Get rid of static Atffiles and let bsd.test.mk generate them on the fly.

branches: 1.1.4;
Add preliminary tests (basically a placeholder for now) for the new modules
framework. At the moment, this just tests for plain load (i.e. no arguments
passed to load) and later unload of the module through the modctl(2) system
call. The tools are not tested yet.

Add tests on the x86 PTEs. We scan the MMU page tables directly and verify
certain properties.

Centralize the base rump libraries into a variable used by all the other
Makefiles so that we can make changes to it centrally as needed and have
less mess. Fixes the sun2 build that needs rumpvfs after librump after
the latest changes.

Re-arrange the ufetchstore tests to look like the other ones.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

Add KCOV - kernel code coverage tracing device

The KCOV driver implements collection of code coverage inside the kernel.
It can be enabled on a per process basis from userland, allowing the kernel
program counter to be collected during syscalls triggered by the same
process.

The device is oriented towards kernel fuzzers, in particular syzkaller.

Currently the only supported coverage type is -fsanitize-coverage=trace-pc.

The KCOV driver was initially developed in Linux. A driver based on the
same concept was then implemented in FreeBSD and OpenBSD.

Documentation is borrowed from OpenBSD and ATF tests from FreeBSD.

This patch has been prepared by Siddharth Muralee, improved by <maxv>
and polished by myself before importing into the mainline tree.

All ATF tests pass.

PR/53908: Alex Raschi: Test that require modules belong in modules; move the
threadpool test from kernel to modules.

branches: 1.13.2; 1.13.4;
PR 52864: add a test for this bug, from sevan@

Define _KERNTYPES for things that need it.

Fix static linking for the tests: -lrump is also used by -lrumpuser,
so we also need -lrump after -lrumpuser. Fixes build for sun2.

branches: 1.10.2; 1.10.8;
Add a helper program that loads a module and returns the errno from modctl(8).

XXX: As noted on tech-kern@, the syscall should be modified to fail
consistently with a given predefined errno on non-MODULAR kernels.

Add a test to ensure that PAGE_SIZE is available in kernel modules.
This test reproduces the error condition in PR port-macppc/46041 and
therefore it is an xfail in this particular platform.

branches: 1.8.6; 1.8.8;
tests for builtin modules

Fix botched update.

Add a new atf test-case to check that recursive module calls actually work.

Get rid of static Atffiles and let bsd.test.mk generate them on the fly.

Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

Convert NetBSD-specific tests that were previously written in C++ to C now
that ATF provides a C-only binding (comes with 0.5).

branches: 1.2.2; 1.2.4;
Add tests for load-time parameter passing to modules, both at the syscall
level through modctl(2) and at the user level through the modload(8)
utility.

branches: 1.1.2;
Add preliminary tests (basically a placeholder for now) for the new modules
framework. At the moment, this just tests for plain load (i.e. no arguments
passed to load) and later unload of the module through the modctl(2) system
call. The tools are not tested yet.

Define _KERNTYPES for things that need it.

add Makefile.inc everywhere so that we can set WARNS=4 by default. Amazing
how many bugs this found :-)

tests: replace deprecated '-s eq:...' with '-s exit:...'

branches: 1.3.2;
For now, skip module tests if modctl(8) fails either with EPERM or ENOSYS.

branches: 1.2.2;
Skip Xen. XXX: There should be a reliable way to detect MODULAR.

Add a test to ensure that PAGE_SIZE is available in kernel modules.
This test reproduces the error condition in PR port-macppc/46041 and
therefore it is an xfail in this particular platform.

Avoid undefined behavior in disabledstat

t_builtin.c:174:16, member access within misaligned address 0x741271c25004
for type 'struct modstat_t'

t_builtin.c:175:4, member access within misaligned address 0x741271c251c4
for type 'struct modstat_t'

Merge the [pgoyette-compat] branch

branches: 1.3.12; 1.3.14;
Don't play with "../.." in includes for h_macros.h; deal with it centrally.
Minor fixes.

branches: 1.2.28;
add Makefile.inc everywhere so that we can set WARNS=4 by default. Amazing
how many bugs this found :-)

tests for builtin modules

branches: 1.10.2;
Drop no longer needed macros KCOV_STORE() KCOV_LOAD() in kcov(4)

Corrently KCOV_STORE() and KCOV_LOAD() are equivalent to x=y.

Obtained from <R3x>

Fix flaky check for the /dev/kcov device in t_kcov

Add a top-level check for the KCOV device, instead of deferring it ot each
thread. The thread-based solution was flaky.

