[PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

>From b9b41b5a546ed0202c099f0d973da4df9aea314a Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 12 Oct 2009 22:44:40 +0200
Subject: [PATCH] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

This looks like a cut-and-paste job.
For example, compare this function to sys_execve in
arch/x86/kernel/process_64.c

and it is almost line by line the same, except the one in x86 nolonger has the
big kernel lock. All of the functions called between the lock are generic
and not specific to blackfin - thus, I believe it is safe to remove the
bkl here.

Signed-off-by: John Kacur <jkacur@redhat.com>
---
 arch/blackfin/kernel/process.c |    6 +-----
 1 files changed, 1 insertions(+), 5 deletions(-)

diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 430ae39..7d9c975 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -215,22 +215,18 @@ copy_thread(unsigned long clone_flags,
 /*
  * sys_execve() executes a new program.
  */
-
 asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __user * __user *envp)
 {
 	int error;
 	char *filename;
 	struct pt_regs *regs = (struct pt_regs *)((&name) + 6);
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
- out:
-	unlock_kernel();
 	return error;
 }
 
-- 
1.6.0.6

[PATCH 2/6 RFC] arch/frv/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

>From 4489496ad43049df06b7ffc37afcea561fdb52b9 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 12 Oct 2009 22:53:25 +0200
Subject: [PATCH] arch/frv/kernel/process.c: Remove the BKL from sys_execve

sys_execve for frv seems to be a copy-and-paste of sys_execve that
no longer requires the bkl. Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
---
 arch/frv/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/frv/kernel/process.c b/arch/frv/kernel/process.c
index 9042559..21d0fd1 100644
--- a/arch/frv/kernel/process.c
+++ b/arch/frv/kernel/process.c
@@ -255,15 +255,12 @@ asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __
 	int error;
 	char * filename;
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, __frame);
 	putname(filename);
- out:
-	unlock_kernel();
 	return error;
 }
 
-- 
1.6.0.6

[PATCH 3/6 RFC] arch/h83000/kernel/process.c: Remove BKL from sys_execve

[John Kacur]

>From 5cd96ce7851d01c1c28da724dd50713c8a1c66d7 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 12 Oct 2009 23:04:11 +0200
Subject: [PATCH] arch/h83000/kernel/process.c: Remove BKL from sys_execve

This looks like a copy-and-paste job for code that no-longer needs the BKL
Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
---
 arch/h8300/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c
index e2f33d0..bd883fa 100644
--- a/arch/h8300/kernel/process.c
+++ b/arch/h8300/kernel/process.c
@@ -218,15 +218,12 @@ asmlinkage int sys_execve(char *name, char **argv, char **envp,int dummy,...)
 	char * filename;
 	struct pt_regs *regs = (struct pt_regs *) ((unsigned char *)&dummy-4);
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
-out:
-	unlock_kernel();
 	return error;
 }
 
-- 
1.6.0.6

[PATCH 4/6 RFC] arch/m68k/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

>From 2dfc4e18d8968d00c0ad16c4c9e7ffd58809944b Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 12 Oct 2009 23:09:24 +0200
Subject: [PATCH] arch/m68k/kernel/process.c: Remove the BKL from sys_execve

This seems like a copy-and-paste from code that no-longer needs the BKL
Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
---
 arch/m68k/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c
index 41230c5..0529659 100644
--- a/arch/m68k/kernel/process.c
+++ b/arch/m68k/kernel/process.c
@@ -317,15 +317,12 @@ asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __
 	char * filename;
 	struct pt_regs *regs = (struct pt_regs *) &name;
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
-out:
-	unlock_kernel();
 	return error;
 }
 
-- 
1.6.0.6

[PATCH 5/6 RFC] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

>From e36c53d296132bc0ddbf6d9fb43ea5ea56dbd4a2 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 12 Oct 2009 23:37:28 +0200
Subject: [PATCH] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve

This looks like a copy-and-paste of functionality that no-longer needs the bkl.
Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
---
 arch/m68knommu/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c
index 8f8f4ab..5c9ecd4 100644
--- a/arch/m68knommu/kernel/process.c
+++ b/arch/m68knommu/kernel/process.c
@@ -352,15 +352,12 @@ asmlinkage int sys_execve(char *name, char **argv, char **envp)
 	char * filename;
 	struct pt_regs *regs = (struct pt_regs *) &name;
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
-out:
-	unlock_kernel();
 	return error;
 }
 
-- 
1.6.0.6

[PATCH 6/6 RFC] arch/mn10300/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

>From f05f8e3a2f5210e5274e758f66e33a9d168eacde Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 12 Oct 2009 23:41:55 +0200
Subject: [PATCH] arch/mn10300/kernel/process.c: Remove the BKL from sys_execve

This looks like a cut-and-paste from functionality that no-longer needs the bkl
Just remove it. Also, rewrite slightly so that it looks closer to sys_execve
on other architectures.

