Re: [PATCH v2 2/8] ima: Switch to ima_hash_algo for boot aggregate

[Mimi Zohar]

On Thu, 2020-02-06 at 12:28 +0000, Roberto Sassu wrote:
> > -----Original Message-----
> > From: Mimi Zohar [mailto:zohar@xxxxxxxxxxxxx]
> > Sent: Thursday, February 6, 2020 1:17 PM
> > To: Roberto Sassu <roberto.sassu@xxxxxxxxxx>;
> > James.Bottomley@xxxxxxxxxxxxxxxxxxxxx;
> > jarkko.sakkinen@xxxxxxxxxxxxxxx
> > Cc: linux-integrity@xxxxxxxxxxxxxxx; linux-security-module@xxxxxxxxxxxxxxx;
> > linux-kernel@xxxxxxxxxxxxxxx; Silviu Vlasceanu
> > <Silviu.Vlasceanu@xxxxxxxxxx>; stable@xxxxxxxxxxxxxxx
> > Subject: Re: [PATCH v2 2/8] ima: Switch to ima_hash_algo for boot
> > aggregate
> > 
> > On Thu, 2020-02-06 at 09:36 +0000, Roberto Sassu wrote:
> > > > Hi Roberto,
> > > >
> > > > On Wed, 2020-02-05 at 11:33 +0100, Roberto Sassu wrote:
> > > >
> > > > <snip>
> > > >
> > > > > Reported-by: Jerry Snitselaar <jsnitsel@xxxxxxxxxx>
> > > > > Suggested-by: James Bottomley
> > > > <James.Bottomley@xxxxxxxxxxxxxxxxxxxxx>
> > > > > Signed-off-by: Roberto Sassu <roberto.sassu@xxxxxxxxxx>
> > > > > Cc: stable@xxxxxxxxxxxxxxx
> > > >
> > > > Cc'ing stable resulted in Sasha's automated message.ÂÂIf you're going
> > > > to Cc stable, then please include the stable kernel release (e.g. Cc:
> > > > stable@xxxxxxxxxxxxxxx # v5.3). ÂAlso please include a "Fixes" tag.
> > > > ÂNormally only bug fixes are backported.
> > >
> > > Ok, will add the kernel version. I also thought which commit I should
> > > mention in the Fixes tag. IMA always read the SHA1 bank from the
> > > beginning. I could mention the patch that introduces the new API
> > > to read other banks, but I'm not sure. What do you think?
> > 
> > This patch is dependent on nr_allocated_banks. ÂPlease try applying
> > this patch to the earliest stable kernel with the commit that
> > introduces nr_allocated_banks and test to make sure it works properly.
> 
> It also depends on 879b589210a9 ("tpm: retrieve digest size of unknown"
> algorithms with PCR read") which exported the mapping between TPM
> algorithm ID and crypto ID, and changed the definition of tpm_pcr_read()
> to read non-SHA1 PCR banks. It requires many patches, so backporting it
> is not a trivial task. I think the earliest kernel this patch can be backported to
> is 5.1.

Agreed.  Thank you for checking.

Mimi