Commit Graph

10399 Commits

Author SHA1 Message Date
phk
ead640e967 setsockopt() transports user option data in an mbuf. if the user
data is greater than MLEN, setsockopt is unable to pass it onto
the protocol handler.  Allocate a cluster in such case.

PR:		2575
Reviewed by:	 phk
Submitted by:	Julian Assange proff@iq.org
1998-04-11 20:31:46 +00:00
dyson
9221f256f8 Minor typo in the wd driver. The manifestation of this bug
is a tremendous perf decrease due to the disabling of advanced
features such as DMA, Ultra DMA, and 32bit mode.  This patch
might have been reported by someone else (I seem to remember
it.)
1998-04-11 20:09:39 +00:00
phk
87312aadcc When pmap_pinit0() allocates a page for proc0's page directory,
kernal page table may need to be extended.  But while growing the
kernel page table (pmap_growkernel()), newly allocated kernel page
table pages are entered into every process' page directory. For
proc0, the page directory is not allocated yet, and results in a
page fault.  Eventually, the machine panics with "lockmgr: not
holding exclusive lock".

PR:		5458
Reviewed by:	phk
Submitted by:	Luoqi Chen <luoqi@luoqi.watermarkgroup.com>
1998-04-11 17:24:06 +00:00
sos
a2dbd182f5 Do not clobber "heads" by &= in wdcommand use only &.
Spotted by: bde
1998-04-10 08:00:24 +00:00
alex
dd07a29831 Grammar police. 1998-04-10 00:09:04 +00:00
sos
8846bf7678 Add the new LBA mode support in the wd? config examples. 1998-04-09 22:28:57 +00:00
sos
42733ffc16 Try to use the "right" CHS translations of a LBA device.
Drives bigger than 8.4G is still in question until I get a drive
to test on...
1998-04-09 17:46:45 +00:00
nate
dc4e7fd7a9 - Whoops, better have the structure created before trying to use it in
the patch I just submitted.

Noticed by:     phk
1998-04-09 14:01:13 +00:00
sos
a5c9554f61 Add LBA mode support for large drives.
Use config flags 0x1000 to enable LBA mode. It should be enabled in
the BIOS too to avoid geometry confusion.

One catch though, I'm not sure all BIOS's uses the 64head/63secs
translation, all mine does but....
1998-04-08 20:04:39 +00:00
sos
6ee5cb0947 Fix a minor bug (|| instead of |) 1998-04-08 20:00:03 +00:00
wosch
e3aed30232 New mount option nosymfollow. If enabled, the kernel lookup()
function will not follow symbolic links on the mounted
file system and return EACCES (Permission denied).
1998-04-08 18:31:59 +00:00
nate
4c738b0efc - Fix bug I introduced a few months ago. If a driver fails the probe,
correctly unregister the interrupt from the system.

[ My fix is much simpler than the one provided in the PR ]

PR:		6249
1998-04-08 15:00:02 +00:00
phk
c641c2b9ee Minor adjustments to the timecounting and proc0.
Mostly Submitted by:	bde
1998-04-08 09:01:53 +00:00
kato
0c131782a5 Sync with sys/i386/boot/biosboot/README.serial revision 1.9. 1998-04-07 15:33:47 +00:00
kato
ab8f0419da Sync with sys/i386/isa/npx.c revision 1.58. 1998-04-07 09:09:41 +00:00
kato
7354b70a08 Sync with sys/i386/conf/files.i386 revision 1.196. 1998-04-07 09:08:27 +00:00
kato
c04e14dfb2 Sync with sys/i386/i386/machdep.c revision 1.294. 1998-04-07 09:07:53 +00:00
peter
33b5f6b63d Today is not my lucky day. Fix missing brace and I got a request
to use EMLINK instead.
1998-04-06 19:32:37 +00:00
peter
b5d3779b73 Fix VM86 compiles. a #include "opt_vm86.h" was missing, and the my_tr
variable was needed in the non-SMP case.

Submitted by:  Jonathan Lemon <jlemon@americantv.com>
1998-04-06 18:59:15 +00:00
peter
db031c687b Use a different errno (ELOOP (as sef mentioned) since the text that goes
with the error sounds ok for the condition) if O_NOFOLLOW gets a link.
1998-04-06 18:43:28 +00:00
peter
b0a513624f Rather than let users get fd's to symlink files, make O_NOFOLLOW cause
an error if it gets a link (like it does if it gets a socket).  The
implications of letting users try and do file operations on symlinks
themselves were too worrying.
1998-04-06 18:25:21 +00:00
peter
14414bee85 Back this out, allowing users to get a fd connected to a symlink is
just too dangerous.
1998-04-06 18:18:50 +00:00
peter
b2396f73c8 Don't panic if a VOP_READ() gets through on a short link, Just Do It
(because we can :-).  This means you can open a link file (or pseudo-file
in the case of short links where the data is stored in the inode rather
than disk blocks) and read the contents.
However, trap any writes from the user as it's difficult to do the right
thing in all cases.  A link may be short and the user may be trying to
extend it beyond the limit and so on.  Although.. being able to re-target
a symlink without deleting it first might have been nice.
This stuff is a bit perverse since symlink() and readlink() calls can
end up actually being implemented as read/write vnode ops.