Signed-off-by: John Kacur <jkacur@redhat.com>
---
 arch/mn10300/kernel/process.c |   12 ++++--------
 1 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/arch/mn10300/kernel/process.c b/arch/mn10300/kernel/process.c
index 892cce8..ec8a21d 100644
--- a/arch/mn10300/kernel/process.c
+++ b/arch/mn10300/kernel/process.c
@@ -275,16 +275,12 @@ asmlinkage long sys_execve(char __user *name,
 	char *filename;
 	int error;
 
-	lock_kernel();
-
 	filename = getname(name);
 	error = PTR_ERR(filename);
-	if (!IS_ERR(filename)) {
-		error = do_execve(filename, argv, envp, __frame);
-		putname(filename);
-	}
-
-	unlock_kernel();
+	if (IS_ERR(filename))
+		return error;
+	error = do_execve(filename, argv, envp, __frame);
+	putname(filename);
 	return error;
 }
 
-- 
1.6.0.6

Re: [PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[Frederic Weisbecker]

On Tue, Oct 13, 2009 at 12:33:21AM +0200, John Kacur wrote:
> From b9b41b5a546ed0202c099f0d973da4df9aea314a Mon Sep 17 00:00:00 2001
> From: John Kacur <jkacur@redhat.com>
> Date: Mon, 12 Oct 2009 22:44:40 +0200
> Subject: [PATCH] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve
> 
> This looks like a cut-and-paste job.
> For example, compare this function to sys_execve in
> arch/x86/kernel/process_64.c
> 
> and it is almost line by line the same, except the one in x86 nolonger has the
> big kernel lock. All of the functions called between the lock are generic
> and not specific to blackfin - thus, I believe it is safe to remove the
> bkl here.
> 
> Signed-off-by: John Kacur <jkacur@redhat.com>
> ---
>  arch/blackfin/kernel/process.c |    6 +-----
>  1 files changed, 1 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
> index 430ae39..7d9c975 100644
> --- a/arch/blackfin/kernel/process.c
> +++ b/arch/blackfin/kernel/process.c
> @@ -215,22 +215,18 @@ copy_thread(unsigned long clone_flags,
>  /*
>   * sys_execve() executes a new program.
>   */
> -
>  asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __user * __user *envp)
>  {
>  	int error;
>  	char *filename;
>  	struct pt_regs *regs = (struct pt_regs *)((&name) + 6);
>  
> -	lock_kernel();
>  	filename = getname(name);
>  	error = PTR_ERR(filename);
>  	if (IS_ERR(filename))
> -		goto out;
> +		return error;
>  	error = do_execve(filename, argv, envp, regs);
>  	putname(filename);
> - out:
> -	unlock_kernel();
>  	return error;
>  }
>  
> -- 
> 1.6.0.6
> 
> 


Yeah, there seem to be nothing to protect there.

Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>

Re: [PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[Mike Frysinger]

On Mon, Oct 12, 2009 at 18:33, John Kacur wrote:
> This looks like a cut-and-paste job.
> For example, compare this function to sys_execve in
> arch/x86/kernel/process_64.c
>
> and it is almost line by line the same, except the one in x86 nolonger has the
> big kernel lock. All of the functions called between the lock are generic
> and not specific to blackfin - thus, I believe it is safe to remove the
> bkl here.

and it still boots/runs fine, so np here
Acked-by: Mike Frysinger <vapier@gentoo.org>
-mike

Re: [PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

On Tue, 13 Oct 2009, Mike Frysinger wrote:

> On Mon, Oct 12, 2009 at 18:33, John Kacur wrote:
> > This looks like a cut-and-paste job.
> > For example, compare this function to sys_execve in
> > arch/x86/kernel/process_64.c
> >
> > and it is almost line by line the same, except the one in x86 nolonger has the
> > big kernel lock. All of the functions called between the lock are generic
> > and not specific to blackfin - thus, I believe it is safe to remove the
> > bkl here.
> 
> and it still boots/runs fine, so np here
> Acked-by: Mike Frysinger <vapier@gentoo.org>
> -mike

Cool - are you testing on one of those nifty smp (or smp-like) blackfins?

