On Mon, Sep 21, 2020 at 12:03:56PM +0200, Borislav Petkov wrote:
> On Tue, Sep 15, 2020 at 02:28:30PM +0300, Jarkko Sakkinen wrote:
> > +static int sgx_validate_secs(const struct sgx_secs *secs)
> > +{
> > + u64 max_size = (secs->attributes & SGX_ATTR_MODE64BIT) ?
> > + sgx_encl_size_max_64 : sgx_encl_size_max_32;
> > +
> > + if (secs->size < (2 * PAGE_SIZE) || !is_power_of_2(secs->size))
> > + return -EINVAL;
> > +
> > + if (secs->base & (secs->size - 1))
> > + return -EINVAL;
> > +
> > + if (secs->miscselect & sgx_misc_reserved_mask ||
> > + secs->attributes & sgx_attributes_reserved_mask ||
> > + secs->xfrm & sgx_xfrm_reserved_mask)
> > + return -EINVAL;
> > +
> > + if (secs->size > max_size)
> > + return -EINVAL;
> > +
> > + if (!(secs->xfrm & XFEATURE_MASK_FP) ||
> > + !(secs->xfrm & XFEATURE_MASK_SSE) ||
> > + (((secs->xfrm >> XFEATURE_BNDREGS) & 1) !=
> > + ((secs->xfrm >> XFEATURE_BNDCSR) & 1)))
>
> Let that last line stick out so that you have each statement on a single line.
>
> > + return -EINVAL;
> > +
> > + if (!secs->ssa_frame_size)
> > + return -EINVAL;
> > +
> > + if (sgx_calc_ssa_frame_size(secs->miscselect, secs->xfrm) >
> > + secs->ssa_frame_size)
>
> Let that stick out.
>
> > + return -EINVAL;
> > +
> > + if (memchr_inv(secs->reserved1, 0, sizeof(secs->reserved1)) ||
> > + memchr_inv(secs->reserved2, 0, sizeof(secs->reserved2)) ||
> > + memchr_inv(secs->reserved3, 0, sizeof(secs->reserved3)) ||
> > + memchr_inv(secs->reserved4, 0, sizeof(secs->reserved4)))
> > + return -EINVAL;
> > +
> > + return 0;
> > +}
> > +
> > +static int sgx_encl_create(struct sgx_encl *encl, struct sgx_secs *secs)
> > +{
> > + unsigned long encl_size = secs->size + PAGE_SIZE;
>
> You're still using secs->size before validation. I know, we will return
> early if sgx_validate_secs() fails but pls move that addition after the
> validation call.
Is this appropriate:
/* The extra page in swap space goes to SECS. */
encl_size = secs->size + PAGE_SIZE;
backing = shmem_file_setup("SGX backing", encl_size + (encl_size >> 5),
VM_NORESERVE);
if (IS_ERR(backing)) {
ret = PTR_ERR(backing);
goto err_out_shrink;
}
> ...
>
> > +/**
> > + * sgx_ioc_enclave_create - handler for %SGX_IOC_ENCLAVE_CREATE
> > + * @filep: open file to /dev/sgx
>
> Dammit, how many times do I have to type this comment here?!
>
> "That's
>
> @encl: enclave pointer
>
> or so."
>
> There's no filep - there is an encl!
I'm not actually sure what has happened. As you can easily grep, the
rename is done in five other sites. I also see a similar problem in
EINIT, which I will fix.
git grep "enclave pointer" arch/x86/kernel/cpu/sgx/ | wc -l
5
> > + * @arg: userspace pointer to a struct sgx_enclave_create instance
> > + *
> > + * Allocate kernel data structures for a new enclave and execute ECREATE
> > after
> > + * verifying the correctness of the provided SECS.
>
> ... which is done in sgx_validate_secs()."
>
> Yes, spell it out, otherwise one has to wonder where that validation is
> happening in the function *below* because the comment is over it - not
> over sgx_validate_secs().
>
> And yes, you need to spell stuff like that out because this SGX crap is
> complex and it better be properly documented!
I agree with this but I also think it would make sense to rephrase
"verifying the correctness of the provided SECS" with something more
informative.
I would rephrase as:
"... after checking that the provided data for SECS meets the expectations
of ENCLS[ECREATE] for an unitialized enclave and size of the address
space does not surpass the platform expectations. This validation is
executed by sgx_validate_secs()."
Is this sufficient for you, or do you have further suggestions?
> --
> Regards/Gruss,
> Boris.
>
> https://people.kernel.org/tglx/notes-about-netiquette
/Jarkko
