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

Add a fast, platform independent hash function to libc.
The algorithm used is the Jenkins hash. The name (mi_vector_hash)
reflects the nature of the hash function.
Add glue for libc ATF tests and include a test case to make sure that
(mis)alignment and endianess are handled correctly.

Bump libc minor to 169.

tests/lib/Makefile: Hook up libstdc++.

branches: 1.35.2;
lua libm API from Phil Rulon

add some basic tests for libossaudio

since there are already tests for audio this is focused on making
sure the ioctls translate properly and implement the necessary OSS
compat quirks.

right now this only covers the dsp ioctls, it should also cover the
mixer ioctls, although that's significantly harder.

Add tests for USER_LDT.

new tests

branches: 1.31.2;
Add initial tests for libnvmm's Mem Assist, with 8 test cases.

branches: 1.30.2; 1.30.4;
PR lib/52007

Move libevent from being a test playing sub-directory, to a groupy,
just hanging around, hoping someone will notice it, and throw it
a bone... (mixed metaphors?)

branches: 1.29.2;
Remove MKCRYPTO option.

Originally, MKCRYPTO was introduced because the United States
classified cryptography as a munition and restricted its export. The
export controls were substantially relaxed fifteen years ago, and are
essentially irrelevant for software with published source code.

In the intervening time, nobody bothered to remove the option after
its motivation -- the US export restriction -- was eliminated. I'm
not aware of any other operating system that has a similar option; I
expect it is mainly out of apathy for churn that we still have it.
Today, cryptography is an essential part of modern computing -- you
can't use the internet responsibly without cryptography.

The position of the TNF board of directors is that TNF makes no
representation that MKCRYPTO=no satisfies any country's cryptography
regulations.

My personal position is that the availability of cryptography is a
basic human right; that any local laws restricting it to a privileged
few are fundamentally immoral; and that it is wrong for developers to
spend effort crippling cryptography to work around such laws.

As proposed on tech-crypto, tech-security, and tech-userlevel to no
objections:

https://mail-index.netbsd.org/tech-crypto/2017/05/06/msg000719.html
https://mail-index.netbsd.org/tech-security/2017/05/06/msg000928.html
https://mail-index.netbsd.org/tech-userlevel/2017/05/06/msg010547.html

P.S. Reviewing all the uses of MKCRYPTO in src revealed a lot of
*bad* crypto that was conditional on it, e.g. DES in telnet... That
should probably be removed too, but on the grounds that it is bad,
not on the grounds that it is (nominally) crypto.

libpthread_dbg(3) deletion from the base distribution

libpthread_dbg(3) is a remnant library from the M:N thread model
(pre-NetBSD-5.0) API to introspect threads within a process and for use
of debuggers.

Currently in the 1:1 model it's not used in GDB neither in LLDB and it's
not either planned to be used. It's current function to read pthread_t
structures is realizable within a regular debugger capable to
instrospect objects within a tracee (GDB, LLDB...).

Remaining users of this API can still use this library from
pkgsrc/devel/libpthread_dbg.

Sponsored by <The NetBSD Foundation>

branches: 1.27.2;
Add new test-suite t_dummy for libpthread_dbg

At the moment this test does nothing except reports failure from td_open()
for overloaded (implemented) dummy1_proc_lookup() (.proc_lookup from
td_proc_callbacks_t) of the following form:

static int
dummy1_proc_lookup(void *arg, const char *sym, caddr_t *addr)
{
return TD_ERR_ERR;
}

This file and directory with tests is placeholder for new ones, without
further need to alter mtree and distribution sets.

The libpthread_dbg interface and library is used by gdb(1) to handle
threads in applications.

Sponsored by <The NetBSD Foundation>

Add some tests for librefuse

branches: 1.25.2;
tests for libusbhid (many of which fail)

Don't build tests that depend on RUMP if BSD_MK_COMPAT_FILE is defined.

Hook Lutok into the build. This is all protected by the MKKYUA guard.

Build libbpfjit test to the build.

Build sljit test when MKSLJIT != no and set MKSLJIT to yes on amd64 and i386.

branches: 1.20.2;
Exclude tests that use rump

one too many e's

hook in libexecinfo

Add a test for crypt salts. From:
http://cvsweb.openwall.com/cgi/cvsweb.cgi/Owl/packages/glibc/crypt_freesec.c
From Solar Designer

Hook in the libtre tests

branches: 1.15.2;
Build and install ppath(3) tests.

Add libcurses test

link libbluetooth tests to the build

Add test which checks rumpclient does not use fds 0-2 for its
internal purposes.

Add a simple test for the purpose of making sure rumphijack works
on -current in addition to 5.x. The test serves a simple index.html
on a hijacked bozohttpd and checks the file can be retrieved.

branches: 1.10.2;
Move the only regress/lib/libm test to the new atf format

Convert the libdes tests to atf.

