41e6fbc398
Delint somewhat.
2002-05-07 04:52:49 +00:00
a19e4685f7
Make CLKF_INTR() slightly less broken. The previous definition always
...
tested true if the CPU was running in the kernel at the time of the clock
interrupt, therefore all such time (kernel and idle) was counted as interrupt
time. The new definition simply punts on trying to count interrupt time,
which allows us to see kernel and idle time.
2002-05-07 04:01:59 +00:00
3a3df0d1d2
Fix reading values mapped in P1 via /dev/kmem (problem uncovered by
...
uch's new pmap).
2002-05-07 03:28:25 +00:00
e5555e5c26
Change struct ipqe to use TAILQ's instead of LIST's (primarily for TCP's
...
benefit currently). Rework tcp_reass code to optimize the 4 most likely causes
of out-of-order packets: first OoO pkt, next OoO pkt in seq, OoO pkt is part
of new chuck of OoO packets, and the OoO pkt fills the first hole. Add evcnts
to instrument tcp_reass (enabled by the options TCP_REASS_COUNTERS). This is
part 1/2 of tcp_reass changes.
2002-05-07 02:59:38 +00:00
2baa6f8e86
Drop SUN4_PGSHIFT and SUN4CM_PGSHIFT. They are already known to .S
...
files from sys/param.h. The problem is that lint stubs are generated
with -C (keep comments) and gcc complains about macro redefinition,
since "assym.h" doesn't have the comments that "sys/param.h" has.
2002-05-07 02:38:44 +00:00
c4e1385f55
Fetch the right page from a file even if it is mapped from middle of it.
...
This makes `tail -<N> <FILE> | cat > file' correctly, where <FILE> is
a regular file larger than 10Mbytes (makes tail to map part of file)
and <N> is big enough to produce output larger than 8kbytes (makes pipe
to use page loan facility). Problem reported by FUKAUMI Naoki on japanese
local mailing list.
2002-05-07 02:29:52 +00:00
91a4b7da7f
Use "NOMAN=" instead of "MKMAN=no"; per lukem.
2002-05-07 02:13:51 +00:00
79cc6dcb79
Remove a "NOMAN="; Makefile.booters now has MKMAN=no.
2002-05-07 02:04:26 +00:00
67c33ffcda
Remove two shadowed variables, one of which was set to what it was
...
shadowing(!) and the other a temporary variable (as was it's shadow).
Code generated with "gcc -O2" is the same before and after.
2002-05-07 02:02:41 +00:00
c38dba996a
lint target depends on ${LINTSTUBS}, not on (nonexistent) ${KERNLINTSTUBS}.
2002-05-07 02:00:17 +00:00
8107633e20
bus space methods should be "static __inline__", not "extern __inline__".
...
Otherwise an unoptimized kernel cannot be linked. Reviewed by pk.
2002-05-07 00:37:47 +00:00
2a4ecb6335
Add a flag for ofpci to let psycho chose between pci and ofpci.
2002-05-06 22:41:10 +00:00
d87653c0d2
Switch to ofpci.
2002-05-06 22:34:40 +00:00
91c28a3626
Add code to attach ofpci.
2002-05-06 22:29:22 +00:00
c419c08421
Regenerate for Netgear FA411.
2002-05-06 22:28:18 +00:00
b72d0938bb
Add in Netgear FA411.
2002-05-06 22:26:48 +00:00
ba4db702d5
Recognize `ofpci' as a PCI bus.
2002-05-06 22:20:57 +00:00
19fcb9e555
Fix printing during attach and get rid of pci_cd that's provided by the linker.
2002-05-06 22:20:07 +00:00
24954ac8ae
Add ofpci_make_tag().
2002-05-06 22:18:51 +00:00
2ec18b07b9
Provide pci_cd if needed.
2002-05-06 22:12:43 +00:00
822aeae880
Note ofpci and ofppb.
2002-05-06 22:11:50 +00:00
05d0665f1c
rename memory_disc_size to md_root_size, so that kernels without
...
MEMORY_DISK_ROOT_SIZE option link
also make local md_root_size size_t
XXX is the load_memory_disc_from_floppy() stuff actually still being used?
2002-05-06 21:18:25 +00:00
1805506872
Add ofpci and ofppb.
2002-05-06 19:48:05 +00:00
e60dcd0a0b
Add ofppb to match ofpci.
2002-05-06 19:40:20 +00:00
509fa464af
Add the ofpci driver.