PR kern/54064 by Andreas Gustafsson

Introduce enhancements to the kcov(4) code

Add new tests verifying dup2(2) scenarios:
- kcov_dup2
- kcov_basic_dup2_pc
- kcov_basic_dup2_cmp

The dup2(2) trick is used by syzkaller and assert that it works.
All new tests pass.

While there add minor non-functional cleanup changes.

Add support for trace type selection in kcov(4)

Allow to specify mode in KCOV_IOC_ENABLE synchronizing the functionality
with Linux, FreeBSD and OpenBSD. As a NetBSD (and OpenBSD) specific of
the ioctl(2) interface, the mode argument has to be specified as &value
rather than value.

There are 3 modes available:
1. KCOV_MODE_NONE -- no trace specified, useful for testing purposes
2. KCOV_MODE_TRACE_PC -- trace the kernel program counter
3. KCOV_MODE_TRACE_CMP -- trace comparison instructions and switch statements

Adapt the ATF tests and documentation for new API.

The KCOV_MODE_TRACE_CMP mode is implemented but still awaits for the
GCC 8.x upgrade or selection of Clang/LLVM as the kernel compiler.

Obtained from OpenBSD and adapted for NetBSD by myself.

Fix bug in kcov_multiple_threads in t_kcov

Spawn the expected number of threads rather than hardcoding one value for
all tests.

Add support for multiple threads in kcov(4)

Reuse the fd_clone() API to associate kcov descriptors (KD) with a file
descriptor. Each fd (/dev/kcov) can be reused for a single LWP.

Add new ATF regression tests and cleanup existing code there. All tests
pass.

Refresh the kcov(4) man page documentation.

Developed with help from <maxv>.

Fix build of t_kcov

Correct the syntax that used to work in earlier uncommitted version.

Add KCOV_LOAD() and KCOV_STORE() - new helper macros

New macros prefer 64-bit atomic operations whenever accessible.

As a fallback they use volatile move operations that are not known
to have negative effect in KCOV even if interrupted in the middle of
operation.

Enable kcov_basic and kcov_thread tests on targets without
__HAVE_ATOMIC64_OPS.

Fix build of kcov tests on CPUs without 64-bit atomics

Restrict the 64-bit atomics to ports defining __HAVE_ATOMIC64_OPS.

Using 64-bit atomics is still good for i586 fuzzing, but the tests are
build for earlier CPUs. This makes this code to be disabled in i386 builds.

There is a similar situation with few other ports that offer 64-bit atomics
in certain CPU models and ABIs.

Add KCOV - kernel code coverage tracing device

The KCOV driver implements collection of code coverage inside the kernel.
It can be enabled on a per process basis from userland, allowing the kernel
program counter to be collected during syscalls triggered by the same
process.

The device is oriented towards kernel fuzzers, in particular syzkaller.

Currently the only supported coverage type is -fsanitize-coverage=trace-pc.

The KCOV driver was initially developed in Linux. A driver based on the
same concept was then implemented in FreeBSD and OpenBSD.

Documentation is borrowed from OpenBSD and ATF tests from FreeBSD.

This patch has been prepared by Siddharth Muralee, improved by <maxv>
and polished by myself before importing into the mainline tree.

All ATF tests pass.

tests: replace deprecated '-s eq:...' with '-s exit:...'

delete attribution per request of contributor

Skip the test on non-modular kernels

PR 52864: add a test for this bug, from sevan@

Update the t_modctl test to ensure that static evcnts are added.

While here, remove the explicit call to sysctl_setup() routine,
since the module infrastructure already invokes such routines
automatically.

Avoid undefined behavior in get_modstat_info

t_modctl.c:114:16, member access within misaligned address 0x71bf5bcede84
for type 'struct modstat_t'

t_modctl.c:116:13, load of misaligned address 0x7e81bc3c9104 for type
'struct modstat_t' which requires 8 byte alignment

wether -> whether

Merge the [pgoyette-compat] branch

branches: 1.12.30; 1.12.32;
Fix stupid thinko when checking for the availability of options MODULAR
and permission to load modules.

Be more verbose if we get unexpected error codes from modctl.

Make use of the new MODCTL_EXISTS check to query kernel wether (and why
not) we can load modules.

Avoid zero-length format string.

Add __printflike attribution to use vprintf and friends with an argument
as format string.

A modctl(MODCTL_STAT, ...) always succeeds, even in non-modular kernels such
as Xen domUs. Thus, refactor the code to detect non-modular kernels; when the
other operations (MODCTL_LOAD or MODCTL_UNLOAD) fail specifically with ENOSYS,
this is taken as a hint about non-modularity and the test is skipped.

Skip the tests even if modctl(2) does not fail specifically with ENOSYS.

branches: 1.5.6;
add Makefile.inc everywhere so that we can set WARNS=4 by default. Amazing
how many bugs this found :-)

Add a new atf test-case to check that recursive module calls actually work.

