199396 Commits

Author SHA1 Message Date
joerg
928f301be9 Rework TLS initialisation:
- Update TCB for the initial thread in pthread__initthread, not
  pthread__init to get it valid as soon as possible.
- Don't overwrite the pt_tls field in pthread__initthread.
- Don't deallocate pt_tls in pthread__scrubthread. This worked more by
  chance than by design.
- Handle freeing the TLS area in pthread_create after removing the
  thread instance from the dead queue.
2011-03-30 00:03:26 +00:00
jym
26f424e994 Typo fix. 2011-03-29 23:51:32 +00:00
agc
a2e4cd88b7 fix a repeated typo 2011-03-29 21:43:17 +00:00
joerg
8fa6585514 Block signals when using the exclusive lock. 2011-03-29 20:56:35 +00:00
plunky
1a7622f6a7 tidy up include pathname to exclude undefined behaviour
"If the characters ', \, ", //, or /* occur in the sequence between
  the < and > delimiters, the behavior is undefined."
2011-03-29 20:10:31 +00:00
jruoho
cb0f983e60 Improve -width. 2011-03-29 18:54:54 +00:00
riz
4c4709acf2 Whoops, merge two lines about iscsi initiator. 2011-03-29 18:07:14 +00:00
sjg
be03bd0379 Check for extra commands as soon as we reach CWD,
so we return faster (if oodate) and give a more accurate debug message.
2011-03-29 17:29:20 +00:00
sjg
ede3eaf4a5 CondDoExists: Avoid use after free() 2011-03-29 17:19:22 +00:00
jmmv
e58081aa53 Halt the httpd server _before_ rump. Per pooka@'s suggestion. 2011-03-29 16:03:32 +00:00
jmmv
5bd1bce108 Make httpd create a pid file and use that to kill the server in a controlled
manner (instead of indirectly through rump.halt).  Suggested by pooka@.
2011-03-29 15:43:45 +00:00
jmmv
255271589f Extend pidfile(3) to support creating pid files in arbitrary locations.
If the argument provided to pidfile(3) contains a '/', then the value is
considered to be an absolute/relative path and the pid file is created
in the given location.

Otherwise, pidfile(3) behaves as before and treats the provided value as
a basename to construct a pid file in /var/run/<basename>.pid.  This means
that to create a pid file named "foo.pid" in the current directory, one
must specify "./foo.pid".
2011-03-29 13:55:37 +00:00
mbalmer
c9dadfa0d7 Don't comment out printer reset on open, if this causes a problem, use the
/dev/ulpnX device node which sets a flag that prevents the reset.
2011-03-29 07:48:13 +00:00
jmmv
83bb438905 Add pid file support: if the new -P option is provided, it specifies the
location of the pid file to create.

OKed by mrg@.
2011-03-29 07:22:31 +00:00
dyoung
149dd44b66 __HAVE_DEVICE_REGISTER_POSTCONFIG and __HAVE_DEVICE_REGISTER
are no more, so don't use them here.
2011-03-28 22:23:39 +00:00
riz
d4bd4b1f37 In-kernel iscsi initiator is looking good for 6.0 2011-03-28 19:42:11 +00:00
para
8b166785aa Added myself. 2011-03-28 18:48:21 +00:00
fair
99c0cae956 Add dholland's comment about units not warning users about non-sensical
conversions to the BUGS section.
2011-03-28 16:09:10 +00:00
tsutsui
0b9b85d98b Use CALLFRAME_SP(sp) instead of CALLFRAME_S0(sp) to save s0.
At least PROM's _getenv() on 3MIN doesn't preserve CALLFRAME_S0(s0)
and prom_findcons() returns bogus console after mips64 merge.

XXX: I wonder if it's safe to use CALLFRAME_S0(sp) (== 0(sp)) on O32/O64..
2011-03-28 14:40:00 +00:00
ahoka
1a7f8f91f0 Fix manpage wrt reversed compare functions. 2011-03-28 13:46:14 +00:00
tsutsui
397c14a1c3 Use REG_L (not REG_S) to restore a register value. 2011-03-28 12:26:59 +00:00
joerg
6b0a585bda Catch up with library roulette in lli. Allow LLVM_DEVELOPER to be set in
mk.conf. From Vladimir Kirillov.
2011-03-28 11:41:48 +00:00
martin
ecf8c9b0f0 ABI police: avoid cluttering global registers %g2 and %g7. Not quite there
yet for %g2, but the target is not very well defined. %g7 should be clean
now.
2011-03-28 11:19:12 +00:00
plunky
b700da1ddc move building of Xorg server into a SUBDIR since it is created from
the other subdir contents, ok by mrg@

(this makes "make all" and "make dependall" do the same thing)
2011-03-28 10:08:06 +00:00
joerg
e49f915c10 Refine locking scheme around init/fini to not hold the exclusive lock.
Use a simple generation count instead and restart looking for work if it
changed (e.g. due to an dlopen call from an init function).
Leave the possible dlclose() race for now.
2011-03-28 00:37:40 +00:00
dholland
75f2c9abd2 fix typo 2011-03-28 00:14:51 +00:00
dholland
8cf15952a4 Update radioactivity-related units from http://physics.nist.gov/cuu/Units/.
The prior definition of sievert was, as far as I can tell, entirely wrong.

