7848004c46
Do the I/O mapping in the main attach routine.
...
Also, we don't need to bother cleaning up on a failed attach -- we keep enough
state for the detach routine to do it.
2004-08-09 19:34:00 +00:00
bc7b11e4ff
Add newlines to the messages...
2004-08-09 19:33:07 +00:00
b7a991ceb6
Whoops, nuke an unused variable.
2004-08-09 19:09:06 +00:00
65d47754b3
Assume that io_map/unmap and intr_establish/disestablish will not be called
...
with the function enabled. If this happens, issue a warning. (This is only
really important for multifunction cards.)
This allows me to get rid of a bunch of extra code.
2004-08-09 19:08:19 +00:00
1710cea8cf
Enable DDB history.
2004-08-09 19:04:32 +00:00
f65dd54152
Correct the intr_establish order here.
2004-08-09 18:51:32 +00:00
7b96c587dc
Fix the intr_establish/disestablish order here too.
2004-08-09 18:41:36 +00:00
9092e2c97e
Make pcmcia_function_disable() primarily responsible for disabling the CCR
...
bits. pcmcia_intr_disestablish() continues to do this as a stopgap.
2004-08-09 18:30:51 +00:00
2ae40d3e4f
Do our io_map()s and intr_establish()es earlier.
2004-08-09 18:11:01 +00:00
af19b73069
Do io_map() and intr_establish() before enabling the function, always. This
...
prevents "stray interrupts" during attach.
2004-08-09 17:00:53 +00:00
deca9207b3
If the function is not enabled, do not attempt to immediately update the CCR.
...
XXX This is a stopgap until all drivers are converted to do their io_maps and
intr_establishes before function_enable.
2004-08-09 16:59:10 +00:00
9a46d5ed7d
Do an extra read of the IIR during initialization, just to ack any pending
...
interrupt.
2004-08-09 16:57:14 +00:00
c74553663e
Add some code to deal with the interrupt mask register on early CEM cards.
...
Doesn't seem to work yet, but it's marginal progress.
2004-08-09 16:05:00 +00:00
ed0b81628a
move DBG settings to CPPFLAGS and COPTS as appropriate
...
so that we can use the default DBG now.
2004-08-09 16:01:37 +00:00
c8180f49fd
Use PCMCIA_WIDTH_AUTO, not PCMCIA_WIDTH_IO16, since we (sometimes) do 8-bit
...
access.
2004-08-09 15:40:56 +00:00
b03f5b32c7
Pass a valid alignment to pcmcia_io_alloc().
2004-08-09 14:24:10 +00:00
6d0330a781
Add {spc,xirc,mhzc}@pcmcia. All tested on a Multia.
2004-08-09 14:16:02 +00:00
4e92489373
Add detach support for spc@pcmcia, modeled after aic@pcmcia.
2004-08-09 14:07:57 +00:00
f5eacc00f9
Don't loop forever if a card went away.
2004-08-09 13:52:21 +00:00
10121549dd
Add a hack that makes the address register programming more reliable on my
...
CE2 cards.
2004-08-09 13:30:16 +00:00
49b8d40924
#if SYSCTL_INCLUDE_DESCR -> #ifdef SYSCTL_INCLUDE_DESCR
2004-08-09 08:35:23 +00:00
0889de87d0
xi -> xirc,com,xi
2004-08-09 05:23:52 +00:00
eaae9fcf2b
The "savedpage" stuff is nonsense. Every place we touch the device is
...
protected by splnet(), so there's no reason to even try.
2004-08-09 05:11:33 +00:00
a271b4e6ab
Go online in xi_init(), not xi_full_reset(). Enable/disable receive in
...
xi_init()/xi_stop().
2004-08-09 04:47:40 +00:00
0e261f0935
Fix a debugging printf().
2004-08-09 02:10:25 +00:00
105470da05
Do not set INTRACK ourselves.
2004-08-09 02:01:16 +00:00
cafa5cc2d1
Implement the correct (I think) interrupt acknowledgement sequence.
2004-08-09 01:49:26 +00:00
00a9c911e6
Multifunction fixes:
...
* Don't call handlers for functions that are not enabled.
* Always call the handler if the function doesn't support the STATUS register.
While I'm here, divide the CCR numbers by 2.
2004-08-09 01:32:04 +00:00
213233a551
Access to this device is strictly 16-bit, so use PCMCIA_WIDTH_IO16.
2004-08-09 00:33:17 +00:00
16ac152cca
Do barriers on the ASIC space without touching the NIC space.
2004-08-09 00:26:15 +00:00
f0e2cfd7fc
Eliminate most 8-bit access. Just the data in/out path remains.
2004-08-09 00:25:33 +00:00
c52e6e82ec
Create two windows iff the config table asks for it. Try to make them
...
contiguous. Map windows with mode AUTO since we have mixed size access.
2004-08-09 00:00:36 +00:00
8231a1a314
Copyright maintenance.
2004-08-08 23:19:59 +00:00
ebed70c6e6
Remove the "offset" and "size" arguments to pcmcia_io_map(). In the singular
...
case (ne@pcmcia) where we were using these to create a subregion, it is better
handled by calling bus_space_subregion().
Now there is a 1:1 mapping between I/O spaces in the config table and windows
mapped in the function. Rework the multifunction mapping code to take
advantage of this by using both I/O base addresses if necessary.
2004-08-08 23:17:12 +00:00
b43cbaf291
fix two bugs in previous:
...
* flags passed from fork1() are FORK_*, not CLONE_*; thus, correct is to check
for FORK_SHAREVM, not CLONE_VM
* fix a reference counting bug
2004-08-08 19:52:37 +00:00
7e7559f105
by popular demand, bind the scroll keys by default.
2004-08-08 19:24:49 +00:00
959186b32c
disable the quirk on the console so that it can scroll.
2004-08-08 19:21:27 +00:00
ca7dc95cab
Enable DDB history by default.
2004-08-08 16:38:50 +00:00
76344241eb
Look in the correct register for the BSR detection value.
...
Also, do this early, before writing any registers.
2004-08-08 16:20:43 +00:00
fa9798d4c2
Search all the configuration entries.
2004-08-08 15:44:13 +00:00
91ad7491a9
AFAICT, there's no reason not to use PCMCIA_WIDTH_AUTO here.
2004-08-08 15:12:20 +00:00
df8fa882b3
Add UVMHIST tracking for data and prefetch aborts.
2004-08-08 14:21:29 +00:00
96e7bb0952
ath_init1: return EIO rather than -1 on a failure of ath_hal_reset.
...
-1 isn't appropriate for if_init.
2004-08-08 13:01:03 +00:00
9aba366e0b
Linux enforces CLONE_VM if CLONE_SIGHAND in clone(2) is specified,
...
follow the suit - this is intended to be Linux-compatible call
2004-08-08 11:02:10 +00:00
142def6b36
pass the fork flags down the emulation proc fork hook
2004-08-08 11:00:05 +00:00
c525acb34a
VT1616i support
2004-08-08 10:41:57 +00:00
28b17ac69e
in_control: fix address leaks on error, which causes a panic
...
("no domain for AF 0") on if_detach.
- SIOCAIFADDR, SIOCSIFADDR: free an address on error.
- SIOCSIFNETMASK, SIOCSIFDSTADDR: reject operations for an interface which
has no AF_INET addresses.
partly from OpenBSD and FreeBSD.
reviewed by Christos Zoulas on tech-net@.
2004-08-08 09:52:41 +00:00
9cc3176930
kvtopte: use a correct base addr for LARGEPAGES.
2004-08-08 09:47:05 +00:00
a714ac8294
linux processes sharing VM space (via clone() call) must also
...
share same 'break' value used for brk()/sbrk(), otherwise application SIGSEGVs
quickly once different threads try to adjust data segment size
this fixes linux Mozilla crashes with SuSE 9.1 libraries, and possibly
other linux applications using real threads
2004-08-08 09:40:50 +00:00
a37fba2a04
correct VAs in a comment.
2004-08-08 09:40:48 +00:00
b606f3e999
reduce gratuitous whitespace/comment differences from i386.
2004-08-08 09:38:50 +00:00
e4250cc050
Nuke an extra printf().
2004-08-08 09:23:01 +00:00
958716ead8
xi -> xirc,com,xi
2004-08-08 08:54:43 +00:00
74436be135
pass the fork flags down to the emulation fork hook, so that emulation
...
code can use the information for setup
2004-08-08 08:42:03 +00:00
bff77eedb1
Don't try to use the MII media status if we're not using MII.
2004-08-08 07:48:46 +00:00
019f8a0d02
* Some rearrangement of where the power control hooks are called.
...
* Don't splnet() in xi_init() -- the callers do that.
* Don't do the enable call in xi_init() -- it should be done only in a process
context so it can sleep.
* Call xi_start() at the end of xi_init() to restart output.
* If the interface is already up, we only need to call xi_set_address() to
change state on SIOCSIFFLAGS.
2004-08-08 07:25:20 +00:00
0cc7712b97
Hey, my LEDs work now.
2004-08-08 06:37:17 +00:00
ea155978ee
Use the correct data port, for compatibility with old cards. Without this,
...
some of my CE2 cards are unable to transfer data successfully; they send
bogus packets and every received packet is all 0s.
"Special."
2004-08-08 06:22:50 +00:00
d643fea8d1
Set xidebug to 0 by default.
2004-08-08 05:56:50 +00:00
2d5fe6edb4
Split "xi" into a "xirc" frontend (similar to mhzc; it attaches two child
...
devices) and a "xi" backend.
My CE2 and CEM2 cards are now probed correctly. However, there are still
some problems with one model of the CE2, and the CEM2 doesn't seem to get
any modem interrupts.
While I'm at it, fix several bugs:
* The tuple scan for the MAC address was broken in multiple ways.
* xi_intr() did not deal well with a shared interrupt.
* We were setting the wrong page number to look at the receive status. (How
did this work???)
* Remove the xi_full_reset() from xi_reset(). Move the parts of
xi_full_reset() needed to undo the effect of xi_stop() into xi_init(). This
allows a stop/init pair to DTRT, and much quicker, as used by various
ioctl()s.
* Set the TRS register before reading the TSO register, as the Linux driver
does. While I'm at it, fix the name.
* Fix numerous problems with xi_set_address(). "Where do I begin?" There's a
chance that multicast works now, but I haven't tested it.
* Explicitly clear the MSR register, and also force SELECT_MII to 0 if we
didn't find any PHYs.
* Clean up some cruft that appears to be bogus.
Probably needs more work, but it's a start.
2004-08-08 05:56:08 +00:00
961619ae0e
pcmcia_socket_{enable,disable}() weren't used. Reimplement them as public
...
functions taking a pcmcia* device structure.
XXX This is a method of last resort for dealing with stupid/insane cards that
we need to probe harder before we can choose a config entry. It should not be
used by most drivers.
2004-08-08 05:33:04 +00:00
fae6b577a8
Recognize the VIA Nehemiah by name.
...
Probe the VIA cache.
Add to, don't step on, feature flags when looking at extended feature
flags in VIA's 0x800000001 CPUID response.
2004-08-08 05:21:01 +00:00
bd2376263b
VIA C3 cache info.
2004-08-08 05:16:16 +00:00
824fd59f6d
correct the mask value for CMI9739
2004-08-08 03:52:19 +00:00
6f7bcb397d
recognize CMI9739 codec.
2004-08-08 03:50:56 +00:00
7732d6db9d
handle T_CONDITION traps (by sending SIGFPE).
...
fix T_EXCEPTION code to handle the other FP major opcodes.
2004-08-07 21:47:05 +00:00
43d9a77774
remove the alignment attribute from the typedef of __cpu_simple_lock_t
...
since it doesn't have any affect and lint chokes on it.
2004-08-07 21:42:10 +00:00
1974de97c5
stub out the ldcw stuff for now, since gcc -O2 chokes on it.
2004-08-07 21:41:44 +00:00
2e5e9111d9
clean up some register naming.
2004-08-07 21:41:18 +00:00
5dd3057b5c
replace some spinlock/tsleep code with a lockmgr lock.
2004-08-07 21:40:47 +00:00
272b73cf0a
use CALL() instead of "bl" in a few more places.
