strings.h: Spruce up for POSIX 2024.

1. Add ffsl, ffsll.
2. Hide bcmp, bcopy, bzero, index, rindex for POSIX>=2024.
3. Expose ffs only for NetBSD or POSIX>=2008 with XSI option.
4. Hide popcount* NetBSD extensions for any POSIX.
5. Sprinkle __constfunc on ffs*.

Add tests for ffs/ffsl/ffsll in tests/include/sys/t_bitops next to
ffs32/ffs64 for convenience.

XXX Still missing strcasecmp_l, strncasecmp_l, and locale_t.

PR lib/58802: missing ffsl(), ffsll() functions from POSIX 2024

branches: 1.16.8;
Mask NULL + 0 LLVM UBSan reports in the ATF test: t_pslist.c

Pass -fno-delete-null-pointer-checks for Clang for the
MKSANITIZER/MKLIBCSANITIZER build

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

Get rid of all the -lrumpdev and -lrumpvfs that are no longer needed
after moving rump's mainbus from rumpdev to rumpkern.

Produces the same atf-run results as before.

branches: 1.13.4;
Add a new minimalistic test for LIST_MOVE to verify that the list's
first entry's prev pointer correctly points to the listhead.

This test will fail until a fix for LIST_MOVE is checked in (soon).

branches: 1.12.6;
We now need librumpdev to resolve rumpns_config_cfdriver_attach which
is now used in a bunch of network modules/components

Introduce pserialize-safe linked lists.

These are like LIST_* from queue(3), but issue the appropriate memory
barriers for pserialize readers and writers.

Discussed on tech-kern:

https://mail-index.netbsd.org/tech-kern/2016/04/03/msg020365.html

Make this list diff- and sort-friendlier.

No functional change.

Define _KERNTYPES for things that need it.

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

Exclude tests that use rump

test the __type macros

branches: 1.5.2;
Move the 'syscall/t_cmsg' test to 'include/sys/t_socket'.

Try to mirror the real source with the test directory structure (otherwise
the latter carries a risk of becoming a difficult mess to manage.)

Follow the real tree with the test directory structure.

Add a simple test file for <sys/bitops.h>. For now, only ilog2(3) is tested.

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

strings.h: Spruce up for POSIX 2024.

1. Add ffsl, ffsll.
2. Hide bcmp, bcopy, bzero, index, rindex for POSIX>=2024.
3. Expose ffs only for NetBSD or POSIX>=2008 with XSI option.
4. Hide popcount* NetBSD extensions for any POSIX.
5. Sprinkle __constfunc on ffs*.

Add tests for ffs/ffsl/ffsll in tests/include/sys/t_bitops next to
ffs32/ffs64 for convenience.

XXX Still missing strcasecmp_l, strncasecmp_l, and locale_t.

PR lib/58802: missing ffsl(), ffsll() functions from POSIX 2024

branches: 1.21.2;
t_bitops: Show the bad results on failure.

Avoid undefined behavior in an ATF test: t_bitops

Do not change the signedness bit with a left shift operation.
Switch to unsigned integer to prevent this.

t_bitops.c:189:9, left shift of 1 by 31 places cannot be represented in type 'int'

Detected with micro-UBSan in the user mode.

branches: 1.19.14; 1.19.16;
one more white space -> tab.

Rewrite ilog2's test. PR lib/49745.
- Reorganize ilog2_basic to ilog2_32bit, ilog2_64bit and ilog2_const.
ilog2_const is compile-time test for __builtin_constant_p() part of
current ilog2() implementation.
- Remove fully meaningless ilog2_log2. So this part of PR misc/44767
is no longer present.

white space -> tab.

use __BITMAP_TYPE

Move the bitmap(3) test to the "right" place. Note it in bitops(3). Xrefs.

branches: 1.14.2; 1.14.8;
No need to keep on trying when this fails in the expected manner. We
really don't need to have 32k failures.


OK jruoho@

remove stray underscore.

disable test on the vax until we fix it.