Kernel modules (kmod files) are now installed in a subdirectory per module.
Make these tests aware of this so that they work again.

branches: 1.2.4; 1.2.8;
- Add a bsd.kmodule.mk to build new style modules.
- Set the file suffix to .kmod

Convert NetBSD-specific tests that were previously written in C++ to C now
that ATF provides a C-only binding (comes with 0.5).

Convert NetBSD-specific tests that were previously written in C++ to C now
that ATF provides a C-only binding (comes with 0.5).

Convert TNF licenses to new 2 clause variant

branches: 1.3.2; 1.3.4;
Add tests for load-time parameter passing to modules, both at the syscall
level through modctl(2) and at the user level through the modload(8)
utility.

branches: 1.2.2;
Add more detailed checks for each of the primary modctl commands.

Add preliminary tests (basically a placeholder for now) for the new modules
framework. At the moment, this just tests for plain load (i.e. no arguments
passed to load) and later unload of the module through the modctl(2) system
call. The tools are not tested yet.

tests: replace deprecated '-s eq:...' with '-s exit:...'

For now, skip module tests if modctl(8) fails either with EPERM or ENOSYS.

The k_helper3 program does not need to be (and shouldn't be) installed
in its own subdirectory.

Add a helper program that loads a module and returns the errno from modctl(8).

XXX: As noted on tech-kern@, the syscall should be modified to fail
consistently with a given predefined errno on non-MODULAR kernels.

Skip Xen. XXX: There should be a reliable way to detect MODULAR.

branches: 1.9.4;
Prevent failures from cleanup routines. Do so by only attempting to do the
cleanup if the test case did not succeed (as, when it succeeds, the cleanup
has already happened).

Adjusts tests after import of atf-0.12:

- The use.fs property is gone.
- Mark the tests/fs/t_create:attrs test as broken when using the default
unprivileged-user:_atf setting. This probably deserves a fix somehow
but I'm not sure at this point.

Properly mark some test cases as having a cleanup routine. Stupidity of
the API? Most likely; will revise it.

Fixes atf-run breaking when running t_psshfs. This does not resolve the
underlying issue though, which is atf-run getting confused trying to
unmount the temporary mount point by itself (I think). (I'm now wondering
if atf should be bothering about unmounting stuff at all. Maybe not. It is
a tricky and uncommon thing.)

atf-0.9 introduces an use.fs test-case property to allow tests to write to
their work directory. The purpose is to be able to know which tests intend
to touch the file system and to allow a minor optimization in atf-run.

Define use.fs=true for all those tests requiring it. (This highlights that
some tests currently require modifying the file system but conceptually
they shouldn't be... which leaves room for further improvements/cleanups
later :-)

Adjust tests to work with the new API in ATF 0.6.

Kernel modules (kmod files) are now installed in a subdirectory per module.
Make these tests aware of this so that they work again.

branches: 1.3.6;
- Add a bsd.kmodule.mk to build new style modules.
- Set the file suffix to .kmod

Convert TNF licenses to new 2 clause variant

branches: 1.1.2; 1.1.4; 1.1.6;
Add tests for load-time parameter passing to modules, both at the syscall
level through modctl(2) and at the user level through the modload(8)
utility.

fix a/an grammar in obvious cases

branches: 1.2.2;
tests: replace deprecated '-s eq:...' with '-s exit:...'

branches: 1.1.2; 1.1.4;
PR/53908: Alex Raschi: One more file needs to move.

branches: 1.1.2;
Re-arrange the ufetchstore tests to look like the other ones.

Split in sub-tests for clarity, and add a new test, marked as expected
failure for now.

Add tests on the x86 PTEs. We scan the MMU page tables directly and verify
certain properties.

KMODULEDIR is no longer absolute (${DESTDIR} is added to it by the system
makefile because we need to install the module in two places the debug part
and the regular part)

Get rid of static Atffiles and let bsd.test.mk generate them on the fly.

Fix two bugs to make MKMODULAR=yes builds will work again, for the
first time.

share/mk/bsd.kmodule.mk: only set _INST_DIRS if KMODULEDIR is not
set. That we we avoid installing some bogus directories in
the DESTDIR and in $DESTDIR/METALOG.

tests/modules/k_helper/Makefile: add missing .include <bsd.own.mk>,
and set the KMODULEDIR that the author seems to have intended.

branches: 1.3.2;
Fix MKMODULAR=yes builds by resolving installation problems of the k_helper
test module after the introduction of bsd.kmodule.mk. The files list was
inconsistent with the new module structure and the Makefile did not use the
correct variable to specify the installation of the module. Hi ad@!

