q3k/linux
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
linux/tools/testing/ktest/sample.conf
Steven Rostedt 35ce5952e6 ktest: Add prompt to use OUTPUT_MIN_CONFIG 
If the defined OUTPUT_MIN_CONFIG in the make_min_config test exists,
then give a prompt to ask the user if they want to use that config
instead, as it is very often the case, especially when the test has been
interrupted. The OUTPUT_MIN_CONFIG is usually the config that one wants
to use to continue the test where they left off.

But if START_MIN_CONFIG is defined (thus the MIN_CONFIG is not the
default), then do not prompt, as it will be annoying if the user has
this as one of many tests, and the test pauses waiting for input, while
the user is sleeping.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2011-07-15 21:57:25 -04:00

881 lines
30 KiB
Text
Raw Blame History

#
# Config file for ktest.pl
#
# Note, all paths must be absolute
#
# Options set in the beginning of the file are considered to be
# default options. These options can be overriden by test specific
# options, with the following exceptions:
#
# LOG_FILE
# CLEAR_LOG
# POWEROFF_ON_SUCCESS
# REBOOT_ON_SUCCESS
#
# Test specific options are set after the label:
#
# TEST_START
#
# The options after a TEST_START label are specific to that test.
# Each TEST_START label will set up a new test. If you want to
# perform a test more than once, you can add the ITERATE label
# to it followed by the number of times you want that test
# to iterate. If the ITERATE is left off, the test will only
# be performed once.
#
# TEST_START ITERATE 10
#
# You can skip a test by adding SKIP (before or after the ITERATE
# and number)
#
# TEST_START SKIP
#
# TEST_START SKIP ITERATE 10
#
# TEST_START ITERATE 10 SKIP
#
# The SKIP label causes the options and the test itself to be ignored.
# This is useful to set up several different tests in one config file, and
# only enabling the ones you want to use for a current test run.
#
# You can add default options anywhere in the file as well
# with the DEFAULTS tag. This allows you to have default options
# after the test options to keep the test options at the top
# of the file. You can even place the DEFAULTS tag between
# test cases (but not in the middle of a single test case)
#
# TEST_START
# MIN_CONFIG = /home/test/config-test1
#
# DEFAULTS
# MIN_CONFIG = /home/test/config-default
#
# TEST_START ITERATE 10
#
# The above will run the first test with MIN_CONFIG set to
# /home/test/config-test-1. Then 10 tests will be executed
# with MIN_CONFIG with /home/test/config-default.
#
# You can also disable defaults with the SKIP option
#
# DEFAULTS SKIP
# MIN_CONFIG = /home/test/config-use-sometimes
#
# DEFAULTS
# MIN_CONFIG = /home/test/config-most-times
#
# The above will ignore the first MIN_CONFIG. If you want to
# use the first MIN_CONFIG, remove the SKIP from the first
# DEFAULTS tag and add it to the second. Be careful, options
# may only be declared once per test or default. If you have
# the same option name under the same test or as default
# ktest will fail to execute, and no tests will run.
#
#### Config variables ####
#
# This config file can also contain "config variables".
# These are assigned with ":=" instead of the ktest option
# assigment "=".
