Commit Graph

549 Commits

Author SHA1 Message Date
David Greenman
e9857eee2b Rewrote memory sizing code to generally deal with holes in extended memory.
This code change should allow certain Compaq machines with a 128K hole
at 16MB to work.
1995-07-19 06:37:12 +00:00
Peter Wemm
1174c7d121 This fixes a compiler warning, and a cosmetic problem with the linux
emul code when compiling with "options KTRACE".
ktrsyscall() was expecting an array of integers, this was passing the
address of a structure containing an array of integers..
The cosmetic problem was that it was calling the "enter syscall"
trace hook twice - this looks like a cut/paste error/typo.
1995-07-16 14:10:55 +00:00
Poul-Henning Kamp
8a7580c3fe Reviewed by: phk
Submitted by:	Andrew McRae <andrew@mega.com.au>

Some initial commits from the pcmcia stuff, to make life easier for the
testers.

We will use the name "pccard" since that is really the buzzword at present.
1995-07-16 10:45:06 +00:00
Poul-Henning Kamp
7a2dada545 Make the bootinfo structure visible from sysctl.
This can be used in libdisk to guess a better bios-geometry.
1995-07-16 10:33:38 +00:00
Joerg Wunsch
446cee6e6d Include ``options POWERFAIL_NMI'' for owners of older (non-apm)
notebooks where a powerfail condition (external power drop; battery
state low) is signalled by an NMI.  Makes it beep instead of panicing.

Reviewed by:	davidg
1995-07-16 10:31:26 +00:00
David Greenman
9e951f36f1 Truncate the fault address to a page boundry when calling vm_fault(). The
last change to fix the fault-twice bug with page tables wasn't quite
complete.
1995-07-16 05:39:22 +00:00
David Greenman
4a67eb7121 Fixed bug that caused page tables to be faulted twice instead of once.
Submitted by:	John Dyson
1995-07-14 09:25:51 +00:00
David Greenman
24a1cce34f NOTE: libkvm, w, ps, 'top', and any other utility which depends on struct
proc or any VM system structure will have to be rebuilt!!!

Much needed overhaul of the VM system. Included in this first round of
changes:

1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages,
   haspage, and sync operations are supported. The haspage interface now
   provides information about clusterability. All pager routines now take
   struct vm_object's instead of "pagers".

2) Improved data structures. In the previous paradigm, there is constant
   confusion caused by pagers being both a data structure ("allocate a
   pager") and a collection of routines. The idea of a pager structure has
   escentially been eliminated. Objects now have types, and this type is
   used to index the appropriate pager. In most cases, items in the pager
   structure were duplicated in the object data structure and thus were
   unnecessary. In the few cases that remained, a un_pager structure union
   was created in the object to contain these items.

3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now
   be removed. For instance, vm_object_enter(), vm_object_lookup(),
   vm_object_remove(), and the associated object hash list were some of the
   things that were removed.

4) simple_lock's removed. Discussion with several people reveals that the
   SMP locking primitives used in the VM system aren't likely the mechanism
   that we'll be adopting. Even if it were, the locking that was in the code
   was very inadequate and would have to be mostly re-done anyway. The
   locking in a uni-processor kernel was a no-op but went a long way toward
   making the code difficult to read and debug.

5) Places that attempted to kludge-up the fact that we don't have kernel
   thread support have been fixed to reflect the reality that we are really
   dealing with processes, not threads. The VM system didn't have complete
   thread support, so the comments and mis-named routines were just wrong.
   We now use tsleep and wakeup directly in the lock routines, for instance.

6) Where appropriate, the pagers have been improved, especially in the
   pager_alloc routines. Most of the pager_allocs have been rewritten and
   are now faster and easier to maintain.

7) The pagedaemon pageout clustering algorithm has been rewritten and
   now tries harder to output an even number of pages before and after
   the requested page. This is sort of the reverse of the ideal pagein
   algorithm and should provide better overall performance.

8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup
   have been removed. Some other unnecessary casts have also been removed.

9) Some almost useless debugging code removed.