Convert the libobjc tests to atf. While doing this, make the tests actually
check that things work.

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

Convert the libpthread tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

I think that this, together with the previous conversion of librt, obsoletes
the tests in the semaphore/ directory. Will investigate later.

Convert the libposix, libprop, librt, libskey and libutil tests to ATF.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

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

Add tests for semaphores. Note: the "unlink" one fails because our
implementation is broken. I'll file a PR shortly.

add ATF tests for libevent

Add a fast, platform independent hash function to libc.
The algorithm used is the Jenkins hash. The name (mi_vector_hash)
reflects the nature of the hash function.
Add glue for libc ATF tests and include a test case to make sure that
(mis)alignment and endianess are handled correctly.

Bump libc minor to 169.

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

tests/lib/csu, tests/libexec/ld.elf_so: Apply CTFMERGE=: workaround.

The workaround for

PR toolchain/59364: ctf tools needs update

required an update after fixing

PR toolchain/57241: mips64el--netbsd-install core dumps randomly

by splitting the recipes for ${PROG} and ${PROG}.debug into an
intermediate ${PROG}.link to avoid overwriting ${PROG} inside the
recipe for ${PROG}.debug.

This is kludgey (writing the `.link' suffix into a Makefile isn't
great) but I see only two cases of it so this'll do for now.

tests/lib/csu/Makefile: Use PROGS_CXX, not PROG_CXX.

There are multiple programs here.

Should fix:

nbmake: /home/riastradh/netbsd/current/src/share/mk/bsd.prog.mk:624: warning: Invalid character " " in variable name "OBJS.h_initfini1 h_initfini2 h_initfini3"

centralize pack-relative-relocs handling

Fix the clang build which complains about pack-relative-relocs ignored.

lib/csu: Test a static PIE with RELR relocations.

PR bin/59360: ld.elf_so(8): missing RELR support

tests/lib/csu: Build h_hello.o with -fPIE, not -fpie.

Static libraries like libc may overflow the bounds assumed by -fpie.
Resolves:

h_hello.o: in function `main':
h_hello.c:(.text.startup+0x30): relocation truncated to fit: R_SPARC_GOT13 against symbol `__sF' defined in .data.rel section in /tmp/build/2025.04.27.10.22.23-sparc64/destdir/usr/lib/libc.a(findfp.o)

PR lib/59359: static pies are broken

lib/csu: Add tests of {static, dynamic} x {pie, non-pie}.

Main goal here is to test static pies, but let's take advantage of
this to test the same code several ways, since the csu logic has some
differences for static vs dynamic builds too like the ordering of
_libc_init calls.

PR lib/59359: static pies are broken

branches: 1.7.12;
Add an initfini_array test that does not depend on the linker putting it
there automatically.

branches: 1.6.2;
Add ifunc support for statically linked applications on x86, ppc, sparc
and ARM.

branches: 1.5.14;
Replace LDADD with LDSTATIC.

Whitespace.

Test alignment of constructor / destructor calls as well as the stack
of new threads. Currently implement for i386 and AMD64.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

branches: 1.1.6; 1.1.12;
Test alignment of constructor / destructor calls as well as the stack
of new threads. Currently implement for i386 and AMD64.

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

branches: 1.1.4;
lib/csu: Add tests of {static, dynamic} x {pie, non-pie}.

Main goal here is to test static pies, but let's take advantage of
this to test the same code several ways, since the csu logic has some
differences for static vs dynamic builds too like the ordering of
_libc_init calls.

PR lib/59359: static pies are broken

Support ifunc on aarch64. The tests pass at least.

Wrap long #if defined(...) || ... and sort. NFCI.

Restore arm in the list of architectures supporting ifunc, accidently
lost in the OABI removal.

Good bye oabi! Tested building earmv7hf.

PR/56230: Jan-Benedict Glaw: arm oabi does not and will not have ifunc support.

branches: 1.2.2; 1.2.12;
Stub out on platforms without ifunc support in the linker.

Add ifunc support for statically linked applications on x86, ppc, sparc
and ARM.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Rename h_initfini3 dso now that it is a simple module.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

branches: 1.1.2; 1.1.4;
Add an initfini_array test that does not depend on the linker putting it
there automatically.

Test alignment of constructor / destructor calls as well as the stack
of new threads. Currently implement for i386 and AMD64.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Add an initfini_array test that does not depend on the linker putting it
there automatically.

branches: 1.4.38; 1.4.40;
Test LD_PRELOAD for stack alignment as well. This exposes issues on
AMD64.

branches: 1.3.6;
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.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

branches: 1.3.4;
lib/csu: Add support for RELR relocations in static PIEs.

Same logic as in ld.elf_so, but baked into the executable itself for
static PIEs.

