux1

net/built-in.o: In function `__vcc_seq_open':
/usr/src/mm/net/atm/proc.c:195: undefined reference to `try_atm_clip_ops'
make: *** [.tmp_vmlinux1] Error 1

Remi

#1 lore raw

Review

Quoted 6 lines · In message <[email protected]>,Remi Colinet writes:
       state->family = family;
(quoted diff) +2 -0
+#if defined(CONFIG_ATM_CLIP) || defined(CONFIG_ATM_CLIP_MODULE)
       state->clip_info = try_atm_clip_ops();
+#endif

instead of doing this, it would probably be cleaner to put the ifdef inside the clip header file and just return 0 when !CLIP.

dave, can you apply this to 2.6?

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas:
# ChangeSet 1.1446 -> 1.1447
# net/atm/proc.c 1.29 -> 1.30
# include/net/atmclip.h 1.2 -> 1.3
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/09/23 [email protected] 1.1447
# [ATM]: if !CLIP try_atm_clip_ops() should always fail
# --------------------------------------------
#

include/net/atmclip.h +8 -0
diff -Nru a/include/net/atmclip.h b/include/net/atmclip.h
@@ -67,7 +67,15 @@
 };
 
 void atm_clip_ops_set(struct atm_clip_ops *);
+#if defined(CONFIG_ATM_CLIP) || defined(CONFIG_ATM_CLIP_MODULE)
 int try_atm_clip_ops(void);
+#else
+static inline int try_atm_clip_ops(void)
+{
+	return 0;
+}
+#endif
+
 
 extern struct neigh_table *clip_tbl_hook;
 extern struct atm_clip_ops *atm_clip_ops;
net/atm/proc.c +0 -2
diff -Nru a/net/atm/proc.c b/net/atm/proc.c
@@ -32,10 +32,8 @@
 #include "common.h" /* atm_proc_init prototype */
 #include "signaling.h" /* to get sigd - ugly too */
 
-#if defined(CONFIG_ATM_CLIP) || defined(CONFIG_ATM_CLIP_MODULE)
 #include <net/atmclip.h>
 #include "ipcommon.h"
-#endif
 
 #if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE)
 #include "lec.h"
#2 lore raw
Quoted 2 lines · chas williams

instead of doing this, it would probably be cleaner to put the ifdef inside the clip header file and just return 0 when !CLIP.

What are your plans with mine and Mitchell's stuff that removes all this ops crap altogether ?

The patch series works, as far as I know.

regards
john

Signature · 2 lines
Khendon's Law:
If the same point is made twice by the same person, the thread is over.
#3 lore raw
Quoted 2 lines · In message <[email protected]>,John Levon writes:

What are your plans with mine and Mitchell's stuff that removes all this ops crap altogether ?

working on testing it. sorry but i havent had much time with the end of the gov't fiscal year coming up. in the meantime this short little patch should solve things in the meantime.