39fd9b490a
Make "regress" depend on the binary, so "make regress" builds it if necessary.
2001-03-11 16:40:40 +00:00
82f3142780
fix NetBSD RCS id tags
2001-02-27 19:04:39 +00:00
d7666f6c7f
use getprogname()
2001-02-20 23:22:49 +00:00
65a1026492
convert to use getprogname()
2001-02-19 22:43:42 +00:00
13dbf7c5c5
Add a very, very simple test of XDR handling.
...
It just checks decoding of doubles, which NetBSD/arm32 seems to have got
wrong forever.
2001-02-18 21:57:16 +00:00
d5822f49d1
add test 41: sort -f < emptyfile bombs.
2001-02-15 17:23:20 +00:00
c1ff96b0ae
go through and look for all of them, too
2001-02-14 06:03:37 +00:00
76914ab972
add IDs in the object
2001-02-14 05:32:19 +00:00
a98e831e03
Add a program to test the functionality of hsearch(). Our current
...
hsearch() falls right over (because it's unredeemably broken). That's
about to change, though.
2001-02-14 05:20:10 +00:00
df1e7da042
40B: just symlink in and out, don't bother copying it
...
use -f for the sort
2001-01-19 12:34:45 +00:00
51a5f86386
don't use $s in test #39 , that test doesn't depend on presence of -s
...
add two new regressions tests for long lines as test #40 , using
/usr/share/dict/words; in-tree sort currently crashes during 40B, going
to fix it soon
2001-01-15 18:52:32 +00:00
fdbaf840be
don't print any note - stest prints its own
2001-01-15 16:13:59 +00:00
439b247426
Add the "sort the dictionary" test that detected the recent lossage.
2001-01-14 14:47:56 +00:00
0f41b74720
remove duplicate target empty.
2001-01-14 06:35:57 +00:00
09b8ab7adf
protect against multiple inclusion; bsd.prog.mk will re-include this file
2001-01-14 06:24:53 +00:00
96731d46b9
use $S for stuff calling sort directly, not via xsort
2001-01-12 22:35:07 +00:00
a8b321c587
print a disclaimer on startup about this test suite - other sort implementations
...
may fail some tests, yet conform to standard too; for non-stable sort,
quite a few tests have actually more possible 'correct' results, yuck!
check for presence of -S and use it to switch to non-stable sort - the tests
are not written for stable sort
Fix test 37 - for 'sort -m -r' to actually work, the input files have
to be reverse sorted as well. Now the in-tree BSD sort passes this test ;-)
print the used sort command and parameters for 'failed' messages
other small sleanups, remove accidentaly added debug stuff
2001-01-12 21:45:17 +00:00
2ec37dafdc
do not mess with $o if misplaced -o is supported - it's orhogonal to +POS/-POS support
2001-01-11 22:26:55 +00:00
4502c34907
do not display the actuall command executing stests
2001-01-11 22:19:29 +00:00
9d7f88dae6
change so that test 35 print warning about missing -g just once
...
s/not present/not supported/
2001-01-11 22:18:38 +00:00
50b80dfd9f
use shell functions instead of dynamically created external scripts
...
warn when test is skipped due to missing feature
2001-01-11 22:10:27 +00:00
0664ed5c26
make objdir safe
2001-01-11 14:12:14 +00:00
699a13760c
Add regression tests for div/ldiv/lldiv.
2001-01-08 04:41:31 +00:00
d7d808597e
Add regression tests for div/ldiv/lldiv.
...
Inspired by port-i386/11914 from Luke Mewburn.
2001-01-08 04:39:31 +00:00
b2779a2f25
test for dlopen(3) and the toolchain generating the files involved
2000-12-08 19:24:29 +00:00
44e89b161e
test program for dlopen(3)
2000-12-08 19:22:51 +00:00
08a3f0ae6e
test cases for dynamically loaded objects: constructor/destructor call for
...
static objects, C++ runtime support
2000-12-08 19:21:28 +00:00
cbfaeab846
"make benchmark" will run a tiny benchmark test.
2000-11-08 06:02:01 +00:00
b7192d086c
move WARNS to ../Makefile.inc
2000-11-08 05:14:56 +00:00
2ce9195af1
set WARNS?=1. fix error trap in rijndael test. sync with kame
2000-11-08 05:08:55 +00:00
5aadf3060b
put in NetBSD RCS id
...
put it comment that this was derived from stuff written by
Peter McIlroy - it's not clear now that it's separate from
the rest of new BSD-licenced sort(1)
2000-11-06 14:42:40 +00:00
76a9ffd207
sync calling sequence with kernel code. sync with kame
2000-11-05 03:15:57 +00:00
9681f1c1b6
use fmtcheck(), not __fmtcheck()
...
Fixes misc/11409 by Richard Earnshaw.
2000-11-02 18:06:30 +00:00
077d6e9048
visit crypto if MKCRYPTO!=no
2000-11-01 15:34:28 +00:00
985a5fe39c
regression test for kernel crypto tree. it helps us to verify those
...
logic are working right, on various platforms. from KAME
2000-11-01 15:33:21 +00:00
374e0e8fff
move usr.bin/sort/TEST/stest to regress/usr.bin/sort, add to
...
regression tests
2000-10-26 23:51:31 +00:00
efba0ca36a
descend into bin/expr
2000-10-26 23:32:59 +00:00
224d8e7e7e
add two tests for expr
2000-10-26 23:32:14 +00:00
06dc227e73
Regression tests for fmtcheck(3).
2000-10-20 18:59:53 +00:00
289ebac6b2
wrap libcrypto with ${MKCRYPTO} == "yes"
2000-10-04 00:49:44 +00:00
35acecee2a
regression test for libcrypto.
2000-10-03 15:13:21 +00:00
d3b573e04a
Add md5sha subdir
2000-08-19 17:53:24 +00:00
7869a64f4e
Add string.h
2000-08-19 17:52:13 +00:00
639160ce5f
Quick & dirty sha1/md5 regression test.
2000-08-19 17:51:19 +00:00
7d792964e5
Fix obvious pasto resulting in third test series never being done.
2000-07-31 12:22:39 +00:00
f4a3ee2726
add NetBSD RCS id
2000-07-30 09:16:06 +00:00
0c8766beed
make the parent kill off it's children after it's main loop is finished.
...
This solves problem reported privately by Thomas Klausner, where the
regress test would take very very long to complete.
2000-07-30 08:48:31 +00:00
eec8346517
ptrace(PT_DETACH,...) behaves same way as PT_CONTINUE WRT addr & data;
...
specifically, if we pass NULL addr, we cause the child to jump
to address zero and SIGSEGV immediatelly; pass (caddr_t) 1 instead
while here, improve some error messages and slighly delint
2000-07-27 21:37:39 +00:00
823d32f2c2
DPADD ${LIBSKEY} as well as LDADDing -lskey
2000-07-27 00:25:05 +00:00
d72583bda6
LDFLAGS -> LDADD, so this compiles statically.
2000-07-24 00:00:21 +00:00
bjh21