Addendum to proposal on tech-userlevel:
https://mail-index.NetBSD.org/tech-userlevel/2025/04/27/msg014727.html

PR bin/59360: ld.elf_so(8): missing RELR support

lib/csu: Test a static PIE with RELR relocations.

PR bin/59360: ld.elf_so(8): missing RELR support

lib/csu: Add tests of {static, dynamic} x {pie, non-pie}.

Main goal here is to test static pies, but let's take advantage of
this to test the same code several ways, since the csu logic has some
differences for static vs dynamic builds too like the ordering of
_libc_init calls.

PR lib/59359: static pies are broken

test: fix typo and improve wording

Support ifunc on aarch64. The tests pass at least.

Move from uname -m to uname -p when checking if this target does support
ifunc - someone with a powerpc machine please double check!

branches: 1.1.2; 1.1.4;
Add ifunc support for statically linked applications on x86, ppc, sparc
and ARM.

branches: 1.1.4; 1.1.8;
Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

branches: 1.1.4; 1.1.8;
Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

branches: 1.1.4; 1.1.8;
Check stack alignment on hppa.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

Simplify

Actually add the check_stack_alignment implementation for x86.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

branches: 1.2.4; 1.2.8;
Fix comment: ar.bsp is the RSE backing store base pointer, not ar.bspstore

Add stack align test stub for ia64 native csu

branches: 1.1.4; 1.1.8;
Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

check the lower bits of the stack pointer,
not the lower bits of the upper half of the stack pointer.

branches: 1.1.4; 1.1.8; 1.1.12; 1.1.16;
Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

riscv: tests/lib/csu: Fix typo in check_stack_alignment()

Now, all tests in tests/lib/csu successfully pass for riscv64.

RISC-V support that works on QEMU with a single hart.

Thanks for Simon Burge for plic(4).

branches: 1.1.4; 1.1.8;
Add stack alignement test for sparc

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

branches: 1.1.4; 1.1.8;
Add stack alignement checking for sparc64

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

branches: 1.1.4; 1.1.8;
Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

Simplify

Actually add the check_stack_alignment implementation for x86.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Convert the csu tests to atf.
Initial work from the GSoC 2008 project by Lukasz Strzygowski.

Test alignment of constructor / destructor calls as well as the stack
of new threads. Currently implement for i386 and AMD64.

Simplify.

TESTSBASE is provided by bsd.own.mk now.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Test alignment of constructor / destructor calls as well as the stack
of new threads. Currently implement for i386 and AMD64.

Replace csu test case with a C++ based version. This makes the various
assembler routines obsolete. Be more exhaustive by testing dynamically
linked, statically linked and dynamically loaded.

XXX currently hard-codes /usr/tests due to limitations of bsd.test.mk

Kill file that should be pulled in from a different directory.
Fixes regression test for i386.

Test alignment of constructor / destructor calls as well as the stack
of new threads. Currently implement for i386 and AMD64.

tests/lib/libarchive: Link against libarchive statically.

The tests reach into libarchive guts, so they don't work with only
the public symbols, which are all that libarchive.so exports now.

PR lib/58829: libarchive exports excess symbols

makemandb(8): Fix static build with PROGDPLIBS.

XXX This shouldn't have to mention transitive dependencies -- we
should have a mechanism for upstream build goo for libarchive to
transmit the information automatically to downstream build goo for
makemandb.

Hook zstd to the build and enable it for libarchive and file.

Add more files, sort

branches: 1.12.2;
Add new test files for new libarchive version

One more no-longer-existing file we cannot install

More libarchive update fallout. There are no _2x files any more.

More libarchive build fallout. With these changes, "build.sh build"
works for me (amd64). Testing "build.sh release" in parallel with this
commit.

make it build (should add more tests)

Don't try to build test_compat_pax_libarchive_2x or test_read_pax_schily_xattr

Both of those were removed in the recent libarchive update:

Removed Files:
src/external/bsd/libarchive/dist: CONTRIBUTING.md
src/external/bsd/libarchive/dist/libarchive/test:
test_compat_pax_libarchive_2x.c
test_compat_pax_libarchive_2x.tar.Z.uu test_read_pax_schily_xattr.c
test_read_pax_schily_xattr.tar.uu

bsd.own.mk: rename GCC_NO_* to CC_WNO_*

Rename compiler-warning-disable variables from
GCC_NO_warning
to
CC_WNO_warning
where warning is the full warning name as used by the compiler.

GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

branches: 1.5.6;
tests/lib/libarchive: Suppress -Werror=stringop-truncation error

This logic correctly uses strncpy(3) to fully initialize a fixed-width field, and also ensures
NUL-termination on the next line as other users of the field expect.

Add -Werror=stringop-truncation to prevent build failure, when run with MKSANITIZER=yes.

