Patchwork [v3] tpm-dev-common: Reject too short writes

login
register
mail settings
Submitter Alexander Steffen
Date Sept. 8, 2017, 3:21 p.m.
Message ID <20170908152132.8728-1-Alexander.Steffen@infineon.com>
Download mbox | patch
Permalink /patch/334829/
State New
Headers show

Comments

Alexander Steffen - Sept. 8, 2017, 3:21 p.m.
tpm_transmit() does not offer an explicit interface to indicate the number
of valid bytes in the communication buffer. Instead, it relies on the
commandSize field in the TPM header that is encoded within the buffer.
Therefore, ensure that a) enough data has been written to the buffer, so
that the commandSize field is present and b) the commandSize field does not
announce more data than has been written to the buffer.

This should have been fixed with CVE-2011-1161 long ago, but apparently
a correct version of that patch never made it into the kernel.

Cc: stable@vger.kernel.org
Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com>
---
v2:
- Moved all changes to tpm_common_write in a single patch.
v3:
- Access data copied from user space (priv->data_buffer) instead of user
  space data directly (buf).
- Changed return code to EINVAL.

 drivers/char/tpm/tpm-dev-common.c | 6 ++++++
 1 file changed, 6 insertions(+)
Jarkko Sakkinen - Sept. 8, 2017, 9:37 p.m.
On Fri, Sep 08, 2017 at 05:21:32PM +0200, Alexander Steffen wrote:
> tpm_transmit() does not offer an explicit interface to indicate the number
> of valid bytes in the communication buffer. Instead, it relies on the
> commandSize field in the TPM header that is encoded within the buffer.
> Therefore, ensure that a) enough data has been written to the buffer, so
> that the commandSize field is present and b) the commandSize field does not
> announce more data than has been written to the buffer.
> 
> This should have been fixed with CVE-2011-1161 long ago, but apparently
> a correct version of that patch never made it into the kernel.
> 
> Cc: stable@vger.kernel.org
> Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com>
> ---
> v2:
> - Moved all changes to tpm_common_write in a single patch.
> v3:
> - Access data copied from user space (priv->data_buffer) instead of user
>   space data directly (buf).
> - Changed return code to EINVAL.
> 
>  drivers/char/tpm/tpm-dev-common.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/char/tpm/tpm-dev-common.c b/drivers/char/tpm/tpm-dev-common.c
> index 610638a..461bf0b 100644
> --- a/drivers/char/tpm/tpm-dev-common.c
> +++ b/drivers/char/tpm/tpm-dev-common.c
> @@ -110,6 +110,12 @@ ssize_t tpm_common_write(struct file *file, const char __user *buf,
>  		return -EFAULT;
>  	}
>  
> +	if (in_size < 6 ||
> +	    in_size < be32_to_cpu(*((__be32 *) (priv->data_buffer + 2)))) {
> +		mutex_unlock(&priv->buffer_mutex);
> +		return -EINVAL;
> +	}
> +
>  	/* atomic tpm command send and result receive. We only hold the ops
>  	 * lock during this period so that the tpm can be unregistered even if
>  	 * the char dev is held open.
> -- 
> 2.7.4
> 

I'm not gonna fight about that "in_size < 6" check. I think it is not
needed, I understand your point but still disagree but it is something
where I can live with having it.

I kind of disagree also with allowing messages longer than the command
size but it does not have to be in the scope of this commit and actually
should be a separate discussion if we ever going to do something about
it.

Thanks for the patience!

Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>

/Jarkko

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
Jarkko Sakkinen - Sept. 9, 2017, 12:53 p.m.
On Sat, Sep 09, 2017 at 12:37:39AM +0300, Jarkko Sakkinen wrote:
> On Fri, Sep 08, 2017 at 05:21:32PM +0200, Alexander Steffen wrote:
> > tpm_transmit() does not offer an explicit interface to indicate the number
> > of valid bytes in the communication buffer. Instead, it relies on the
> > commandSize field in the TPM header that is encoded within the buffer.
> > Therefore, ensure that a) enough data has been written to the buffer, so
> > that the commandSize field is present and b) the commandSize field does not
> > announce more data than has been written to the buffer.
> > 
> > This should have been fixed with CVE-2011-1161 long ago, but apparently
> > a correct version of that patch never made it into the kernel.
> > 
> > Cc: stable@vger.kernel.org
> > Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com>
> > ---
> > v2:
> > - Moved all changes to tpm_common_write in a single patch.
> > v3:
> > - Access data copied from user space (priv->data_buffer) instead of user
> >   space data directly (buf).
> > - Changed return code to EINVAL.
> > 
> >  drivers/char/tpm/tpm-dev-common.c | 6 ++++++
> >  1 file changed, 6 insertions(+)
> > 
> > diff --git a/drivers/char/tpm/tpm-dev-common.c b/drivers/char/tpm/tpm-dev-common.c
> > index 610638a..461bf0b 100644
> > --- a/drivers/char/tpm/tpm-dev-common.c
> > +++ b/drivers/char/tpm/tpm-dev-common.c
> > @@ -110,6 +110,12 @@ ssize_t tpm_common_write(struct file *file, const char __user *buf,
> >  		return -EFAULT;
> >  	}
> >  
> > +	if (in_size < 6 ||
> > +	    in_size < be32_to_cpu(*((__be32 *) (priv->data_buffer + 2)))) {
> > +		mutex_unlock(&priv->buffer_mutex);
> > +		return -EINVAL;
> > +	}
> > +
> >  	/* atomic tpm command send and result receive. We only hold the ops
> >  	 * lock during this period so that the tpm can be unregistered even if
> >  	 * the char dev is held open.
> > -- 
> > 2.7.4
> > 
> 
> I'm not gonna fight about that "in_size < 6" check. I think it is not
> needed, I understand your point but still disagree but it is something
> where I can live with having it.
> 
> I kind of disagree also with allowing messages longer than the command
> size but it does not have to be in the scope of this commit and actually
> should be a separate discussion if we ever going to do something about
> it.
> 
> Thanks for the patience!
> 
> Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> 
> /Jarkko

Without your fix:

$ python -m unittest -v tpm2_smoke.SmokeTest.test_too_short_cmd
test_too_short_cmd (tpm2_smoke.SmokeTest) ... FAIL

======================================================================
FAIL: test_too_short_cmd (tpm2_smoke.SmokeTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "tpm2_smoke.py", line 157, in test_too_short_cmd
    self.assertEqual(rejected, True)
AssertionError: False != True

----------------------------------------------------------------------
Ran 1 test in 2.108s

FAILED (failures=1)

The test case expects to get a posix error, which it doesn't get.

With your fix:

$ python -m unittest -v tpm2_smoke.SmokeTest.test_too_short_cmd
test_too_short_cmd (tpm2_smoke.SmokeTest) ... ok

----------------------------------------------------------------------
Ran 1 test in 2.099s

OK

So looks good to me.

Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>

Can you test the master branch with SPI TPM? I had to tinker your
commits a bit because of merge conflicts with Arnd's commit. I'll
put everything back to next as soon as I hear from you. Thanks