10) Terminology of shadow objects vs. backing objects straightened out.
    The fact that the vm_object data structure escentially had this
    backwards really confused things. The use of "shadow" and "backing
    object" throughout the code is now internally consistent and correct
    in the Mach terminology.

11) Several minor bug fixes, including one in the vm daemon that caused
    0 RSS objects to not get purged as intended.

12) A "default pager" has now been created which cleans up the transition
    of objects to the "swap" type. The previous checks throughout the code
    for swp->pg_data != NULL were really ugly. This change also provides
    the rudiments for future backing of "anonymous" memory by something
    other than the swap pager (via the vnode pager, for example), and it
    allows the decision about which of these pagers to use to be made
    dynamically (although will need some additional decision code to do
    this, of course).

13) (dyson) MAP_COPY has been deprecated and the corresponding "copy
    object" code has been removed. MAP_COPY was undocumented and non-
    standard. It was furthermore broken in several ways which caused its
    behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will
    continue to work correctly, but via the slightly different semantics
    of MAP_PRIVATE.

14) (dyson) Sharing maps have been removed. It's marginal usefulness in a
    threads design can be worked around in other ways. Both #12 and #13
    were done to simplify the code and improve readability and maintain-
    ability. (As were most all of these changes)

TODO:

1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing
   this will reduce the vnode pager to a mere fraction of its current size.

2) Rewrite vm_fault and the swap/vnode pagers to use the clustering
   information provided by the new haspage pager interface. This will
   substantially reduce the overhead by eliminating a large number of
   VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be
   improved to provide both a "behind" and "ahead" indication of
   contiguousness.

3) Implement the extended features of pager_haspage in swap_pager_haspage().
   It currently just says 0 pages ahead/behind.

4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps
   via a much more general mechanism that could also be used for disk
   striping of regular filesystems.

5) Do something to improve the architecture of vm_object_collapse(). The
   fact that it makes calls into the swap pager and knows too much about
   how the swap pager operates really bothers me. It also doesn't allow
   for collapsing of non-swap pager objects ("unnamed" objects backed by
   other pagers).
1995-07-13 08:48:48 +00:00
Stefan Eßer
d2a2d5ec41 The PCI config mechanism 1 test failed for the Intel Aries.
Make it less strict ...

Submitted by:	NIIMI Satoshi <sa2c@and.or.jp>
1995-06-30 16:11:42 +00:00
Stefan Eßer
0847c06d2e PCI configuration mechanism now determined by a method, that doesn't
fail on new hardware (Compaq Prolinea and Compaq Prosignea), and that
doesn't erroneously identify old mech. 2 chip sets as using mech. 1.
(See section 3.6.4.1.1 of the PCI bus specs rev. 2.0)
1995-06-28 15:54:57 +00:00
David Greenman
338cd8f6be Killed redundant vnode_pager_umount() call. This is already done at
FS unmount time.
1995-06-28 04:46:11 +00:00
David Greenman
53b5ed936e Make path to kernel absolute if it is passed in relative. This fixes
a related bug in some of the new 'foo'boot bootstrap code that has been
added over the past months. This change makes it no longer necessary
for the bootstrap to fix up the path (i.e. it can be removed).
1995-06-28 04:42:25 +00:00
Bruce Evans
943c18018b Fix standards conformance bugs in <signal.h>:
include/signal.h:
There was massive namespace pollution from including <sys/types.h>.
POSIX functions were declared even when _ANSI_SOURCE is defined.

sys.sys/signal.h:
NSIG was declared even if _ANSI_SOURCE or _POSIX_SOURCE is defined.
sig_atomic_t wasn't declared if _POSIX_SOURCE is defined.
Declare a typedef for signal handling functions and use it to
unobfuscate declarations and to avoid half-baked function types
that cause unwanted compiler warnings at certain warning levels.
Fix confusing comment about SA_RESTART.

sys/i386/include/signal.h:
This has to be included to get the declaration of sig_atomic_t even
when _ANSI_SOURCE is defined, so be more careful about polluting
the ANSI namespace.