Error was reported when build.sh was run with MKSANITIZER=yes flag.

Reviewed by: kamil@

branches: 1.4.4;
Fix static linking (sun2)

not require .d files for the build.

Fix build. Three list.h files are required to generate .d files

new tests

PR kern/55272: do not skip this test on single cpu machines - it is not
the only test causing the watchdog starvation and we better find and
fix the real issue.

PR/55612: Martin Husemann: libarchive tests pollute /tmp w/o cleanup
Set $TMPDIR to where we are.

Improve sentence

PR kern/55272: skip this test on uniprocessor machines, it is too dangerous
and can kill the host kernel if a userland watchdog is running

Skip this test on machines where it would cause serious swapping.

branches: 1.3.2;
Bump timeout to 10m

Bump timeout to 3600 - the libarchive tests take quite a while to
complete (on a nearly 1 GHz dual armv7 machine it takes more than 700s)

new tests

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

whitespace police

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

whitespace police

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

whitespace police

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

whitespace police

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

test that data element lists are not placed by sdp_put_attr(), as
an attribute/value pair should only contain a single data element

whitespace police

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

whitespace police

add tests for the following libbluetooth functions

bluetooth(3)
bt_aton()
bt_ntoa()

sdp_data(3)
sdp_data_type()
sdp_data_size()
sdp_get_data()
sdp_get_attr()
sdp_get_uuid()
sdp_get_bool()
sdp_get_uint()
sdp_get_int()
sdp_get_seq()
sdp_get_alt()
sdp_get_str()
sdp_get_url()
sdp_match_uuid16()
sdp_put_data()
sdp_put_attr()
sdp_put_uuid()
sdp_put_uuid16()
sdp_put_uuid2(3)
sdp_put_uuid128()
sdp_put_bool()
sdp_put_uint()
sdp_put_uint8()
sdp_put_uint16()
sdp_put_uint2(3)
sdp_put_uint64()
sdp_put_int()
sdp_put_int8()
sdp_put_int16()
sdp_put_int2(3)
sdp_put_int64()
sdp_put_seq()
sdp_put_alt()
sdp_put_str()
sdp_put_url()
sdp_set_bool()
sdp_set_uint()
sdp_set_int()
sdp_set_seq()
sdp_set_alt()

[all pass on i386]

Disable PAX mprotect to make just-in-time-compile tests work again.

Ok: Christos Zoulas

Add t_extmem test.

Add t_cop test.

libarch on sparc isn't -l${MACHINE_CPU}. Don't link to it.

Link with -l${MACHINE_CPU}.

branches: 1.1.2; 1.1.4; 1.1.10;
Add libbpfjit test.

avoid uninitialised variable.

found by GCC 12.

Add two more libbpfjit_jmp_jeq_x_noinit_XX tests.

Replace accidentally committed test with libbpfjit_jmp_jeq_x_noinit_ax test.

Improve libbpfjit_jmp_jeq_x test.

Add libbpfjit_jmp_ja_overflow test.

Add libbpfjit_jmp_ja_invalid test.

Add libbpfjit_ret_k and libbpfjit_bad_ret_k tests.

Add BPF_MOD tests. Plus one tiny change.

Add BPF_XOR tests.

Rename bpfjit tests to libbpfjit, don't test zero buflen and a couple of
cosmetic changes.

Implement copfuncs and external memory in bpfjit.

Update copyright year.

Test Array Bounds Check Elimination (ABC) and test bpf programs from bpf(4).

branches: 1.2.2;
- Add bpf_args_t and convert bpf_filter_ext() to use it. This allows the
caller to initialise (and re-use) the memory store.
- Add bpf_jit_generate() and bpf_jit_freecode() wrappers.

branches: 1.1.2; 1.1.4;
Add libbpfjit test.

branches: 1.4.2; 1.4.6;
Add bpfjit_cop_copx and bpfjit_copx_cop tests.

Rename bpfjit tests to libbpfjit.

Fix copyright. These tests are based on tests created in 2013 on github.com.

https://github.com/alnsn/bpfjit , files test/test_cop{,x}.c.

Add BPF_COP and BPF_COPX tests.

branches: 1.3.2; 1.3.6;
Add a link to ../../net/bpfjit/t_extmem.c.

Rename bpfjit tests to libbpfjit.

Add external memory tests.

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

remove "string" directory for now, it isn't installed

Add a regression test for the SHA2 family that also checks for unaligned
accesses.

Add popcount(3) and the long and long long version. Name is inspired by
gnulib, the implementation goes back to the AMD Software Optimizer
guide. A number of platforms will want to replace the C version with
assembler code using native instructions.

Add a fast, platform independent hash function to libc.
The algorithm used is the Jenkins hash. The name (mi_vector_hash)
reflects the nature of the hash function.
Add glue for libc ATF tests and include a test case to make sure that
(mis)alignment and endianess are handled correctly.

