Commit Graph

43200 Commits

Author SHA1 Message Date
rumble
db8f31d6dd Comment out tl(4) and nsphy(4) (used by the Set Engineering board) as it
appears to have an unresolved problem with bulk transmits.
2006-09-01 03:36:30 +00:00
rumble
67e8288ba5 Remove the 'FIFO Full' hack for interrupt vector 0.
It is unclear that this interrupt fails to latch as int2_wait_fifo()
depends on it doing so and appears to work. Furthermore, we were previously
unconditionally unmasking the interrupt 0 vector, which led to more
'unexpected interrupt' messages on the console, rather than avoiding them.

Approved by sekiya@.
2006-09-01 03:33:41 +00:00
matt
4139dfe2aa Move a variable to prevent a "unused" variable warning. 2006-09-01 03:24:50 +00:00
matt
28ee68ae9b Make this compile. The "fixed" bus_space_tag_t is horrible. 2006-09-01 03:24:08 +00:00
matt
aa6c194477 Remove an unneeded volatile. 2006-09-01 03:22:58 +00:00
matt
fc28199c6e Fix a -Wshadow error. 2006-09-01 03:22:26 +00:00
bjh21
1323fd42da This file isn't used any more. 2006-08-31 23:16:28 +00:00
freza
fd156a0bc9 * move the "cheating" conditional into unmapiodev() implementation
itself (instead of memio_unmap()), as it differs between OEA and
  ibm4xx.

OK by matt@
2006-08-31 23:14:38 +00:00
freza
8f274ce5a8 * implement unmapiodev(), make mapiodev() aware of ppc4xx_tlb_mapiodev()
* we now don't rely on having console linear-mapped (that was terribly
  broken, really)
* also, this can be used as generic ibm4xx mapiodev()/unmapiodev()

OK by matt@
2006-08-31 22:53:40 +00:00
freza
ca97defaa7 * ppc4xx_tlb_reserve(): allocate "reserved" TLB entries dynamically
* ppc4xx_tlb_mapiodev(): resolve pa to va from reserved TLB entries

OK by matt@

XXX we'll keep TLB_NRESERVED defined until we fix explora to use new API
2006-08-31 22:13:51 +00:00
freza
60d1041835 * add PVR values for Xilinx 405 cores
* don't try to decode vendor-specific PVR, print raw value instead.
* panic() if we see cache wasn't probed, we'd crash later anyway.
* rework the way PVR gets translated to core name.
* while there, normalize printf format ("%s: ...", device_xname(self), ...).

OK by matt@
2006-08-31 21:32:27 +00:00
dyoung
8cd106d3d2 Per discussion on tech-kern and tech-userlevel, move the bit-twiddling
macros, __BIT, __BITS, SHIFTIN, SHIFTOUT, and __arraycount() from
lib/libkern/libkern.h to sys/cdefs.h.  Add a __-prefix to SHIFTIN
and SHIFTOUT, and add a manual page for the bit-twiddling macros,
bits(3).

Make the __BIT and __BITS macros "widthless," as best I can, by
changing their type to uintmax_t from uint32_t.  XXX The manual
page lags this change by a bit.

Define __PRIxBIT and __PRIxBITS printf(3) format strings.
2006-08-31 19:24:37 +00:00
tsutsui
a714ccc7c2 volatile void func(decl) -> void func(decl) __attribute__((__noreturn__)) 2006-08-31 19:23:44 +00:00
matt
77ef4246de Call unmapiodev when unmapping bus_space to possibly reclaim the VA. 2006-08-31 18:28:47 +00:00
matt
3d324b035e Add unmapiodev(vaddr_t, vsize_t) (to be used by bus_space_unmap) 2006-08-31 18:23:40 +00:00
matt
25159b5f42 Add unmapiodev. 2006-08-31 18:18:17 +00:00
matt
6a157410d2 Increase SYMTAB_SPACE 2006-08-31 18:12:42 +00:00
matt
00185e0248 Make these compile again. (#include <arm/ep93xx/epgpioreg.h> and
do the proc -> lwp changeover).
2006-08-31 17:53:19 +00:00
matt
d2fa9e5329 Make cpu_coredump dependent on COREDUMP 2006-08-31 16:49:21 +00:00
matt
34827b737d More PTRACE fallout. 2006-08-31 16:39:13 +00:00
matt
5322f324e6 Make this console now that vconsole is done and with GCC4. 2006-08-31 05:11:07 +00:00
xtraeme
6b60608a48 Execute k8_powernow_init() if it's a K8 CPU and is able to change FID/VID
(previously the test was always false).