Uniformize idempotency ifdefs.
1995-06-28 02:14:13 +00:00
Bruce Evans
4ebf8117df Partially fix `sysctl machdep.console_device'. The fix will be complete
when syscons stops mapping the console to minor MAXCONS.  There is
usually no corresponding device in /dev, and the correct device has
minor 0.

cons.c:
Initialize cn_tty properly, so that CPU_CONSDEV can work.
Comment about too many variants of the console tty pointer.

machdep.c:
Return device NODEV and not error EFAULT when there is no console device.
1995-06-26 07:39:52 +00:00
Bruce Evans
6f5014b462 Convert to ANSI C: change #endif THING to #endif /* THING */.
Fix one such THING in code to match comment.
Sort IO_GSC* into numeric order and update comments about the gaps.
Sort common SCSI addresses into alphabetical order.
Remove bogus comments about com ports having i/o size 4.
Uniformize whitespace.
Uniformize case in hex digits.

This file is very incomplete.  In particular, it doesn't mention any
network cards.  This doesn't matter much for the base addresses, but
it means that the comments about which addresses are free are mostly
bogus.  The i/o sizes are unreliable because of split address ranges
for many devices (VGA, wd).  The i/o sizes are incomplete.  In
particular, there are no sizes for SCSI controllers.  The bt driver
still returns a truth value instead of a size.
1995-06-14 07:38:31 +00:00
Rodney W. Grimes
d3628763db Merge RELENG_2_0_5 into HEAD 1995-06-11 19:33:05 +00:00
Rodney W. Grimes
9b2e535452 Remove trailing whitespace. 1995-05-30 08:16:23 +00:00
David Greenman
ac8d676972 Fix setdumpdev():
- the major number wasn't checked, so accesses beyond the end of bdevsw[]
  were possible.  Bogus major numbers are easy to get because `sysctl -w'
  doesn't handle dev_t's reasonably - it doesn't convert names to dev_t's
  and it converts the number 1025 to the dev_t 0x35323031.
- Driver d_psize() functions return -1 to indicate error ENXIO or ENODEV
  (the interface is too braindamaged to say which).  -1 was interpreted
  as a size and resulted in the bogus error ENOSPC.
- it was possible to set the dumpdev for devices without a d_psize()
  function.  This is equivalent to setting the dumpdev to NODEV except
  it confuses sysctl.
- change a 512 to DEV_BSIZE.  There is an official macro dtoc() for
  converting "pages" to disk blocks but it is never used in /usr/src/sys.
  There is much confusion between PAGE_SIZE sized pages and NBPG sized
  pages.  Maxmem consists of both.

Not fixed:
- there is nothing to invalidate the dumpdev if the media goes away.
  This reduces the benefits of the early calculation of dumplo.  Bounds
  checking in the dump routines is relied on to reduce the risk of
  damage and little would be lost by relying on the dump routines to
  calculate dumplo.
- no attempt is made to stay away from the start of the device to
  avoid clobbering labels.

Fix wrong && anachronistic comment about the type of bootdev.

Reviewed by:	davidg
Submitted by:	Bruce Evans
1995-05-29 04:08:13 +00:00
David Greenman
b64b660cd3 Made "NMBCLUSTERS" calculation dynamic and fixed bogus use of "NMBCLUSTERS"
in machdep.c (it should use the global nmbclusters). Moved the calculation
of nmbclusters into conf/param.c (same place where nmbclusters has always
been assigned), and made the calculation include an extra amount based
on "maxusers". NMBCLUSTERS can still be overrided in the kernel config
file as always, but this change will make that generally unnecessary. This
fixes the "bug" reports from people who have misconfigured kernels seeing
the network hang when the mbuf cluster pool runs out.

Reviewed by:	John Dyson
1995-05-25 07:41:28 +00:00
David Greenman
5eb46edfb0 Added "BROKEN_KEYBOARD_RESET" option to disable using the keyboard reset
in cpu_reset(). Some MBs don't deal with this properly.

