Commit Graph

13852 Commits

Author SHA1 Message Date
sakamoto
8c2e671594 Fix PR/4679:missing cast, from SUNAGAWA Keiki. 1998-01-27 05:46:59 +00:00
thorpej
5d3fc3e8bc Implement bus_dmamap_load_mbuf(). 1998-01-27 03:31:54 +00:00
thorpej
5ac7e550d1 Implement bus_dmamap_load_mbuf(). 1998-01-27 02:35:58 +00:00
gwr
995291fbac Deal with the fact that PROM driver descriptors are allowed to have
zero-length list of "standard addresses" by ignoring d_devbytes if
there are no addresses provided.  In these cases, the PROM will map
the device without our help.
1998-01-26 22:02:17 +00:00
gwr
faf4390c38 Save the "bustype" from our attach args in the controller "softc"
and use that instead of hard-coding BUS_VME32 everywhere.
(The Sun3X needs a more specific bustype value.)
1998-01-26 21:03:43 +00:00
gwr
3ba1e07ef5 Simplify the colormap and blanking stuff a little. 1998-01-26 20:53:17 +00:00
gwr
ac930c3c4d Moved back to sun3/sun3 1998-01-25 22:41:07 +00:00
pk
72f0ab269f These drivers are now in dev/vme.
The `ie' driver has been split in a VME and OBIO front-end. The latter
lives on in this directory as `if_ie_obio.c'
1998-01-25 20:33:54 +00:00
pk
3d46658e38 disk_establish(): add xd & xy cases (XXX - need more thought). 1998-01-25 19:59:14 +00:00
pk
f937c21c08 VME device declaration changes:
- `vmes,vmel' => `vme'
	- interrupt locator is the VME interrupt priority (`pri')
1998-01-25 19:56:30 +00:00
pk
531a4b2d77 sparc/obio bus front-end to the i82586 driver. 1998-01-25 19:50:50 +00:00
pk
8bc346d14a All VME stuff is gone from here. 1998-01-25 19:44:43 +00:00
pk
7a39d28b24 bus_map() & bus_untmp() are only used for sun4/obio devices. 1998-01-25 16:51:16 +00:00
pk
d44f401c95 bus_tmp() => obio_bus_tmp() 1998-01-25 16:49:33 +00:00
pk
bcb062819f The VME bus is now called `vme'. 1998-01-25 16:47:21 +00:00
pk
333b6319fc Convert to new VME interface. 1998-01-25 16:38:01 +00:00
pk
4efb52487a Split si & sw attach routines, and make the si part use the new VME interface. 1998-01-25 16:37:08 +00:00
pk
f8ba1e0bd9 sparc/VME bus implementation. 1998-01-25 16:06:25 +00:00
mycroft
63162f08ec Add a svr4_setregs() (that just calls setregs() right now). 1998-01-24 18:26:09 +00:00
mycroft
15ae963bea When dumping, print out the device number as major,minor. 1998-01-24 16:46:23 +00:00
pk
8a5023e38c After writing to the IOMMU flush registers, read something back from
IOMMU control space to a flush register to drain internal write buffers (?).
This avoids utter lossage on some machines (SS4s & SS5s) where our caller
would see some of its local (`%lx') registers trashed. This is probably
caused by a silicon bug allowing interference on internal data paths..
1998-01-24 16:22:47 +00:00
mycroft
1f21545d93 Correct an oversight in the startup changes that caused the wrong exception
to be reported in one case.
1998-01-24 15:50:42 +00:00
mycroft
8661bee5de Minor change. 1998-01-24 14:22:44 +00:00
ragge
eda1eecede Remove __BROKEN_INDIRECT_CONFIG and change all drivers to use
config_search instead.
1998-01-24 14:16:11 +00:00
mycroft
d664bf39b8 Add some comments about what the control words do. 1998-01-24 14:00:36 +00:00
mycroft
a963fe5864 Store the initial NPX control word appropriate for the emulation in the saved
FPU state, and use it when initializing the FPU on demand.