- Add a bsd.kmodule.mk to build new style modules.
- Set the file suffix to .kmod

branches: 1.1.4; 1.1.6;
Add preliminary tests (basically a placeholder for now) for the new modules
framework. At the moment, this just tests for plain load (i.e. no arguments
passed to load) and later unload of the module through the modctl(2) system
call. The tools are not tested yet.

Update the t_modctl test to ensure that static evcnts are added.

While here, remove the explicit call to sysctl_setup() routine,
since the module infrastructure already invokes such routines
automatically.

branches: 1.6.32;
Fix build - delete extra &

branches: 1.5.6;
add Makefile.inc everywhere so that we can set WARNS=4 by default. Amazing
how many bugs this found :-)

Add a new atf test-case to check that recursive module calls actually work.

Remove clause 3 and 4 from TNF licenses

branches: 1.2.2; 1.2.4;
Add tests for load-time parameter passing to modules, both at the syscall
level through modctl(2) and at the user level through the modload(8)
utility.

branches: 1.1.2;
Add preliminary tests (basically a placeholder for now) for the new modules
framework. At the moment, this just tests for plain load (i.e. no arguments
passed to load) and later unload of the module through the modctl(2) system
call. The tools are not tested yet.

KMODULEDIR is no longer absolute (${DESTDIR} is added to it by the system
makefile because we need to install the module in two places the debug part
and the regular part)

Add a new atf test-case to check that recursive module calls actually work.

add Makefile.inc everywhere so that we can set WARNS=4 by default. Amazing
how many bugs this found :-)

Add a new atf test-case to check that recursive module calls actually work.

branches: 1.4.2;
The k_helper3 program does not need to be (and shouldn't be) installed
in its own subdirectory.

Let's get it right this time.

Put k_helper3 in the correct location.

Add a helper program that loads a module and returns the errno from modctl(8).

XXX: As noted on tech-kern@, the syscall should be modified to fail
consistently with a given predefined errno on non-MODULAR kernels.

While we need an absolute path when loading a module (including the .kmod
extension), we want the shortname w/o extension for the unload.

Adjust.

branches: 1.1.2;
Add a helper program that loads a module and returns the errno from modctl(8).

XXX: As noted on tech-kern@, the syscall should be modified to fail
consistently with a given predefined errno on non-MODULAR kernels.

KMODULEDIR is no longer absolute (${DESTDIR} is added to it by the system
makefile because we need to install the module in two places the debug part
and the regular part)

branches: 1.1.2; 1.1.4;
Add a test to ensure that PAGE_SIZE is available in kernel modules.
This test reproduces the error condition in PR port-macppc/46041 and
therefore it is an xfail in this particular platform.

branches: 1.1.2; 1.1.4;
Add a test to ensure that PAGE_SIZE is available in kernel modules.
This test reproduces the error condition in PR port-macppc/46041 and
therefore it is an xfail in this particular platform.

KMODULEDIR is no longer absolute (${DESTDIR} is added to it by the system
makefile because we need to install the module in two places the debug part
and the regular part)

branches: 1.1.2; 1.1.4;
PR/53908: Alex Raschi: Test that require modules belong in modules; move the
threadpool test from kernel to modules.

branches: 1.1.2; 1.1.4;
PR/53908: Alex Raschi: Test that require modules belong in modules; move the
threadpool test from kernel to modules.

KMODULEDIR is no longer absolute (${DESTDIR} is added to it by the system
makefile because we need to install the module in two places the debug part
and the regular part)

branches: 1.3.2;
Re-arrange the ufetchstore tests to look like the other ones.

fix TESTSDIR.
http://releng.netbsd.org/builds/HEAD/201904150550Z/evbarm-earm.build.failed

Add missing Makefile.

Grabbed from thorpej's github repository.
XXX not actually built tested...

Re-arrange the ufetchstore tests to look like the other ones.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

branches: 1.1.2;
Re-arrange the ufetchstore tests to look like the other ones.

branches: 1.4.2;
Add a test to make sure an overflowing-into-kernel-space fetch/store near
vm_max_address fails with EFAULT.

Revert previous. arc port was fixed correctly.

Avoid redefining UADDR for arc.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

branches: 1.1.2;
Re-arrange the ufetchstore tests to look like the other ones.

KMODULEDIR is no longer absolute (${DESTDIR} is added to it by the system
makefile because we need to install the module in two places the debug part
and the regular part)

Add tests on the x86 PTEs. We scan the MMU page tables directly and verify
certain properties.

requires pmap_private.h now.

Split in sub-tests for clarity, and add a new test, marked as expected
failure for now.

Add tests on the x86 PTEs. We scan the MMU page tables directly and verify
certain properties.