Submitted by:	Rod Grimes
1995-05-18 09:17:07 +00:00
David Greenman
30fd0561cd Added apersand constraint to make sure that the source and destination
registers aren't combined.

Reviewed by:	Bruce Evans and David Greenman
Submitted by:	John Dyson
1995-05-14 22:25:11 +00:00
David Greenman
a401ebbe32 Changed swap partition handling/allocation so that it doesn't
require specific partitions be mentioned in the kernel config
file ("swap on foo" is now obsolete).

From Poul-Henning:

The visible effect is this:

As default, unless
        options "NSWAPDEV=23"
is in your config, you will have four swap-devices.
You can swapon(2) any block device you feel like, it doesn't have
to be in the kernel config.

There is a performance/resource win available by getting the NSWAPDEV right
(but only if you have just one swap-device ??), but using that as default
would be too restrictive.

The invisible effect is that:

Swap-handling disappears from the $arch part of the kernel.
It gets a lot simpler (-145 lines) and cleaner.

Reviewed by:	John Dyson, David Greenman
Submitted by:	Poul-Henning Kamp, with minor changes by me.
1995-05-14 03:00:10 +00:00
Jordan K. Hubbard
c25cfd0b0d "1 easy fix in 10 excrutiating steps"
A phone call from Manfred quickly pointed up the fact that I got the conflict
check backwards.  NOW we implement the conflict checking correctly!  Wheesh!
1995-05-13 00:09:38 +00:00
Garrett Wollman
5bb4f73817 The death of `options NODUMP'. Now the dump area can be dynamically
configured (and unconfigured) on the fly.  A sysctl(3) MIB variable is
provided to inspect and modify the dump device setting.
1995-05-12 19:17:31 +00:00
Rodney W. Grimes
b2b795f07c Fix -Wformat warnings from LINT kernel. 1995-05-11 19:26:53 +00:00
Bruce Evans
d81fba0cda Add variable `idelayed' and macros setdelayed() and schedsofttty()
to access it.  setdelayed() actually ORs the bits in `idelayed' into
`ipending' and clears `idelayed'.

Call setdelayed() every (normal) clock tick to convert delayed
interrupts into pending ones.

