platform_system_core/llkd
Mark Salyzyn afd66f2fd3 llkd: bootstat: propagate detailed livelock canonical boot reason
Report kernel_panic,sysrq,livelock,<state> reboot reason via last
dmesg (pstore console).  Add ro.llk.killtest property, which will
allow reliable ABA platforms to drop kill test and go directly
to kernel panic.  This should also allow some manual unit testing
of the canonical boot reason report.

New canonical boot reasons from llkd are:
- kernel_panic,sysrq,livelock,alarm llkd itself locked up (Hail Mary)
- kernel_panic,sysrq,livelock,driver uninterrruptible D state
- kernel_panic,sysrq,livelock,zombie uninterrruptible Z state

Manual test assumptions:
- llkd is built by the platform and landed on system partition
- unit test is built and landed in /data/nativetest (could
  land in /data/nativetest64, adjust test correspondingly)
- llkd not enabled, ro.llk.enable and ro.llk.killtest
  are not set by platform allowing test to adjust all the
  configuration properties and start llkd.
- or, llkd is enabled, ro.llk.enable is true, and killtest is
  disabled, ro.llk.killtest is false, setup by the platform.
  This breaks the go/apct generic operations of the unit test
  for llk.zombie and llk.driver as kernel panic results
  requiring manual intervention otherwise.  If test moves to
  go/apct, then we will be forced to bypass these tests under
  this condition (but allow them to run if ro.llk.killtest
  is "off" so specific testing above/below can be run).

for i in driver zombie; do
        adb shell su root setprop ro.llk.killtest off
        adb shell /data/nativetest/llkd_unit_test/llkd_unit_test --gtest_filter=llkd.${i}
        adb wait-for-device
        adb shell su root setprop ro.llk.killtest off
        sleep 60
        adb shell getprop sys.boot.reason
        adb shell /data/nativetest/llkd_unit_test/llkd_unit_test --gtest_filter=llkd.${i}
done

Test: llkd_unit_test (see test assumptions)
Bug: 33808187
Bug: 72838192
Change-Id: I2b24875376ddfdbc282ba3da5c5b3567de85dbc0
2018-04-18 14:02:16 -07:00
..
include llkd: bootstat: propagate detailed livelock canonical boot reason 2018-04-18 14:02:16 -07:00
tests llkd: bootstat: propagate detailed livelock canonical boot reason 2018-04-18 14:02:16 -07:00
Android.bp llkd: add live-lock daemon 2018-04-18 14:01:56 -07:00
OWNERS llkd: add live-lock daemon 2018-04-18 14:01:56 -07:00
README.md llkd: bootstat: propagate detailed livelock canonical boot reason 2018-04-18 14:02:16 -07:00
libllkd.cpp llkd: bootstat: propagate detailed livelock canonical boot reason 2018-04-18 14:02:16 -07:00
llkd.cpp llkd: add live-lock daemon 2018-04-18 14:01:56 -07:00
llkd.rc llkd: bootstat: propagate detailed livelock canonical boot reason 2018-04-18 14:02:16 -07:00

README.md

Android Live-LocK Daemon

Introduction

Android Live-LocK Daemon (llkd) is used to catch kernel deadlocks and mitigate.

Code is structured to allow integration into another service as either as part of the main loop, or spun off as a thread should that be necessary. A default standalone implementation is provided by llkd component.

The 'C' interface from libllkd component is thus:

#include "llkd.h"
bool llkInit(const char* threadname) /* return true if enabled */
unsigned llkCheckMillseconds(void)   /* ms to sleep for next check */

If a threadname is provided, a thread will be automatically spawned, otherwise caller must call llkCheckMilliseconds in its main loop. Function will return the period of time before the next expected call to this handler.

Operations

If a thread is in D or Z state with no forward progress for longer than ro.llk.timeout_ms, or ro.llk.[D|Z].timeout_ms, kill the process or parent process respectively. If another scan shows the same process continues to exist, then have a confirmed live-lock condition and need to panic. Panic the kernel in a manner to provide the greatest bugreporting details as to the condition. Add a alarm self watchdog should llkd ever get locked up that is double the expected time to flow through the mainloop. Sampling is every ro.llk_sample_ms.

Default will not monitor init, or [kthreadd] and all that [kthreadd] spawns. This reduces the effectiveness of llkd by limiting its coverage. If there is value in covering [kthreadd] spawned threads, the requirement will be that the drivers not remain in a persistent 'D' state, or that they have mechanisms to recover the thread should it be killed externally (this is good driver coding hygiene, a common request to add such to publicly reviewed kernel.org maintained drivers). For instance use wait_event_interruptible() instead of wait_event(). The blacklists can be adjusted accordingly if these conditions are met to cover kernel components.

An accompanying gTest set have been added, and will setup a persistent D or Z process, with and without forward progress, but not in a live-lock state because that would require a buggy kernel, or a module or kernel modification to stimulate. The test will check that llkd will mitigate first by killing the appropriate process. D state is setup by vfork() waiting for exec() in child process. Z state is setup by fork() and an un-waited for child process. Should be noted that both of these conditions should never happen on Android on purpose, and llkd effectively sweeps up processes that create these conditions. If the test can, it will reconfigure llkd to expedite the test duration by adjusting the ro.llk.* Android properties. Tests run the D state with some scheduling progress to ensure that ABA checking prevents false triggers. If 100% reliable ABA on platform, then ro.llk.killtest can be set to false; however this will result in some of the unit tests to panic kernel instead of deal with more graceful kill operation.

Android Properties

Android Properties llkd respond to (_ms parms are in milliseconds):

ro.config.low_ram

default false, if true do not sysrq t (dump all threads).

ro.llk.enable

default false, allow live-lock daemon to be enabled.

llk.enable

default ro.llk.enable, and evaluated for eng.

ro.khungtask.enable

default false, allow [khungtask] daemon to be enabled.

khungtask.enable

default ro.khungtask.enable and evaluated for eng.

ro.llk.mlockall

default false, enable call to mlockall().

ro.khungtask.timeout

default value 12 minutes, [khungtask] maximum timelimit.

ro.llk.timeout_ms

default 10 minutes, D or Z maximum timelimit, double this value and it sets the alarm watchdog for llkd.

ro.llk.D.timeout_ms

default ro.llk.timeout_ms, D maximum timelimit.

ro.llk.Z.timeout_ms

default ro.llk.timeout_ms, Z maximum timelimit.

ro.llk.check_ms

default 2 minutes samples of threads for D or Z.

ro.llk.blacklist.process

default 0,1,2 (kernel, init and [kthreadd]) plus process names init,[kthreadd],[khungtaskd],lmkd,lmkd.llkd,llkd,watchdogd, [watchdogd],[watchdogd/0],...,[watchdogd/<get_nprocs-1>].

ro.llk.blacklist.parent

default 0,2 (kernel and [kthreadd]).

ro.llk.blacklist.uid

default , comma separated list of uid numbers or names.

Architectural Concerns

  • Create kernel module and associated gTest to actually test panic.
  • Create gTest to test out blacklist (ro.llk.blacklist. generally not be inputs). Could require more test-only interfaces to libllkd.
  • Speed up gTest using something else than ro.llk., which should not be inputs.