Re: Patch Upstream: module: Remove module size limit

Re: Patch Upstream: module: Remove module size limit

[Greg KH]

For the patch below, what are the "real world use cases" that would
cause this to be needed in the stable kernel trees?  Is there an
in-kernel module that is affected by this limitation?

Also, when was "a while ago" exactly?

thanks,

greg k-h


On Wed, Mar 28, 2012 at 06:01:21PM -0400, Gregs git-bot wrote:
> commit: f946eeb9313ff1470758e171a60fe7438a2ded3f
> From: Sasha Levin <levinsasha928@gmail.com>
> Date: Mon, 30 Jan 2012 23:07:22 -0500
> Subject: module: Remove module size limit
> 
> Module size was limited to 64MB, this was legacy limitation due to vmalloc()
> which was removed a while ago.
> 
> Limiting module size to 64MB is both pointless and affects real world use
> cases.
> 
> Cc: Tim Abbott <tim.abbott@oracle.com>
> Cc: stable@vger.kernel.org
> Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
> ---
>  kernel/module.c |    3 +--
>  1 files changed, 1 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/module.c b/kernel/module.c
> index 294692d..78ac6ec 100644
> --- a/kernel/module.c
> +++ b/kernel/module.c
> @@ -2411,8 +2411,7 @@ static int copy_and_check(struct load_info *info,
>  		return -ENOEXEC;
>  
>  	/* Suck in entire file: we'll want most of it. */
> -	/* vmalloc barfs on "unusual" numbers.  Check here */
> -	if (len > 64 * 1024 * 1024 || (hdr = vmalloc(len)) == NULL)
> +	if ((hdr = vmalloc(len)) == NULL)
>  		return -ENOMEM;
>  
>  	if (copy_from_user(hdr, umod, len) != 0) {
> -- 
> 1.7.3.4

Re: Patch Upstream: module: Remove module size limit

[Sasha Levin]

On Thu, Mar 29, 2012 at 12:41 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> For the patch below, what are the "real world use cases" that would
> cause this to be needed in the stable kernel trees?  Is there an
> in-kernel module that is affected by this limitation?

There are no in-kernel modules which are affected, the "real world"
scenario is debug modules generated by Ksplice.

> Also, when was "a while ago" exactly?

Uncertain, but according to my tests for this patch vmalloc is fine
with allocating large blocks on a 2.6.18 kernel.

> thanks,
>
> greg k-h

Thanks.

Re: Patch Upstream: module: Remove module size limit

[Greg KH]

On Thu, Mar 29, 2012 at 12:57:47AM +0200, Sasha Levin wrote:
> On Thu, Mar 29, 2012 at 12:41 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> > For the patch below, what are the "real world use cases" that would
> > cause this to be needed in the stable kernel trees? �Is there an
> > in-kernel module that is affected by this limitation?
> 
> There are no in-kernel modules which are affected, the "real world"
> scenario is debug modules generated by Ksplice.

So why is this needed for the stable kernels?  This really looks like a
feature not a bugfix to me, how about you?

greg k-h

Re: Patch Upstream: module: Remove module size limit

[Sasha Levin]

On Thu, Mar 29, 2012 at 1:09 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> On Thu, Mar 29, 2012 at 12:57:47AM +0200, Sasha Levin wrote:
>> On Thu, Mar 29, 2012 at 12:41 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
>> > For the patch below, what are the "real world use cases" that would
>> > cause this to be needed in the stable kernel trees?  Is there an
>> > in-kernel module that is affected by this limitation?
>>
>> There are no in-kernel modules which are affected, the "real world"
>> scenario is debug modules generated by Ksplice.
>
> So why is this needed for the stable kernels?  This really looks like a
> feature not a bugfix to me, how about you?

I guess it depends on whether you consider the inability to load
modules bigger than 64mb as a bug or not.