Drivers can set bits in `idelayed' at any time to schedule an interrupt
at the next clock tick.  This is more efficient than calling timeout().
Currently only software interrupts can be scheduled.
1995-05-11 07:44:40 +00:00
Bruce Evans
0ee893eb32 Add loadandclear(). It atomically loads a value from memory, clears the
value in memory and returns the original value.
1995-05-11 07:24:35 +00:00
Jordan K. Hubbard
da8d0268df Pass me the pointed chapeau - this typo somehow got through my testing. 1995-05-11 05:20:43 +00:00
Bill Paul
2c739e6439 If you config a kernel with 'config kernel swap generic' and try to
boot diskless with it, you get a panic because setconf() is only
called for mountroot == ffs_mountroot. It really needs to be called
no matter what manner of rootfs we have. I can't really say if
swapgeneric will work with a CD-ROM though. (I get the feeling I'm
the only one who uses swapgeneric these days anyway.)
1995-05-11 02:50:11 +00:00
Jordan K. Hubbard
9cc347481f Remove all vestiges of the ALLOW_CONFLICT_FOO evil and replace it with
something slightly less evil - a per device conflict flag.
1995-05-11 02:15:55 +00:00
Garrett Wollman
c9cd353a18 Delete two debugging printfs that mistakenly crept in. 1995-05-11 00:16:44 +00:00
Garrett Wollman
748e0b0acc Make networking domains drop-ins, through the magic of GNU ld. (Some day,
there may even be LKMs.)  Also, change the internal name of `unixdomain'
to `localdomain' since AF_LOCAL is now the preferred name of this family.
Declare netisr correctly and in the right place.
1995-05-11 00:13:26 +00:00
David Greenman
85eaa94715 Correct the definition for the (unused) cpu_setstack(). 1995-05-04 07:50:06 +00:00
David Greenman
6486b7f194 Added a memcpy() routine. 1995-05-02 05:20:26 +00:00
Poul-Henning Kamp
807afe4fa5 A missing 'and', probably my fault.
Submitted by:	Ed Hudson <elh@p5.spnet.com>
1995-05-02 04:40:06 +00:00
John Dyson
641a27e68b Fixed a problem that can cause left-over pv_entries and as
as side-effect, removed some legacy code that was necessary
when we called vm_fault inside of vm_fault_quick instead of using
the kernel/user space byte move routines.
1995-05-01 23:32:30 +00:00
Rodney W. Grimes
2320728f46 Add outb to keyboard controller to do a cpu_reset, this fixes 2 known
cases of motherboards that failed to reboot.
1995-04-26 07:38:35 +00:00
Poul-Henning Kamp
58e81aafbb Add support for MFS root filesystem. 1995-04-25 03:41:12 +00:00
Poul-Henning Kamp
d3f4d460f8 Added "bio" to matcd. 1995-04-24 05:33:59 +00:00
Julian Elischer
6944abf1d7 hmm spotted a difference resulting from a merge I didn't examine close enough 1995-04-23 09:17:24 +00:00
Julian Elischer
73bac981dd include hooks for EISA configuration (possibly wrong :) 1995-04-23 09:13:08 +00:00
Poul-Henning Kamp
94624c5572 Forgot this commit the other day. The receiving end of the "boot -C" option. 1995-04-23 04:14:41 +00:00
Bill Paul
31e0d8b925 Tiny printf formatting change: if we have no cpu_vendor or cpu_id info,
don't generate a newline. (Yeah, I'm picking nits, but that empty line
I get on my 386 just looks dumb, okay? :)
1995-04-22 03:58:46 +00:00
Poul-Henning Kamp
804e33a872 Add wd2 and wd3 as swap-devices too. 1995-04-20 06:05:17 +00:00
Rodney W. Grimes
781434c1fc Reapply my fix for this:
Output the CPU features line during the probe on a seperate line, for
folks with lots of features the output use to wrap and look ugle.
1995-04-18 23:55:26 +00:00
Poul-Henning Kamp
8425684b90 Print the BIOS geometries in a human-readable format. 1995-04-17 16:49:50 +00:00
David Greenman
857f4d44be Remove gratuitous waste of 2K of memory for BIOS variables. We never load
the kernel at 0-640k; we haven't had the ability to do that since before
2.0R. Furthermore, I fail to see how putting an instruction at 0 and then
doing a .org 0x500 is going to prevent the stuff from getting clobbered
in the first place; a.out is just too stupid to know about sparse address
spaces.
1995-04-16 10:12:16 +00:00
Bruce Evans
7897b3ce69 Don't waste time sending an EOI to ICU1 if option AUTO_EOI_1 is defined.
Previously, this worked right if both AUTO_EOI_1 and AUTO_EOI_2 are
defined, but not if AUTO_EOI_1 is defined and AUTO_EOI_2 is not defined.
The latter case should be the default.  DUMMY_NOPS should be the default
too.  Currently there are only two NOPs slowing down rtcin() (although
there are no delays in writertc()) and several FASTER_NOPs slowing down
interrupt handling in vector.s.

Fix stack offsets for the (previously) unused untested
FAST_INTR_HANDLER_USES_ES case.
1995-04-15 21:32:18 +00:00
Peter Dufault
8db34c841e Add scsi target. Add "after config" call to autoconf so that scsi
targets will be configured after all scsi busses have been configured.
1995-04-14 15:13:26 +00:00
Bill Paul
09a901eb60 Hopefully I won't get flamed for this: insert a few more #if defined(I486_CPU)
and #if defined (I586_CPU) thingies into identifycpu() so that we only
compile in what's actually needed for a given CPU. So far as I can tell,
none of my 386 machines generate a cpu_vendor code, so I made the extra vendor
and feature line conditional on I486_CPU and I586_CPU. (Otherwise we
print out a blank line which looks silly.)
1995-04-14 02:06:00 +00:00