tsutsui
99adff7a25
Remove db_machine_init(), which was removed from mips/db_interface.c.
2001-02-05 13:34:42 +00:00
ragge
1af765a6dd
Oops, forgot this file.
2001-02-05 13:20:10 +00:00
tsutsui
de70ac347a
db_machine_init() is no longer needed.
...
From Shigeji Ishikawa <psybear@ogre.rim.or.jp> in port-cobalt/12123.
2001-02-05 13:18:28 +00:00
tsutsui
869fdfbf12
Fix typo on pci_intr_map() changes.
...
Reported by Shigeji Ishikawa <psybear@ogre.rim.or.jp> in port-cobalt/12121.
2001-02-05 13:14:21 +00:00
mrg
22a68e0e29
oops, kill some debugging messages.
2001-02-05 13:13:47 +00:00
tsutsui
8bc87999fc
Remove obsolete cpu_exec_ecoff_hook().
2001-02-05 13:10:07 +00:00
mrg
2d7ba2f0e7
turn off UVMHIST. turn on LKM. makeoptions DEBUG=-g
2001-02-05 12:49:11 +00:00
mrg
05752997ef
use sunos32_machdep.c on 64 bit kernels.
2001-02-05 12:47:47 +00:00
mrg
e1ac1f31b9
make sure 32 bit addresses are truncated properly.
2001-02-05 12:45:38 +00:00
chs
f889fd6ff9
remove signal trampoline stuff since it'll never be used in the kernel.
...
clean up the ELF support.
2001-02-05 12:37:33 +00:00
chs
5f244b1055
conditionalize the ddb hook on DDB && DEBUG_FPE rather than DDB && DEBUG,
...
since that's the condition under which we include the header file that
allows it to compile.
2001-02-05 12:35:13 +00:00
chs
755889bc0b
clear pcb_onfault before calling uvm_fault() and restore it afterward.
...
if the fault handler generates a pagefault, we'd like to panic rather
than invoking the onfault handler.
2001-02-05 11:12:05 +00:00
chs
09cb38f22b
expose the definitions of MIN() and MAX() in sys/param.h to the kernel
...
and use those in favor of a dozen copies scattered around the source tree.
2001-02-05 10:42:40 +00:00
eeh
db133b2424
Add signal trampoline for 64-bit svr4 emulations.
2001-02-05 06:56:45 +00:00
itojun
d0bc7a3c3d
s/unix/unit/ for ld* on mlx* :-)
2001-02-05 04:16:52 +00:00
jmc
ece49e13a9
Back out last change. m68k.h is now in cpu.h (which got updated in between
...
my noticing this problem here on a sun3).
2001-02-04 23:53:31 +00:00
ragge
2c1941b3eb
Giant update from Michael Kukat (michael@unixiron.org).
...
Fixes support for:
MicroVAX 3100m90
MicroVAX 3100m95
VAX 4000/100 (tested)
VAX 4000/105A (tested)
VAX 4000/108
VAX 4000/400 (tested)
VAX 4000/500 (tested)
VAX 4000/600
2001-02-04 20:36:26 +00:00
briggs
c8ef73b71f
Support for the Motorola Sandpoint reference platform using the
...
MPC8240 processor module. See the README for more information.
2001-02-04 18:32:10 +00:00
briggs
863805b4d8
Correct previous. Call openpic_init() correctly.
2001-02-04 17:46:33 +00:00
briggs
522ec180e9
Build with -Wall -Wetc.
2001-02-04 17:38:10 +00:00
briggs
e560b86e0b
Adapt to new openpic_init() parameters.
2001-02-04 17:36:08 +00:00
perry
65e351a124
keep joystick lines together
2001-02-04 17:36:03 +00:00
briggs
ace7a75bd5
Remove dependancy on ICU_LEN m.d. define. Thanks, Tsubai.
2001-02-04 17:35:28 +00:00
briggs
cbb39b8875
defopt openpic
2001-02-04 17:34:11 +00:00
briggs
0ce1220246
For now, all macppc will get OpenPIC
2001-02-04 17:33:42 +00:00
ad
8d75ab98d3
Add a driver for the Mylex DAC960 family (including DEC SWXCR).
2001-02-04 17:05:11 +00:00
jmc
7ce76feecc
In the LKM section add some things the LKM's need (since they aren't
...
defining _SUN3_ or _SUN3X_ generally to pickup the full vmparam.h
Added MAXDSIZ (set to sun3's limit) and VM_NFREELIST
2001-02-04 10:16:03 +00:00
jmc
36f1b05b1f
Include m68k/m68k.h to pickup definition for fputype.
2001-02-04 10:13:08 +00:00
jmc
e1fb29e236
Make targets proper so this builds both as a make with no args, make all
...
or make dependall (without the latter snapshot builds break).
Also, change the build layout for the 2 standalone objs.
XXX - Add them to LOBJS. (We're not building lint libraries, but having them
in LOBJS and MKLINT=no results in objects building anyways. This needs to be
cleaner but nothing in the standard mk files jumps out as "other object files")
2001-02-04 08:59:57 +00:00
jmc
17a976fa32
Make targets proper so this builds both as a make with no args, make all
...
or make dependall (without the latter snapshot builds break).
Also, add udp.c into the SRCS list. rpc.c calls routines from it.
2001-02-04 08:57:52 +00:00
jmc
243eb5180a
Make targets proper so this builds both as a make with no args, make all
...
or make dependall (without the latter snapshot builds break).
2001-02-04 08:57:21 +00:00
takemura
596e28b8c2
Marge hpcmips/conf/platid.def and hpcsh/conf/platid.def
...
into hpc/conf/platid.def.
2001-02-04 05:19:14 +00:00
tsutsui
20c4de8a40
Move cachectl1() declaration from sun3/machdep.h to include/cpu.h;
...
it is refered by sys/compat/linux/arch/m68k/linux_machdep.c.
XXX: all declarations in machdep.h should be moved into cpu.h?
2001-02-04 01:31:54 +00:00
tsutsui
a031d05fa8
Add -D__m68k__ to XDEF.
...
It is required by sys/compat/linux/common/linux_exec.h.
XXX: -undef is still required?
2001-02-04 01:26:44 +00:00
marcus
0d6830073e
Added subunit scan.
2001-02-03 23:25:51 +00:00
marcus
3f60497b85
Increased HZ to 100.
2001-02-03 23:24:30 +00:00
is
e36ad02480
gspa can directly create the C language array.
2001-02-03 20:23:59 +00:00
bjh21
74591d814f
Add a definition of MID_MACHINE -- kernel cores still need it.
2001-02-03 17:18:30 +00:00
tsutsui
4f047d1990
Bring FPU probe code from atari/fpu.c and remove unneeded includes.
2001-02-03 14:30:42 +00:00
tsutsui
b26a441f6a
Remove reg_dump() declaration, which is in <m68k/m68k.h>
2001-02-03 13:02:20 +00:00
tsutsui
8ae17a2c61
Use macro in <m68k/m68k.h> for cputype and mmutype.
2001-02-03 12:59:48 +00:00
tsutsui
497bb4f2ea
Remove declaration of copypage() and zeropage(); they are in <m68k/m68k.h>
2001-02-03 12:57:05 +00:00
tsutsui
5f9454752b
m68k/m68k.h is now pulled via machine/cpu.h.
2001-02-03 12:55:35 +00:00
tsutsui
fdddc67af0
Remove declaration of getdfc(), getsfc() and fputype.
...
They are in <m68k/m68k.h>
2001-02-03 12:53:23 +00:00
tsutsui
31a0a141cb
Include <m68k/m68k.h> to use m68k common definitions.
2001-02-03 12:50:53 +00:00
tsutsui
a3cd82caf9
Remove declaration of cold; it's in <sys/kernel.h>
2001-02-03 12:46:04 +00:00
tsutsui
53e0e33ba4
Remove some of broken cf_unit usage.
2001-02-03 12:40:58 +00:00
is
92fd3fa589
Initial, still incomplete wscons support by Jukka Andberg, PR 11068
2001-02-02 21:52:11 +00:00
thorpej
c8cdb12913
Add some comments about P/USEG
2001-02-02 18:25:04 +00:00
fvdl
fbd66cd154
Add COMPAT_15
2001-02-02 10:53:48 +00:00
sato
dcdef16dcd
Copyright year.
2001-02-02 08:31:04 +00:00
mrg
c084c19f60
add "wd" to chrtoblktbl
2001-02-02 07:39:28 +00:00
mrg
2a9fceab8d
enable the kbd/ms. enable some compat options for compat_sunos
2001-02-02 07:33:42 +00:00
mrg
9954eaa78b
include compat/sunos32/sunos32.files
2001-02-02 07:32:45 +00:00
mrg
04cb3c57d6
bump NMBLUSTERS now this machine has NFS server tasks.
2001-02-02 07:32:20 +00:00
marcus
5592504080
Fixed type mismatch in debug output...
2001-02-02 06:27:44 +00:00
briggs
9150a401ba
No longer needed
2001-02-02 06:12:34 +00:00
briggs
044caccfff
Split OpenPIC support into powerpc so ports only need minimal custom
...
configuration. openpic_init() now takes a single argument that is the
base of the OpenPIC register space.
2001-02-02 06:11:52 +00:00
thorpej
4fab177f7e
Don't make the reporting of the keyboard type a parenthetical.
2001-02-02 03:09:16 +00:00
thorpej
2ae9f715f6
Report the VGA/NTSC RGB/composite mode of the display.
2001-02-02 03:07:29 +00:00
thorpej
ef2315d256
Rewrite how mod/ref emulation is done. In particular, make
...
modified attribute checking actually work. Previously, a
page could be reported as modified even if it wasn't! This
is very bad, and e.g. caused the VM system to attempt to
"clean" pages of read-only files.
Only lip-services is paid to referenced attribute emulation.
This could be improved.
2001-02-02 02:28:18 +00:00
leo
50c553e643
Announce page-size to the VM-system. Thanks to Izumi Tsutsui for pointing
...
this out.
2001-02-01 21:22:55 +00:00
thorpej
6195ca3474
Add some info to the dmamap_sync assertion messages.
2001-02-01 19:56:44 +00:00
marcus
fc7cc52392
Fixed the last raw accesses to use bus_space. Also added DMA mem size field.
2001-02-01 19:35:04 +00:00
leo
3d1167e274
Fix fallout from lastest tty changes.
2001-02-01 08:59:45 +00:00
leo
9b569c3abe
No longer update the PAGE_* variables. They have turned into compile time
...
constants.
2001-02-01 08:58:03 +00:00
tsutsui
94ebf6d0d7
Update for the install floppy:
...
- Add newer devices (wscons etc.)
- Add -mmemcpy to COPTS
- Reduce maxusers
- Remove KERNFS
- Disable some COMPAT_* options
- Add MEMORY_DISK_SERVER=0
- Enable options NFS_V2_ONLY and VNODE_OP_NOINLINE
2001-02-01 06:23:54 +00:00
thorpej
f436156dd7
Use the BOLD8x16 font for the console, and remove the RCONS_16BPP
...
option (no longer needed).
2001-02-01 01:27:59 +00:00
thorpej
f16d219c1b
Config glue for rasops/wsfont.
2001-02-01 01:27:04 +00:00
thorpej
fbfc0872d8
New PowerVR display driver, based on the TURBOchannel "cfb"
...
driver by Tohru Nisimura (a very good framework for a FB driver,
BTW!), using bits of Marcus's original driver for the hardware
access.
The main advantage of this version is that is uses rasops (which
is faster than the rcons stuff) and wsfont, so you can select the
console font in the kernel config file. Using the BOLD8x16 font,
you get an 80x30 console (vast improvement compared to the Gallant12x22
font).
2001-02-01 01:25:56 +00:00
thorpej
41b0cdf3e5
Add configuration info for the SEGA Broadband Adapter,
...
and enable the NFS file system and DHCP NFS root support.
2001-02-01 01:09:12 +00:00
thorpej
26cf6921bc
GAPS PCI bridge support. Low-level stuff from Marcus, bus_dma
...
implementation from me.
2001-02-01 01:04:55 +00:00
thorpej
e2c0abd45d
Implement {read,write}_region_1, and sprinkle some __volatile
...
around for sanity.
2001-02-01 01:01:50 +00:00
marcus
f2358f8d43
Basic G2 locking implemented.
2001-01-31 21:58:37 +00:00
thorpej
d729ba2b1f
Support for masking/unmasking IRQ 11 (used by the G2/GAPS bridge).
...
From Marcus.
2001-01-31 18:44:07 +00:00
thorpej
a42d651b53
G2/GAPS PCI configuration glue, for the Broadband Adapter.
2001-01-31 18:39:46 +00:00
thorpej
9342211052
PCI glue, from Marcus.
2001-01-31 18:38:26 +00:00
thorpej
7135afee31
New bus_space and bus_dma framework, based on the Alpha implemetation.
2001-01-31 18:37:11 +00:00
thorpej
63885cd020
G2 bus support. Low level stuff by Marcus, bus_space stuff by me.
2001-01-31 18:33:24 +00:00
pk
25a91d5c1e
_bus_dma_valloc_skewed(): return a vaddr_t, not error codes; make missing
...
boundary check return an error, so the caller can decide how to respond.
2001-01-31 15:17:42 +00:00
thorpej
c625c7ee25
Set the keyboard layout based on what type of keyboard
...
we probe, defaulting to Japanese if we don't know what
kind of keyboard we have.
XXX This should be done differently, if we ever want to
support multiple keyboards.
2001-01-31 03:21:03 +00:00
thorpej
37aca7f3f0
Add a US keyboard map. This may need to be reworked, as this
...
uses the Japanese keyboard map as the base.
2001-01-31 03:19:50 +00:00
thorpej
b81a17e100
Make the "maple" bus a direct config bus. Now, when a device is
...
not configured into the kernel, you know about it, e.g.:
Dreamcast Controller at maple0 port 0 not configured
2001-01-31 00:14:29 +00:00
thorpej
5c8624a602
Recognize the US keyboard.
2001-01-31 00:08:35 +00:00
tshiozak
d3545308ee
add umidi.
2001-01-30 23:32:34 +00:00
tsutsui
677243711e
Add a version file and print version strings generated by newvers_stand.sh.
2001-01-30 14:29:47 +00:00
tsutsui
9af4874133
Umm, fix botch in the previous;
...
include bsd.own.mk at the top, and include bsd.prog.mk at the bottom.
2001-01-30 14:21:49 +00:00
tsutsui
75e4c0c1be
- bzero() -> memset()
...
- Remove unneeded includes
- WARNSfy
2001-01-30 14:11:00 +00:00
tsutsui
85e5565f13
Remove (unused) -DBOOT_DEBUG.
2001-01-30 12:48:34 +00:00
tsutsui
2b258b5e2b
Include bsd.prog.mk at the top to pull in ${PRINTOBJDIR}.
2001-01-30 12:42:43 +00:00
msaitoh
61831529f2
add some macros for 8bit PCMCIA access
2001-01-30 04:15:10 +00:00
msaitoh
e0b71cc821
fix last commit.
2001-01-30 03:01:45 +00:00
msaitoh
e094a52b1f
fix pmap_* arguments when SH4_PCMCIA is defined.
2001-01-30 02:39:57 +00:00
msaitoh
69ababe7ba
more use _C_LABEL()
2001-01-30 01:32:56 +00:00
msaitoh
14bda6deb9
use macro for CCR register
2001-01-29 22:19:19 +00:00
chuck
9f82193c0e
ignore garbage left in registers by some APM bios (IBM thinkpad) when
...
event is a duplicate.
gets rid of annoying "APM get event: unknown error code (0x530b)" messages.
PR#12067 (YAMAMOTO Takashi)
2001-01-29 19:26:36 +00:00
leo
dcee338286
Fix y2001 problem in the initial read of the RTC.
2001-01-29 08:32:23 +00:00
uch
67c95e9ab7
remove old platid_gen
2001-01-29 04:23:01 +00:00
mrg
6d444f5960
kill the ad clause in my license.
2001-01-29 01:37:56 +00:00
itohy
a9cd6b907b
Performance hack.
...
Mainly for the Neptune-X Ethernet adaptor,
and improves ftp(1) speed by ~10%.
get: 219 KB/s -> 246 KB/s
put: 226 KB/s -> 247 KB/s
(X68030, MC68030 25MHz)
2001-01-29 01:09:29 +00:00
is
6cec6799ac
SIR_CLOCK is gone.
2001-01-28 22:31:53 +00:00
is
5d4912dc4b
Add NetBSD rcsid, clean up comments.
2001-01-28 22:18:17 +00:00
is
f7aa87a4a5
Jens Schönfeld seems to recall the name is ISDN link.
2001-01-28 21:43:38 +00:00
ragge
e4d9af4b42
A bunch of new CPUs that are supported, from Michael Kukat
...
(michael@unixiron.org ). Among them are 4000/400, 4000/500 and 4000/105A.
2001-01-28 21:01:52 +00:00
pk
5538893eda
sun4_dmamap_load(): Check if boundary condition can be satisfied instead
...
of panicing blindly.