Re: Patch Upstream: module: Remove module size limit

[Greg KH]

On Thu, Mar 29, 2012 at 01:22:23AM +0200, Sasha Levin wrote:
> On Thu, Mar 29, 2012 at 1:09 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> > On Thu, Mar 29, 2012 at 12:57:47AM +0200, Sasha Levin wrote:
> >> On Thu, Mar 29, 2012 at 12:41 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> >> > For the patch below, what are the "real world use cases" that would
> >> > cause this to be needed in the stable kernel trees? �Is there an
> >> > in-kernel module that is affected by this limitation?
> >>
> >> There are no in-kernel modules which are affected, the "real world"
> >> scenario is debug modules generated by Ksplice.
> >
> > So why is this needed for the stable kernels? �This really looks like a
> > feature not a bugfix to me, how about you?
> 
> I guess it depends on whether you consider the inability to load
> modules bigger than 64mb as a bug or not.

As there is no in-kernel user with this problem, no, I don't think it's
a bug that meets the rules documented in
Documentation/stable_kernel_rules.txt, do you?

So I'll drop it, sorry.

greg k-h

Re: Patch Upstream: module: Remove module size limit

[Rusty Russell]

On Wed, 28 Mar 2012 16:32:52 -0700, Greg KH <gregkh@linuxfoundation.org> wrote:
> On Thu, Mar 29, 2012 at 01:22:23AM +0200, Sasha Levin wrote:
> > On Thu, Mar 29, 2012 at 1:09 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> > > On Thu, Mar 29, 2012 at 12:57:47AM +0200, Sasha Levin wrote:
> > >> On Thu, Mar 29, 2012 at 12:41 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> > >> > For the patch below, what are the "real world use cases" that would
> > >> > cause this to be needed in the stable kernel trees?  Is there an
> > >> > in-kernel module that is affected by this limitation?
> > >>
> > >> There are no in-kernel modules which are affected, the "real world"
> > >> scenario is debug modules generated by Ksplice.
> > >
> > > So why is this needed for the stable kernels?  This really looks like a
> > > feature not a bugfix to me, how about you?
> > 
> > I guess it depends on whether you consider the inability to load
> > modules bigger than 64mb as a bug or not.
> 
> As there is no in-kernel user with this problem, no, I don't think it's
> a bug that meets the rules documented in
> Documentation/stable_kernel_rules.txt, do you?

Your call, but it has been my policy to support ksplice, such as making
sure we export symbols it needs.  And this really is a low-risk change.

Thanks,
Rusty.
-- 
  How could I marry someone with more hair than me?  http://baldalex.org

Re: Patch Upstream: module: Remove module size limit

[Greg KH]

On Thu, Mar 29, 2012 at 03:25:20PM +1030, Rusty Russell wrote:
> On Wed, 28 Mar 2012 16:32:52 -0700, Greg KH <gregkh@linuxfoundation.org> wrote:
> > On Thu, Mar 29, 2012 at 01:22:23AM +0200, Sasha Levin wrote:
> > > On Thu, Mar 29, 2012 at 1:09 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> > > > On Thu, Mar 29, 2012 at 12:57:47AM +0200, Sasha Levin wrote:
> > > >> On Thu, Mar 29, 2012 at 12:41 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> > > >> > For the patch below, what are the "real world use cases" that would
> > > >> > cause this to be needed in the stable kernel trees? �Is there an
> > > >> > in-kernel module that is affected by this limitation?
> > > >>
> > > >> There are no in-kernel modules which are affected, the "real world"
> > > >> scenario is debug modules generated by Ksplice.
> > > >
> > > > So why is this needed for the stable kernels? �This really looks like a
> > > > feature not a bugfix to me, how about you?
> > > 
> > > I guess it depends on whether you consider the inability to load
> > > modules bigger than 64mb as a bug or not.
> > 
> > As there is no in-kernel user with this problem, no, I don't think it's
> > a bug that meets the rules documented in
> > Documentation/stable_kernel_rules.txt, do you?
> 
> Your call, but it has been my policy to support ksplice, such as making
> sure we export symbols it needs.  And this really is a low-risk change.

Ok, but note, it really would be good if we could get the ksplice code
into the kernel tree, I hate supporting out-of-tree modules...

thanks,

greg k-h