Re: [PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[Mike Frysinger]

On Tue, Oct 13, 2009 at 04:27, John Kacur wrote:
> On Tue, 13 Oct 2009, Mike Frysinger wrote:
>> On Mon, Oct 12, 2009 at 18:33, John Kacur wrote:
>> > This looks like a cut-and-paste job.
>> > For example, compare this function to sys_execve in
>> > arch/x86/kernel/process_64.c
>> >
>> > and it is almost line by line the same, except the one in x86 nolonger has the
>> > big kernel lock. All of the functions called between the lock are generic
>> > and not specific to blackfin - thus, I believe it is safe to remove the
>> > bkl here.
>>
>> and it still boots/runs fine, so np here
>> Acked-by: Mike Frysinger <vapier@gentoo.org>
>
> Cool - are you testing on one of those nifty smp (or smp-like) blackfins?

no, this was a uniproc.  i could try a SMP Blackfin if you think it'll
make a difference.
-mike

Re: [PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

On Tue, 13 Oct 2009, Mike Frysinger wrote:

> On Tue, Oct 13, 2009 at 04:27, John Kacur wrote:
> > On Tue, 13 Oct 2009, Mike Frysinger wrote:
> >> On Mon, Oct 12, 2009 at 18:33, John Kacur wrote:
> >> > This looks like a cut-and-paste job.
> >> > For example, compare this function to sys_execve in
> >> > arch/x86/kernel/process_64.c
> >> >
> >> > and it is almost line by line the same, except the one in x86 nolonger has the
> >> > big kernel lock. All of the functions called between the lock are generic
> >> > and not specific to blackfin - thus, I believe it is safe to remove the
> >> > bkl here.
> >>
> >> and it still boots/runs fine, so np here
> >> Acked-by: Mike Frysinger <vapier@gentoo.org>
> >
> > Cool - are you testing on one of those nifty smp (or smp-like) blackfins?
> 
> no, this was a uniproc.  i could try a SMP Blackfin if you think it'll
> make a difference.
> -mike
> 

Well - hopefully it won't make a difference, but if there are problems, 
then that is where they will show up.

Thanks

Re: [PATCH 5/6 RFC] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve

[Greg Ungerer]

Hi John,

On 10/13/2009 08:34 AM, John Kacur wrote:
>> From e36c53d296132bc0ddbf6d9fb43ea5ea56dbd4a2 Mon Sep 17 00:00:00 2001
> From: John Kacur<jkacur@redhat.com>
> Date: Mon, 12 Oct 2009 23:37:28 +0200
> Subject: [PATCH] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve
>
> This looks like a copy-and-paste of functionality that no-longer needs the bkl.
> Just remove it.
>
> Signed-off-by: John Kacur<jkacur@redhat.com>

Tested and runs fine on m68knommu.

Acked-by: Greg Ungerer <gerg@uclinux.org>

Do you want me to push this into the m68knommu git tree,
for eventual merging into Linus' tree?  Or are you going
to push the whole set of patches together?

Regards
Greg



> ---
>   arch/m68knommu/kernel/process.c |    5 +----
>   1 files changed, 1 insertions(+), 4 deletions(-)
>
> diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c
> index 8f8f4ab..5c9ecd4 100644
> --- a/arch/m68knommu/kernel/process.c
> +++ b/arch/m68knommu/kernel/process.c
> @@ -352,15 +352,12 @@ asmlinkage int sys_execve(char *name, char **argv, char **envp)
>   	char * filename;
>   	struct pt_regs *regs = (struct pt_regs *)&name;
>
> -	lock_kernel();
>   	filename = getname(name);
>   	error = PTR_ERR(filename);
>   	if (IS_ERR(filename))
> -		goto out;
> +		return error;
>   	error = do_execve(filename, argv, envp, regs);
>   	putname(filename);
> -out:
> -	unlock_kernel();
>   	return error;
>   }
>

-- 
------------------------------------------------------------------------
Greg Ungerer  --  Principal Engineer        EMAIL:     gerg@snapgear.com
SnapGear Group, McAfee                      PHONE:       +61 7 3435 2888
825 Stanley St,                             FAX:         +61 7 3891 3630
Woolloongabba, QLD, 4102, Australia         WEB: http://www.SnapGear.com

