Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
123,083 Commits 606 Branches 0 Tags 3.1 GiB
986c9703a5
Commit Graph

580 Commits

Author SHA1 Message Date
martin
9bff0c1a82 Remove previous makefile hack (disabling optimization) and add 
__insn_barrier()s instead.
 2004-01-07 19:42:22 +00:00
martin
266b7c2701 Avoid optimizations that move common code out of the RAS. 
The compiler does not understand RAS, so it makes wrong assumptions
about the lifetime of register values.
 2004-01-06 21:40:57 +00:00
lukem
c926d8d22d support MKCRYPTO=no (des.h is not present) 2004-01-04 00:25:52 +00:00
lukem
7138133de7 support MKSKEY=no 2004-01-03 15:05:37 +00:00
martin
0d9a131abe On archictectures not supporting RAS make this regression tests 
print a message to that effect and let them pass.
 2004-01-02 22:16:43 +00:00
cl
ab4f8b76c0 enable sigmask1 and siglongjmp1 2004-01-02 19:27:46 +00:00
cl
f44e43cd85 Regression test for siglongjmp out of a signal handler back into 
its thread.
 2004-01-02 19:27:06 +00:00
cl
ec7c3305d4 Regression test for pthread_sigmask when SA upcalls aren't started yet. 2004-01-02 19:26:24 +00:00
itojun
6d206bfa62 skip ai_canonname from testing, as it can be vary by /etc/hosts setting. 
pointed out by Martin Husemann
 2004-01-02 15:07:39 +00:00
martin
572d0edcf8 Descend into siginfo on cleandir always. 
From David P. Reese Jr. in PR toolchain/23936.
 2003-12-31 11:14:25 +00:00
junyoung
b0b7ac330d Fix regression failure due to missing dependency. 
From Jorge Acereda via PR bin/23664.
 2003-12-23 19:29:11 +00:00
veego
8bf78d509a Update the checksum to match src/lib/libc/stdlib/random.c rev 1.23. 2003-12-19 18:39:04 +00:00
jmc
4b5c99bea9 Don't check for HAVE_SIGINFO if make(obj) (as more than likely the compiler 
doesn't exist yet and the subdir is always descended on make obj)
 2003-12-17 05:36:43 +00:00
matt
d0d3605346 .include <bsd.own.mk> so the proper have of CC is defined before we use it. 2003-12-11 22:45:06 +00:00
drochner
7e4c34eca1 Add a check whether the stack is executable when needed, and one whether 
the stack is non-executable if not explicitely enabled.
The latter one is not enabled in the Makefile yet because it cannot
succeed on platforms where the non-exec stack is not implemented.
Both checks rely on the fact that gcc puts trampoline onto the stack
for nested functions. On platforms where no such trampolines are used,
no meaningful results can be expected.
While not perfect, this is much easier than hand-coding trampolines for
each CPU type.
 2003-12-10 13:24:59 +00:00
itojun
c81f32fe6c comment from niels provos; 
- seed2 is necessary, but use it as "seed2 + x" not "seed2 ^ x".
- skipping number is not needed, so disable it for 16bit generator (makes
  the repetition period to 30000)
 2003-12-10 05:22:18 +00:00
christos
7bcd27fa9a Add another regression test that does not work right now. 2003-12-02 18:30:35 +00:00
yamt
f1e9f5f0af less magic constant. 2003-11-29 19:09:21 +00:00
simonb
a7fbae904d Fix up abort test so that we abort the first time we see a small enough 
period, not the second time.
Use UINT64_MAX (from Klaus Klein).
Remove trailing blank line.