Caution: while "gray" and "sievert" have the same dimensionality,
they're not interchangeable -- you need to multiply by a fudge factor
that varies depending on the type of radiation and the tissue it's
affecting. (Dimensional analysis is often not a substitute for knowing
what you're doing.)

It would be nice if units had a way to warn users when they're trying
to do something that doesn't make sense, since there are lots of ways
to do so, but it doesn't, and it wouldn't be easy to arrange in the
general case.
2011-03-28 00:08:47 +00:00
wiz
7091eebfa9 Whitespace and punctuation. 2011-03-27 22:55:07 +00:00
wiz
971b386cde Oh no -- a space! get rid of it, quick. 2011-03-27 22:51:35 +00:00
wiz
93cb656370 Add RCS Id. 2011-03-27 22:47:40 +00:00
wiz
d6a06aa7e9 Fix a typo and a punctuation nit. 2011-03-27 22:44:47 +00:00
joerg
7435c9ba49 exit, not enter 2011-03-27 22:20:51 +00:00
joerg
23a805d314 Handle _rtld_exit as full entry point since other threads may still be
running at the time.
2011-03-27 21:58:50 +00:00
spz
2fbfb83e1b fix compiling with IPSEC_DEBUG:
it's authsize not authlen in struct auth_hash
2011-03-27 21:56:57 +00:00
riz
678d404217 Reorganize this file a bit, and move some stuff to "will not make it" to
give at least some semblance of realism.
2011-03-27 21:41:08 +00:00
agc
d337d7fa82 Sync NetBSD-6 plans with reality - update on features that have been
completed.
2011-03-27 21:21:50 +00:00
riz
d16ddb6294 Don't try to kmem_alloc() 0 bytes. Without this change, some trivial
kernel modules were not loadable by rump_server.
2011-03-27 21:16:52 +00:00
sjg
1e444cd2f2 Use curdirOk as the token 2011-03-27 19:47:46 +00:00
sjg
dd0a10eeff When .MAKE.MODE inlcudes 'meta'; 'curdir=true' enables creating .meta
files in .CURDIR - such as when running make in .OBJDIR with a generated
makefile.
2011-03-27 19:39:21 +00:00
phx
0e41517647 Fixed PHY access.
Support 8169SC/8110SC (as found on QNAP V200 boards).
Make frame receiving work (FRAMELEN <-> FRAMESIZE).
Driver works now, but not the first time after cold start.
2011-03-27 19:09:43 +00:00
martin
22699d7467 On second thought do not use bit 13 as a flag for syscall numbers, userland
typically sets those from a 13bit signed integer immediate field in the
instruction, so would need to jump through hoops (ok, small hoops) to
avoid sign extension.
Use a combination of the existing syscall flags instead.
2011-03-27 18:47:08 +00:00
bouyer
e5fa3d981b Make sure to not report quotas for a class if we didn't get valid datas
from kernel.
2011-03-27 17:15:17 +00:00
jruoho
1b871e25a1 Add a test case for pthread_cond_timedwait(3) failures reported by
Sad Clouds in PR lib/44756. This was discussed also in:

	http://mail-index.netbsd.org/tech-userlevel/2011/03/17/msg004689.html
2011-03-27 16:45:15 +00:00
plunky
bbe1bca01a if not including <bsd.prog.mk> then we need to clean up our own mess 2011-03-27 14:22:02 +00:00
ahoka
9e1e81e9f1 use nand_init_interface 2011-03-27 13:36:50 +00:00
ahoka
1d3d5c9972 Use nand_init_interface, which is required now. 2011-03-27 13:35:39 +00:00
ahoka
c68f5a8379 Add support for redifining page read and program functions by drivers.
Some controllers implement read/write in one step, so this is required
to support those.
2011-03-27 13:33:04 +00:00
joerg
2feaef06e3 Add some debug messages for explicit rtld entry points 2011-03-27 13:15:34 +00:00
joerg
0bc9be2143 Locking around dlclose() 2011-03-27 13:14:42 +00:00
njoly
119aff4da3 Fix sigprocmask section (3 -> 2). 2011-03-27 12:53:16 +00:00