2004-08-07 21:40:05 +00:00
c5b0509d65
Clear the "card type" bit when powering up the socket, to make it more robust
...
for memory cards.
2004-08-07 20:11:39 +00:00
ef638869ae
Copy a bug fix from spc. If the target transferred too much, we would return
...
a bogus residual count.
2004-08-07 20:05:25 +00:00
0553444712
Fix a printf() format for Alpha.
2004-08-07 20:04:44 +00:00
6742fd43f4
Add Alpha to the list of platforms that use bus_space (for nca@pcmcia).
2004-08-07 20:01:48 +00:00
665e1f0611
Attempt to support the 688 as well. It looks vaguely right from inspecting
...
esl.
2004-08-07 19:32:56 +00:00
9e25bf4c82
AD1888 support.
...
PR kern/25949
2004-08-07 17:37:31 +00:00
5e289bdf22
Don't pass in a MAC address -- yet.
2004-08-07 17:27:30 +00:00
9918238ca5
Regen.
2004-08-07 17:18:20 +00:00
7e01821c61
Clean up the Lucent (and clone) entries a bit.
...
Add the Hermes II (doesn't work yet, but it's harmless).
2004-08-07 17:18:09 +00:00
900254667d
Don't pass a MAC address to wi_attach() -- yet.
2004-08-07 17:13:27 +00:00
1afe1f83af
Be much more robust with RID values during attach:
...
* If the device returns a length too short to even contain the RID number,
flag it as not supported and return EOPNOTSUPP, rather than returning a
length of -2 and trying to use the value anyway.
* Check the returned lengths to see if we actually got anything.
* If we get no rate list, fail the attach, so we don't just blow up later.
Allow the frontend to pass in a MAC address.
2004-08-07 17:12:44 +00:00
1c7464b5ca
recognize ALC655/658/850
2004-08-07 16:59:54 +00:00
3dce04c6cd
- Don't attempt to clear the CELV bit by setting it high like the
...
FIFOE/BCIS/LVBCI bits; it's is cleared automatically by the hardware.
- Separate AUICH_DEBUG out from AUDIO_DEBUG and make it possible to debug
interrupt handling separately.
- A little KNF.
2004-08-07 16:12:57 +00:00
f506c2b14b
Bah! Fix tyop CORE_UNKOWN_HANDLER -> CORE_UNKNOWN_HANDLER
2004-08-07 12:20:20 +00:00
eb40e03a12
Use SYSTEM_COPROC.
2004-08-07 12:07:08 +00:00
2d02f2af34
Add SYSTEM_COPROC (cp15).
2004-08-07 12:05:54 +00:00
548ccf50bd
Use FPA_COPROC instead of a manifest constant.
2004-08-07 12:04:05 +00:00
3de3493771
FP_COPROC* ->FPA_COPROC*
2004-08-07 12:02:36 +00:00
9264ede43d
Changing the interrupt mask is a scheduling barrier.
2004-08-07 11:47:58 +00:00
b0a3a4309a
Don't overload the unknown instruction handling for the core onto CP0.
...
Instead, add a seventeenth 'co-processor' specifically for the core.
Add support for ARMv5 unknown instructions in the 'NV' space.
2004-08-07 11:45:41 +00:00
951ec8627f
Don't use vector_page before it's been initialized.
2004-08-07 11:20:53 +00:00
da975d6d89
Fix pasto error in comment.
2004-08-07 11:13:09 +00:00
5b0328df32
Add barriers to prevent scheduling of critical code outside of the
...
SPL-protected region.
2004-08-07 11:08:58 +00:00
0163f96e43
Add fixed configs for fxp0 and sd0 booting.
2004-08-07 11:00:06 +00:00
0e811ff437
Support probing the compact PCI bus.
2004-08-07 10:54:54 +00:00
7c8434c371
Set CPUFLAGS not COPTS
2004-08-07 10:47:19 +00:00
f71b2096d2
On manual transfer via TEMP register, data should be read
...
before asserting ACK.
2004-08-07 07:17:09 +00:00
5ef4913a65
Remove extra space.
2004-08-07 05:46:39 +00:00
c4fbc53cad
Clean up attach messages.
2004-08-07 05:27:39 +00:00
mycroft