Re: [PATCH 5/6 RFC] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve

[John Kacur]

On Tue, 13 Oct 2009, Greg Ungerer wrote:

> Hi John,
> 
> On 10/13/2009 08:34 AM, John Kacur wrote:
> > > From e36c53d296132bc0ddbf6d9fb43ea5ea56dbd4a2 Mon Sep 17 00:00:00 2001
> > From: John Kacur<jkacur@redhat.com>
> > Date: Mon, 12 Oct 2009 23:37:28 +0200
> > Subject: [PATCH] arch/m68knommu/kernel/process.c: Remove the BKL from
> > sys_execve
> > 
> > This looks like a copy-and-paste of functionality that no-longer needs the
> > bkl.
> > Just remove it.
> > 
> > Signed-off-by: John Kacur<jkacur@redhat.com>
> 
> Tested and runs fine on m68knommu.
> 
> Acked-by: Greg Ungerer <gerg@uclinux.org>
> 
> Do you want me to push this into the m68knommu git tree,
> for eventual merging into Linus' tree?  Or are you going
> to push the whole set of patches together?
> 
> Regards
> Greg
 
@Thomas - do you want to pick this up for your proposed kill bkl tree and
push it to Linus during this next merge window - or in this case is it 
more appropriate for Greg as the m68knommu maintainer to do so?

Thanks

John 

 
> > ---
> >   arch/m68knommu/kernel/process.c |    5 +----
> >   1 files changed, 1 insertions(+), 4 deletions(-)
> > 
> > diff --git a/arch/m68knommu/kernel/process.c
> > b/arch/m68knommu/kernel/process.c
> > index 8f8f4ab..5c9ecd4 100644
> > --- a/arch/m68knommu/kernel/process.c
> > +++ b/arch/m68knommu/kernel/process.c
> > @@ -352,15 +352,12 @@ asmlinkage int sys_execve(char *name, char **argv,
> > char **envp)
> >   	char * filename;
> >   	struct pt_regs *regs = (struct pt_regs *)&name;
> > 
> > -	lock_kernel();
> >   	filename = getname(name);
> >   	error = PTR_ERR(filename);
> >   	if (IS_ERR(filename))
> > -		goto out;
> > +		return error;
> >   	error = do_execve(filename, argv, envp, regs);
> >   	putname(filename);
> > -out:
> > -	unlock_kernel();
> >   	return error;
> >   }
> > 
> 
> -- 
> ------------------------------------------------------------------------
> Greg Ungerer  --  Principal Engineer        EMAIL:     gerg@snapgear.com
> SnapGear Group, McAfee                      PHONE:       +61 7 3435 2888
> 825 Stanley St,                             FAX:         +61 7 3891 3630
> Woolloongabba, QLD, 4102, Australia         WEB: http://www.SnapGear.com
> 

Re: [PATCH 2/6 RFC] arch/frv/kernel/process.c: Remove the BKL from sys_execve

[David Howells]

John Kacur <jkacur@redhat.com> wrote:

> From 4489496ad43049df06b7ffc37afcea561fdb52b9 Mon Sep 17 00:00:00 2001
> From: John Kacur <jkacur@redhat.com>
> Date: Mon, 12 Oct 2009 22:53:25 +0200
> Subject: [PATCH] arch/frv/kernel/process.c: Remove the BKL from sys_execve
> 
> sys_execve for frv seems to be a copy-and-paste of sys_execve that
> no longer requires the bkl. Just remove it.
> 
> Signed-off-by: John Kacur <jkacur@redhat.com>

Acked-by: David Howells <dhowells@redhat.com>

Re: [PATCH 6/6 RFC] arch/mn10300/kernel/process.c: Remove the BKL from sys_execve

[David Howells]

John Kacur <jkacur@redhat.com> wrote:

> From f05f8e3a2f5210e5274e758f66e33a9d168eacde Mon Sep 17 00:00:00 2001
> From: John Kacur <jkacur@redhat.com>
> Date: Mon, 12 Oct 2009 23:41:55 +0200
> Subject: [PATCH] arch/mn10300/kernel/process.c: Remove the BKL from sys_execve
> 
> This looks like a cut-and-paste from functionality that no-longer needs the bkl
> Just remove it. Also, rewrite slightly so that it looks closer to sys_execve
> on other architectures.
> 
> Signed-off-by: John Kacur <jkacur@redhat.com>

