d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
26,579 Commits 4 Branches 49 Tags 2 GiB
0fc319ac44
Commit Graph

26579 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Steve Price
694ad0a9b1 Fix a couple of operator precedence bugs. 
PR:		5450
Submitted by:	Sakari Jalovaara <sja@tekla.fi>
 1998-01-25 17:25:41 +00:00
KATO Takenori
31e6b13443 Added cpuid instruction. 1998-01-25 17:02:00 +00:00
KATO Takenori
2f5a8c876a Execute cpuid if BIOS disables cpuid instruction of Cyrix 6x86MX CPU, 
and store its result into cpu_id and cpu_feature variables.

Tested by:	Simon Coggins <chaos@ultra.net.au>
 1998-01-25 17:01:31 +00:00
Steve Price
9e7a1d01b0 Use -literal to make for a more pleasing list. 
PR:		5547
Submitted by:	Matthew Hunt <mph@pobox.com>
 1998-01-25 16:38:42 +00:00
Steve Price
2ccbe549bc Fix a couple of typos. 
PR:		5561
Submitted by:	Kazuo Horikawa <k-horik@yk.rim.or.jp>
 1998-01-25 16:22:22 +00:00
Kazutaka YOKOTA
65b8196101 Fix typo. 
PR: docs/5562
 1998-01-25 15:49:00 +00:00
KATO Takenori
88a92ff10e Even though BIOS writer's guide recommends cpuid instruction of Cyrix 
6x86MX CPU is enabled (BIOS should not disable it), some BIOS disables
it via CCR4.  In this case, cpu variable becomes CPU_486 and
identblue() is called.  Because Cyrix 6x86MX has MSR and doesn't have
MSR1002, wrmsr instruction generates general protection fault.

Tested by:	Simon Coggins <chaos@ultra.net.au>
 1998-01-25 12:01:38 +00:00
John Birrell
44626f8fec Make machine dependent asm code machine dependent. 8-) 1998-01-25 09:52:32 +00:00
John Birrell
3b306d567c Change MACHINE -> MACHINE_ARCH so that the design make sense on those 
machines where the processor chip determines the compiler, and where
multiple machines use the same architecture.
 1998-01-25 09:49:00 +00:00
KATO Takenori
40f6a786db Sync with sys/i386/conf/options.i386 revision 1.67. 1998-01-25 09:45:38 +00:00
John Birrell
2a4a5158b6 Modify for MACHINE_ARCH support instead of assuming just x86. 
Added alpha and m68k stuff.
 1998-01-25 09:45:00 +00:00
KATO Takenori
9758931d62 Fixed typo in comment. 1998-01-25 09:44:33 +00:00
John Birrell
155ad7517d signal() returns an pointer, so must be cast to long, not int before 
comparing < 0.
 1998-01-25 09:16:55 +00:00
John Birrell
e3f5655fab read() and write() return ssize_t not int, so the first arg to atomic() 
should be ssize_t too.
 1998-01-25 09:15:10 +00:00
John Birrell
b0ed821446 Add a sysmk -> /dev/null definition ready for a coming change to 
makesyscalls.sh.
 1998-01-25 09:09:48 +00:00
John Birrell
c1e8822ae4 Cast a size_t to a long to avoid a gcc warning. 1998-01-25 08:59:08 +00:00
John Dyson
33b90a70cd Various NFS fixes: 
Make vfs_bio buffer mgmt work better.
	Buffers were being used after brelse.
	Make nfs_getpages work independently of other NFS
		interfaces.  This eliminates some difficult
		recursion problems and decreases pagefault
		overhead.
	Remove an erroneous vfs_unbusy_pages.
	Fix a reentrancy problem, with nfs_vinvalbuf when
		vnode is already being rundown.
	Reassignbuf wasn't being called when needed under
		certain circumstances.

	(Thanks to Bill Paul for help.)
 1998-01-25 06:24:09 +00:00
John Dyson
8447a0aed7 Various NFS fixes: 
Make vfs_bio buffer mgmt work better.
	Buffers were being used after brelse.
	Make nfs_getpages work independently of other NFS
		interfaces.  This eliminates some difficult
		recursion problems and decreases pagefault
		overhead.
	Remove an erroneous vfs_unbusy_pages.
	Fix a reentrancy problem, with nfs_vinvalbuf when
		vnode is already being rundown.
	Reassignbuf wasn't being called when needed under
		certain circumstances.

	(Thanks for help from Bill Paul.)
 1998-01-25 06:14:26 +00:00
Eivind Eklund
bed196b4e2 Turn CLK_USE_TSC_CALIBRATION into a new-style option. 1998-01-25 04:37:47 +00:00
Eivind Eklund
92252381f3 Make TCP_COMPAT_42 a new style option. 1998-01-25 04:23:33 +00:00
Eivind Eklund
f71bb3a160 Turn NSWAPDEV into a new-style option. 1998-01-25 04:13:25 +00:00
Eivind Eklund
c007c0bc64 Remove the FDSEEKWAIT option and description. 
The functionality was present for two days in october/november 1994
before being backed out; I don't think we can consider it really
critical ;-)
 1998-01-25 03:55:47 +00:00
Eivind Eklund
455bb2116c Make MD5 a new-style option. This one was eeeeeaaaaaasssssyyyyy ;-) 1998-01-25 03:34:17 +00:00
KATO Takenori
3d2db2b0dc Sync with sys/i386/conf/GENERIC revision 1.105. 1998-01-25 02:22:04 +00:00
John Birrell
2b3f4eadb0 Fix a cast from a pointer to a long instead of an int which was enough 
to ruin a 64-bit day.
 1998-01-24 20:57:38 +00:00
Jordan K. Hubbard
99e6f3d3e3 Add script for properly breaking out the doc dist. 1998-01-24 17:45:14 +00:00
Kazutaka YOKOTA
1f2ea3a434 Add brief note on error message from the driver. 1998-01-24 12:14:13 +00:00
Kazutaka YOKOTA
a364145b79 Small adjustment in sync check code. 1998-01-24 12:12:32 +00:00
Jordan K. Hubbard
b59d8897c6 Clean up doc bootstrap ports afterwards. 
Small stylistic cleanup.
 1998-01-24 07:30:55 +00:00
KATO Takenori
55bcdaeda5 Sync with sys/i386/isa/wd.c revision 1.148. 1998-01-24 06:54:17 +00:00
KATO Takenori
19eba5818f Sync with sys/i386/i386/machdep.c revision 1.283. 1998-01-24 06:53:32 +00:00
Steve Price
86fb910326 Random data is no longer based on the time of day, rather the random 
number device.  Also clearly state that sequential data is the default.

PR:		5553
Submitted by:	Jonathan Hanna <pangolin@rogers.wave.ca>
 1998-01-24 03:32:33 +00:00
KATO Takenori
7ebb9221f6 Sync with sys/i386/i386/machdep.c revision 1.282. 1998-01-24 03:15:37 +00:00
Eivind Eklund
7b778b5e61 Make all file-system (MFS, FFS, NFS, LFS, DEVFS) related option new-style. 
This introduce an xxxFS_BOOT for each of the rootable filesystems.
(Presently not required, but encouraged to allow a smooth move of option *FS
to opt_dontuse.h later.)

LFS is temporarily disabled, and will be re-enabled tomorrow.
 1998-01-24 02:54:56 +00:00
Eivind Eklund
7f4ba71355 LKM-support for converting all file-system options to new-style options. 1998-01-24 02:21:53 +00:00
John Dyson
50ce7ff499 Add better support for larger I/O clusters, including larger physical 
I/O.  The support is not mature yet, and some of the underlying implementation
needs help.  However, support does exist for IDE devices now.
 1998-01-24 02:01:46 +00:00
Brian Somers
4147817ab9 Fix error message. 
Don't complicate the test compilation mentioned at the start of
the file.
 1998-01-24 00:03:14 +00:00
Brian Somers
2e8ab8d16a Stop bogus warnings. 1998-01-23 22:29:16 +00:00
Brian Somers
813bfe4d88 Re-implement the proxy arp sockaddr_dl search routine (again). 
It now works :-) and searches all interface aliases.
 1998-01-23 21:37:27 +00:00
Jordan K. Hubbard
e97fdb2506 Bring back handbook and FAQ by adding a mechanism for bootstrap ports 
to the build.  This will need more review as we go into it, but it should
do the basic job for now.
 1998-01-23 20:15:54 +00:00
Jordan K. Hubbard
5f19805b10 Add FAQ and handbook back to doc subtree. 1998-01-23 20:11:46 +00:00
Jordan K. Hubbard
c4be71c254 DISTRIBUTION=bin - we separate it out later. 1998-01-23 20:09:21 +00:00
Jordan K. Hubbard
feb308afd5 In !dists case, invoke the custom dist menu rather than the toplevel 
dist menu.  This makes a good deal more sense and should have been
this way all along.
 1998-01-23 07:53:44 +00:00
Jordan K. Hubbard
e0e1133124 spelling and style police. 1998-01-23 07:48:54 +00:00
Brian Somers
0c2bc41c96 Don't allow manual dialing unless in auto or interactive mode. 1998-01-23 04:36:42 +00:00
Brian Somers
005f91820b Ask spammers to mail postmaster@$m, not postmaster@FreeBSD.ORG. 1998-01-23 01:58:49 +00:00
Brian Somers
c5227e754c Just list one of each domain that was rejected 
(although only by Scheck_relay or Scheck_mail).
Suggested by: ache
 1998-01-23 01:58:18 +00:00
Jordan K. Hubbard
5dcabf2d07 Add wfd to all target, reorder devs to put like with like since we lost 
all pretense of an alphabetical ordering some time back.
 1998-01-22 21:26:22 +00:00
Jordan K. Hubbard
c922e45175 MF22: match entries for wfd 1998-01-22 21:14:31 +00:00
Søren Schmidt
f5dee08294 Use /bin/pwd instead of a potentially shell builtin that does 
not return the "real" path.
This makes it possible to have the .o files put in the /usr/obj
hierachy instead of in the /usr/src one.
 1998-01-22 19:34:31 +00:00