Bump libc minor to 169.

Add a new test for PR kern/56713 and set to expected_failure for now.

membar_ops(3): Add some automatic tests.

These tests run two threads for five seconds each to try to trigger
races in the event of broken memory barriers. They run only on
machines with at least two CPUs; on uniprocessor systems there's no
point -- the membars can correctly just be (instruction barrier)
no-ops.

Add tests for missing libc catalog entries

Add ATF tests for atomic_ops(3).

Register micro-UBSan ATF tests in the distribution

Populate distrib files, mtree lists and add the entry in Makefile to
include the new code.

branches: 1.47.24; 1.47.26;
Add a link-time test for __sync_* primitives (see PR 48368) - this
will allow us to notice missing functions during the build, instead
of when compiling arbitrary pkgsrc stuff later.

rename the only use of HAS_SSP in the tree to HAVE_SSP.

Don't include bsd.sys.mk; like sys.mk it is supposed to be used only
internally.

Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Add regression test for cdbr(3) and cdbw(3).

branches: 1.42.2; 1.42.8;
Remove ieeefp subdirectory.

Turn warns on for all tests and fix all the bugs.

branches: 1.40.2;
Add the new 'termios' subdirectory.

atf-ify the various locale tests

Add TLS support infrastructure. For dynamic binaries, ld.elf_so exports
_rtld_tls_allocate and _rtld_tls_free. libpthread uses this functions to
setup the thread private area of all new threads. ld.elf_so is
responsible for setting up the private area for the initial thread.
Similar functions are called from _libc_init for static binaries, using
dl_iterate_phdr to access the ELF Program Header.

Add test cases to exercise the different TLS storage models. Test cases
are compiled and installed on all platforms, but are skipped on
platforms not marked for TLS support.

This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.

It is inspired by the TLS support in FreeBSD by Doug Rabson and the
clean ups of the DragonFly port of the original FreeBSD modifications.

Moving t_hsearch - belongs in stdlib/

Another test moving: lib/libc/t_inet --> lib/libc/inet/t_inet_network

Last one for today: move t_ptm test from lib/libc/ to lib/libc/ttyio/

Continuing the (re)organization of the lib/libc atf regression tests

The nsdispatch test belongs in libc/net and not in libc

Move t_mktime and t_strptime to their own ..../time/ subdirectory.

(Concurrence from christos@)

Put the atexit test where it belongs - in libc/stdlib/

Move the servent and protoent tests to src/tests/lib/libc/net/ (where
they should have gone initially).

By request, put the getaddrinfo test in tests/lib/libc/net/

Atf-ify the getaddrinfo test, with updated "golden" output files.

test for sigqueue

Atf-ify the regex test

Atf-ify the rpc/t_xdr test

Atf-ify the db tests.

XXX Note that the original regress version of this test did not run the
XXX btree_delete test; that test is broken, and is disabled for now.

Atf-ify a couple more tests

Atf-ify test for mktime(3)

Atf-ify the nsdispath test.

Update the ieeefp tests so that they can be built on all architectures.
This enables us to avoid the set-list mess.

Build tested on amd64, i386, sun2, and vax!

Ooopppsss - locale/ tests are not yet ready - don't SUBIDR yet

Atf-ify the atexit test

Atf-ify servent and protoent tests, and the remaining hash test.

Atf-ify several of the ieeefp tests.

Note that t_round is currently mostly disabled, due to PR/44293. Also,
testfloat has not been converted.

Moving more tests out of regress and into atf mainstream

Ooopppsss - ieefp tests have not been atf-ified, so remove the stuff
that got committed by mistake.

Migrate a few miscellaneous tests from the old regress to atf

Move the various setjmp tests from regress to atf

Don't try to build the SSP tests if the target doesn't have SSP.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate J.T.Conklin's public-domain str* tests from regress to atf.

While here, do some clean-up and knf.

disable string again

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

glue glob tests

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

Disable population count test for now, please re-enable once
PR toolchain/42885 is fixed

Add a regression test for the SHA2 family that also checks for unaligned
accesses.

Add popcount(3) and the long and long long version. Name is inspired by
gnulib, the implementation goes back to the AMD Software Optimizer
guide. A number of platforms will want to replace the C version with
assembler code using native instructions.

Add a fast, platform independent hash function to libc.
The algorithm used is the Jenkins hash. The name (mi_vector_hash)
reflects the nature of the hash function.
Add glue for libc ATF tests and include a test case to make sure that
(mis)alignment and endianess are handled correctly.

Bump libc minor to 169.

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

Put the atexit test where it belongs - in libc/stdlib/

Atf-ify the atexit test

The nsdispatch test belongs in libc/net and not in libc

Atf-ify the nsdispath test.

