The previous tail -1 broke with commit 7ff525712a ("kbuild: fake the
"Entering directory ..." message more simply")
Link: http://lkml.kernel.org/r/20141022194408.GA20989@pobox.suse.cz
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If git bisect reply is being used in the bisect tests, don't bother
doing the git bisect good or git bisect bad calls. The git bisect
reply will override them anyway, and that's called immediately
after the other two. Going the git bisect (good|bad) is just a
waste of time.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The reboot function when rebooting back to a good kernel has a check
to make sure that a new kernel was indeed booted. But that check
uses a timeout value, which when calling the monitor will still
return success if the timeout is hit (no bug was found). It should
return an error to let the reboot code know that a new kernel was
not reached. Only the reboot code checks the return value of the
monitor.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When doing a manual bisect, a build can fail or a test can be inconclusive.
In these cases it would be helpful to be able to skip the test entirely.
Link: http://lkml.kernel.org/r/1409164021-2136-1-git-send-email-chris.j.arges@canonical.com
Reviewed-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
Signed-off-by: Chris J Arges <chris.j.arges@canonical.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Add a way to run a patchcheck test on the commits that are in one branch
but not in another. This uses git cherry to find a list of commits to
test each one with.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
After the rewrite of the config bisect, there were several unused
functions that can be removed.
One of the unused functions printed out the failed config nicer than
what the rewrite did, so I kept that and used it to output the
bad config.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The new rewrite left out the CONFIG_BISECT_CHECK, which allows the
user to test that their "bad" config still is bad and their "good"
config still is good. This is especially important as the configs
are passed through a "make oldconfig" to update them with the lastest
kernel. Things could change that causes a bad config to work, or a
good config to break. The check is done after the configs have run
through the oldconfig processing.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
I never liked the way config-bisect worked. I would assume the bad config
had some config that broke the system. But it would not work if the bad
config just happened to be missing something that the good config had.
I rewrote the config-bisect to do this properly. It does a diff of the two
configs, and sets half of the configs that are in one and not the other.
The way it works is that when it "sets", it really just makes one copy
what the other has. That is, a "set" can be setting a:
# CONFIG_FOO is not set
Basically, it looks at the differences between the two files and makes
them similar until it comes down to one config that makes it work or
not work depending on if it is set or not.
Note, if more than one config change makes the bad config not work, it
will only find one of them. But this is true with all bisect logic.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Some cleanup for improving readability as follows.
- Initialize $ktest_config at its definition.
- Put parentheses around the `config-file' argument in the usage message
because it's a optional one.
- Rename get_ktest_config{,s} to more descriptive get_mandatory_config{,s}.
Link: http://lkml.kernel.org/r/87fvmr30kb.wl%satoru.takeuchi@gmail.com
Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If we'd like to set the redirect target file of run_command(),
we should define $redirect before calling this function and should undef it
after calling this function. Since it's user-unfriendly, add 2nd parameter of
run_command() for this purpose.