Acked-by: David Howells <dhowells@redhat.com>

Re: [PATCH 1/6 RFC] arch/blackfin/kernel/process.c: Remove the BKL from sys_execve

[Mike Frysinger]

On Tue, Oct 13, 2009 at 04:41, John Kacur wrote:
> On Tue, 13 Oct 2009, Mike Frysinger wrote:
>> On Tue, Oct 13, 2009 at 04:27, John Kacur wrote:
>> > On Tue, 13 Oct 2009, Mike Frysinger wrote:
>> >> On Mon, Oct 12, 2009 at 18:33, John Kacur wrote:
>> >> > This looks like a cut-and-paste job.
>> >> > For example, compare this function to sys_execve in
>> >> > arch/x86/kernel/process_64.c
>> >> >
>> >> > and it is almost line by line the same, except the one in x86 nolonger has the
>> >> > big kernel lock. All of the functions called between the lock are generic
>> >> > and not specific to blackfin - thus, I believe it is safe to remove the
>> >> > bkl here.
>> >>
>> >> and it still boots/runs fine, so np here
>> >> Acked-by: Mike Frysinger <vapier@gentoo.org>
>> >
>> > Cool - are you testing on one of those nifty smp (or smp-like) blackfins?
>>
>> no, this was a uniproc.  i could try a SMP Blackfin if you think it'll
>> make a difference.
>
> Well - hopefully it won't make a difference, but if there are problems,
> then that is where they will show up.

it seems that the Blackfin SMP code is currently broken during boot,
so i'll have to wait until that is fixed ;)

i would just queue this in the mean time since, as you said, the code
is generic (sounds like a case for moving sys_execve to common code)
and there shouldnt be any issues.
-mike

Re: [PATCH 5/6 RFC] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve

[Thomas Gleixner]

On Tue, 13 Oct 2009, Greg Ungerer wrote:

> Hi John,
> 
> On 10/13/2009 08:34 AM, John Kacur wrote:
> > > From e36c53d296132bc0ddbf6d9fb43ea5ea56dbd4a2 Mon Sep 17 00:00:00 2001
> > From: John Kacur<jkacur@redhat.com>
> > Date: Mon, 12 Oct 2009 23:37:28 +0200
> > Subject: [PATCH] arch/m68knommu/kernel/process.c: Remove the BKL from
> > sys_execve
> > 
> > This looks like a copy-and-paste of functionality that no-longer needs the
> > bkl.
> > Just remove it.
> > 
> > Signed-off-by: John Kacur<jkacur@redhat.com>
> 
> Tested and runs fine on m68knommu.
> 
> Acked-by: Greg Ungerer <gerg@uclinux.org>
> 
> Do you want me to push this into the m68knommu git tree,
> for eventual merging into Linus' tree?  Or are you going
> to push the whole set of patches together?

I'm collecting and pushing the BKL stuff, but if you want to take it
just let me know and I drop it.
 
Thanks,

	tglx

[tip:bkl/arch] blackfin: Remove the BKL from sys_execve

[tip-bot for John Kacur]

Commit-ID:  25708a5fe7467dcc69d9b92c1701aad4a0c71887
Gitweb:     http://git.kernel.org/tip/25708a5fe7467dcc69d9b92c1701aad4a0c71887
Author:     John Kacur <jkacur@redhat.com>
AuthorDate: Mon, 12 Oct 2009 22:44:40 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 14 Oct 2009 17:18:14 +0200

blackfin: Remove the BKL from sys_execve

This looks like a cut-and-paste job. For example, compare this
function to sys_execve in arch/x86/kernel/process_64.c and it is
almost line by line the same, except the one in x86 nolonger has the
big kernel lock. All of the functions called between the lock are
generic and not specific to blackfin - thus, I believe it is safe to
remove the bkl here.

