Bug 1632786

Summary:	rngd lists no entropy sources, but still reads from TPM2
Product:	Red Hat Enterprise Linux 7	Reporter:	Vilém Maršík <vmarsik>
Component:	rng-tools	Assignee:	Neil Horman <nhorman>
Status:	CLOSED ERRATA	QA Contact:	Vilém Maršík <vmarsik>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	7.6	CC:	nhorman, vmarsik
Target Milestone:	rc	Keywords:	Reopened
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-08-06 13:18:44 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Vilém Maršík 2018-09-25 14:26:16 UTC

Description of problem:
rngd on a TPM2 machine lists no entropy sources, but secretly feeds TPM2 entropy into system.
Found while testing Bug 1627822 - rngd: failed to init TPM2 RNG.

Version-Release number of selected component (if applicable):
rng-tools-6.3.1-3.el7.x86_64
kernel-3.10.0-952.el7.x86_64

How reproducible:
100%

Steps to Reproduce:
1. log in to a machine with working TPM2
2. # rngd -f -d -x 1 -x 2 -x 5
3. read from /dev/random

Actual results:
rngd does not list any available source:
# rngd -f -d -x 1 -x 2 -x 5
Disabling 1: TPM RNG Device
Disabling 2: Intel RDRAND Instruction RNG
Disabling 5: JITTER Entropy generator
Initalizing available sources

rngd still feeds entropy, running "time cat /dev/random | rngtest -c 1" finishes in about 2.5s, without rngd it never finished for me and was killed after minutes

tracing shows that rngd reads from TPM:
# echo *tpm* > /sys/kernel/debug/tracing/set_ftrace_filter
# echo *locality* >> /sys/kernel/debug/tracing/set_ftrace_filter
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
# time cat /dev/random | rngtest -c 1
(...)
real    0m2.391s
(...)
# cat /sys/kernel/debug/tracing/trace
# tracer: function_graph
#
# CPU  DURATION                  FUNCTION CALLS
# |     |   |                     |   |   |   |
   6)               |  tpm_hwrng_read() {
   6)               |    tpm_get_random() {
   6)               |      tpm_chip_find_get() {
   6)   0.142 us    |        tpm_try_get_ops();
   6)   0.774 us    |      }
   6)               |      tpm2_get_random() {
   6)               |        tpm_transmit_cmd() {
   6)               |          tpm_transmit() {
   6)   0.077 us    |            tpm_tis_clkrun_enable();
   6)               |            request_locality() {
   6)               |              check_locality() {
   6)   1.445 us    |                tpm_tcg_read_bytes();
   6)   1.867 us    |              }
   6)   0.046 us    |              tpm_tcg_write_bytes();
   6)               |              check_locality() {
   6)   1.393 us    |                tpm_tcg_read_bytes();
   6)   1.808 us    |              }
   6)   4.838 us    |            }
(...)
   6)               |  tpm_hwrng_read() {
   6)               |    tpm_get_random() {
   6)               |      tpm_chip_find_get() {
   6)   0.045 us    |        tpm_try_get_ops();
   6)   0.363 us    |      }
   6)               |      tpm2_get_random() {
   6)               |        tpm_transmit_cmd() {
   6)               |          tpm_transmit() {
   6)   0.034 us    |            tpm_tis_clkrun_enable();
   6)               |            request_locality() {
   6)               |              check_locality() {
   6)   1.410 us    |                tpm_tcg_read_bytes();
   6)   1.744 us    |              }
   6)   0.034 us    |              tpm_tcg_write_bytes();
   6)               |              check_locality() {
   6)   1.479 us    |                tpm_tcg_read_bytes();
   6)   1.819 us    |              }
   6)   4.560 us    |            }
(...)

Expected results:
TPM2 listed in the available sources, when being used

Additional info:

Comment 3 Vilém Maršík 2018-09-25 14:43:28 UTC

And I wonder how to disable this TPM2 source with the '-x' option.

