Having just seen a report of using "fips=1" on the kernel command line,
I could not find it documented anywhere, so add some help for it.
Signed-off-by: Randy Dunlap <[email protected]>
Cc: Dexuan Cui <[email protected]>
Cc: [email protected]
Cc: Eric Biggers <[email protected]>
Cc: Herbert Xu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: [email protected]
---
Updates/corrections welcome.
Documentation/admin-guide/kernel-parameters.txt | 15 ++++++++++++++
1 file changed, 15 insertions(+)
--- linux-next-20210329.orig/Documentation/admin-guide/kernel-parameters.txt
+++ linux-next-20210329/Documentation/admin-guide/kernel-parameters.txt
@@ -1370,6 +1370,21 @@
See Documentation/admin-guide/sysctl/net.rst for
fb_tunnels_only_for_init_ns
+ fips= Format: { 0 | 1}
+ Use to disable (0) or enable (1) FIPS mode.
+ If enabled, any process that is waiting on the
+ 'fips_fail_notif_chain' will be notified of fips
+ failures.
+ This setting can also be modified via sysctl at
+ /proc/sysctl/crypto/fips_enabled, i.e.,
+ crypto.fips_enabled.
+ If fips_enabled = 1, some crypto tests are skipped.
+ It can also effect which ECC curve is used.
+ If fips_enabled = 1 and a test fails, it will cause a
+ kernel panic.
+ If fips_enabled = 1, RSA test requires a key size of
+ 2K or larger.
+
floppy= [HW]
See Documentation/admin-guide/blockdev/floppy.rst.
On 3/29/21 9:37 PM, Herbert Xu wrote:
> On Mon, Mar 29, 2021 at 09:00:01PM -0700, Randy Dunlap wrote:
>>
>> + If fips_enabled = 1, some crypto tests are skipped.
>
> I don't think any tests are skipped. It does however disable
> many algorithms by essentially failing them at the testing stage.
That statement was based on crypto/testmgr.c (in 4 places):
if (fips_enabled && template[i].fips_skip)
continue;
and
if (fips_enabled && vec->fips_skip)
return 0;
and
if (fips_enabled && !alg_test_descs[i].fips_allowed)
goto non_fips_alg;
and
if (fips_enabled && ((i >= 0 && !alg_test_descs[i].fips_allowed) ||
(j >= 0 && !alg_test_descs[j].fips_allowed)))
goto non_fips_alg;
so it appears (at least to me) that there are some methods (infrastructure)
for tests to be skipped, but maybe none are actually using that possiblilty.
In any case, I don't mind dropping that part of the documentation.
thanks.
--
~Randy
On Mon, Mar 29, 2021 at 10:00:45PM -0700, Randy Dunlap wrote:
> On 3/29/21 9:37 PM, Herbert Xu wrote:
> > On Mon, Mar 29, 2021 at 09:00:01PM -0700, Randy Dunlap wrote:
> >>
> >> + If fips_enabled = 1, some crypto tests are skipped.
> >
> > I don't think any tests are skipped. It does however disable
> > many algorithms by essentially failing them at the testing stage.
>
> That statement was based on crypto/testmgr.c (in 4 places):
>
> if (fips_enabled && template[i].fips_skip)
> continue;
By skipping the test, the algorithm effectively fails the self-test
and therefore is disabled.
Cheers,
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt