Commit graph

85051 commits

Author SHA1 Message Date
Maxime Henrion 1d8dc7e4a3 Don't try to free() map in bus_dmamap_destroy() when it's
set to &nobounce_dmamap.  A similar bug was fixed by wpaul
in revision 1.19 of sys/alpha/alpha/busdma_machdep.c.
2003-01-18 18:33:56 +00:00
Sebastien Gioria 411359429f Allow French releases notes to be build correctly
Hat To:  Hiroki SATO  <hrs@eos.ocn.ne.jp>
2003-01-18 18:03:01 +00:00
Udo Erdelhoff 8b521c8f4a MFbed: translation updates and fixes
early-adopter/article.sgml:	fixes only
errata/article.sgml:		1.17 -> 1.18
2003-01-18 16:42:18 +00:00
Poul-Henning Kamp e9fe7d1f7f Inline now trivial functions getccdbuf() and putccdbuf().
Fix another trivial memory-leak.
2003-01-18 12:23:49 +00:00
Dag-Erling Smørgrav 991a7df22b Don't truncate lines if an error occurred. 2003-01-18 12:15:34 +00:00
Dag-Erling Smørgrav ea54c2afef Indentation cleanup. 2003-01-18 12:12:49 +00:00
Peter Grehan 8a7f89713e Removed unnecessary includes and brought up to date with ata
common code by adding lock functions.
2003-01-18 11:46:50 +00:00
Poul-Henning Kamp 6b267654ab Fix minor memory-leak. 2003-01-18 11:33:06 +00:00
Peter Grehan f48649e525 Stub profile.h, required for userland builds.
Approved by:  Benno
2003-01-18 11:31:58 +00:00
Peter Grehan 4e84a94d6b <machine/ieee.h>, taken from sparc64
Approved by:  Benno
2003-01-18 11:30:59 +00:00
Peter Grehan 8e9238c604 Fix bugs with operand ordering and unnecessary sync/eieio ops. Mostly
obtained from Alpha atomic.h

Approved by:  Benno
2003-01-18 11:28:36 +00:00
Peter Grehan add0e67094 Allow the MD frame definition to be seen in. Required for truss/ptrace.
Approved by:  Benno
2003-01-18 11:25:21 +00:00
Peter Grehan 5cbdd8139a RAIDframe requires LONG_BIT
Approved by:  Benno
2003-01-18 11:23:42 +00:00
Peter Grehan 0a9b03cb65 Prepended underscores to macro local vars, avoiding gcc "declaration
shadows global" warning

Approved by:  benno
2003-01-18 11:20:06 +00:00
Peter Grehan 0212856f3c Change definition of int64 to avoid gcc3.2.1 complaints. Taken from i386
Approved by:  benno
2003-01-18 11:18:42 +00:00
Poul-Henning Kamp b51ea35631 Use the M_CCD malloc bucket instead of M_DEVBUF.
Don't keep a private freelist of a low number of trivially small structures.
2003-01-18 11:04:41 +00:00
Poul-Henning Kamp 360d71d1b5 Inline trivial function ccdintr() into its one caller ccdiodone().
Only call ccdfind() once in ccdiodone() and cache the result.
2003-01-18 10:44:17 +00:00
Poul-Henning Kamp b02a19f2b7 Fix two errorchecks to check for negative error returns. 2003-01-18 10:29:24 +00:00
Poul-Henning Kamp 3b1746df8b Sanitize the copyright section a bit: We do not need two copies of the
four-clause BSD license in the file, one will do.
2003-01-18 10:08:27 +00:00
Yoshihiro Takahashi bb0e369994 Remove unneeded entries. 2003-01-18 08:29:10 +00:00
Yoshihiro Takahashi 1f23083ab9 Fixed comment. 2003-01-18 08:26:42 +00:00
Juli Mallett e0328ede47 Consistentify output whitespace. 2003-01-18 06:29:15 +00:00
Bill Fumerola 22c530b927 MFCrypto: update RFC assigned telnet options (r1.9) 2003-01-18 06:13:57 +00:00
Bill Fumerola a3bb66de04 add more RFC defined telnet options
Reviewed by:	ps
2003-01-18 06:10:21 +00:00
Mike Makonnen 276ae160e6 Make the debug messages, for when a script rolls its own start routine,
more usefull. Currently it just says 'xxx_cmd'; whereas with this
change it will name the routine that's being run.

Approved by: markm (mentor)
2003-01-18 05:25:59 +00:00
Juli Mallett 063a27dc9b Fix typo. 2003-01-18 05:06:07 +00:00
Matthew N. Dodd 95ca467a68 - Removed unecesary code.
- Correct smapi32_new() asm (still doesn't work.)
- Attach to mainboard not isa.
2003-01-18 04:36:12 +00:00
Juli Mallett 49b2a6863b Nuke dumb error reporting code, people can just use disk::d_error. Unify the
DEBUG and d_error initialisation into an ERROR macro, which can both trace and
set the d_error field.  Much a more meaningful thing, I should say.
2003-01-18 04:22:14 +00:00
Juli Mallett cbd866f411 Simplify conditional. 2003-01-18 04:20:20 +00:00
Juli Mallett 775bba9fd8 Refer to the process label as proclabel, as there is a function called label,
and that's what these locals were called before.
2003-01-18 03:31:30 +00:00
Robert Watson eef9bd4c3f Caution programmers not to confuse M_DONTWAIT with M_NOWAIT. They
are not the same.

Suggested by:	Hiten Pandya <hiten@unixdaemons.com>
2003-01-18 03:25:40 +00:00
Marcel Moolenaar da5e9a5bd6 MFp4: Add support for memory mapped UARTs, but don't add any devices
yet that depend on it because sio(4) needs support for it before it
can be used. There's no reason why zs(4) couldn't attach to puc(4)
in the (near?) future (in principle), so don't make memory mapped I/O
support in sio(4) a precondition for this change.
2003-01-18 02:54:16 +00:00
Marcel Moolenaar 5f85bb36ee If we're going to build and install this on ia64, we might as well
teach it about ia64 specific section types, dynamic tags and machine
type. This is a mostly insignificant change given the amount of
work that this tool obviously needs...
2003-01-18 02:33:49 +00:00
Takeshi Shibagaki 5770575cfc Add BUFFALO LPC-CF-CLT(10Base-T Compact Flash Ether Card). 2003-01-18 02:14:23 +00:00
Takeshi Shibagaki d7c38c4ccc Regen. 2003-01-18 02:13:10 +00:00
Takeshi Shibagaki 708d4a51e6 Add Accton EN2216. 2003-01-18 01:14:48 +00:00
Takeshi Shibagaki 4779186a05 Regen. 2003-01-18 01:14:01 +00:00
Jeffrey Hsu f5c5746047 Fix long-standing bug predating FreeBSD where calling connect() twice
on a raw ip socket will crash the system with a null-dereference.
2003-01-18 01:10:55 +00:00
Tim J. Robbins d1841903a0 Fake up a struct componentname to pass to VOP_WHITEOUT instead of passing
NULL. union_whiteout() expects the componentname argument to be non-NULL.
Fixes a NULL dereference panic when an existing union mount becomes the
upper layer of a new union mount.
2003-01-18 01:01:20 +00:00
Poul-Henning Kamp 6fb6c30600 Remove subr_diskslice.c and subr_diskmbr.c which I can see no traces off
a need for in the x86_64 files.  Not compile tested.
2003-01-17 23:02:13 +00:00
Poul-Henning Kamp 03cb40281e Remove subr_diskslice.c and subr_diskmbr.c which I can find no trace of
why should be needed in the powerpc files.  Not compile tested.
2003-01-17 23:01:09 +00:00
Matthew N. Dodd 146cc87269 - Style fixes.
- llabs() -> qabs(); long long isn't valid for the kernel.

Submitted by:	 bde
2003-01-17 22:58:32 +00:00
Poul-Henning Kamp f527e9857e ia64 didn't need subr_diskslice.c and subr_diskmbr.c either. 2003-01-17 22:09:12 +00:00
Poul-Henning Kamp 123bf099ea We do not need subr_diskslice.c and subr_diskmbr.c any more on alpha. 2003-01-17 20:02:46 +00:00
Poul-Henning Kamp ffffe9203f Move alpha_fix_srm_checksum() from subr_diskmbr.c to subr_disklabel.c 2003-01-17 19:37:55 +00:00
Poul-Henning Kamp 40f683a443 Remove the unused DSO_* options. 2003-01-17 19:36:14 +00:00
Thomas Moestl 1902ba0c83 Document that listen() can return EINVAL now. 2003-01-17 19:25:27 +00:00
Thomas Moestl 6f7cab9301 Disallow listen() on sockets which are in the SS_ISCONNECTED or
SS_ISCONNECTING state, returning EINVAL (which is what POSIX mandates
in this case).
listen() on connected or connecting sockets would cause them to enter
a bad state; in the TCP case, this could cause sockets to go
catatonic or panics, depending on how the socket was connected.

Reviewed by:	-net
MFC after:	2 weeks
2003-01-17 19:20:00 +00:00
Poul-Henning Kamp 5269d202cc Sparc64 does not need subr_diskslice.c and subr_mbr.c now. 2003-01-17 19:15:26 +00:00
Tony Finch 00c9c61ae5 Style fixes: brackets around the argument of return, and removal of
excess space.

Obtained from: OpenBSD
2003-01-17 19:12:02 +00:00