#
# The difference between ktest options and config variables
# is that config variables can be used multiple times,
# where each instance will override the previous instance.
# And that they only live at time of processing this config.
#
# The advantage to config variables are that they can be used
# by any option or any other config variables to define thing
# that you may use over and over again in the options.
#
# For example:
#
# USER := root
# TARGET := mybox
# TEST_CASE := ssh ${USER}@${TARGET} /path/to/my/test
#
# TEST_START
# MIN_CONFIG = config1
# TEST = ${TEST_CASE}
#
# TEST_START
# MIN_CONFIG = config2
# TEST = ${TEST_CASE}
#
# TEST_CASE := ssh ${USER}@${TARGET} /path/to/my/test2
#
# TEST_START
# MIN_CONFIG = config1
# TEST = ${TEST_CASE}
#
# TEST_START
# MIN_CONFIG = config2
# TEST = ${TEST_CASE}
#
# TEST_DIR := /home/me/test
#
# BUILD_DIR = ${TEST_DIR}/linux.git
# OUTPUT_DIR = ${TEST_DIR}/test
#
# Note, the config variables are evaluated immediately, thus
# updating TARGET after TEST_CASE has been assigned does nothing
# to TEST_CASE.
#
# As shown in the example, to evaluate a config variable, you
# use the ${X} convention. Simple $X will not work.
#
# If the config variable does not exist, the ${X} will not
# be evaluated. Thus:
#
# MAKE_CMD = PATH=/mypath:${PATH} make
#
# If PATH is not a config variable, then the ${PATH} in
# the MAKE_CMD option will be evaluated by the shell when
# the MAKE_CMD option is passed into shell processing.
#### Using options in other options ####
#
# Options that are defined in the config file may also be used
# by other options. All options are evaulated at time of
# use (except that config variables are evaluated at config
# processing time).
#
# If an ktest option is used within another option, instead of
# typing it again in that option you can simply use the option
# just like you can config variables.
#
# MACHINE = mybox
#
# TEST = ssh root@${MACHINE} /path/to/test
#
# The option will be used per test case. Thus:
#
# TEST_TYPE = test
# TEST = ssh root@{MACHINE}
#
# TEST_START
# MACHINE = box1
#
# TEST_START
# MACHINE = box2
#
# For both test cases, MACHINE will be evaluated at the time
# of the test case. The first test will run ssh root@box1
# and the second will run ssh root@box2.
#### Mandatory Default Options ####
# These options must be in the default section, although most
# may be overridden by test options.
# The machine hostname that you will test
#MACHINE = target
# The box is expected to have ssh on normal bootup, provide the user
# (most likely root, since you need privileged operations)
#SSH_USER = root
# The directory that contains the Linux source code
#BUILD_DIR = /home/test/linux.git
# The directory that the objects will be built
# (can not be same as BUILD_DIR)
#OUTPUT_DIR = /home/test/build/target
# The location of the compiled file to copy to the target
# (relative to OUTPUT_DIR)
#BUILD_TARGET = arch/x86/boot/bzImage
# The place to put your image on the test machine
#TARGET_IMAGE = /boot/vmlinuz-test
# A script or command to reboot the box
#
# Here is a digital loggers power switch example
#POWER_CYCLE = wget --no-proxy -O /dev/null -q --auth-no-challenge 'http://admin:admin@power/outlet?5=CCL'