Signed-off-by: John Kacur <jkacur@redhat.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
LKML-Reference: <alpine.LFD.2.00.0910130007240.3658@localhost.localdomain>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/blackfin/kernel/process.c |    6 +-----
 1 files changed, 1 insertions(+), 5 deletions(-)

diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 430ae39..7d9c975 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -215,22 +215,18 @@ copy_thread(unsigned long clone_flags,
 /*
  * sys_execve() executes a new program.
  */
-
 asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __user * __user *envp)
 {
 	int error;
 	char *filename;
 	struct pt_regs *regs = (struct pt_regs *)((&name) + 6);
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
- out:
-	unlock_kernel();
 	return error;
 }
 

[tip:bkl/arch] frv: Remove the BKL from sys_execve

[tip-bot for John Kacur]

Commit-ID:  b69975a35a7332624b6977a5676a6fc320070360
Gitweb:     http://git.kernel.org/tip/b69975a35a7332624b6977a5676a6fc320070360
Author:     John Kacur <jkacur@redhat.com>
AuthorDate: Mon, 12 Oct 2009 22:53:25 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 14 Oct 2009 17:18:14 +0200

frv: Remove the BKL from sys_execve

sys_execve for frv seems to be a copy-and-paste of sys_execve that no
longer requires the bkl. Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: David Howells <dhowells@redhat.com>
LKML-Reference: <alpine.LFD.2.00.0910130008320.3658@localhost.localdomain>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/frv/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/frv/kernel/process.c b/arch/frv/kernel/process.c
index 9042559..21d0fd1 100644
--- a/arch/frv/kernel/process.c
+++ b/arch/frv/kernel/process.c
@@ -255,15 +255,12 @@ asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __
 	int error;
 	char * filename;
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, __frame);
 	putname(filename);
- out:
-	unlock_kernel();
 	return error;
 }
 

[tip:bkl/arch] h83000: Remove BKL from sys_execve

[tip-bot for John Kacur]

Commit-ID:  e3c1a6b35c84795f7cb31cb7f4748166904cc0bc
Gitweb:     http://git.kernel.org/tip/e3c1a6b35c84795f7cb31cb7f4748166904cc0bc
Author:     John Kacur <jkacur@redhat.com>
AuthorDate: Mon, 12 Oct 2009 23:04:11 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 14 Oct 2009 17:18:14 +0200

h83000: Remove BKL from sys_execve

This looks like a copy-and-paste job for code that no-longer needs the
BKL Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
LKML-Reference: <alpine.LFD.2.00.0910130010000.3658@localhost.localdomain>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/h8300/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c
index e2f33d0..bd883fa 100644
--- a/arch/h8300/kernel/process.c
+++ b/arch/h8300/kernel/process.c
@@ -218,15 +218,12 @@ asmlinkage int sys_execve(char *name, char **argv, char **envp,int dummy,...)
 	char * filename;
 	struct pt_regs *regs = (struct pt_regs *) ((unsigned char *)&dummy-4);
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
-out:
-	unlock_kernel();
 	return error;
 }
 

[tip:bkl/arch] m68k: Remove the BKL from sys_execve

[tip-bot for John Kacur]

Commit-ID:  2624167cd8b4965d0ef548e84767a30214edd8d4
Gitweb:     http://git.kernel.org/tip/2624167cd8b4965d0ef548e84767a30214edd8d4
Author:     John Kacur <jkacur@redhat.com>
AuthorDate: Mon, 12 Oct 2009 23:09:24 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 14 Oct 2009 17:18:14 +0200

m68k: Remove the BKL from sys_execve

This seems like a copy-and-paste from code that no-longer needs the
BKL Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <alpine.LFD.2.00.0910130011180.3658@localhost.localdomain>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/m68k/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c
index 41230c5..0529659 100644
--- a/arch/m68k/kernel/process.c
+++ b/arch/m68k/kernel/process.c
@@ -317,15 +317,12 @@ asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __
 	char * filename;
 	struct pt_regs *regs = (struct pt_regs *) &name;
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
-out:
-	unlock_kernel();
 	return error;
 }
 

[tip:bkl/arch] m68knommu: Remove the BKL from sys_execve

[tip-bot for John Kacur]

Commit-ID:  eb7371d4fdcc40a390e8bdf90c99b5541213ca45
Gitweb:     http://git.kernel.org/tip/eb7371d4fdcc40a390e8bdf90c99b5541213ca45
Author:     John Kacur <jkacur@redhat.com>
AuthorDate: Mon, 12 Oct 2009 23:37:28 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 14 Oct 2009 17:18:14 +0200

m68knommu: Remove the BKL from sys_execve

This looks like a copy-and-paste of functionality that no-longer needs
the bkl.  Just remove it.

