Commit Graph

33947 Commits

Author SHA1 Message Date
Alan Cox
7f0bf36a2e o Extend the scope of the page queues lock in contigmalloc1().
o Replace vm_page_sleep_busy() with vm_page_sleep_if_busy()
   in vm_contig_launder().
2002-08-04 07:07:34 +00:00
Alan Cox
4453ada654 o Convert a vm_page_sleep_busy() into a vm_page_sleep_if_busy()
with appropriate page queue locking.
2002-08-04 06:27:37 +00:00
Alan Cox
ea5e5b13f8 o Request a wired page from vm_page_grab() in _pmap_allocpte(). 2002-08-04 04:55:31 +00:00
Jake Burkholder
7746cd480a Print out the strings in vers.o instead of hardcoding the loader banner. 2002-08-04 01:30:26 +00:00
Matthew N. Dodd
9ccba881d9 Kernel modifications necessary to allow to follow fork()ed children.
PR:		 bin/25587 (in part)
MFC after:	 3 weeks
2002-08-04 01:07:02 +00:00
Alan Cox
b9c51c912e o Ask for a prezeroed page in pmap_pinit() for the page directory page. 2002-08-03 20:24:51 +00:00
Alan Cox
3327872297 o Convert two instances of vm_page_sleep_busy() to vm_page_sleep_if_busy()
with appropriate page queue locking.
2002-08-03 18:59:19 +00:00
Luigi Rizzo
ea779ff36c Fix handling of packets which matched an "ipfw fwd" rule on the input side. 2002-08-03 14:59:45 +00:00
Alan Cox
5da2d6a46d o Don't set PG_MAPPED on the page allocated and mapped in _pmap_allocpte().
(Only set this flag if the mapping has a corresponding pv list entry,
   which this mapping doesn't.)
2002-08-03 06:42:30 +00:00
Peter Wemm
af932613d7 Ignore memory above 4GB for now due to unpleasant pci issues. 2002-08-03 03:35:06 +00:00
Alan Cox
aa9b1d9412 o Remove the setting of PG_MAPPED from vm_page_wire() and
vm_page_alloc(VM_ALLOC_WIRED).
2002-08-03 01:29:52 +00:00
Peter Wemm
8f1586dd65 Take advantage of the fact that there is a small 1MB direct mapped region
on x86 in between KERNBASE and the kernel load address.  pmap_mapdev()
can return pointers to this for devices operating in the isa "hole".
2002-08-03 01:02:37 +00:00
Peter Wemm
1af04fadd1 Take a shot at fixing a nasty bug in the pmap changes that I did. I
missed the pmap_kenter/kremove in this file, which leads to read()/write()
of /dev/mem using stale TLB entries. (gah!)  Fortunately, mmap of /dev/mem
wasn't affected, so it wasn't as bad as it could have been.  This throws
some light on the 'X server affects stability' thread....

Pointed out by:	bde
2002-08-03 00:59:15 +00:00
Warner Losh
a0cd6004a9 We now have opt_wi.h, so go ahead and add it. Default it to 1. 2002-08-03 00:21:29 +00:00
Warner Losh
073eef8ca2 Add code that will download firmware to a Symbol LA4100-series of CF
cards.  Since the firmware is hard coded into the kernel, I've made it
a kernel option (WI_SYMBOL_FIRMWARE).

Note: This only downloads into the RAM of these cards.  It doesn't
download into FLASH, and is somewhat limited.  There needs to be a
better way to deal, but this works for now.  My Symbol LA4132 CF card
works now.

Obtained from: NetBSD
2002-08-03 00:19:58 +00:00
Warner Losh
b8182b24d7 While PCI interrupts are shareable, this should not have been committed just
yet.
2002-08-02 22:26:03 +00:00
Maxime Henrion
f2b17113cf Make the consumers of the linker_load_file() function use
linker_load_module() instead.

This fixes a bug where the kernel was unable to properly locate and
load a kernel module in vfs_mount() (and probably in the netgraph
code as well since it was using the same function).  This is because
the linker_load_file() does not properly search the module path.

Problem found by:	peter
Reviewed by:		peter
Thanks to:		peter
2002-08-02 20:56:07 +00:00
Robert Watson
4b32dfdcd7 Introduce support for Mandatory Access Control and extensible
kernel access control.

When generating nd6 output on an interface, label the packet
appropriately.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-02 20:49:14 +00:00
Robert Watson
e316463a86 When preserving the IP header in extra mbuf in the IP forwarding
case, also preserve the MAC label.  Note that this mbuf allocation
is fairly non-optimal, but not my fault.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-02 20:45:27 +00:00
Robert Watson
01e990be89 Improve ordering of MAC entry points in mac_policy_op structure.
Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-02 20:33:05 +00:00
Alan Cox
00f9e8b421 o Convert two instances of vm_page_sleep_busy() into vm_page_sleep_if_busy()
with appropriate page queue locking.
2002-08-02 18:55:29 +00:00
Poul-Henning Kamp
14a3a6ea0b Move a prototype to the least wrong place.
Suggested by:	bde
2002-08-02 18:45:43 +00:00
Peter Wemm
6a905efed3 ptvmmap does not exist on alpha. 2002-08-02 18:40:01 +00:00
Robert Watson
09a555cbf9 Work to fix LINT build.
Reported by:	phk
2002-08-02 18:08:14 +00:00
Poul-Henning Kamp
5f9dcc4468 SYSINIT needs to be SI_SUB_PSEUDO. Add a printf to tell we are here. 2002-08-02 16:47:29 +00:00
Poul-Henning Kamp
007054391b typo. 2002-08-02 15:55:18 +00:00
Poul-Henning Kamp
927b6b099d Add the minimalist elan-mmcr device driver.
This driver allows a userland program to mmap the MMCR of the AMD
Elan sc520 CPU.
2002-08-02 15:53:04 +00:00
Mark Murray
052c7c89c2 Modernise the cdevsw WRT to (unused) kqueue. 2002-08-02 11:24:43 +00:00
Scott Long
8db4c2f20c Calculate the correct physical block number for files that are
embedded into their file_entry descriptor.  This is more for
correctness, since these files cannot be bmap'ed/mmap'ed anyways.
Enforce this restriction.