#
# Here is an example to reboot a virtual box on the current host
# with the name "Guest".
#POWER_CYCLE = virsh destroy Guest; sleep 5; virsh start Guest
# The script or command that reads the console
#
# If you use ttywatch server, something like the following would work.
#CONSOLE = nc -d localhost 3001
#
# For a virtual machine with guest name "Guest".
#CONSOLE = virsh console Guest
# Required version ending to differentiate the test
# from other linux builds on the system.
#LOCALVERSION = -test
# The grub title name for the test kernel to boot
# (Only mandatory if REBOOT_TYPE = grub)
#
# Note, ktest.pl will not update the grub menu.lst, you need to
# manually add an option for the test. ktest.pl will search
# the grub menu.lst for this option to find what kernel to
# reboot into.
#
# For example, if in the /boot/grub/menu.lst the test kernel title has:
# title Test Kernel
# kernel vmlinuz-test
#GRUB_MENU = Test Kernel
# A script to reboot the target into the test kernel
# (Only mandatory if REBOOT_TYPE = script)
#REBOOT_SCRIPT =
#### Optional Config Options (all have defaults) ####
# Start a test setup. If you leave this off, all options
# will be default and the test will run once.
# This is a label and not really an option (it takes no value).
# You can append ITERATE and a number after it to iterate the
# test a number of times, or SKIP to ignore this test.
#
#TEST_START
#TEST_START ITERATE 5
#TEST_START SKIP
# Have the following options as default again. Used after tests
# have already been defined by TEST_START. Optionally, you can
# just define all default options before the first TEST_START
# and you do not need this option.
#
# This is a label and not really an option (it takes no value).
# You can append SKIP to this label and the options within this
# section will be ignored.
#
# DEFAULTS
# DEFAULTS SKIP
# The default test type (default test)
# The test types may be:
# build - only build the kernel, do nothing else
# boot - build and boot the kernel
# test - build, boot and if TEST is set, run the test script
# (If TEST is not set, it defaults back to boot)
# bisect - Perform a bisect on the kernel (see BISECT_TYPE below)
# patchcheck - Do a test on a series of commits in git (see PATCHCHECK below)
#TEST_TYPE = test
# Test to run if there is a successful boot and TEST_TYPE is test.
# Must exit with 0 on success and non zero on error
# default (undefined)
#TEST = ssh user@machine /root/run_test
# The build type is any make config type or special command
# (default randconfig)
# nobuild - skip the clean and build step
# useconfig:/path/to/config - use the given config and run
# oldconfig on it.
# This option is ignored if TEST_TYPE is patchcheck or bisect
#BUILD_TYPE = randconfig
# The make command (default make)
# If you are building a 32bit x86 on a 64 bit host
#MAKE_CMD = CC=i386-gcc AS=i386-as make ARCH=i386
# Any build options for the make of the kernel (not for other makes, like configs)
# (default "")
#BUILD_OPTIONS = -j20
# If you need an initrd, you can add a script or code here to install
# it. The environment variable KERNEL_VERSION will be set to the
# kernel version that is used. Remember to add the initrd line
# to your grub menu.lst file.
#
# Here's a couple of examples to use:
#POST_INSTALL = ssh user@target /sbin/mkinitrd --allow-missing -f /boot/initramfs-test.img $KERNEL_VERSION
#
# or on some systems:
#POST_INSTALL = ssh user@target /sbin/dracut -f /boot/initramfs-test.img $KERNEL_VERSION