Move the servent and protoent tests to src/tests/lib/libc/net/ (where
they should have gone initially).

Atf-ify servent and protoent tests, and the remaining hash test.

Move the servent and protoent tests to src/tests/lib/libc/net/ (where
they should have gone initially).

Atf-ify servent and protoent tests, and the remaining hash test.

Put the atexit test where it belongs - in libc/stdlib/

Atf-ify the atexit test

t_cdb: No need for weird padding any more.

cdbw_output never needed it at runtime, and the declaration no longer
makes gcc angry about not having it.

cdbw_output()'s 3rd argument must be a 16-byte char array, so pad the
string literals passed to it with enough nuls to make this true.

found by GCC 12.

need <sys/stat.h>

branches: 1.1.2; 1.1.4; 1.1.16;
Add regression test for cdbr(3) and cdbw(3).

Continuing the (re)organization of the lib/libc atf regression tests

Do close fd 4 in cerror_64 testcase too.

Atf-ify a couple more tests

Continuing the (re)organization of the lib/libc atf regression tests

Moving more tests out of regress and into atf mainstream

Continuing the (re)organization of the lib/libc atf regression tests

Moving more tests out of regress and into atf mainstream

Move unsigned int/long conversion test to their own testcases, to
avoid having the whole marked as skipped. While here update testcases
names.

Turn warns on for all tests and fix all the bugs.

Do not print sizeof values in skip message.

Remove unneeded prototypes.

Remove newlines from various messages.

Do not call exit() after atf_tc_fail().

branches: 1.1.4;
Migrate a few miscellaneous tests from the old regress to atf

Add missing include <stdlib.h>

make the test more precise.

branches: 1.2.2; 1.2.8;
Avoid printf(3) directive interpretation in long_format testcase
description.

branches: 1.1.4;
Migrate a few miscellaneous tests from the old regress to atf

Moving t_hsearch - belongs in stdlib/

Moving more tests out of regress and into atf mainstream

Another test moving: lib/libc/t_inet --> lib/libc/inet/t_inet_network

Moving more tests out of regress and into atf mainstream

Move t_mktime and t_strptime to their own ..../time/ subdirectory.

(Concurrence from christos@)

Atf-ify test for mktime(3)

The nsdispatch test belongs in libc/net and not in libc

Atf-ify the nsdispath test.

Move the servent and protoent tests to src/tests/lib/libc/net/ (where
they should have gone initially).

Now that PR/44311 is fixed we can remove the expected-failure from this
test.

Atf-ify servent and protoent tests, and the remaining hash test.

Last one for today: move t_ptm test from lib/libc/ to lib/libc/ttyio/

Moving more tests out of regress and into atf mainstream

Continuing the (re)organization of the lib/libc atf regression tests

Migrate a few miscellaneous tests from the old regress to atf

Move the servent and protoent tests to src/tests/lib/libc/net/ (where
they should have gone initially).

Atf-ify servent and protoent tests, and the remaining hash test.

Move t_mktime and t_strptime to their own ..../time/ subdirectory.

(Concurrence from christos@)

Moving more tests out of regress and into atf mainstream

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

aarch64 support per page execute protection

branches: 1.1.4; 1.1.24; 1.1.26;
Preliminary files for AARCH64 (64-bit ARM) support.
Enough for a distribution build.

branches: 1.1.4;
Preliminary files for AARCH64 (64-bit ARM) support.
Enough for a distribution build.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Implement helper function for alpha

branches: 1.1.34;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Make this real even though it isn't yet used.

branches: 1.1.2; 1.1.8;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Avoid plabels on hppa to make some tests not crash

whitespace including space before delay slot instruction.

implement return_one for hppa, mips, ppc64, and vax.

branches: 1.1.42;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

PR port-ia64/51261

Fix for ia64

branches: 1.1.24;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

provide a test function, avoids crashing tests with SIGSEGV.

branches: 1.1.42;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Use .set noreorder and use reg names.

atf- lib/libc/sys/t_mprotect mprotect_mremap_exec is a bit happier now.

implement return_one for hppa, mips, ppc64, and vax.

branches: 1.1.42;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

New files for OR1K support

New files for OR1K support

Allow testing of exec pages on PowerPC BookE.
Make return_one actually do the right thing.

branches: 1.1.2; 1.1.4;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

add return_one_start

Allow testing of exec pages on PowerPC BookE.
Make return_one actually do the right thing.

branches: 1.1.2; 1.1.4;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

implement return_one for hppa, mips, ppc64, and vax.

branches: 1.1.42;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

New files for Userland support of UCB RISC-V (both 32-bit and 64-bit)

Update for new RISCV ABI

New files for Userland support of UCB RISC-V (both 32-bit and 64-bit)

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Don't execute random garbage. Fixes mprotect_mremap_exec test.

branches: 1.1.34;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Sync sparc with sparc64

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Sync sparc with sparc64

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

All sparc64 CPUs support per page execute protection.

branches: 1.1.24;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Don't define a label twice.

