dyson [Sat, 9 Aug 1997 10:13:32 +0000 (10:13 +0000)]
Modify the scheduling policy to take into account disk I/O waits
as chargeable CPU usage. This should mitigate the problem of processes
doing disk I/O hogging the CPU. Various users have reported the
problem, and test code shows that the problem should now be gone.
dyson [Sat, 9 Aug 1997 01:43:15 +0000 (01:43 +0000)]
Add our doscmd to the tree. This is a result of work from BSDI, and
a group of dos emulator developers.
Submitted by: Jonathan Lemon <jlemon@americantv.com>
Obtained from: BSDI
dyson [Sat, 9 Aug 1997 01:38:03 +0000 (01:38 +0000)]
Add the code that represents most of the interface between the VM86
pseudo-machine and the rest of the FreeBSD kernel.
Submitted by: Jonathan Lemon <jlemon@americantv.com>
julian [Sat, 9 Aug 1997 01:25:54 +0000 (01:25 +0000)]
Use up 4 precious bytes to give the kernel a hook to
support hardware watchdogs. The actual functions would be supplied in an LKM
or a linked file, but they need to hang off something.
dyson [Sat, 9 Aug 1997 00:04:06 +0000 (00:04 +0000)]
VM86 kernel support.
Work done by BSDI, Jonathan Lemon <jlemon@americantv.com>,
Mike Smith <msmith@gsoft.com.au>, Sean Eric Fagan <sef@kithrup.com>,
and probably alot of others.
Submitted by: Jnathan Lemon <jlemon@americantv.com>
phk [Fri, 8 Aug 1997 21:11:40 +0000 (21:11 +0000)]
Add ID's for 5 VLSI chips. They're not very friendly, so this info was
found by taking my HP800CT apart, perusing HPs (Very good!) service
manual and inference from a bad gif file I found in Finland.
Sigh... But it's a nice machine :-)
alex [Fri, 8 Aug 1997 14:36:29 +0000 (14:36 +0000)]
Support interface names up to 15 characters in length. In order to
accommodate the expanded name, the ICMP types bitmap has been
reduced from 256 bits to 32.
A recompile of kernel and user level ipfw is required.
To be merged into 2.2 after a brief period in -current.
steve [Thu, 7 Aug 1997 21:37:39 +0000 (21:37 +0000)]
Remove remaining Lite1 stuff from the man page and uphold the
precedence in changing sccsid to rcsid as set by Phillippe
Charnier in his previous 100 or so commits.
julian [Thu, 7 Aug 1997 02:41:15 +0000 (02:41 +0000)]
don't lose track of how many packets we've sent
if we get 'host unreachable'. (or any other errors than ENOBUFS)
makes -c work again even if you can't get there..
This really needs a rework..
Remove two features that have been reported to cause problems with
certain variants of the NCR chip from FE_CACHE_SET: FE_CLSE (enable
cache-line size register) and FE_ERMP (enable read-multiple). They
will be re-enabled, if a fix for the underlying problem (a restriction
in the memory to memory move logic of some chips) has been implemented.
dyson [Wed, 6 Aug 1997 04:58:05 +0000 (04:58 +0000)]
Add exposure of some vm_zone allocation stats by sysctl. Also, change
the initialization parameters of some zones in VM map. This contains
only optimizations and not bugfixes.
dyson [Tue, 5 Aug 1997 23:03:24 +0000 (23:03 +0000)]
Fixed the commit botch that was causing crashes soon after system
startup. Due to the error, the initialization of the zone for
pv_entries was missing. The system should be usable again.
dyson [Tue, 5 Aug 1997 22:07:27 +0000 (22:07 +0000)]
Fix some bugs, document vm_zone better. Add copyright to vm_zone.h. Use
the new zone code in pmap.c so that we can get rid of the ugly ad-hoc
allocations in pmap.c.
asami [Tue, 5 Aug 1997 03:49:49 +0000 (03:49 +0000)]
Merge buildworld/installworld patch to Makefile from RELENG_2_2. Patch
a couple *.mk files to enable -current world building on really old
machines (e.g., 2.1.5).
Reviewed by: too many many people to list here, special thanks to bde
dyson [Tue, 5 Aug 1997 00:02:08 +0000 (00:02 +0000)]
Get rid of the ad-hoc memory allocator for vm_map_entries, in lieu of
a simple, clean zone type allocator. This new allocator will also be
used for machine dependent pmap PV entries.
fsmp [Mon, 4 Aug 1997 17:31:43 +0000 (17:31 +0000)]
Eliminate frequent silo overflows by restoring the TEST_LOPRIO code.
This code was eliminated when the PEND_INTS algorithm was added. But it was
discovered that PEND_INTS only worsen latency for FAST_INTR() routines,
which can't be marked pending.
Noticed & debugged by: dave adkins <adkin003@gold.tc.umn.edu>
dyson [Mon, 4 Aug 1997 07:43:28 +0000 (07:43 +0000)]
Fix a problem with the vfs vnode caching that it doesn't grow quickly
enough and can cause some strange performance problems. Specifically, at
or near startup time is when the problem is worst. To reproduce
the problem, run "lat_syscall stat" from the alpha lmbench code right
after bootup. A positive side effect of this mod is that the name
cache can be set to grow again by sysctl. A noticable positive
performance impact is realized due to a larger namecache being available
as needed (or tuned.)