Signed-off-by: John Kacur <jkacur@redhat.com>
Reviewed: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Greg Ungerer <gerg@uclinux.org>
LKML-Reference: <alpine.LFD.2.00.0910130014520.3658@localhost.localdomain>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/m68knommu/kernel/process.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c
index 8f8f4ab..5c9ecd4 100644
--- a/arch/m68knommu/kernel/process.c
+++ b/arch/m68knommu/kernel/process.c
@@ -352,15 +352,12 @@ asmlinkage int sys_execve(char *name, char **argv, char **envp)
 	char * filename;
 	struct pt_regs *regs = (struct pt_regs *) &name;
 
-	lock_kernel();
 	filename = getname(name);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
-		goto out;
+		return error;
 	error = do_execve(filename, argv, envp, regs);
 	putname(filename);
-out:
-	unlock_kernel();
 	return error;
 }
 

[tip:bkl/arch] mn10300: Remove the BKL from sys_execve

[tip-bot for John Kacur]

Commit-ID:  8c0daee204f794d095ae301f408c5f9f40e4547d
Gitweb:     http://git.kernel.org/tip/8c0daee204f794d095ae301f408c5f9f40e4547d
Author:     John Kacur <jkacur@redhat.com>
AuthorDate: Mon, 12 Oct 2009 23:41:55 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 14 Oct 2009 17:18:15 +0200

mn10300: Remove the BKL from sys_execve

This looks like a cut-and-paste from functionality that no-longer
needs the bkl Just remove it. Also, rewrite slightly so that it looks
closer to sys_execve on other architectures.

Signed-off-by: John Kacur <jkacur@redhat.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: David Howells <dhowells@redhat.com>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
LKML-Reference: <alpine.LFD.2.00.0910130016540.3658@localhost.localdomain>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/mn10300/kernel/process.c |   12 ++++--------
 1 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/arch/mn10300/kernel/process.c b/arch/mn10300/kernel/process.c
index 892cce8..ec8a21d 100644
--- a/arch/mn10300/kernel/process.c
+++ b/arch/mn10300/kernel/process.c
@@ -275,16 +275,12 @@ asmlinkage long sys_execve(char __user *name,
 	char *filename;
 	int error;
 
-	lock_kernel();
-
 	filename = getname(name);
 	error = PTR_ERR(filename);
-	if (!IS_ERR(filename)) {
-		error = do_execve(filename, argv, envp, __frame);
-		putname(filename);
-	}
-
-	unlock_kernel();
+	if (IS_ERR(filename))
+		return error;
+	error = do_execve(filename, argv, envp, __frame);
+	putname(filename);
 	return error;
 }
 

Re: [PATCH 5/6 RFC] arch/m68knommu/kernel/process.c: Remove the BKL from sys_execve

[Greg Ungerer]

Hi Thomas,

Thomas Gleixner wrote:
> On Tue, 13 Oct 2009, Greg Ungerer wrote:
>> On 10/13/2009 08:34 AM, John Kacur wrote:
>>>> From e36c53d296132bc0ddbf6d9fb43ea5ea56dbd4a2 Mon Sep 17 00:00:00 2001
>>> From: John Kacur<jkacur@redhat.com>
>>> Date: Mon, 12 Oct 2009 23:37:28 +0200
>>> Subject: [PATCH] arch/m68knommu/kernel/process.c: Remove the BKL from
>>> sys_execve
>>>
>>> This looks like a copy-and-paste of functionality that no-longer needs the
>>> bkl.
>>> Just remove it.
>>>
>>> Signed-off-by: John Kacur<jkacur@redhat.com>
>> Tested and runs fine on m68knommu.
>>
>> Acked-by: Greg Ungerer <gerg@uclinux.org>
>>
>> Do you want me to push this into the m68knommu git tree,
>> for eventual merging into Linus' tree?  Or are you going
>> to push the whole set of patches together?
> 
> I'm collecting and pushing the BKL stuff, but if you want to take it
> just let me know and I drop it.

Would make more sense to keep it all together I think.
So you take it.

Regards
Greg


------------------------------------------------------------------------
Greg Ungerer  --  Principal Engineer        EMAIL:     gerg@snapgear.com
SnapGear Group, McAfee                      PHONE:       +61 7 3435 2888
825 Stanley St,                             FAX:         +61 7 3891 3630
Woolloongabba, QLD, 4102, Australia         WEB: http://www.SnapGear.com