2002-05-06 19:39:51 +00:00
5d47d866f7
Only put interrupts into the list for that interrupt level if the handler
...
is not shared. Otherwise put the shared interrupt dispatcher in the list.
2002-05-06 19:19:48 +00:00
f23186dae7
Don't return an error for "Not Ready To Ready Transition (Medium May Have
...
Changed)" and "Power On, Reset, or Bus Device Reset Occurred" senses, if
we're trying to open the device.
2002-05-06 13:43:57 +00:00
8178b68188
Added ukphy to use USB ethernet adopters.
2002-05-06 13:42:29 +00:00
b26614a4af
s,<dev/sun/sun_boot.h>,<sys/bootblock.h>,
2002-05-06 13:35:17 +00:00
3777b83a3f
<sys/bootblock.h> is a better name than <sys/bootinfo.h>
2002-05-06 13:32:19 +00:00
e8bbb685d9
Replace <dev/sun/sun_boot.h> with <sys/bootinfo.h>, which is intended
...
to be more generic than ``bbinfo definitions for Sun-based systems''.
Other platforms can store bbinfo-style information here, and possibly
other platform-specific boot information that needs to be accessible
by foriegn platforms in tools such as /usr/sbin/installboot.
2002-05-06 05:27:53 +00:00
59a79c7928
Remove wrong assertion in previous commit.
2002-05-06 03:20:54 +00:00
73843c4a73
Comment change only. Update the version of GENERIC this config is
...
synced with.
2002-05-06 02:04:14 +00:00
e27890b7d9
fix previous more correctly; MAXBLOCKNUM was replaced by SUN68K_BBINFO_MAXBLOCKS
2002-05-06 01:25:11 +00:00
5d6cd5914c
Put back a missing brace.
2002-05-06 00:48:58 +00:00
8876669f4c
Since npages may includes number of pages behind us, we can't use it to
...
update current offset. Instead, use the last page in the run of pages
to calculate new offset.
2002-05-06 00:42:22 +00:00
e6513c283e
Stylistic change; introduce new local variable and use it instead of
...
sprinkling different expression to test if we're pagedaemon.
2002-05-06 00:18:15 +00:00
cfa571e2e2
The per nfsnode n_commitlock is a sleep lock, but we can't sleep if
...
PGO_LOCKED getpages request. So, just make the lock fail and tell
the caller that there is no pages available if we can't acquire it.
The caller will call us again soon without PGO_LOCKED. Reviewed by chuq.
2002-05-06 00:07:51 +00:00
5a1f9fc69c
400-series machines only have DIO framebuffers. From discussion
...
with Michael Lorenz and Bernd Sieker.
2002-05-05 22:55:49 +00:00
f1a07a4001
no manpages here
2002-05-05 22:09:05 +00:00
093d8bdbdd
Put the contents inside #ifdef _KERNEL, similarily to mac68k one,
...
so that the innards are not exposed to userland.
2002-05-05 22:07:59 +00:00
628a2b8665
Get ctrlsp.S via .PATH directive, rather than adding full path
...
to SRCS. This should fix build problem with ro source tree.
2002-05-05 20:38:48 +00:00
9eed164863
remove redundant (and wrong) main() prototype
2002-05-05 20:11:37 +00:00
92896ac2a1
Add a null asm statement in adb_read_date_time()
...
to avoid wrong reordering by gcc-2.95.3 with -fgcse.
XXX Should we disable -fgcse everywhere?
2002-05-05 18:36:03 +00:00
87ebd3218c
Backout previous. It is not timing problem, but gcc optimization bug.
2002-05-05 18:29:06 +00:00
8b225b8ff7
also don't make the machine symlinks for 'clean' target
...
move the CLEANFILES update out of the .if, so that the files get removed
for 'clean' or 'cleandir'
2002-05-05 17:11:41 +00:00
e926e6ec99
use the correct size when zeroing an array.
2002-05-05 17:01:41 +00:00
dcc6963777
for softdep vnodes, always write together the pages for any block that
...
might have a dependency , since the accounting doesn't work otherwise.
fixes PRs 15364 16336 16448.
2002-05-05 17:00:06 +00:00
f2f12a240b
Update to md(4) changes: memory_disk_size is now md_root_size, and
...
type is size_t
2002-05-05 16:26:30 +00:00
988df8394c
look in the right flags field for PQ_INACTIVE.
...
make uvmpd_scan_inactive() return void since its return value is ignored.
2002-05-05 16:26:17 +00:00
uwe