All sparc64 CPUs support per page execute protection.

branches: 1.1.24;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

implement return_one for hppa, mips, ppc64, and vax.

branches: 1.1.42;
Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

branches: 1.2.2;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

branches: 1.3.2;
Revert previous hack.
Now all libc have __sync_and_and_fetch functions.

Make the _and_and_ have-nots compile.

Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

t___sync_compare_and_swap: Mark tests xfail on armv5.

PR port-arm/56839: GCC emits wrong codes for compare_and_swap_1
bultins on armv5 (el & eb)

tests/lib/libc/atomic: Test subword compare-and-swap explicitly.

Make sure the sign doesn't bleed into an adjacent word.

We already had failing tests, but this should make the failures a
little more obvious and perhaps check for more adjacent problems.

PR port-arm/56839: GCC emits wrong codes for compare_and_swap_1
bultins on armv5 (el & eb)

branches: 1.1.2; 1.1.14;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

t___sync_lock: Try to make this work on VAX.

Add note about why it _isn't_ also special on HPPA even though it may
seems like it ought to be.

PR port-vax/59260: t___sync_lock:__sync_lock_release_N tests are failing

branches: 1.1.2; 1.1.14;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

branches: 1.1.2;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

branches: 1.1.2;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

branches: 1.1.2;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

branches: 1.1.2;
Add ATF tests for __sync_* functions instead of all_sync_ops_linkable.c

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

branches: 1.1.2;
Add ATF tests for atomic_ops(3).

Remove the incomplete define.

branches: 1.1.2; 1.1.4; 1.1.32;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

avoid gcc warning.

Add more tests for absolute paths with different fds. (from Svivraj)

branches: 1.3.24;
PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.2.12;
Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

tests/fchmodat: add missing space in test failure message

correct test conditions.

tests/libc: replace some more '=' with the intended '=='

Fix = -> ==. Pointed out by Shivraj

branches: 1.3.14; 1.3.22; 1.3.24;
PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.2.12;
Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.3.12;
Remove unnecessary cleanup routines.

branches: 1.2.2;
Mark a few tests that require root

branches: 1.1.2;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51805: Ngie Cooper: Fix file descriptor leaks

branches: 1.2.12;
fexecve is not implemented, so mark the test as an expected failure.

While doing this, clean this whole thing: do not define a useless cleanup
routine and wait for the subprocess to finish instead of using sleep.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.2.12;
Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Add mkfifo{,at}(2) mode in mknod{,at}(2) as requested by POSIX

mknod with mode & S_IFIFO and dev=0 shall behave like mkfifo.

Update the documentation to reflect this state.

Add ATF tests.

This is an in-kernel implementation as typically user-space programs use
mkfifo(2) directly, however whenever there is need to bypass libc (like in
valgrind) then portable POSIX software calls the mknod syscall.

Noted on tech-kern@ by Greg Troxel.

branches: 1.4.12;
PR/51868: Ngie Cooper: mkfifo does not return an fd

PR/51805: Ngie Cooper: Fix file descriptor leaks

branches: 1.2.12;
Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51805: Ngie Cooper: Fix file descriptor leaks

branches: 1.3.12;
Remove unnecessary cleanup routines.

branches: 1.2.2;
Mark a few tests that require root

branches: 1.1.2;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

O_SEARCH tests: expect revoke +x failure on NFS

Patch by Kyle Evans (FreeBSD)

PR misc/54946

Add a few O_SEARCH tests, currently only run on FreeBSD.
Patch from Kyle Evans.

PR misc/54940: do not combine O_SEARCH with O_RDWR, from Kyle Evans.

PR misc/54939: fix file descriptor leak, patch from Kyle Evans.

PR 54902: fix octal numbers accidently spelled as decimal.

branches: 1.5.12;
need <sys/stat.h>

branches: 1.4.12;
Remove unnecessary cleanup routines.

branches: 1.3.2;
Revert defective O_SEARCH implementation committed by manu@ along with
the *at system calls on November 18th of last year. Reasons to revert
it include:
- it is incorrect in a whole variety of ways (but fortunately, one
of them is that the missing and improper permission checks have
no net effect);
- it was committed without review or discussion;
- core ruled that all the new O_* flags pertaining to the *at calls
needed to wait until their semantics could be clarified.

manu was asked to revert it on these grounds but has ignored the request.

I have left O_SEARCH defined and visible and made open() explicitly
ignore it. This way, most code that tries to use it will continue to
build and run. I've also arranged lib/libc/c063/t_o_search.c so that
the tests that make use of the O_SEARCH semantics will disappear until
O_SEARCH comes back, and fixed some mistakes and/or incorrect hacks
that were causing some of these to succeed despite the broken O_SEARCH
implementation.

