Commit Graph

205363 Commits

Author SHA1 Message Date
jruoho
719ceba55f IEEE corner case tests for the pow(3) family (incl. PR lib/45372). 2011-09-17 08:15:43 +00:00
jruoho
ca5cfe7128 Reduce the powers. 2011-09-17 06:21:19 +00:00
minskim
05d81a59ae Add back a line accidentally deleted in the previous commit. 2011-09-17 03:09:51 +00:00
christos
369862dcef Never invoke utmp_update() recursively if we are root. Instead close the
read-only file so it can be opened later. It can happen when a setuid
program utmp_update, seteuid(ruid) -> getutxent() -> seteuid(0) -> pututxent()
2011-09-17 01:52:29 +00:00
christos
8017e7237a log to both syslog and stderr. 2011-09-17 01:50:54 +00:00
christos
3145fd5cc6 Add utmp/utmpx support.
XXX: Should we make the top pty (tmux) hide?
XXX: Should we add an option to not create utmp entries for a tmux terminal?
2011-09-17 01:50:08 +00:00
christos
7aee9ef7a1 Add utmp support. 2011-09-17 01:48:46 +00:00
jym
b8647b8e93 Update list of algorithm identifiers available in opencrypto(9). 2011-09-16 22:27:36 +00:00
reinoud
981eb029d6 Fix sigactsunshare() to do what the manpage and the commit tells it to do!
I.e. unshare a process its sigacts by creating a new one BUT maintaining all
signal state.

The old code would CLEAR all the signal state. It used to work only due to the
fact that sigactunshare() is only used in execsigs() and the fact that
SIG_DFL was NULL.
2011-09-16 22:07:17 +00:00
christos
963ca07b86 new file(1) 2011-09-16 21:08:32 +00:00
christos
79bb278aa6 merge changes 2011-09-16 21:06:21 +00:00
reinoud
12590f04bf Improve diagnostics message on entry point too low/high checking 2011-09-16 21:02:28 +00:00
christos
1594e4f0e7 restore from the dead. 2011-09-16 20:38:12 +00:00
christos
f2668aff30 from ftp.astron.com:
- elf seeking fixes
- improvements in cdf parsing
- waitpid only for our child
- magic fixes and additions
2011-09-16 20:37:23 +00:00
christos
cc9ecc5e39 PR/45370: Takehiko NOZAKI: termcap emulation tget{flag,num,str} should
work with non-NUL terminated strings.
2011-09-16 18:51:44 +00:00
christos
025e844498 don't update access/change times on open. 2011-09-16 18:43:44 +00:00
joerg
ba0add90b0 Disable noreturn checks for remaining external code 2011-09-16 16:41:20 +00:00
reinoud
5da488838d Remove the blockage of SIGALRM here too 2011-09-16 16:30:51 +00:00
joerg
301722b440 Do not care about dead functions for test cases, in this case it is
often enough just noise.
2011-09-16 16:30:18 +00:00
reinoud
f99c069b55 Reduce the size of the memory base of the testmachine even more! 2011-09-16 16:29:47 +00:00
joerg
2ae99cf42b There is no such thing as WARNS=5. 2011-09-16 16:29:12 +00:00
reinoud
c9c81fefa4 Move the debug printf's to the new dprintf_debug() and disable the SIGALRM
blocking since we queue them now.
2011-09-16 16:29:11 +00:00
reinoud
7d70ede9ed Print out the memory information on verbose booting too, its too handy to have
around without all the goo of debug printouts
2011-09-16 16:27:39 +00:00
joerg
c6e6ace444 Use __dead 2011-09-16 16:27:36 +00:00
reinoud
8e95c015a7 Split the two debug sections 2011-09-16 16:26:19 +00:00
reinoud
9923cc8437 Allthough in normal practice at most two are queued, with all the debug info
spounting outi, a lot can get queued in the first phase of kernel startup.
especially the clock.
2011-09-16 16:25:44 +00:00
reinoud
609efbfe19 Like the others, let the clock signal restart calls 2011-09-16 16:24:01 +00:00
joerg
3c101252bb Use __dead 2011-09-16 16:13:41 +00:00
plunky
2b8aaed8cd NULL does not need a cast, here 2011-09-16 16:13:16 +00:00
joerg
ce5205bcbb Use __dead 2011-09-16 16:09:03 +00:00
joerg
6b5273200c Use __dead for the small build too 2011-09-16 16:06:23 +00:00
joerg
67f518f496 Use __dead 2011-09-16 16:05:58 +00:00
joerg
bd80b77cd9 Use __dead. Save a pointer. 2011-09-16 15:42:56 +00:00
joerg
6818646ac8 Use __dead 2011-09-16 15:39:25 +00:00
joerg
84700a0331 Use __dead consistently. If it doesn't exist, define it away. 2011-09-16 15:38:03 +00:00
joerg
a85aba86f0 Use __dead. 2011-09-16 15:36:18 +00:00
joerg
6a878ae49f Reapply formatting cleanup 2011-09-16 15:36:00 +00:00
apb
9438a85a85 More corrections to format compatibility between stat(1), find(1) -ls,
and ls(1) -lT.
2011-09-16 14:32:53 +00:00
wiz
8438689645 Remove unnecessary Ns. 2011-09-16 13:33:13 +00:00
jmcneill
d4bcf1a5bb push TEXTADDR option into kernel makefile 2011-09-16 11:28:36 +00:00
apb
229948abdb Anothe difference in output format: stat -f "%Z" does not add any
space in "major,minor" for device nodes, whereas both "ls -l" and
"find -ls" add at least one space.
2011-09-16 11:06:27 +00:00
reinoud
568419374b Reduce the amount of memory needed to run by default 2011-09-16 11:05:42 +00:00
rkujawa
86a4d8f318 Fix typos (make this compile again...) 2011-09-16 10:23:01 +00:00
blymn
2494728ef8 - Update the check files due to a bug fix in the slave code 2011-09-16 08:05:28 +00:00
jruoho
6ceed68f62 Adjust to see where this fails on i386/qemu. 2011-09-16 04:54:46 +00:00
christos
87e56634fd This is no place to attach the primary cpu. Things go wrong from here because
for example it is missing its name.
2011-09-16 01:03:52 +00:00
christos
98f06b6095 check that we have valid event strings on attach too, so that the error
is obvious instead of catching it when vmstat -e runs!
2011-09-16 01:03:02 +00:00
reinoud
4e15406e15 Use vfprintf() to make it compile under 5.1 again 2011-09-15 20:25:23 +00:00
apb
7f5f7e4b90 Add examples of formats for stat(1) to simulate "find -ls" and "ls -lTd". 2011-09-15 19:56:35 +00:00
reinoud
970d3494e4 Check reentry to prevent double use of the aiocb structure 2011-09-15 19:32:28 +00:00