Re: [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

Re: [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Marcos Paulo de Souza]

On Wed, Mar 29, 2017 at 09:31:14AM +0200, Greg KH wrote:
> On Tue, Mar 28, 2017 at 11:14:06PM -0300, Marcos Paulo de Souza wrote:
> > head_up parameter is marked with __user attribute, tmp is filled
> > by a copy_from_user from next, that is also marked as __user, so
> > tmp.next needs to be "casted" as __user to make sparse happy.
> 
> But is it the correct change?

I don't know, it's my first sparse patch, so I tried to fix this
warning.

> 
> You also have a typo in your subject :(

Sorry, didn't noticed yesterday :(

> 
> > Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
> > ---
> > 
> >  this is mt first patch addressing an issue of sparse, so let me know
> >  if I misunderstood the error message
> > 
> >  drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > index c6a683b..fb7ad74 100644
> > --- a/drivers/staging/lustre/lnet/selftest/conrpc.c
> > +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > @@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
> >  				   sizeof(struct list_head)))
> >  			return -EFAULT;
> >  
> > -		if (tmp.next == head_up)
> > +		if ((struct list_head __user *)tmp.next == head_up)
> 
> Aer you sure this is correct?  __user changes for lustre is not
> trivial...
> 
> How did you test this?

I didn't tested, it just removed the warning. Is this a false positive?

> 
> thanks,
> 
> greg k-h

-- 
Thanks,
	Marcos

Re: [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Marcos Paulo de Souza]

On Wed, Mar 29, 2017 at 12:34:05PM +0200, Greg KH wrote:
> On Thu, Mar 23, 2017 at 04:09:03PM -0300, Marcos Paulo de Souza wrote:
> > On Wed, Mar 29, 2017 at 09:31:14AM +0200, Greg KH wrote:
> > > On Tue, Mar 28, 2017 at 11:14:06PM -0300, Marcos Paulo de Souza wrote:
> > > > head_up parameter is marked with __user attribute, tmp is filled
> > > > by a copy_from_user from next, that is also marked as __user, so
> > > > tmp.next needs to be "casted" as __user to make sparse happy.
> > > 
> > > But is it the correct change?
> > 
> > I don't know, it's my first sparse patch, so I tried to fix this
> > warning.
> > 
> > > 
> > > You also have a typo in your subject :(
> > 
> > Sorry, didn't noticed yesterday :(
> > 
> > > 
> > > > Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
> > > > ---
> > > > 
> > > >  this is mt first patch addressing an issue of sparse, so let me know
> > > >  if I misunderstood the error message
> > > > 
> > > >  drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
> > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > 
> > > > diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > > index c6a683b..fb7ad74 100644
> > > > --- a/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > > +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > > @@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
> > > >  				   sizeof(struct list_head)))
> > > >  			return -EFAULT;
> > > >  
> > > > -		if (tmp.next == head_up)
> > > > +		if ((struct list_head __user *)tmp.next == head_up)
> > > 
> > > Aer you sure this is correct?  __user changes for lustre is not
> > > trivial...
> > > 
> > > How did you test this?
> > 
> > I didn't tested, it just removed the warning. Is this a false positive?
> 
> I don't know, it's up to you to prove to me that you know this change is
> correct.  You have to justify your changes, and "because checkpatch.pl
> complained" isn't a valid justification for something like this :)

Fair enough, I'll take in another sparse report to work on. Thanks!

> 
> thanks,
> 
> greg k-h

-- 
Thanks,
	Marcos

[PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Marcos Paulo de Souza]

head_up parameter is marked with __user attribute, tmp is filled
by a copy_from_user from next, that is also marked as __user, so
tmp.next needs to be "casted" as __user to make sparse happy.

Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
---

 this is mt first patch addressing an issue of sparse, so let me know
 if I misunderstood the error message

 drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
index c6a683b..fb7ad74 100644
--- a/drivers/staging/lustre/lnet/selftest/conrpc.c
+++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
@@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
 				   sizeof(struct list_head)))
 			return -EFAULT;
 
-		if (tmp.next == head_up)
+		if ((struct list_head __user *)tmp.next == head_up)
 			return 0;
 
 		next = tmp.next;
-- 
2.9.3

[lustre-devel] [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Greg KH]

On Tue, Mar 28, 2017 at 11:14:06PM -0300, Marcos Paulo de Souza wrote:
> head_up parameter is marked with __user attribute, tmp is filled
> by a copy_from_user from next, that is also marked as __user, so
> tmp.next needs to be "casted" as __user to make sparse happy.

But is it the correct change?

You also have a typo in your subject :(

> Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
> ---
> 
>  this is mt first patch addressing an issue of sparse, so let me know
>  if I misunderstood the error message
> 
>  drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
> index c6a683b..fb7ad74 100644
> --- a/drivers/staging/lustre/lnet/selftest/conrpc.c
> +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
> @@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
>  				   sizeof(struct list_head)))
>  			return -EFAULT;
>  
> -		if (tmp.next == head_up)
> +		if ((struct list_head __user *)tmp.next == head_up)

Aer you sure this is correct?  __user changes for lustre is not
trivial...