# If there is a script that you require to run before the build is done
# you can specify it with PRE_BUILD.
#
# One example may be if you must add a temporary patch to the build to
# fix a unrelated bug to perform a patchcheck test. This will apply the
# patch before each build that is made. Use the POST_BUILD to do a git reset --hard
# to remove the patch.
#
# (default undef)
#PRE_BUILD = cd ${BUILD_DIR} && patch -p1 < /tmp/temp.patch
# To specify if the test should fail if the PRE_BUILD fails,
# PRE_BUILD_DIE needs to be set to 1. Otherwise the PRE_BUILD
# result is ignored.
# (default 0)
# PRE_BUILD_DIE = 1
# If there is a script that should run after the build is done
# you can specify it with POST_BUILD.
#
# As the example in PRE_BUILD, POST_BUILD can be used to reset modifications
# made by the PRE_BUILD.
#
# (default undef)
#POST_BUILD = cd ${BUILD_DIR} && git reset --hard
# To specify if the test should fail if the POST_BUILD fails,
# POST_BUILD_DIE needs to be set to 1. Otherwise the POST_BUILD
# result is ignored.
# (default 0)
#POST_BUILD_DIE = 1
# Way to reboot the box to the test kernel.
# Only valid options so far are "grub" and "script"
# (default grub)
# If you specify grub, it will assume grub version 1
# and will search in /boot/grub/menu.lst for the title $GRUB_MENU
# and select that target to reboot to the kernel. If this is not
# your setup, then specify "script" and have a command or script
# specified in REBOOT_SCRIPT to boot to the target.
#
# The entry in /boot/grub/menu.lst must be entered in manually.
# The test will not modify that file.
#REBOOT_TYPE = grub
# The min config that is needed to build for the machine
# A nice way to create this is with the following:
#
# $ ssh target
# $ lsmod > mymods
# $ scp mymods host:/tmp
# $ exit
# $ cd linux.git
# $ rm .config
# $ make LSMOD=mymods localyesconfig
# $ grep '^CONFIG' .config > /home/test/config-min
#
# If you want even less configs:
#
# log in directly to target (do not ssh)
#
# $ su
# # lsmod | cut -d' ' -f1 | xargs rmmod
#
# repeat the above several times
#
# # lsmod > mymods
# # reboot
#
# May need to reboot to get your network back to copy the mymods
# to the host, and then remove the previous .config and run the
# localyesconfig again. The CONFIG_MIN generated like this will
# not guarantee network activity to the box so the TEST_TYPE of
# test may fail.
#
# You might also want to set:
# CONFIG_CMDLINE="<your options here>"
# randconfig may set the above and override your real command
# line options.
# (default undefined)
#MIN_CONFIG = /home/test/config-min
# Sometimes there's options that just break the boot and
# you do not care about. Here are a few:
# # CONFIG_STAGING is not set
# Staging drivers are horrible, and can break the build.
# # CONFIG_SCSI_DEBUG is not set
# SCSI_DEBUG may change your root partition
# # CONFIG_KGDB_SERIAL_CONSOLE is not set
# KGDB may cause oops waiting for a connection that's not there.
# This option points to the file containing config options that will be prepended
# to the MIN_CONFIG (or be the MIN_CONFIG if it is not set)
#
# Note, config options in MIN_CONFIG will override these options.
#
# (default undefined)
#ADD_CONFIG = /home/test/config-broken
# The location on the host where to write temp files
# (default /tmp/ktest/${MACHINE})
#TMP_DIR = /tmp/ktest/${MACHINE}
# Optional log file to write the status (recommended)
# Note, this is a DEFAULT section only option.
# (default undefined)
#LOG_FILE = /home/test/logfiles/target.log
# Remove old logfile if it exists before starting all tests.
# Note, this is a DEFAULT section only option.
# (default 0)
#CLEAR_LOG = 0
# Line to define a successful boot up in console output.
# This is what the line contains, not the entire line. If you need
# the entire line to match, then use regural expression syntax like:
# (do not add any quotes around it)
#
# SUCCESS_LINE = ^MyBox Login:$
#
# (default "login:")
#SUCCESS_LINE = login:
# In case the console constantly fills the screen, having
# a specified time to stop the test after success is recommended.
# (in seconds)
# (default 10)
#STOP_AFTER_SUCCESS = 10
# In case the console constantly fills the screen, having
# a specified time to stop the test after failure is recommended.
# (in seconds)
# (default 60)
#STOP_AFTER_FAILURE = 60
# In case the console constantly fills the screen, having
# a specified time to stop the test if it never succeeds nor fails
# is recommended.
# Note: this is ignored if a success or failure is detected.
# (in seconds)
# (default 600, -1 is to never stop)
#STOP_TEST_AFTER = 600
# Stop testing if a build fails. If set, the script will end if
# a failure is detected, otherwise it will save off the .config,
# dmesg and bootlog in a directory called
# MACHINE-TEST_TYPE_BUILD_TYPE-fail-yyyymmddhhmmss
# if the STORE_FAILURES directory is set.
# (default 1)
# Note, even if this is set to zero, there are some errors that still
# stop the tests.
#DIE_ON_FAILURE = 1
# Directory to store failure directories on failure. If this is not
# set, DIE_ON_FAILURE=0 will not save off the .config, dmesg and
# bootlog. This option is ignored if DIE_ON_FAILURE is not set.
# (default undefined)
#STORE_FAILURES = /home/test/failures
# Build without doing a make mrproper, or removing .config
# (default 0)
#BUILD_NOCLEAN = 0
# As the test reads the console, after it hits the SUCCESS_LINE
# the time it waits for the monitor to settle down between reads
# can usually be lowered.
# (in seconds) (default 1)
#BOOTED_TIMEOUT = 1
# The timeout in seconds when we consider the box hung after
# the console stop producing output. Be sure to leave enough
# time here to get pass a reboot. Some machines may not produce
# any console output for a long time during a reboot. You do
# not want the test to fail just because the system was in
# the process of rebooting to the test kernel.
# (default 120)
#TIMEOUT = 120
# In between tests, a reboot of the box may occur, and this
# is the time to wait for the console after it stops producing
# output. Some machines may not produce a large lag on reboot
# so this should accommodate it.
# The difference between this and TIMEOUT, is that TIMEOUT happens
# when rebooting to the test kernel. This sleep time happens
# after a test has completed and we are about to start running
# another test. If a reboot to the reliable kernel happens,
# we wait SLEEP_TIME for the console to stop producing output
# before starting the next test.
# (default 60)
#SLEEP_TIME = 60
# The time in between bisects to sleep (in seconds)
# (default 60)
#BISECT_SLEEP_TIME = 60
# The time in between patch checks to sleep (in seconds)
# (default 60)
#PATCHCHECK_SLEEP_TIME = 60
# Reboot the target box on error (default 0)
#REBOOT_ON_ERROR = 0
# Power off the target on error (ignored if REBOOT_ON_ERROR is set)
# Note, this is a DEFAULT section only option.
# (default 0)
#POWEROFF_ON_ERROR = 0
# Power off the target after all tests have completed successfully
# Note, this is a DEFAULT section only option.
# (default 0)
#POWEROFF_ON_SUCCESS = 0
# Reboot the target after all test completed successfully (default 1)
# (ignored if POWEROFF_ON_SUCCESS is set)
#REBOOT_ON_SUCCESS = 1
# In case there are isses with rebooting, you can specify this
# to always powercycle after this amount of time after calling
# reboot.
# Note, POWERCYCLE_AFTER_REBOOT = 0 does NOT disable it. It just
# makes it powercycle immediately after rebooting. Do not define
# it if you do not want it.
# (default undefined)
#POWERCYCLE_AFTER_REBOOT = 5
# In case there's isses with halting, you can specify this
# to always poweroff after this amount of time after calling
# halt.
# Note, POWEROFF_AFTER_HALT = 0 does NOT disable it. It just
# makes it poweroff immediately after halting. Do not define
# it if you do not want it.
# (default undefined)
#POWEROFF_AFTER_HALT = 20
# A script or command to power off the box (default undefined)
# Needed for POWEROFF_ON_ERROR and SUCCESS
#
# Example for digital loggers power switch:
#POWER_OFF = wget --no-proxy -O /dev/null -q --auth-no-challenge 'http://admin:admin@power/outlet?5=OFF'
#
# Example for a virtual guest call "Guest".
#POWER_OFF = virsh destroy Guest
# The way to execute a command on the target
# (default ssh $SSH_USER@$MACHINE $SSH_COMMAND";)
# The variables SSH_USER, MACHINE and SSH_COMMAND are defined
#SSH_EXEC = ssh $SSH_USER@$MACHINE $SSH_COMMAND";
