On Wed, Aug 11, 2021 at 08:59:14AM -0700, Elliot Berman wrote:
If rcu_read_lock_sched tracing is enabled, the tracing subsystem can
perform a jump which needs to be checked by CFI. For example, stm_ftrace
source is enabled as a module and hooks into enabled ftrace events. This
can cause an recursive loop where find_shadow_check_fn ->
rcu_read_lock_sched -> (call to stm_ftrace generates cfi slowpath) ->
find_shadow_check_fn -> rcu_read_lock_sched -> ...
To avoid the recursion, either the ftrace codes needs to be marked with
__no_cfi or CFI should not trace. Use the "_notrace" in CFI to avoid
tracing so that CFI can guard ftrace.
Signed-off-by: Elliot Berman <quic_eberman@xxxxxxxxxxx>
Thanks for this patch! While applying it I noticed that the DKIM
signature failed. This is actually even visible in the lore archive:
https://lore.kernel.org/lkml/20210811155914.19550-1-quic_eberman@xxxxxxxxxxx/raw
(DKIM_INVALID)
$ b4 am -tls https://lore.kernel.org/lkml/20210811155914.19550-1-quic_eberman@xxxxxxxxxxx/
Grabbing thread from lore.kernel.org/lkml/20210811155914.19550-1-quic_eberman%40quicinc.com/t.mbox.gz
Analyzing 2 messages in the thread
Checking attestation on all messages, may take a moment...
---
✗ [PATCH 1/1] cfi: Use rcu_read_{un}lock_sched_notrace
+ Reviewed-by: Sami Tolvanen <samitolvanen@xxxxxxxxxx> (✓ DKIM/google.com)
+ Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx>
+ Link: https://lore.kernel.org/r/20210811155914.19550-1-quic_eberman@xxxxxxxxxxx
---
✗ BADSIG: DKIM/quicinc.com
Do you know if qualcomm is mangling outbound emails? (i.e. was the
trailing body suffix added after calculating the DKIM signature?)
Thanks!
-Kees