2001-01-28 20:48:31 +00:00
ragge
747983bb5b
Fix late "constifying" in ddb.
2001-01-28 20:47:45 +00:00
ragge
630d0523f6
These are not used anymore.
2001-01-28 20:04:50 +00:00
ragge
8032849729
Rename dz_ibus to dz_vsbus.
2001-01-28 19:26:34 +00:00
uch
17b8c0e553
setup platid and config_hook.
2001-01-28 03:40:25 +00:00
uch
2f13dfd80c
s/EVBSH3/HPCSH/
2001-01-28 03:31:30 +00:00
uch
584cfa7669
Switch to MI md_root.c.
2001-01-28 03:09:40 +00:00
uch
37e1f36fec
add hpc dir if MACHINE is hpcmips or hpcsh.
2001-01-28 02:58:19 +00:00
uch
6be22c0c6f
hpcmips hpcsh common files are moved to arch/hpc.
2001-01-28 02:52:15 +00:00
martin
a7206004fc
Faster error handling for iommu_dmamap_load_raw; avoid loading the map
...
and unloading on error when we clearly can't get out without an error.
2001-01-28 01:26:57 +00:00
nathanw
88fbd1edd4
Add joy at pnpbios.
2001-01-28 00:50:23 +00:00
nathanw
5028590f77
PNPBIOS joystick attachment.
2001-01-28 00:37:52 +00:00
eeh
6c3bf28ffd
Fix dvma problem caused by previous. Seems like the dvma segment calculation
...
in iommu_dmamap_unload() was not quite right.
2001-01-27 03:40:39 +00:00
lukem
952413677f
add options NTP (ok-ed by thorpej)
2001-01-27 03:02:52 +00:00
is
facd2d0feb
Support for the Zeus Development(?) ISDN board (2189/3).
2001-01-26 21:49:04 +00:00
minoura
42c425801b
Use IOCS_S_READEXT only when reading the partition boot from
...
higher part of the disk than 0x200000.
2001-01-26 17:30:15 +00:00
tsutsui
9f8af7c171
Sync with sun3/machdep.c:
...
> Require the machine-dependant DDB commands to be in db_machine_command_table[]
> and link it directly to db_command_table[] so that it's not necessary
> to do this at runtime. Make db_machine_command_table[] const on all ports.
> g/c now unneded stuff, like db_machine_commands_install(), db_machine_init()
>
> Patch written by enami.
2001-01-26 16:12:12 +00:00
aymeric
a85be0a99c
process_machdep.c is actually procfs_machdep.c
2001-01-26 10:31:29 +00:00
sato
8211e0e08b
* support more than 16 ids.
...
* check the event ids are in the table.
2001-01-26 03:44:33 +00:00
sato
18ae54300c
* renumber config_hook event type.
...
* change event id of old event type(POWERCONTROL,BUTTONEVRNT)
to same as event id of new event type(EVENT,SET,GET).
2001-01-26 03:42:30 +00:00
is
d2ace86ad7
Two suggestions by jhawk:
...
- be more verbose about what file is created
- adjust copied $NetBSD...$ from $ NetBSD to NetBSD.
2001-01-25 22:34:20 +00:00
is
3f25157bea
Add support for ISDN Surfer (ISDN only, for the time being).
2001-01-25 22:22:15 +00:00
martin
f4eec8f31f
iommu_dvmamap_load_raw: add a check for creating maps larger than
...
expected by the creator of the dma map and fail apropriately. This
makes the syssrc/dev/pci/if_vr.c driver work.
2001-01-25 21:41:10 +00:00
fvdl
f2b255d7c2
Don't use pmap_enter in vmapbuf. This is essentially a reversal of
...
revision 1.76. It avoids problems where an I/O interrupt for physio,
using a bounce buffer, would find the destination address mapped
read-only because the syncer process hit.
Suggested by Chuck Cranor.
2001-01-25 20:14:28 +00:00
tsutsui
08f4daf2ce
Add drivers for keyboard and mouse.
...
(but no working framebuffer yet..)
2001-01-25 14:33:28 +00:00
tsubai
d0a3070c40
Generate key press event and key release event when capslock is pressed.
...
Now capslock key works correctly on X11.
2001-01-25 14:08:55 +00:00
wiz
ec35b8d060
Make output selection work -- you can use lineout now instead of the
...
internal loudspeakers.