XXX: 12000 still isn't right - values around 11800 have been observed.
 2003-11-29 00:11:18 +00:00
simonb
db92efc5c8 Print out a little more info in the debug case. 2003-11-27 01:03:13 +00:00
itojun
d94d0c139d uint64_t may not be "unsigned long long", so use of ULLONG_MAX is not 
appropriate.
adjust max period (see randomid.3 1.2 -> 1.3)
 2003-11-25 23:14:48 +00:00
nathanw
374a535730 Add condcancel1. 2003-11-21 19:26:05 +00:00
nathanw
d2ac4c0565 RCS ID 2003-11-21 19:25:50 +00:00
nathanw
2d3513cfd8 Regression test for destroying a condition variable that had a cancelled 
wait.
 2003-11-21 19:24:01 +00:00
uwe
e77146bf27 Actually print self in the "Thread %p got semaphore" message. 
While here make whitespace only lines empty.
 2003-11-19 00:40:03 +00:00
christos
1f9b81f68b kjk reported this does not work on the vax. 
- don't include <ieeefp.h> if vax
- never include <sys/float_ieee754.h>
 2003-11-18 23:21:38 +00:00
dsainty
4c758a9f41 Descend into bin/dd 2003-11-15 10:50:59 +00:00
dsainty
9adae06087 Regression tests for dd(1). Not currently comprehensive, but tests for the bug reported in PR bin/8521. 2003-11-15 10:48:53 +00:00
simonb
bafaeff914 Descend into the randomid directory. 2003-11-14 23:11:20 +00:00
simonb
1fd22c15cc Add a regression test for randomid(3). 2003-11-14 23:10:48 +00:00
christos
6d0baf7772 Remove <frame.h> 2003-11-14 14:13:31 +00:00
simonb
967d9a12d5 Add NetBSD RCS id's. 2003-11-13 06:48:31 +00:00
christos
ae4155f2d9 Annotate changes and remove non-existing files. 2003-11-07 23:10:00 +00:00
augustss
03d49d9b1b Remove more troublesome files for now. 2003-11-07 21:10:33 +00:00
augustss
91cf4290f4 Comment out some files that makes this break on some platforms. 2003-11-07 20:57:26 +00:00
christos
abd55fcd3e no more disklabel_mbr.h 2003-11-07 17:41:52 +00:00
christos
5054637087 don't include varargs.h for gcc-3 2003-11-07 17:40:38 +00:00
christos
7264e1f642 add okheaders to the test 2003-11-07 15:48:09 +00:00
lukem
f85d2d1c14 Use ${HOST_SH} instead of `sh'. 
If necessary, pull in <bsd.sys.mk> to get the definition of HOST_SH;
Makefiles that pull in one of (most of) <bsd.*.mk> will get this anyway.
 2003-10-26 07:25:33 +00:00
lukem
857ab1bdd0 use ${HOST_SH} instead of /bin/sh 2003-10-26 03:50:07 +00:00
kleink
d0d137fcf0 Check HUGE_VALF and HUGE_VALL as well. Also add some cross checking for 
floating type promotion and demotion of these.
 2003-10-25 22:38:19 +00:00
kleink
d039cc06d0 Lift the vax restriction on the nan test. 
None of the tests in this directory will even build on the vax, and it's
disabled in the parent Makefile for that reason.
 2003-10-24 21:25:20 +00:00
kleink
d66a9fb0bc Make this test the currently libc-internal isinfl() as well. 
While I'm here, turn these into assert(3)-based tests.
 2003-10-24 16:35:08 +00:00
kleink
d805d82f20 Make this test the currently libc-internal isnanl() as well. 
While I'm here, turn these into assert(3)-based tests.
 2003-10-24 15:54:46 +00:00
he
f9e944d1a0 While the intention of picking up the target's include files was fine, 
those files are not available in certain phases in the build process,
notably the "obj" phase, so unconditionally decend into siginfo during
that pass.  This should make read-only source trees build again when
cross-compiling for a target supporting siginfo on a host not supporting
siginfo.
 2003-10-23 21:59:35 +00:00
kleink
aa20eb60ec Fix a comment. 2003-10-22 23:57:04 +00:00
itojun
73704c4ce4 NULL/integer mixup 2003-10-21 09:42:07 +00:00
agc
492f508473 Don't use NULL inappropriately. 2003-10-21 09:32:54 +00:00
agc
0833278c93 Don't compare the return value of strncmp(3) against NULL - it's an integer. 2003-10-21 08:46:15 +00:00
he
4bdc6bf1cb Make sure to pick up the target's include files in the test for whether 
we have SA_SIGINFO defined.
 2003-10-20 00:07:47 +00:00