Reviewed by: phk
1998-04-06 17:44:40 +00:00
peter
d5ab1c3759 Implement a new open(2) flag: O_NOFOLLOW. This will instruct open
to not follow symlinks, but to open a handle on the link itself(!).
As strange as this might sound, it has several useful applications
safe race-free ways of opening files in hostile areas (eg: /tmp, a mode
1777 /var/mail, etc).  It also would allow things like fchown() to work
on the link rather than having to implement a new syscall specifically for
that task.

Reviewed by: phk
1998-04-06 17:38:43 +00:00
peter
6e3ec235ff curproc is initialized in locore at the same time for both SMP and UP now. 1998-04-06 15:51:22 +00:00
peter
93c5e9e460 remove #ifdef declaration of npxproc, use globals.s and the extern always. 1998-04-06 15:50:29 +00:00
peter
3a2f3b0303 add globals.s for data that is treated differently on SMP. 1998-04-06 15:49:35 +00:00
peter
32a92c71b6 Use real types for the SMP pages being allocated rather than arrays of
ints.  Remove some no longer needed casts.  Initialize the per-cpu
global data area using the structs rather than knowing too much about
layout, alignment, etc.
1998-04-06 15:48:30 +00:00
peter
1a4b1c6550 clean up #ifdefs, define the variables that have to be per-cpu on SMP
in globals.s only and use externs always.
1998-04-06 15:46:17 +00:00
peter
5164f19dff _curpcb is always defined in globals.s instead of here in #ifdefs 1998-04-06 15:44:31 +00:00
peter
7f380682ae Bogus casts 1998-04-06 15:43:29 +00:00
peter
1e48b96ed9 Defunct, now part of globals.s 1998-04-06 15:42:54 +00:00
peter
62b8130425 Rather than filling this file up with SMP .sets, use those from
globals.s instead.
Initialize curproc in the same place for both UP and SMP.
1998-04-06 15:42:26 +00:00
peter
a31fe28469 Generate #defines that the asm code can access for the per-cpu data
structures.
1998-04-06 15:40:10 +00:00
peter
d7e0d95d1e generate .sets for variables used in asm and C that are stored in per-cpu
space under SMP.
1998-04-06 15:39:09 +00:00
peter
531b8c0e84 A pair of C structures used for laying out the SMP per-cpu data space. 1998-04-06 15:37:21 +00:00
peter
45ae9cd4b8 some missing symbols 1998-04-06 15:33:02 +00:00
phk
45ced96af0 Use getmicrotime() for if_lastchange, 10msec is plenty precision. 1998-04-06 11:43:12 +00:00
phk
fe94bc8288 Use random() to find our initial xid. 1998-04-06 11:41:07 +00:00
phk
4b6dfe92bd Use read_random() 1998-04-06 11:40:42 +00:00
phk
1e2405b5f1 Use random() for seq numbers and read_random for CHAP challenge. 1998-04-06 11:40:17 +00:00
phk
5394eb1aa3 Use random() rather then than homegrown stuff. 1998-04-06 11:39:04 +00:00
phk
b4d5f00838 Remove stuff related to microtime.s, which is gone. 1998-04-06 11:38:18 +00:00
phk
86ff9d1a6e Use getmicrotime insted of microtime, we only use the second part. 1998-04-06 11:37:17 +00:00
phk
ab5541db4c Make read_random() take a (void *) argument instead of (char *) 1998-04-06 09:30:42 +00:00
phk
3c122bd961 Make a kernel version of the timer* functions called timerval* to be
more consistent.

OK'ed by:	bde
1998-04-06 08:26:08 +00:00
ahasty
011360bdd9 take out opt_bktr.h from brooktree848.c 1998-04-06 07:57:55 +00:00
phk
e9827cb58f Remove the last traces of TUBA.
Inspired by:	PR kern/3317
1998-04-06 06:52:47 +00:00
peter
631390e349 Really make this compile..
Pointed out by: bde
1998-04-06 05:11:42 +00:00
kato
81a8571091 Sync with sys/i386/isa/clock.c revision 1.119. 1998-04-06 03:38:18 +00:00