2001-01-25 13:39:22 +00:00
tsutsui
b2457f9a76
Move NEWS keymap file into MI place to share with news68k.
2001-01-25 13:13:59 +00:00
marcus
7df82ad12b
Changed implementation of partitions, multisession discs now work better.
2001-01-25 01:41:47 +00:00
jmc
59d9cb6e67
Spelling police
2001-01-24 19:38:44 +00:00
jmc
5fe51991eb
Enable MFS. This will allow /dev to be built on mfs.
2001-01-24 18:08:26 +00:00
sato
3689556739
* fix typo.
...
* change PWR_* handling to PWR_SOFT* handling, if these routines need to sleep.
2001-01-24 09:34:26 +00:00
sato
2e9c446398
* dopowerhooks(PWR_SOFTSUSPEND) in apm_suspend().
...
* dopowerhooks(PWR_SOFTSTANDBY) in apm_standby().
* dopowerhooks(PWR_SOFTRESUME) in apm_resume().
* dopowerhooks(PWR_RESUME), dopowerhooks(PWR_STANDBY),
dopowerhooks(PWR_SUSPEND) run in splhigh() state.
These changes sync with arch/i386/i386/apm.c.
2001-01-24 09:29:25 +00:00
jmc
c97a56563e
Turn on NFS_V2_ONLY as the kernel has gotten too big for the ramdisk otherwise.
2001-01-24 06:10:09 +00:00
sato
fbf168d909
delete unneed debug message.
2001-01-24 03:52:16 +00:00
sato
a164b74b53
revert one of 1.34 change.
...
> * emul/mapping infomation turn to per screen.
(my misunderstanding)
currently per device.
2001-01-24 03:50:20 +00:00
sato
4af7f46d47
* support hpcfb_show_screens()'s cb and cbarg arguments. (HPCFB_MULTI)
...
* emul/mapping infomation turn to per screen.
* rename hpcfb_check_scroll() to hpcfb_check_update(). (HPCFB_JUMP)
* rename hpcfb_scroll_update() to hpcfb_update(). (HPCFB_JUMP)
* update experimental HPCFB_JUMP codes. (but not yet fix yet)
2001-01-24 01:38:53 +00:00
bjh21
917e1e8fd2
Set the palette in 4bpp modes. They still don't work right, though.
2001-01-24 00:56:29 +00:00
bjh21
050769744d
cf_unit usage is now fixed.
2001-01-24 00:03:21 +00:00
bjh21
3c3ed6cecd
Flush out cf_unit abuses.
...
Also, most device functions related to IRQ handling no longer take a device
pointer. We make so many assumptions about the machine's layout in irq.c that
this just seemed silly.
2001-01-23 23:58:31 +00:00
bjh21
d21e15aefe
This is no longer used.
2001-01-23 23:19:45 +00:00
bjh21
c59e8d1831
ANSIfy.
2001-01-23 22:59:24 +00:00
bjh21
34bf08cd05
Use a permanent breakpoint to invoke DDB rather than constructing a trapframe
...
manually. Prevents the problem introduced in the trapframe unification
whereby continuing from DDB would cause a panic.
2001-01-23 22:41:16 +00:00
bjh21
f32fc0fc90
Use evcnts properly for interrupt counting.
...
The architecture here follows that of the vax port -- each device has its
evcnt in its softc, but defers actually incrementing it to the IRQ
dispatcher. This way, devices can attach sub-counts (e.g. Rx and Tx counts
for Ethernet interfaces), but don't all have to have code to increment the
counters.
Drivers deliberately call evcnt_attach_dynamic() before establishing their
interrupt handler so that the establish routine can attach a parent event if
that's appropriate. At present, it isn't.
2001-01-23 22:07:59 +00:00
bjh21
8c87715ce0
We abuse cf_unit. The shame!
2001-01-23 22:07:58 +00:00
bjh21
616dad1a1b
Use evcnts properly for interrupt counting.
...
The architecture here follows that of the vax port -- each device has its
evcnt in its softc, but defers actually incrementing it to the IRQ
dispatcher. This way, devices can attach sub-counts (e.g. Rx and Tx counts
for Ethernet interfaces), but don't all have to have code to increment the
counters.
Drivers deliberately call evcnt_attach_dynamic() before establishing their
interrupt handler so that the establish routine can attach a parent event if
that's appropriate. At present, it isn't.
2001-01-23 22:07:57 +00:00