Comment 4 Neil Horman 2018-09-25 14:52:10 UTC

tpm2 exports its entropy as a hwrng (/dev/hwrng), and your command line doesn't disable the hwrng entropy source (-x 0)

Comment 5 Vilém Maršík 2018-09-25 15:22:15 UTC

(In reply to Neil Horman from comment #4)
> tpm2 exports its entropy as a hwrng (/dev/hwrng), and your command line
> doesn't disable the hwrng entropy source (-x 0)

Then why is that source not listed there?

Comment 6 Neil Horman 2018-09-25 15:57:52 UTC

That is an excellent question.  Can you post the console output of rngd here please?

Comment 7 Vilém Maršík 2018-09-26 14:38:47 UTC

See Description:
# rngd -f -d -x 1 -x 2 -x 5
Disabling 1: TPM RNG Device
Disabling 2: Intel RDRAND Instruction RNG
Disabling 5: JITTER Entropy generator
Initalizing available sources

^ i.e. TPM2/HWRNG not listed but used (and the '-x' flags make no difference in what was printed). Now I have found that the source gets listed by the '-l' flag:
# rngd -l
Entropy sources that are available but disabled
1: TPM RNG Device
4: NIST Network Entropy Beacon
Available and enabled entropy sources:
0: Hardware RNG Device
2: Intel RDRAND Instruction RNG
5: JITTER Entropy generator

This is what I had on my mind while reporting this, think about reopening. Any ideas?

Comment 8 Neil Horman 2018-09-26 15:43:32 UTC

I'm not sure what you're driving at.  I'm sorry you didn't realize that rngd -l lists the entropy sources (though that is documented in the man page).  It seems pretty clear to me that source 0 (hwrng) is enabled and available, so thats where your getting entropy from.

What is the ask here.  Are you looking for a message that expressly indicates that the hwrng is being enabled?  If so, sure I can do that, though I'd like to be sure thats what your after first

Comment 9 Vilém Maršík 2018-09-27 09:18:55 UTC

Yes, I ask for message that HWRNG is being used, when it is the case. Otherwise it's confusing:

I run rngd, it says it uses RDRAND + JITTER:
# rngd -f -d
Initalizing available sources
Enabling RDRAND rng support
Enabling JITTER rng support

Now I run it with these sources disabled, rngd runs with no sources listed and does not exit:
# rngd -f -d -x 2 -x 5
Disabling 2: Intel RDRAND Instruction RNG
Disabling 5: JITTER Entropy generator
Initalizing available sources

Strangely, entropy is still increasing, where did it come from?
# rngd -l
(...)
Available and enabled entropy sources:
0: Hardware RNG Device
2: Intel RDRAND Instruction RNG
5: JITTER Entropy generator

Ah, there is a secret source 0, that was hidden before.
Could it have been listed instead?

Comment 10 Neil Horman 2018-09-27 13:54:50 UTC

In fairness, its not unlisted, the list command clearly shows it.

But I take your meaning, clearly calling out all sources as they are initalized is a good idea. I'll take care of that, though It may be too late for 7.6.  I'll try my best

Comment 11 Vilém Maršík 2018-09-27 14:15:53 UTC

I know that the list commands shows that, but still got quite confused that the '-d' did not, before trying to use '-l'. Would like to avoid such confusions for the users.

I don't insist on this going into 7.6, but will do my best if it does.

Thanks

Comment 12 Neil Horman 2018-09-27 15:10:43 UTC

looks like were not going to make it for 7.6 (I need to get blocker approval, and its too late to get that for this as its not function critical).  I've fixed it upstream though, and will backport for 7,7

Comment 14 Vilém Maršík 2019-06-21 17:53:31 UTC

The messages now look correct in rng-tools-6.3.1-4.el7.x86_64:

# rngd -f -d -x 1 -x 2 -x 5
Disabling 1: TPM RNG Device
Disabling 2: Intel RDRAND Instruction RNG
Disabling 5: JITTER Entropy generator

Initalizing available sources

Initalizing entropy source Hardware RNG Device
^ here

Setting verified.

Comment 16 errata-xmlrpc 2019-08-06 13:18:44 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2019:2338