Link: http://lkml.kernel.org/r/87vbvwokq8.wl%satoru.takeuchi@gmail.com
Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
o Add config to modify the signal to terminate console
o Update to documentation (missing some config options)
o Add KERNEL_VERSION variable to use for other configs
o Add '=~' to let configs eval other configs
o Add BISECT_TRIES to run multiple tests per git bisect good
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.15 (GNU/Linux)
iQEcBAABAgAGBQJS3UMjAAoJEKQekfcNnQGuU/YH/RdTp8o6GgHfBans6vG588mZ
CoTOhfYH9loGTAcs8JqKt0sIBaPDmdF1Qur+nthBJ/vqyFSpRtaoCLTTvNTb8Snh
Wia5DEpi4BnfH3TMyMqNVCyXJjO8wkxq7bv5wB1+pqVnIDADlX5xejOVCDms2scm
EUXQeZx38KrFHH2ZvrjzjrGjIboKB/vTriqF8Q/Vmu0lRYo+Lt6HkDUEJEehxeLz
oU+oQOuep5ABo1APMYIml9YZqBeJJuCeqKd75T3QfgWyyCaYLv1ggHXG/bymBROX
uTd/RK3lKC/ODkVi9u57a4djptYYUW6mTDkUcLO3lZhqhVKqSiOQt0B9Bw/JTdQ=
=+mEf
-----END PGP SIGNATURE-----
Merge tag 'ktest-v3.14' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest
Pull ktest updates from Steven Rostedt:
"Here's some basic updates to ktest.pl. They include:
- add config to modify the signal to terminate console
- update to documentation (missing some config options)
- add KERNEL_VERSION variable to use for other configs
- add '=~' to let configs eval other configs
- add BISECT_TRIES to run multiple tests per git bisect good"
* tag 'ktest-v3.14' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest:
ktest: Add BISECT_TRIES to bisect test
ktest: Add eval '=~' command to modify variables in config file
ktest: Add special variable ${KERNEL_VERSION}
ktest: Add documentation of CLOSE_CONSOLE_SIGNAL
ktest: Make the signal to terminate the console configurable
For those cases that it takes several tries to hit a bug, it would be
useful for ktest.pl to try a test multiple times before it considers
the test as a pass. To accomplish this, BISECT_TRIES ktest config
option has been added. It is default to one, as most of the time a
bisect only needs to try a test once. But the user can now up this
to make ktest run a given test multiple times. The first failure
that is detected will set a bisect bad. It only repeats on success.
Note, as with all race bugs, there's no guarantee that if it succeeds,
it is really a good bisect. But it helps in case the bug is somewhat
reliable.
You can set BISECT_TRIES to zero, and all tests will be considered
good, unless you also set BISECT_MANUAL.
Suggested-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
With the added variable ${KERNEL_VERSION}, it is useful to be
able to use parts of it for other variables.
For example, if you want to create a warnings file for each major
kernel version to test sub versions against you can create
your warnings file with like this:
WARNINGS_FILE = warnings-file-${KERNEL_VERSION}
But this may add 3.8.12 or something, and we want all 3.8.* to
use the same file, and 3.10.* to use another file, and so on.
With the eval command we can, by adding:
WARNINGS_FILE =~ s/(-file-\d+\.\d+).*/$1/
Which will chop off the extra characters after the 3.8.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Add a special variable that can be used in other variables called
${KERNEL_VERSION}. This will embed the current kernel version into
the variable. For example:
WARNINGS_FILE = ${OUTPUT_DIR}/warnings-${KERNEL_VERSION}
If the current version is v3.8 then the WARNINGS_FILE will become
${OUTPUT_DIR}/warnings-v3.8
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Currently ktest sends SIGINT to terminate the console.
However, there are consoles which do not exit by this signal, for example,
in my case, "virsh console <guest OS>". In such case, ktest is blocked in
close_console(). It prevents this automate test.
This patch adds new option CLOSE_CONSOLE_SIGNAL which mean the
signal to terminate the console. Since its default value is "INT",
the original behavior isn't changed.
Link: http://lkml.kernel.org/r/87zjol8pl5.wl%satoru.takeuchi@gmail.com
Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Different tests may use a different machine. In such cases, we need to
try to get the current grub menu index. If the same grub menu is used
for two different machines, it may not be at the same index on the
second machine. A search for the index must be performed again.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
To save connecting and searching for a given grub menu for each test,
ktest.pl will cache the grub number it found. The problem is that
different tests might use a different grub menu, but ktest.pl will
ignore it.
Instead, have ktest.pl check if the grub menu it used to cache the
content is the same as when it grabbed the menu. If not, grab it again,
otherwise just return the cached value.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The index of a line where a warning is tested can be returned
differently on different versions of gcc (or same version compiled
differently). That is, a tab + space can give different results. This
causes the warning check to produce a false positive. Removing the
index from the check fixes this issue.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The reboot just wants to get to the next kernel. But if a warning (Call
Trace) appears, the monitor will report an error, and the reboot will
think something went wrong and power cycle the box, even though we
successfully made it to the next kernel.
Ignore warnings during the reboot until we get to the next kernel. It
will still timeout if we never get to the next kernel and then a power
cycle will happen. That's what we want it to do.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Sometimes when a test kernel passed fine, but on reboot it crashed,
ktest could get stuck and not proceed. This would be frustrating if you
let a test run overnight to find out the next morning that it was stuck
on the first test.