/Jarkko

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
Alexander Steffen - Sept. 11, 2017, 10:24 a.m.
> On Sat, Sep 09, 2017 at 12:37:39AM +0300, Jarkko Sakkinen wrote:
> > On Fri, Sep 08, 2017 at 05:21:32PM +0200, Alexander Steffen wrote:
> > > tpm_transmit() does not offer an explicit interface to indicate the
> number
> > > of valid bytes in the communication buffer. Instead, it relies on the
> > > commandSize field in the TPM header that is encoded within the buffer.
> > > Therefore, ensure that a) enough data has been written to the buffer, so
> > > that the commandSize field is present and b) the commandSize field does
> not
> > > announce more data than has been written to the buffer.
> > >
> > > This should have been fixed with CVE-2011-1161 long ago, but apparently
> > > a correct version of that patch never made it into the kernel.
> > >
> > > Cc: stable@vger.kernel.org
> > > Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com>
> > > ---
> > > v2:
> > > - Moved all changes to tpm_common_write in a single patch.
> > > v3:
> > > - Access data copied from user space (priv->data_buffer) instead of user
> > >   space data directly (buf).
> > > - Changed return code to EINVAL.
> > >
> > >  drivers/char/tpm/tpm-dev-common.c | 6 ++++++
> > >  1 file changed, 6 insertions(+)
> > >
> > > diff --git a/drivers/char/tpm/tpm-dev-common.c
> b/drivers/char/tpm/tpm-dev-common.c
> > > index 610638a..461bf0b 100644
> > > --- a/drivers/char/tpm/tpm-dev-common.c
> > > +++ b/drivers/char/tpm/tpm-dev-common.c
> > > @@ -110,6 +110,12 @@ ssize_t tpm_common_write(struct file *file,
> const char __user *buf,
> > >  		return -EFAULT;
> > >  	}
> > >
> > > +	if (in_size < 6 ||
> > > +	    in_size < be32_to_cpu(*((__be32 *) (priv->data_buffer + 2)))) {
> > > +		mutex_unlock(&priv->buffer_mutex);
> > > +		return -EINVAL;
> > > +	}
> > > +
> > >  	/* atomic tpm command send and result receive. We only hold the
> ops
> > >  	 * lock during this period so that the tpm can be unregistered even if
> > >  	 * the char dev is held open.
> > > --
> > > 2.7.4
> > >
> >
> > I'm not gonna fight about that "in_size < 6" check. I think it is not
> > needed, I understand your point but still disagree but it is something
> > where I can live with having it.
> >
> > I kind of disagree also with allowing messages longer than the command
> > size but it does not have to be in the scope of this commit and actually
> > should be a separate discussion if we ever going to do something about
> > it.
> >
> > Thanks for the patience!
> >
> > Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> >
> > /Jarkko
> 
> Without your fix:
> 
> $ python -m unittest -v tpm2_smoke.SmokeTest.test_too_short_cmd
> test_too_short_cmd (tpm2_smoke.SmokeTest) ... FAIL
> 
> ==========================================================
> ============
> FAIL: test_too_short_cmd (tpm2_smoke.SmokeTest)
> ----------------------------------------------------------------------
> Traceback (most recent call last):
>   File "tpm2_smoke.py", line 157, in test_too_short_cmd
>     self.assertEqual(rejected, True)
> AssertionError: False != True
> 
> ----------------------------------------------------------------------
> Ran 1 test in 2.108s
> 
> FAILED (failures=1)
> 
> The test case expects to get a posix error, which it doesn't get.
> 
> With your fix:
> 
> $ python -m unittest -v tpm2_smoke.SmokeTest.test_too_short_cmd
> test_too_short_cmd (tpm2_smoke.SmokeTest) ... ok
> 
> ----------------------------------------------------------------------
> Ran 1 test in 2.099s
> 
> OK
> 
> So looks good to me.
> 
> Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> 
> Can you test the master branch with SPI TPM? I had to tinker your
> commits a bit because of merge conflicts with Arnd's commit. I'll
> put everything back to next as soon as I hear from you. Thanks
> 
> /Jarkko

tpm_tis_spi in master (3897f7c) is broken, it does not transfer any data. I'll send you a fixed patch for the DMA change.

Alexander
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

Patch

diff --git a/drivers/char/tpm/tpm-dev-common.c b/drivers/char/tpm/tpm-dev-common.c
index 610638a..461bf0b 100644
--- a/drivers/char/tpm/tpm-dev-common.c
+++ b/drivers/char/tpm/tpm-dev-common.c
@@ -110,6 +110,12 @@  ssize_t tpm_common_write(struct file *file, const char __user *buf,
 		return -EFAULT;
 	}
 
+	if (in_size < 6 ||
+	    in_size < be32_to_cpu(*((__be32 *) (priv->data_buffer + 2)))) {
+		mutex_unlock(&priv->buffer_mutex);
+		return -EINVAL;
+	}
+
 	/* atomic tpm command send and result receive. We only hold the ops
 	 * lock during this period so that the tpm can be unregistered even if
 	 * the char dev is held open.