Commit Graph

109687 Commits

Author SHA1 Message Date
scottl
6f7ab617d8 Don't force busdma to pre-allocate bounce pages for the parent tag. 2004-11-21 04:02:36 +00:00
tjr
4e61a637dc Remove regression tests for the obsolete rune interface. 2004-11-21 03:21:22 +00:00
tjr
1f410c73aa Fix computation of the 'n' argument to mbrtowc (through XMBRTOWC) to avoid
reading past 'stop' in various places when converting multibyte characters.
Reading too far caused truncation to not be detected when it should have
been, eventually causing regexec() to loop infinitely in with certain
combinations of patterns and strings in multibyte locales.

PR:		74020
MFC after:	4 weeks
2004-11-21 03:14:32 +00:00
mlaier
0ccf8f0d91 Commit ALTQ-patch for ed(4).
Requested and tested by:	pav
MFC after:			1 week
2004-11-21 02:42:09 +00:00
marcel
7a1c331305 o Support for the i8251 wasn't there. Remove the files.
o  Remove the headers with IC register definitions. The headers are
   now taken from sys/dev/ic
2004-11-21 01:51:37 +00:00
marcel
835ad23df6 Include the header with the register definitions from sys/dev/ic. They
are shared now.
2004-11-21 01:43:27 +00:00
marcel
1fa26ac061 This file was repocopied from sys/dev/uart/uart_dev_z8530.h. 2004-11-21 01:34:15 +00:00
marcel
725f195a21 This file was repocopied from sys/dev/uart/uart_dev_sab82532.h. 2004-11-21 01:33:39 +00:00
rwatson
47cc295ab6 Word choice fix; use .Fn for non-.Xr functions.
Submitted by:	ru
2004-11-21 00:24:21 +00:00
peter
62072eecc4 In the amd64 hybrid libraries case, move the kerberos5 tools to before
building the kerberos5 includes.  This is not the same patch that
Bjoern A. Zeeb came up with, but the credit still goes to him for finding
the problem.  Thanks!
2004-11-20 23:41:25 +00:00
marcel
5d458a18cb Stop building uart_dev_i8251.c. It was copied from uart_dev_ns8250.c
without ever being changed to actually work with an i8251. Nobody is
working on this either at the moment, so it's not about to change
soon.
When the code necessary to support the i8251 is committed, this can
be reverted again.
2004-11-20 23:38:48 +00:00
marcel
a0d6670526 Include the common <dev/ic/ns16550.h> instead of the private
<dev/uart_dev_ns8250.h>. The latter can be removed now.
2004-11-20 23:22:04 +00:00
marcel
b27f556929 o Remove the com_thr, com_rhr, com_isr and com_lctl defines. They are
not used and aliases for other defines.
o  Add REG_DATA as an alias for com_data. Likewise for other register
   defines.
o  Add LCR_SBREAK and make CFCR_SBREAK an alias for it. Likewise for
   the other LCR register bits that are known with the CFCR prefix.
o  Add MCR_IE and make MCR_IENABLE an alias for it.
o  Add LSR_TEMT and make LSR_TSRE an alias for it.
o  Add LSR_THRE and make LSR_TXRDY as alias for it.
o  Add FCR_ENABLE and make FIFO_ENABLE as alias for it. Likewise for
   the other FCR register bits that are known with the FIFO prefix.
o  Add EFR_CTS and make EFR_AUTOCTS an alias for it.
o  Add EFR_RTS and make EFR_AUTORTS an alias for it.

This is a first step in cleaning up the definitions in this file.
2004-11-20 23:19:42 +00:00
marcel
5819df2a6e o Remove opt_ddb.h from SRCS.
o  Add opt_gdb.h and opt_kdb.h to SRCS.
2004-11-20 23:04:32 +00:00
das
6175c08488 Remove local definitions of RANGEOF() and use __rangeof() instead.
Also remove a few bogus casts.
2004-11-20 23:00:59 +00:00
cognet
feeee6f8b5 Remove useless code. 2004-11-20 16:52:30 +00:00
cognet
75354c7e52 Implement enough to be able to enter and leave DDB. 2004-11-20 16:52:10 +00:00
cognet
c2bbd5fdeb Get the kernel stack right now that the u-area is gone. 2004-11-20 16:51:32 +00:00
delphij
2958d0c63c Forced commit to clarify that the last commit has enforced a
more strict check against PPP packets, but does not involve
a DoS vulnerability.

The ``DoS Vulnerability'' in last commit message (rev. 1.5) is
actually a false positive, because pppd(8) will fork for every
connection and thus the issue closed in rev. 1.5 will not block
subsequent connections, nor affect connections of other users.

Pointed out by: josef
2004-11-20 15:49:29 +00:00
fjoe
89f1126514 Use M_ZERO to not panic in mtx_init when INVARIANTS enabled.
Submitted by:	simokawa
MFC after:	1 week
2004-11-20 13:10:04 +00:00
marcel
f92a530b7d Sort MFILES. 2004-11-20 06:09:11 +00:00
brueffer
7500d809d9 The ixgb(4) driver actually first appeared in 5.3 (was merged to RELENG_4
after 4.10).