To fix this, I made reboot check for the REBOOT_SUCCESS_LINE. If the
line was not detected, then it would power cycle the box.
What it didn't cover was if the REBOOT_SUCCESS_LINE wasn't defined or if
a 'good' kernel did not display the line. Instead have it search for the
Linux banner "Linux version". The reboot just needs to get to the start
of the next kernel, it does not need to test if the next kernel makes it
to a boot prompt.
After we find the next kernel has booted, then we just wait for either
the REBOOT_SUCCESS_LINE to appear or the timeout.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Although the patchcheck test checks for warnings in the files that were
changed, this check does not catch warnings that were caused by header
file changes and the warnings appear in C files not touched by the
commit.
Add a new option called WARNINGS_FILE. If this option is set, then the
file it points to is read before bulid, and the file should contain a
list of known warnings. If a warning appears in the build, this file is
checked, and if the warning does not exist in this file, then it fails
the build showing the new warning.
If the WARNINGS_FILE points to a file that does not exist, this will
cause any warning in the build to fail.
A new test is also added called "make_warnings_file". This test will
create do a build and record any warnings it finds into the
WARNINGS_FILE. This test is something that can be run before other tests
to build a warnings file of "known warnings", ie, warnings that were
there before your changes.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Options are allowed to use other options, for example:
LOG_FILE = ${OUTPUT_DIR}/${MACHINE}.log
where the option LOG_FILE used the options OUTPUT_DIR and MACHINE.
But if a test option were to use a default option, it will not get
substituted:
OUTPUT_DIR = ${THIS_DIR}/${MACHINE}
TEST_START
OUTPUT_DIR = ${OUTPUT_DIR}/t1
For the above test, OUTPUT_DIR will stay literally "${OUTPUT_DIR}/t1"
and not be converted to "${THIS_DIR}/${MACHINE}/t1". When the test runs,
it will pass the ${OUTPUT_DIR} to the shell, which would probaly
interpret it as "", and the output directory will end up as "/t1".
Change the code where if a test option has its own option name in
its defined field, and a default option exists, then substitute the
default option in its place.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The patchcheck test looks at what files are modified for each patch it
checks and makes sure that those files do not produce any warnings.
Unfortunately, when it read the diffstat, the newlines were added on the
files and this made compares miss warnings, and commits that should not
have passed, ktest let pass.
Fix this by using the perl command "chomp" that strips off whitespace at
the end of lines.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If the user is doing a build or install bisect, there's no reason to
have them define CONSOLE, as the console does not need to be read. The
console only needs to be read for boot tests.
CONSOLE is not required for normal build or install tests, let's not
require it for bisect tests with BISECT_TYPE of build or install.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Sometimes a test kernel will crash or hang on reboot (this is even more
apparent when testing a config without CGROUPS on a box running
systemd). When this happens, on the next iteration of installing a
kernel, ktest will fail when it tries to install.
Have ktest do a check to see if the target can be connected to via ssh
before it tries to install. If it can't connect, then reboot again.
This time the reboot will fail because it can't connect and will force a
power cycle.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Commit fb16d891 "kconfig: replace 'oldnoconfig' with 'olddefconfig', and
keep the old name", changed ktest's default config update from
oldnoconfig to olddefconfig without adding oldnoconfig as a backup.
The make oldnoconfig works much better than its backup of:
yes '' | make oldconfig
But due to this change, and the fact that ktest is used to build lots of
older kernels (and for bisects), it forgoes the oldnoconfig completely.
Cc: Adam Lee <adam8157@gmail.com>
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
I installed Fedora 17 which no longer supports grub v1. I worked
with grub2 for a while, but there's so many issues with it and automated
rebooting, that I decided to switch to syslinux. Instead of using
the REBOOT_SCRIPT and add customized changes to get syslinux booted,
I thought it better to make ktest aware of syslinux and add options
to simplify the use of syslinux on a target test box.
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: John 'Warthog9' Hawley <warthog9@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Before rebooting the target, run the sync command, as it seems that
either Grub2 or systemd gets screwed up if you update to reboot a kernel
once and do a reboot without doing a sync.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
As only grub or 'script' is supported for rebooting to a new kernel,
and Fedora 17 has dropped support for grub, I decided to add grub2
support as well (I also plan on adding syslinux/extlinux support too).