# The way to copy a file to the target
# (default scp $SRC_FILE $SSH_USER@$MACHINE:$DST_FILE)
# The variables SSH_USER, MACHINE, SRC_FILE and DST_FILE are defined.
#SCP_TO_TARGET = scp $SRC_FILE $SSH_USER@$MACHINE:$DST_FILE
# The nice way to reboot the target
# (default ssh $SSH_USER@$MACHINE reboot)
# The variables SSH_USER and MACHINE are defined.
#REBOOT = ssh $SSH_USER@$MACHINE reboot
# The way triple faults are detected is by testing the kernel
# banner. If the kernel banner for the kernel we are testing is
# found, and then later a kernel banner for another kernel version
# is found, it is considered that we encountered a triple fault,
# and there is no panic or callback, but simply a reboot.
# To disable this (because it did a false positive) set the following
# to 0.
# (default 1)
#DETECT_TRIPLE_FAULT = 0
#### Per test run options ####
# The following options are only allowed in TEST_START sections.
# They are ignored in the DEFAULTS sections.
#
# All of these are optional and undefined by default, although
# some of these options are required for TEST_TYPE of patchcheck
# and bisect.
#
#
# CHECKOUT = branch
#
# If the BUILD_DIR is a git repository, then you can set this option
# to checkout the given branch before running the TEST. If you
# specify this for the first run, that branch will be used for
# all preceding tests until a new CHECKOUT is set.
#
#
# TEST_NAME = name
#
# If you want the test to have a name that is displayed in
# the test result banner at the end of the test, then use this
# option. This is useful to search for the RESULT keyword and
# not have to translate a test number to a test in the config.
#
# For TEST_TYPE = patchcheck
#
# This expects the BUILD_DIR to be a git repository, and
# will checkout the PATCHCHECK_START commit.
#
# The option BUILD_TYPE will be ignored.
#
# The MIN_CONFIG will be used for all builds of the patchcheck. The build type
# used for patchcheck is oldconfig.
#
# PATCHCHECK_START is required and is the first patch to
# test (the SHA1 of the commit). You may also specify anything
# that git checkout allows (branch name, tage, HEAD~3).
#
# PATCHCHECK_END is the last patch to check (default HEAD)
#
# PATCHCHECK_TYPE is required and is the type of test to run:
# build, boot, test.
#
# Note, the build test will look for warnings, if a warning occurred
# in a file that a commit touches, the build will fail, unless
# IGNORE_WARNINGS is set for the given commit's sha1
#
# IGNORE_WARNINGS can be used to disable the failure of patchcheck
# on a particuler commit (SHA1). You can add more than one commit
# by adding a list of SHA1s that are space delimited.
#
# If BUILD_NOCLEAN is set, then make mrproper will not be run on
# any of the builds, just like all other TEST_TYPE tests. But
# what makes patchcheck different from the other tests, is if
# BUILD_NOCLEAN is not set, only the first and last patch run
# make mrproper. This helps speed up the test.
#
# Example:
# TEST_START
# TEST_TYPE = patchcheck
# CHECKOUT = mybranch
# PATCHCHECK_TYPE = boot
# PATCHCHECK_START = 747e94ae3d1b4c9bf5380e569f614eb9040b79e7
# PATCHCHECK_END = HEAD~2
# IGNORE_WARNINGS = 42f9c6b69b54946ffc0515f57d01dc7f5c0e4712 0c17ca2c7187f431d8ffc79e81addc730f33d128
#
#
#
# For TEST_TYPE = bisect
#
# You can specify a git bisect if the BUILD_DIR is a git repository.
# The MIN_CONFIG will be used for all builds of the bisect. The build type
# used for bisecting is oldconfig.
#
# The option BUILD_TYPE will be ignored.
#
# BISECT_TYPE is the type of test to perform:
# build - bad fails to build
# boot - bad builds but fails to boot
# test - bad boots but fails a test
#
# BISECT_GOOD is the commit (SHA1) to label as good (accepts all git good commit types)
# BISECT_BAD is the commit to label as bad (accepts all git bad commit types)
#
# The above three options are required for a bisect operation.
#
# BISECT_REPLAY = /path/to/replay/file (optional, default undefined)
#
# If an operation failed in the bisect that was not expected to
# fail. Then the test ends. The state of the BUILD_DIR will be
# left off at where the failure occurred. You can examine the
# reason for the failure, and perhaps even find a git commit
# that would work to continue with. You can run:
#
# git bisect log > /path/to/replay/file
#
# The adding:
#
# BISECT_REPLAY= /path/to/replay/file
#
# And running the test again. The test will perform the initial
# git bisect start, git bisect good, and git bisect bad, and
# then it will run git bisect replay on this file, before
# continuing with the bisect.