1998-01-24 13:19:46 +00:00
mycroft
a9a7cc88d1 Need foo_exec.h. 1998-01-24 12:45:16 +00:00
drochner
ae1a2006c5 -take over debugging printf format fixes from sys/lib/libsa/nfs.c rev.1.23
-make error reporting for mount working as in sys/lib/libsa/nfs.c
1998-01-24 12:43:09 +00:00
mycroft
8bd9717302 Add local versions of setregs(), so I can tweak some registers later. 1998-01-24 12:42:22 +00:00
mycroft
4b8a102fbb Fix process_{read,write}_fpregs(). 1998-01-24 12:11:15 +00:00
pk
49d2bfe020 Add a couple of options I forgot in last commit. 1998-01-23 22:47:20 +00:00
mycroft
d15d246d32 Add wdc* at isapnp?. 1998-01-23 20:42:34 +00:00
thorpej
cacd48c0d6 Print the correct error number if nfs_mount() fails. From Chris Demetriou. 1998-01-23 19:13:28 +00:00
mycroft
c1d56d473f Add wdc* at pcmcia. 1998-01-23 01:16:03 +00:00
mycroft
22c1670658 Some rearrangement of the startup code:
* Move some of the memory bootstrapping and the DDB startup earlier.
* Always put the IDT and the initial GDT and LDT in a separate page.
* Various minor changes.
1998-01-23 00:44:02 +00:00
gwr
119b962a9d Add: extern struct cfdriver cgtwo_cd; 1998-01-23 00:01:28 +00:00
gwr
66be7643e5 Use the same names as the sun3x (DVMA_MAP_*) 1998-01-22 23:59:45 +00:00
gwr
5ba5bda641 Use the same names as the sun3x (DVMA_MAP_*) 1998-01-22 23:45:05 +00:00
gwr
b6be0a1994 Use MONEND instead of DVMA_SPACE_START in the "DDB convenience" symbols. 1998-01-22 23:39:03 +00:00
gwr
dd049a163c Add config flags for tty[cd] to set TIOCFLAG_SOFTCAR.
Remove the framebuffer devices (not useful here).
1998-01-22 23:21:57 +00:00
is
53d89006d1 If on the M68060, make cache-inhibited pages imprecise. Reads and writes
are still in-order, but cached reads dont wait for the last write to finish.
Xamiga on a Altais in 8bit-mode became 30% faster servicing xanim (well, 6%
if you count xanim, too).
1998-01-22 23:13:40 +00:00
gwr
2c9c7a74bd Remove the config flags from "si" lines so the driver default will
be used.  That also makes it possible to patch the kernel in case
the default option causes boot failures or something...
1998-01-22 23:00:24 +00:00
gwr
8c3664b216 Remove the config flags from "si" lines so the driver default will
be used.  That also makes it possible to patch the kernel in case
the default option causes boot failures or something...
1998-01-22 22:27:57 +00:00
gwr
c292276dcf Change our DVMA map from the last 16MB to the last 1MB because that is
the only part that is visible to VME masters, and is adequate space.
This allows the "si" driver (on VME) to use DMA transfers.
1998-01-22 22:20:35 +00:00
gwr
56100bea19 When calling iommu_* functions, use IOMMU_VA_MASK. 1998-01-22 22:12:36 +00:00
gwr
0c50bf6487 No need to print the size (it's constant). 1998-01-22 22:09:11 +00:00
gwr
19e7bf4dcd Add: IOMMU_VA_MASK 1998-01-22 22:07:45 +00:00
gwr
cc1b4e4724 In pmap_bootstrap_copyprom, copy only the last PTE from the PROM's
DVMA segment (for the page reserved by the PROM).  If we copy all
the mappings, pmap_enter_kernel may complain about finding valid
PTEs that are not recorded in our PV lists.  Make pmap_remove_pte
do something reasonable if it finds a PTE for a managed page that
has not been recorded in the PV lists.  (Plus other minor stuff).
1998-01-22 22:00:25 +00:00
gwr
58126a3b29 Separate pmap_activate() from _pmap_switch(), so the latter can remain
equivalent to what locore.s:cpu_switch does with in-line assembly when
PMAP_DEBUG is not defined.  Also do splimp/splx in pmap_activate.
1998-01-22 21:48:42 +00:00
gwr
43013bccbb Use MONEND instead of DVMA_SPACE_START in the "DDB convenience" symbols. 1998-01-22 21:35:27 +00:00