Submitted by:	tes@sgi.com
2002-08-02 06:22:20 +00:00
Scott Long
678d5effd3 Check for deleted files in udf_lookup(), not just udf_readdir().
Submitted by:	tes@sgi.com
2002-08-02 06:19:43 +00:00
Alan Cox
1e7ce68ff4 o Lock page queue accesses in nwfs and smbfs.
o Assert that the page queues lock is held in vm_page_deactivate().
2002-08-02 05:23:58 +00:00
Alan Cox
e5b95b1209 o Lock page queue accesses by vm_page_deactivate(). 2002-08-02 04:40:10 +00:00
Alan Cox
64a1b85efa o Lock page queue accesses by vm_page_deactivate(). 2002-08-02 04:14:19 +00:00
Alan Cox
91bb74a88c o Lock page queue accesses by vm_page_deactivate(). 2002-08-02 03:56:31 +00:00
Robert Watson
67d722ed73 Introduce support for Mandatory Access Control and extensible
kernel access control.

Teach devfs how to respond to pathconf() _POSIX_MAC_PRESENT queries,
allowing it to indicate to user processes that individual vnode labels
are available.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-02 03:12:40 +00:00
Robert Watson
bdc2cd1318 Hook up devfs_pathconf() for specfs devfs nodes, not just regular
devfs nodes.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 22:27:57 +00:00
Robert Watson
1dbad969fe Demonstrate that MAC modules can also be linked statically as
well as loaded as modules by hooking up mac_none if
'options MAC_NONE' is defined.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 22:26:38 +00:00
Robert Watson
eddc160e00 Introduce support for Mandatory Access Control and extensible
kernel access control.

Invoke appropriate MAC entry points for a number of VFS-related
operations in the Linux ABI module.  In particular, handle uselib
in a manner similar to open() (more work is probably needed here),
as well as handle statfs(), and linux readdir()-like calls.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 22:23:02 +00:00
Robert Watson
85bb40473a Sample loader.conf lines for various MAC modules. 2002-08-01 22:03:19 +00:00
Jake Burkholder
4fbe520926 Forgot to commit this.
Spotted by:	scottl
2002-08-01 21:39:54 +00:00
Robert Watson
bdb3fa1832 Introduce support for Mandatory Access Control and extensible
kernel access control.

Add MAC support for the UDP protocol.  Invoke appropriate MAC entry
points to label packets that are generated by local UDP sockets,
and to authorize delivery of mbufs to local sockets both in the
multicast/broadcast case and the unicast case.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 21:37:34 +00:00
Robert Watson
8f293a63ce Introduce support for Mandatory Access Control and extensible
kernel access control.

Introduce two ioctls, SIOCGIFMAC, SIOCSIFMAC, which permit user
processes to manage the MAC labels on network interfaces.  Note
that this is part of the user process API/ABI that will be revised
prior to 5.0-RELEASE.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 21:15:53 +00:00
Robert Watson
64c2d84570 Introduce support for Mandatory Access Control and extensible
kernel access control.

Add MAC support for if_ppp.  Label packets as they are removed from
the raw PPP mbuf queue.  Preserve the mbuf MAC label across various
PPP data-munging and reconstitution operations.  Perform access
control checks on mbufs to be transmitted via the interface.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 21:13:47 +00:00
Robert Watson
10722b852b Introduce support for Mandatory Access Control and extensible
kernel access control.

Label packets generated by the gif virtual interface.

Perform access control on packets delivered to gif virtual interfaces.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 21:00:05 +00:00
Robert Watson
18b770b2fb Introduce support for Mandatory Access Control and extensible
kernel access control.

Invoke appropriate MAC framework entry points to authorize readdir()
operations in the native ABI.

Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, NAI Labs
2002-08-01 20:44:52 +00:00
Julian Elischer
67759b33f6 Fix a comment. 2002-08-01 19:10:40 +00:00
Julian Elischer
04774f2357 Slight cleanup of some comments/whitespace.
Make idle process state more consistant.
Add an assert on thread state.
Clean up idleproc/mi_switch() interaction.
Use a local instead of referencing curthread 7 times in a row
(I've been told curthread can be expensive on some architectures)
Remove some commented out code.
Add a little commented out code (completion coming soon)

Reviewed by:	jhb@freebsd.org
2002-08-01 18:45:10 +00:00
Robert Watson
ee0812f320 Since we have the struct file data pointer cached in vp, use that
instead when invoking VOP_POLL().
2002-08-01 18:29:30 +00:00
Alan Cox
46086ddf91 o Acquire the page queues lock before calling vm_page_io_finish().
o Assert that the page queues lock is held in vm_page_io_finish().
2002-08-01 17:57:42 +00:00
Robert Watson
f9d0d52459 Include file cleanup; mac.h and malloc.h at one point had ordering
relationship requirements, and no longer do.

Reminded by:	bde
2002-08-01 17:47:56 +00:00