How did you test this?

thanks,

greg k-h

Re: [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Greg KH]

On Tue, Mar 28, 2017 at 11:14:06PM -0300, Marcos Paulo de Souza wrote:
> head_up parameter is marked with __user attribute, tmp is filled
> by a copy_from_user from next, that is also marked as __user, so
> tmp.next needs to be "casted" as __user to make sparse happy.

But is it the correct change?

You also have a typo in your subject :(

> Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
> ---
> 
>  this is mt first patch addressing an issue of sparse, so let me know
>  if I misunderstood the error message
> 
>  drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
> index c6a683b..fb7ad74 100644
> --- a/drivers/staging/lustre/lnet/selftest/conrpc.c
> +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
> @@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
>  				   sizeof(struct list_head)))
>  			return -EFAULT;
>  
> -		if (tmp.next == head_up)
> +		if ((struct list_head __user *)tmp.next == head_up)

Aer you sure this is correct?  __user changes for lustre is not
trivial...

How did you test this?

thanks,

greg k-h

[lustre-devel] [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Greg KH]

On Thu, Mar 23, 2017 at 04:09:03PM -0300, Marcos Paulo de Souza wrote:
> On Wed, Mar 29, 2017 at 09:31:14AM +0200, Greg KH wrote:
> > On Tue, Mar 28, 2017 at 11:14:06PM -0300, Marcos Paulo de Souza wrote:
> > > head_up parameter is marked with __user attribute, tmp is filled
> > > by a copy_from_user from next, that is also marked as __user, so
> > > tmp.next needs to be "casted" as __user to make sparse happy.
> > 
> > But is it the correct change?
> 
> I don't know, it's my first sparse patch, so I tried to fix this
> warning.
> 
> > 
> > You also have a typo in your subject :(
> 
> Sorry, didn't noticed yesterday :(
> 
> > 
> > > Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
> > > ---
> > > 
> > >  this is mt first patch addressing an issue of sparse, so let me know
> > >  if I misunderstood the error message
> > > 
> > >  drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > index c6a683b..fb7ad74 100644
> > > --- a/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > @@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
> > >  				   sizeof(struct list_head)))
> > >  			return -EFAULT;
> > >  
> > > -		if (tmp.next == head_up)
> > > +		if ((struct list_head __user *)tmp.next == head_up)
> > 
> > Aer you sure this is correct?  __user changes for lustre is not
> > trivial...
> > 
> > How did you test this?
> 
> I didn't tested, it just removed the warning. Is this a false positive?

I don't know, it's up to you to prove to me that you know this change is
correct.  You have to justify your changes, and "because checkpatch.pl
complained" isn't a valid justification for something like this :)

thanks,

greg k-h

Re: [PATCH] staging: lusten: conrpc.c: fix different address space sparse warning

[Greg KH]

On Thu, Mar 23, 2017 at 04:09:03PM -0300, Marcos Paulo de Souza wrote:
> On Wed, Mar 29, 2017 at 09:31:14AM +0200, Greg KH wrote:
> > On Tue, Mar 28, 2017 at 11:14:06PM -0300, Marcos Paulo de Souza wrote:
> > > head_up parameter is marked with __user attribute, tmp is filled
> > > by a copy_from_user from next, that is also marked as __user, so
> > > tmp.next needs to be "casted" as __user to make sparse happy.
> > 
> > But is it the correct change?
> 
> I don't know, it's my first sparse patch, so I tried to fix this
> warning.
> 
> > 
> > You also have a typo in your subject :(
> 
> Sorry, didn't noticed yesterday :(
> 
> > 
> > > Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
> > > ---
> > > 
> > >  this is mt first patch addressing an issue of sparse, so let me know
> > >  if I misunderstood the error message
> > > 
> > >  drivers/staging/lustre/lnet/selftest/conrpc.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > index c6a683b..fb7ad74 100644
> > > --- a/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c
> > > @@ -487,7 +487,7 @@ lstcon_rpc_trans_interpreter(struct lstcon_rpc_trans *trans,
> > >  				   sizeof(struct list_head)))
> > >  			return -EFAULT;
> > >  
> > > -		if (tmp.next == head_up)
> > > +		if ((struct list_head __user *)tmp.next == head_up)
> > 
> > Aer you sure this is correct?  __user changes for lustre is not
> > trivial...
> > 
> > How did you test this?
> 
> I didn't tested, it just removed the warning. Is this a false positive?

I don't know, it's up to you to prove to me that you know this change is
correct.  You have to justify your changes, and "because checkpatch.pl
complained" isn't a valid justification for something like this :)

thanks,

greg k-h