MFC after:	3 days
2004-11-20 03:10:35 +00:00
brueffer
81444750b4 Autogenerate device listings for hptmv(4) 2004-11-20 02:46:02 +00:00
brueffer
0ea90bb7e8 Add hptmv(4) 2004-11-20 02:44:47 +00:00
brueffer
b675dcb5d2 Small correction in the HISTORY section:
The hptmv driver appeared in 5.3, not 5.4.

MFC after:	3 days
2004-11-20 02:43:10 +00:00
das
252b79ceed Remove the declaration of uarea_pages.
Reviewed by:	arch@
2004-11-20 02:33:05 +00:00
das
e02efdf4e6 Remove the p_uarea and p_upages_obj fields from struct proc.
Prototype new routines to allocate, copy, and free pstats.

Reviewed by:	arch@
2004-11-20 02:32:50 +00:00
das
35f1e7649c Remove the uarea column from the DDB 'ps' display, and from grog's gdb
scripts.

Reviewed by:	arch@
2004-11-20 02:32:42 +00:00
das
917ee09309 Update the comment about what NO_SWAPPING does.
Reviewed by:	arch@
2004-11-20 02:32:15 +00:00
das
8d8b5ace18 Maintain the broken state of backwards compatibilty for a.out (and
PECOFF!) core dumps.  None of the old versions of gdb I tried were
able to read a.out core dumps before or after this change.

Reviewed by:	arch@
2004-11-20 02:32:04 +00:00
das
02205f8827 We don't do U area swapping anymore, so update some comments. Also,
prototype the new routine that creates a mock U area for a.out core
dumps.

Reviewed by:	arch@
2004-11-20 02:31:46 +00:00
das
dbdcb7be99 Remove UAREA_PAGES and USPACE definitions. The definitions of
USPACE_SVC_STACK_TOP, USPACE_SVC_STACK_BOTTOM, USPACE_UNDEF_STACK_TOP,
and USPACE_UNDEF_STACK_BOTTOM look wrong to me, so I'm leaving them
alone.

Reviewed by:	arch@
2004-11-20 02:31:37 +00:00
das
799ea28b61 Remove some references to U area here while trying not to break
anything.  Someone with ARM hardware could do a lot more to untangle
this code.

Reviewed by:	arch@
2004-11-20 02:31:23 +00:00
das
901a681d75 Remove references to U area and garbage collect includes.
Reviewed by:	arch@
2004-11-20 02:30:59 +00:00
das
6f19619485 gcore(1) apparently still cares about a.out core dumps, so it still
needs to know that a 1-page U area is part of the dump format.

Reviewed by:	arch@
2004-11-20 02:30:02 +00:00
das
90a65a896e Remove UAREA_PAGES.
Reviewed by:	arch@
2004-11-20 02:29:50 +00:00
das
c072a76d90 U areas are going away, so don't allocate them. It's worrisome that
mp_machdep.c was using UAREA_PAGES to allocate something that isn't a
U area, and that there seems to be an implicit assumption that the PCB
is just past the end of the kernel stack.

Reviewed by:	arch@
2004-11-20 02:29:36 +00:00
das
8375566745 U areas are going away, so don't allocate one for process 0.
Reviewed by:	arch@
2004-11-20 02:29:25 +00:00
das
af608beb40 Disable U area swapping and remove the routines that create, destroy,
copy, and swap U areas.

Reviewed by:	arch@
2004-11-20 02:29:00 +00:00
das
22907ad4ac Malloc p_stats instead of putting it in the U area. We should consider
simply embedding it in struct proc.

Reviewed by:	arch@
2004-11-20 02:28:48 +00:00
das
bab4eb89e3 In preparation to remove U areas, don't refer to p_uarea in libkvm.
Reviewed by:	arch@
2004-11-20 02:28:26 +00:00
das
ec9f9971e7 user.h is included only to get pcb.h, so use the latter directly instead. 2004-11-20 02:28:14 +00:00
jkoshy
efe0d253ae Bring signal(9) in sync with -current.
Reviewed by:	ru (mdoc changes)
2004-11-20 02:11:42 +00:00
cognet
d28f205253 Increase cnt.v_syscall and cnt.v_trap when needed. 2004-11-20 00:55:57 +00:00
ru
339fc9c24a Do not emit a spurious warning when "directory" argument
to "rmdir -p" is absolute, with one or more leading slash.
2004-11-20 00:41:08 +00:00
ru
f5e118fbbd Fixed "rmdir -p" that got broken by rev. 1.15.
(This also fixes "rmdir -v /nonexistent".)
2004-11-20 00:12:08 +00:00
glebius
c52af354ac Remove unused macro.
Approved by:	julian (mentor)
2004-11-19 22:35:53 +00:00
bz
5035152f77 Document NO_NIS variable.
Submitted by:	Pawel Worach <pawel dot worach at telia dot com>
Approved by:	rwatson (mentor)
MFC after:	1 week
2004-11-19 22:23:57 +00:00
glebius
4eeaf10421 Cancel pending timeout before scheduling a new one. This fixes
item leak, I've introduced with previous change.

Approved by:	julian (mentor)
2004-11-19 20:23:46 +00:00
emax
9f3a10cb66 Correct typo. Return ENXIO instead of EIO. 2004-11-19 18:54:54 +00:00