The options GRUB_FILE and GRUB_REBOOT were added to allow the user
to specify where to find the grub.cfg and what tool to use to reboot
into the next kernel respectively.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
In order to decide if ktest should bother installing modules on the
target box, it checks if the config file has CONFIG_MODULES=y. But it
also checks if the '=y' part exists. It only will install modules if the
config exists and is set with '=y'. But as the regex that was used
tests:
/^CONFIG_MODULES(=y)?/
this will also match:
CONFIG_MODULES_USE_ELF_RELA
as the '=y' part was optional and it did not test the rest of the line.
When this happens, ktest will stop checking the rest of the configs but
it will also think that no modules are needed to be installed. What it
should do is only jump out of the loop if it actually found a
CONFIG_MODULES that is set to true.
Otherwise, ktest wont install the necessary modules needed for proper
booting of the test target.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Pull kconfig changes from Michal Marek:
"kconfig in v3.7 is going to
- initialize ncurses only once in menuconfig
- be able to jump to a search result in menuconfig
- change the misnomer oldnoconfig to a more meaningful name
olddefconfig, keeping the old name as alias"
* 'kconfig' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kconfig: replace 'oldnoconfig' with 'olddefconfig', and keep the old name as an alias
menuconfig: Assign jump keys per-page instead of globally
menuconfig: Do not open code textbox scroll up/down
menuconfig: Add jump keys to search results
menuconfig: Extend dialog_textbox so that it can return to a scrolled position
menuconfig: Extend dialog_textbox so that it can exit on arbitrary keypresses
menuconfig: Remove superfluous conditionnal
kconfig: document oldnoconfig to what it really does in conf.c
kconfig/mconf.c: revision of curses initialization.
As 67d34a6a39 said, 'oldnoconfig' doesn't
set new symbols to 'n', but instead sets it to their default values.
So, this patch replaces 'oldnoconfig' with 'olddefconfig', stop making
people confused, and keep the old name 'oldnoconfig' as an alias,
because people already are dependent on its behavior with the
counter-intuitive name.
Signed-off-by: Adam Lee <adam8157@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
The ELSE IF statements do not work as expected if another ELSE statement
follows. This is because the $if_set is not set. If the ELSE IF
condition is true, the following ELSE should be ignored. But because the
$if_set is not set, the following ELSE will also be executed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Add '=~' and '!~' to the list of allowed conditionals for DEFAULT and
TEST_START section if statements.
ie.
TEST_START IF TEST =~ .*test$
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The option IGNORE_ERRORS is used to allow a test to succeed even if a
warning appears from the kernel. Sometimes kernels will produce warnings
that are not associated with a test, and the user wants to test
something else.
The IGNORE_ERRORS works for boot up, but was not preventing test runs to
succeed if the kernel produced a warning.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The min configs are saved in a perl hash called force_configs, and this
hash is used to add configs to the .config file. But it was not being
reset between tests and a min config from a previous test would affect
the min config of the next test causing undesirable results.
Reset the force_config hash at the start of each test.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Usually the target is booted into a dependable kernel when a test
starts. The test will install the test kernel and reboot the box. But
there may be a time that the kernel is running an unreliable kernel and
the reboot may crash.
Have ktest detect crashes on a reboot and force a power-cycle instead.
This can usually happen if a test kernel was installed to run manual
tests, but the user forgot to reboot to the known good kernel.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If the console is constantly outputting content, this can cause ktest
to get stuck waiting on the monitor to settle down.
The option MAX_MONITOR_WAIT is the maximum time (in seconds) for ktest
to wait for the console to flush.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
With a name like 'oldnoconfig' one may think that the config generated
would disable all configs that were not defined (selecting "no" for all
options). But this is not the case. It selects the default. If a config
has a 'default y', then it is added if not specified.
This broke the config bisect, because options not specified by a config
will just use the default, where it expected to turn off. This caused an
option to be enabled that disabled an option that would break the build.
The end result was that we never found the bad config at the end of the
test.
