Skip to contentSkip to conversation

Remove init_mm export as planned for 2.6.26

PATCH Dave Jones 3 participants 4 messages Jul 11, 2008 – Jul 14, 2008

Subsystems (2)
lore
Share & tools
lore

Dave Jones

author
#0 lore raw
Documentation/feature-removal-schedule.txt +0 -12
index 46ece3f..5e89689 100644
@@ -270,18 +270,6 @@ Who:	Michael Buesch <[email protected]>
 
 ---------------------------
 
-What:	init_mm export
-When:	2.6.26
-Why:	Not used in-tree. The current out-of-tree users used it to
-	work around problems in the CPA code which should be resolved
-	by now. One usecase was described to provide verification code
-	of the CPA operation. That's a good idea in general, but such
-	code / infrastructure should be in the kernel and not in some
-	out-of-tree driver.
-Who:	Thomas Gleixner <[email protected]>
-
-----------------------------
-
 What:	usedac i386 kernel parameter
 When:	2.6.27
 Why:	replaced by allowdac and no dac combination
arch/x86/kernel/init_task.c +0 -1
index a4f93b4..d399180 100644
@@ -14,7 +14,6 @@ static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
-EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */
 
 /*
  * Initial thread structure.
Signature · 1 line
http://www.codemonkey.org.uk
#1 lore raw
Quoted 31 lines · On Fri, 11 Jul 2008 16:19:59 -0400 Dave Jones <[email protected]> wrote:
Documentation/feature-removal-schedule.txt +0 -12
index 46ece3f..5e89689 100644
@@ -270,18 +270,6 @@ Who:	Michael Buesch <[email protected]>
 
 ---------------------------
 
-What:	init_mm export
-When:	2.6.26
-Why:	Not used in-tree. The current out-of-tree users used it to
-	work around problems in the CPA code which should be resolved
-	by now. One usecase was described to provide verification code
-	of the CPA operation. That's a good idea in general, but such
-	code / infrastructure should be in the kernel and not in some
-	out-of-tree driver.
-Who:	Thomas Gleixner <[email protected]>
-
-----------------------------
-
 What:	usedac i386 kernel parameter
 When:	2.6.27
 Why:	replaced by allowdac and no dac combination
arch/x86/kernel/init_task.c +0 -0
index a4f93b4..d399180 100644

arch/x86/kernel/init_task.c static struct fs_struct init_fs = INIT_FS; · line 19

Quoted 7 lines
arch/x86/kernel/init_task.c +0 -1
@@ -14,7 +14,6 @@ static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
-EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */
 
 /*
  * Initial thread structure.

grep -r init_mm include/asm-sparc64

This patch used to break stuff. I'm not aware that anything was done to prevent this.

#2 lore raw
Quoted 10 lines · Andrew Morton
Quoted 5 lines

-EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */

/*
* Initial thread structure.

grep -r init_mm include/asm-sparc64

This patch used to break stuff. I'm not aware that anything was done to prevent this.

Dave's patch only unexprts it on x86, which per defintion can't break sparc64 which is known to need this export. It would be nice if someone could do a little audit which architectures require it in addition to sparc64 and remove it fromm all others.

#3 lore raw
Quoted 7 lines · On Sun, 13 Jul 2008 22:37:09 -0400 Christoph Hellwig <[email protected]> wrote:
Quoted 6 lines · Andrew Morton
Quoted 5 lines

-EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */

/*
* Initial thread structure.

Quoted 7 lines
Quoted 4 lines

grep -r init_mm include/asm-sparc64

This patch used to break stuff. I'm not aware that anything was done to prevent this.

Dave's patch only unexprts it on x86, which per defintion can't break sparc64 which is known to need this export.

What, you want beauty _and_ reading comprehension?

Quoted 3 lines

It would be nice if someone
could do a little audit which architectures require it in addition to sparc64 and remove it fromm all others.

Yup. afaik sparc64 is the only one.