#
# BISECT_START = commit (optional, default undefined)
#
# As with BISECT_REPLAY, if the test failed on a commit that
# just happen to have a bad commit in the middle of the bisect,
# and you need to skip it. If BISECT_START is defined, it
# will checkout that commit after doing the initial git bisect start,
# git bisect good, git bisect bad, and running the git bisect replay
# if the BISECT_REPLAY is set.
#
# BISECT_SKIP = 1 (optional, default 0)
#
# If BISECT_TYPE is set to test but the build fails, ktest will
# simply fail the test and end their. You could use BISECT_REPLAY
# and BISECT_START to resume after you found a new starting point,
# or you could set BISECT_SKIP to 1. If BISECT_SKIP is set to 1,
# when something other than the BISECT_TYPE fails, ktest.pl will
# run "git bisect skip" and try again.
#
# BISECT_FILES = <path> (optional, default undefined)
#
# To just run the git bisect on a specific path, set BISECT_FILES.
# For example:
#
# BISECT_FILES = arch/x86 kernel/time
#
# Will run the bisect with "git bisect start -- arch/x86 kernel/time"
#
# BISECT_REVERSE = 1 (optional, default 0)
#
# In those strange instances where it was broken forever
# and you are trying to find where it started to work!
# Set BISECT_GOOD to the commit that was last known to fail
# Set BISECT_BAD to the commit that is known to start working.
# With BISECT_REVERSE = 1, The test will consider failures as
# good, and success as bad.
#
# BISECT_MANUAL = 1 (optional, default 0)
#
# In case there's a problem with automating the bisect for
# whatever reason. (Can't reboot, want to inspect each iteration)
# Doing a BISECT_MANUAL will have the test wait for you to
# tell it if the test passed or failed after each iteration.
# This is basicall the same as running git bisect yourself
# but ktest will rebuild and install the kernel for you.
#
# BISECT_CHECK = 1 (optional, default 0)
#
# Just to be sure the good is good and bad is bad, setting
# BISECT_CHECK to 1 will start the bisect by first checking
# out BISECT_BAD and makes sure it fails, then it will check
# out BISECT_GOOD and makes sure it succeeds before starting
# the bisect (it works for BISECT_REVERSE too).
#
# You can limit the test to just check BISECT_GOOD or
# BISECT_BAD with BISECT_CHECK = good or
# BISECT_CHECK = bad, respectively.
#
# Example:
# TEST_START
# TEST_TYPE = bisect
# BISECT_GOOD = v2.6.36
# BISECT_BAD = b5153163ed580e00c67bdfecb02b2e3843817b3e
# BISECT_TYPE = build
# MIN_CONFIG = /home/test/config-bisect
#
#
#
# For TEST_TYPE = config_bisect
#
# In those cases that you have two different configs. One of them
# work, the other does not, and you do not know what config causes
# the problem.
# The TEST_TYPE config_bisect will bisect the bad config looking for
# what config causes the failure.
#
# The way it works is this:
#
# First it finds a config to work with. Since a different version, or
# MIN_CONFIG may cause different dependecies, it must run through this
# preparation.
#
# Overwrites any config set in the bad config with a config set in
# either the MIN_CONFIG or ADD_CONFIG. Thus, make sure these configs
# are minimal and do not disable configs you want to test:
# (ie. # CONFIG_FOO is not set).
#
# An oldconfig is run on the bad config and any new config that
# appears will be added to the configs to test.
#
# Finally, it generates a config with the above result and runs it
# again through make oldconfig to produce a config that should be
# satisfied by kconfig.
#
# Then it starts the bisect.
#
# The configs to test are cut in half. If all the configs in this
# half depend on a config in the other half, then the other half
# is tested instead. If no configs are enabled by either half, then
# this means a circular dependency exists and the test fails.
#
# A config is created with the test half, and the bisect test is run.
#
# If the bisect succeeds, then all configs in the generated config
# are removed from the configs to test and added to the configs that
# will be enabled for all builds (they will be enabled, but not be part
# of the configs to examine).
#
# If the bisect fails, then all test configs that were not enabled by
# the config file are removed from the test. These configs will not
# be enabled in future tests. Since current config failed, we consider
# this to be a subset of the config that we started with.