Instead of using 'make oldnoconfig', ktest now builds the options it
expects enabled and disabled. When it turns off an option, it will no
longer remove it, but actually set it to:
# CONFIG_FOO is not set.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The config-bisect can take a bad config and bisect it down to find out
what config actually breaks the config. But as all tests will apply a
minconfig (defined by a user) to apply before booting, it is possible
that the minconfig could actually make the bad config work (minconfigs
can disable configs). The end result is that the config bisect test will
not find a config that breaks. This can be rather frustrating to the
user.
The CONFIG_BISECT_CHECK option, when set to 1, will make sure that the
bad config (with the minconfig applied) still fails before trying to
bisect.
And yes, I did get burned by this.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Add the PRE_INSTALL option that will allow a user to specify a shell
command to be executed before the install operation executes.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
In order to let the user add commands before and after ktest runs, the
PRE_KTEST and POST_KTEST options are defined. They hold shell commands
that will execute befor ktest runs its first test, as well as when it
completed its last test.
The PRE_TEST and POST_TEST will be run befor and after (respectively)
for a given test. They can either be global (done for all tests) or
defined by a single test.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If the file that OUTPUT_MIN_CONFIG exists then ktest.pl will prompt the
user and ask them if the OUTPUT_MIN_CONFIG should be used as the
starting point for make_min_config instead of MIN_CONFIG.
This is usually the case, and to allow the user to do so, which is
helpful if the user is creating different min configs based on tests,
and they know one is a superset of another test, they can set
USE_OUTPUT_MIN_CONFIG to one, which will prevent kest.pl from prompting
to use the OUTPUT_MIN_CONFIG and it will just use it.
If USE_OUTPUT_MIN_CONIFG is set to zero, then ktest.pl will continue to
use MIN_CONFIG instead.
The default is that USE_OUTPUT_MIN_CONFIG is undefined.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Add a MIN_CONFIG_TYPE that can be set to 'test' or 'boot'. The default
is 'boot' which is what make_min_config has done previously: makes a
config file that is the minimum needed to boot the target.
But when MIN_CONFIG_TYPE is set to 'test', not only must the target
boot, but it must also successfully run the TEST. This allows the
creation of a config file that is the minimum to boot and also
perform ssh to the target, or anything else a developer wants.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The PRE_BUILD and POST_BUILD options of ktest are added to allow the
user to add temporary patch to the system and remove it on builds. This
is sometimes use to take a change from another git branch and add it to
a series without the fix so that this series can be tested, when an
unrelated bug exists in the series.
The problem comes when a tagged commit is being used. For example, if
v3.2 is being tested, and we add a patch to it, the kernelrelease for
that commit will be 3.2.0+, but without the patch the version will be
3.2.0. This can cause problems when the kernelrelease is determined for
creating the /lib/modules directory. The kernel booting has the '+' but
the module directory will not, and the modules will be missing for that
boot, and may not allow the kernel to succeed.
The fix is to put the creation of the kernelrelease in the POST_BUILD
logic, before it applies the POST_BUILD operation. The POST_BUILD is
where the patch may be removed, removing the '+' from the kernelrelease.
The calculation of the kernelrelease will also stay in its current
location but will be ignored if it was already calculated previously.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The change to let individual tests decide to reboot the machine on
success of the entire test also prevented errors from rebooting
when an error was detected.
The "no_reboot" variable was only cleared if the test had
reboot_on_success set. But the no_reboot variable also prevents the test
rebooting when an error was detected even when REBOOT_ON_ERROR was set.
Add a new "reboot_success" variable that is used to determine if the
test should reboot on success and not touch the no_reboot variable.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When BISECT_REVERSE and BISECT_SKIP are used together with boot or test
testing, build failures are treated as boot or test failures and
'git bisect bad' is executed instead of 'git bisect skip'. This is because
the $ret value of -1 is treated as a build failure, but the $reverse_bisect
logic does not properly handle this.
Simple fix, only invert it if it is positive.
Link: http://lkml.kernel.org/r/1335235380-8509-1-git-send-email-Russ.Dill@ti.com
Signed-off-by: Russ Dill <Russ.Dill@ti.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