Add expected failures for Qemu.

Remove some cruft that is no longer needed.

Remove Xfails that are related to the infamous qemu/amd64 floating point
bugs. It appears to be quite difficult to identify the exact Qemu version
and setup. These do not fail on the TNF's qemu/amd64 setup, which can be
taken as a reference point for expected failures.

Improve the QEMU/amd64 detection.

Add missing __RCSID(3).

Do not skip the QEMU bugs but instead mark these as expected failures.

Bluntly skip the tests that fail under QEMU. XXX: system(3) is used for this?

wnCVS: ----------------------------------------------------------------------

Even these naive test cases caught one (QEMU?) bug; comment PR # 44767.

Add also a basic, naive, test for fast_divide32(3).

Add some naive test cases for the ffs32(3) family of functions.

Add a simple test file for <sys/bitops.h>. For now, only ilog2(3) is tested.

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

rework to avoid new GCC 5.3 errors and improve readability

Fix two tests for typefit: constants larger than LLONG_MAX
do not fit in a signed long long.

branches: 1.2.2; 1.2.4;
Use SCHAR_{MIN,MAX} for signed char, not CHAR_* since not char are signed.

test the __type macros

tests/sys/queue.h: fix mismatch between '=' and '=='

branches: 1.2.2; 1.2.18;
Update recently-added test to adapt to new signature of LIST_MOVE()

Add a new minimalistic test for LIST_MOVE to verify that the list's
first entry's prev pointer correctly points to the listhead.

This test will fail until a fix for LIST_MOVE is checked in (soon).

Adapt <sys/pslist.h> to use atomic_load/store_*.

Changes:

- membar_producer();
*p = v;

=>

atomic_store_release(p, v);

(Effectively like using membar_exit instead of membar_producer,
which is what we should have been doing all along so that stores by
the `reader' can't affect earlier loads by the writer, such as
KASSERT(p->refcnt == 0) in the writer and atomic_inc(&p->refcnt) in
the reader.)

- p = *pp;
if (p != NULL) membar_datadep_consumer();

=>

p = atomic_load_consume(pp);

(Only makes a difference on DEC Alpha. As long as lists generally
have at least one element, this is not likely to make a big
difference, and keeps the code simpler and clearer.)

No other functional change intended. While here, annotate each
synchronizing load and store with its counterpart in a comment.

branches: 1.1.16;
Introduce pserialize-safe linked lists.

These are like LIST_* from queue(3), but issue the appropriate memory
barriers for pserialize readers and writers.

Discussed on tech-kern:

https://mail-index.netbsd.org/tech-kern/2016/04/03/msg020365.html

use memcpy() for non C-strings.

branches: 1.5.14;
Don't play with "../.." in includes for h_macros.h; deal with it centrally.
Minor fixes.

branches: 1.4.2;
2 seconds is too small as timeout on slow machines, bump to 10 (my hppa
needs ~5).

fix unused variable warnings

socket(-1, SOCK_CLOEXEC, 0); followed by the process exiting (i.e.
fd slot does not get initiailized before fd_free()) cases a diagnostic
kernel panic.

branches: 1.1.2; 1.1.8;
Move the 'syscall/t_cmsg' test to 'include/sys/t_socket'.

Try to mirror the real source with the test directory structure (otherwise
the latter carries a risk of becoming a difficult mess to manage.)

Avoid undefined behavior in an ATF test: t_types

Replace UB with implementation defined logic to check whether ssize_t can
wrap to a negative number.

t_types.c:63:7, signed integer overflow: 9223372036854775807 + 1 cannot be represented in type 'long int'

Detected with micro-UBSan in the user mode.

branches: 1.4.30; 1.4.32;
To be on the safe side, use the category/number notation when referring to
PRs (otherwise third-party sed-scripts might miss the references). Also
remove white-space.

this does not fail anymore

branches: 1.2.2;
Make types_signed/types_unsigned testcases use ATF_CHECK instead of
ATF_REQUIRE.

Follow the real tree with the test directory structure.