#
# When we are down to one config, it is considered the bad config.
#
# Note, the config chosen may not be the true bad config. Due to
# dependencies and selections of the kbuild system, mulitple
# configs may be needed to cause a failure. If you disable the
# config that was found and restart the test, if the test fails
# again, it is recommended to rerun the config_bisect with a new
# bad config without the found config enabled.
#
# The option BUILD_TYPE will be ignored.
#
# CONFIG_BISECT_TYPE is the type of test to perform:
# build - bad fails to build
# boot - bad builds but fails to boot
# test - bad boots but fails a test
#
# CONFIG_BISECT is the config that failed to boot
#
# If BISECT_MANUAL is set, it will pause between iterations.
# This is useful to use just ktest.pl just for the config bisect.
# If you set it to build, it will run the bisect and you can
# control what happens in between iterations. It will ask you if
# the test succeeded or not and continue the config bisect.
#
# CONFIG_BISECT_GOOD (optional)
# If you have a good config to start with, then you
# can specify it with CONFIG_BISECT_GOOD. Otherwise
# the MIN_CONFIG is the base.
#
# Example:
# TEST_START
# TEST_TYPE = config_bisect
# CONFIG_BISECT_TYPE = build
# CONFIG_BISECT = /home/test/<2F>onfig-bad
# MIN_CONFIG = /home/test/config-min
# BISECT_MANUAL = 1
#
#
#
# For TEST_TYPE = make_min_config
#
# After doing a make localyesconfig, your kernel configuration may
# not be the most useful minimum configuration. Having a true minimum
# config that you can use against other configs is very useful if
# someone else has a config that breaks on your code. By only forcing
# those configurations that are truly required to boot your machine
# will give you less of a chance that one of your set configurations
# will make the bug go away. This will give you a better chance to
# be able to reproduce the reported bug matching the broken config.
#
# Note, this does take some time, and may require you to run the
# test over night, or perhaps over the weekend. But it also allows
# you to interrupt it, and gives you the current minimum config
# that was found till that time.
#
# Note, this test automatically assumes a BUILD_TYPE of oldconfig
# and its test type acts like boot.
# TODO: add a test version that makes the config do more than just
# boot, like having network access.
#
# To save time, the test does not just grab any option and test
# it. The Kconfig files are examined to determine the dependencies
# of the configs. If a config is chosen that depends on another
# config, that config will be checked first. By checking the
# parents first, we can eliminate whole groups of configs that
# may have been enabled.
#
# For example, if a USB device config is chosen and depends on CONFIG_USB,
# the CONFIG_USB will be tested before the device. If CONFIG_USB is
# found not to be needed, it, as well as all configs that depend on
# it, will be disabled and removed from the current min_config.
#
# OUTPUT_MIN_CONFIG is the path and filename of the file that will
# be created from the MIN_CONFIG. If you interrupt the test, set
# this file as your new min config, and use it to continue the test.
# This file does not need to exist on start of test.
# This file is not created until a config is found that can be removed.
# If this file exists, you will be prompted if you want to use it
# as the min_config (overriding MIN_CONFIG) if START_MIN_CONFIG
# is not defined.
# (required field)
#
# START_MIN_CONFIG is the config to use to start the test with.
# you can set this as the same OUTPUT_MIN_CONFIG, but if you do
# the OUTPUT_MIN_CONFIG file must exist.
# (default MIN_CONFIG)
#
# IGNORE_CONFIG is used to specify a config file that has configs that
# you already know must be set. Configs are written here that have
# been tested and proved to be required. It is best to define this
# file if you intend on interrupting the test and running it where
# it left off. New configs that it finds will be written to this file
# and will not be tested again in later runs.
# (optional)
#
# Example:
#
# TEST_TYPE = make_min_config
# OUTPUT_MIN_CONFIG = /path/to/config-new-min
# START_MIN_CONFIG = /path/to/config-min
# IGNORE_CONFIG = /path/to/config-tested
#
Reference in a new issue View git blame Copy permalink