[PATCH] exec: Fix mem leak in kernel_read_file

[PATCH] exec: Fix mem leak in kernel_read_file

[Thibaut Sautereau]

Commit f612acfae86af7ecad754ae6a46019be9da05b8e upstream
("exec: Fix mem leak in kernel_read_file"), addressing CVE-2019-8980,
should be applied to 4.20, 4.19, 4.14 and 4.9 stable kernels.

Thanks,

-- 
Thibaut Sautereau
CLIP OS developer

Re: [PATCH] exec: Fix mem leak in kernel_read_file

[Greg KH]

On Mon, Mar 04, 2019 at 11:17:38PM +0100, Thibaut Sautereau wrote:
> Commit f612acfae86af7ecad754ae6a46019be9da05b8e upstream
> ("exec: Fix mem leak in kernel_read_file"), addressing CVE-2019-8980,
> should be applied to 4.20, 4.19, 4.14 and 4.9 stable kernels.

Any reason you didn't cc: the authors of that patch?

And as it _just_ went into Linus's tree today, give us a few weeks to
get it backported...

Also, it's just a "normal" syzbot fix, for a very rare case, why is this
a CVE?

thanks,

greg k-h

Re: [PATCH] exec: Fix mem leak in kernel_read_file

[Thibaut Sautereau]

On Tue, Mar 05, 2019 at 07:20:20AM +0100, Greg KH wrote:
> On Mon, Mar 04, 2019 at 11:17:38PM +0100, Thibaut Sautereau wrote:
> > Commit f612acfae86af7ecad754ae6a46019be9da05b8e upstream
> > ("exec: Fix mem leak in kernel_read_file"), addressing CVE-2019-8980,
> > should be applied to 4.20, 4.19, 4.14 and 4.9 stable kernels.
> 
> Any reason you didn't cc: the authors of that patch?

No, I just forgot, sorry for that.

> And as it _just_ went into Linus's tree today, give us a few weeks to
> get it backported...
> 
> Also, it's just a "normal" syzbot fix, for a very rare case, why is this
> a CVE?

I don't know (I'm not the one who requested a CVE), but I saw that this
patch had been backported in Arch Linux's kernels to address
CVE-2019-8980 [1] and that stable@kernel.org hadn't been put in Cc:.

As the fix was already waiting in a pull-request [2] from Al Viro, I
thought it was too late to notice the author about Ccing stable,
therefore I followed option 2 of
Documentation/process/stable-kernel-rules.rst to ensure it would not
fall through the cracks.

If that was the wrong way to do it, please tell me what I should have
done in this case.

Thanks,

[1] https://nvd.nist.gov/vuln/detail/CVE-2019-8980
[2] https://lkml.org/lkml/2019/3/2/230

-- 
Thibaut Sautereau

Re: [PATCH] exec: Fix mem leak in kernel_read_file

[Greg KH]

On Tue, Mar 05, 2019 at 10:00:39AM +0100, Thibaut Sautereau wrote:
> On Tue, Mar 05, 2019 at 07:20:20AM +0100, Greg KH wrote:
> > On Mon, Mar 04, 2019 at 11:17:38PM +0100, Thibaut Sautereau wrote:
> > > Commit f612acfae86af7ecad754ae6a46019be9da05b8e upstream
> > > ("exec: Fix mem leak in kernel_read_file"), addressing CVE-2019-8980,
> > > should be applied to 4.20, 4.19, 4.14 and 4.9 stable kernels.
> > 
> > Any reason you didn't cc: the authors of that patch?
> 
> No, I just forgot, sorry for that.
> 
> > And as it _just_ went into Linus's tree today, give us a few weeks to
> > get it backported...
> > 
> > Also, it's just a "normal" syzbot fix, for a very rare case, why is this
> > a CVE?
> 
> I don't know (I'm not the one who requested a CVE), but I saw that this
> patch had been backported in Arch Linux's kernels to address
> CVE-2019-8980 [1] and that stable@kernel.org hadn't been put in Cc:.
> 
> As the fix was already waiting in a pull-request [2] from Al Viro, I
> thought it was too late to notice the author about Ccing stable,
> therefore I followed option 2 of
> Documentation/process/stable-kernel-rules.rst to ensure it would not
> fall through the cracks.
> 
> If that was the wrong way to do it, please tell me what I should have
> done in this case.

No, this is fine, just next time you should cc: the developers as well.

Also, this needs to go to 5.0.y, now queued up.

greg k-h