Split the test cases where root/non-root makes a difference in two and mark
them apropriately. Exact permission semantics are still under discussion,
this will have to be cleaned up once that discussion is settled.
For now, one test cases fails.

branches: 1.1.2;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.2.12;
Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.3.12;
Remove unnecessary cleanup routines.

branches: 1.2.2;
0-terminate strings returned by readlinkat

branches: 1.1.2;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.2.12;
Remove unnecessary cleanup routines.

branches: 1.1.2; 1.1.4;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

t_utimensat: Don't check atime on file systems mounted noatime.

PR kern/58571: utimensat tests are failing on armv7, aarch64, and
riscv64

t_utimensat: Tidy this up to make failures more obvious.

PR kern/58571: utimensat tests are failing on armv7, aarch64, and
riscv64

tests/libc: replace some more '=' with the intended '=='

branches: 1.6.14; 1.6.22; 1.6.24;
PR/51804: Ngie Cooper: Sort headers; include <sys/stat.h>. On NetBSD,
<sys/stat.h> gets side-loaded from <sys/fcntl.h>. Should be fixed.

branches: 1.5.12;
Remove unnecessary cleanup routines.

branches: 1.4.2;
Compare against modification time instead of creation time - which we did
not set.

Fix off by one in static array acces, use valid timestamps (ns part < 1e9),
just in case a filesystem would need to convert this to some other
representation.

Do not require tv_nsec to be != 0 but instead require equality with the
test value.

branches: 1.1.2;
Add most system calls for POSIX extended API set, part 2, with test cases:
faccessat(2), fchmodat(2), fchownat(2), fstatat(2), mkdirat(2), mkfifoat(2),
mknodat(2), linkat(2), readlinkat(2), symlinkat(2), renameat(2), unlinkat(2),
utimensat(2), openat(2).

Also implement O_SEARCH for openat(2)

Still missing:
- some flags for openat(2)
- fexecve(2) implementation

Avoid plabels on hppa to make some tests not crash

Add a simple test case to check executable mapping rights for mprotect(2).

- provide an exec_prot_support() routine so $ARCH can indicate whether
it supports execution protection or not, and skip test accordingly.

- have a trivial 'return_one' shellcode to copy anywhere in a page, and
call it. The decision to keep the assembly solution is mine, reasons are
twofold:

- all pure-C implementations cannot be controlled easily:
bounds detection (beginning/end) of return_one is unpredictable,
or requires the use of overkill solutions like libelf. Using
dlsym(3) was a good proposal, however I can't use it to know the
end address of the payload. It makes copying of the shellcode a bit
more difficult: using a constant may be too small (code has not been
entirely copied, and can lead to errors that can be erroneously
detected as "test passed"), or too big (depending on where it is mapped
in memory, copying past the end of the function may trigger SIGSEGV).

- I have to ensure that the resulting assembly is the most compact
possible, especially as it will be reused to test other parts of
memory (stack, data, rodata, etc.).

Only i386 and amd64 are implemented so far. Others will come in due time.
FWIW, writing the exec_prot_support() callback and the return_one payload
should be enough. Writing callback requires good knowledge of the platform,
depending on chip revision, CPU, board, MMU... the protection level may vary.

Current files are put under lib/libc/arch/* and lib/libc/common/. Feel free to
move them around the tests/ tree. Keep in mind that the common/ and arch/ code
will be used outside of libc, so please do not hide them too deep in the tree.

I checked a few architectures via build.sh cross-compile, and will keep an
eye on buildbot for potential build breakage. Feel free to contact me in
case you see any, of course.

Add more of the torture tests from the mit kerberos tree.

branches: 1.3.2;
Add more exhaustive db tests that include higher bucket sizes

Add new failing test.

Atf-ify the db tests.

XXX Note that the original regress version of this test did not run the
XXX btree_delete test; that test is broken, and is disabled for now.

Atf-ify the db tests.

XXX Note that the original regress version of this test did not run the
XXX btree_delete test; that test is broken, and is disabled for now.

remove mpool_getf

Add more of the torture tests from the mit kerberos tree.

branches: 1.1.28;
Atf-ify the db tests.

XXX Note that the original regress version of this test did not run the
XXX btree_delete test; that test is broken, and is disabled for now.

Reduce the number of iterations in the bsize_torture test from 65535
to 1000 to make the ATF test suite as a whole take less time. Before
the change, this single test case could take more than two hours to
run on a qemu emulated ARM.

branches: 1.1.8;
Add more exhaustive db tests that include higher bucket sizes

bsize_torture: skip bigger page size tests if space in the database
directory is limited (numbers pulled out of thin air).

btree_weird_page_split: set timeout to 900s, now my landisk tests have
a chance to complete this one.