POWERNOW_K8 now runs in 32-bit mode with amd64 CPUs.
2006-08-31 04:20:21 +00:00
rumble
8b5466f087 Describe the GIO Product Identification Word and add a few macros to
extract other fields.
2006-08-31 00:01:10 +00:00
rumble
c693018067 Add the giopci device and associated ethernet/mii chipsets for the Phobos
and Set Engineering GIO fast ethernet boards.
2006-08-30 23:59:50 +00:00
rumble
ba52409a2e Provide a pseudo-pci bus for the following PCI ethernet boards, which live
behind custom PCI<->GIO bridges:
	- Set Engineering GIO Fast Ethernet (TI ThunderLAN)
	- Phobos G100 (DEC 21140?)
	- Phobos G130 (DEC 21143)
	- Phobos G160 (DEC 21143)

All boards present the chipsets' pci configuration registers at some
defined offset in their slots' address space as well as device registers.
We simply allow the MI pci subsystem to attach the devices.

This has been tested with a G130 board (DEC 21143) and works well on IP20
and IP24. The Set Engineering board attaches, works fine when receiving and
lightly transmitting, but chokes for unknown reasons on heavy transmits.
The tl(4) driver may need some fixing.
2006-08-30 23:58:13 +00:00
rumble
2d1cbf7222 Implement the gio_arb_config() interface for changing the machine-specific
gio bus arbiter parameters via imc(4) or pic(4).

Slots are identified by name: GIO_SLOT_GFX, GIO_SLOT_EXP1, GIO_SLOT_EXP2.

Provide some helper functions for establishing interrupts associated with
each slot and obtaining product descriptions.
2006-08-30 23:48:55 +00:00
rumble
a4d0bfd738 Add pic and imc-specific functions to alter the GIO bus arbiter
configuration. These are to be used exclusively by a higher interface in
gio/gio.c.

Also add functions to enable, disable and query the state of SysAd bus
parity detection since certain gio(4) cards need to be able to disable
it in order to work around their own buggy implementations.
2006-08-30 23:44:52 +00:00
rumble
aa71bc7886 Generalise pci/pci_machdep.c further and move all mace-specific code into
mace/pci_mace.c in preparation for pci devices on the gio(4) bus.
2006-08-30 23:35:10 +00:00
matt
4ba15ad876 Conditionalize cpu_coredump on COREDUMP 2006-08-30 22:24:55 +00:00
matt
2107a6af86 Don't include cpu_coredump unless the COREDUMP option is present. 2006-08-30 21:25:22 +00:00
cube
dc566666d0 Work around an issue in gas when cross-compiling i386 from amd64.
XXX gas should be fixed to not be confused by such statements instead.
2006-08-30 15:35:56 +00:00
cube
c847884087 Let this compile under "no options COREDUMP". 2006-08-30 15:34:31 +00:00
cube
98719e5089 Let this compile under "no options COREDUMP". 2006-08-30 14:01:57 +00:00
cherry
d98d4f0f0f remove redundant #defines 2006-08-30 11:14:23 +00:00
cherry
50ca7431be #include cleanups. 2006-08-30 11:12:04 +00:00
he
2f6928fe68 Use the common <dev/apm/apmio.h> file instead of replicating struct and
ioctl definitions in a local copy.  What remains is local additions.
Makes usr.bin/kdump build for macppc again.
2006-08-29 18:17:33 +00:00
gdamore
f7fbb8d998 Trivial white-space and comment fixup. 2006-08-29 02:35:44 +00:00
bouyer
aeeb0b33ca Some bus_dma(9) fixes for Xen:
- Attempt to gracefully recover from a failed decrease_reservation or
  increase_reservation, by avoiding physical memory loss.
- always store a machine address in ds_addr; this avoids some mistakes
  where machine address would in some case be freed at physical address, or
  mapped as physical address.
2006-08-28 19:58:56 +00:00
he
4c2420029f Add duplicate include protection to <dev/sun/kbio.h>.
Make the x68k port use this include file, so that what remains
is only the port-specific additions.
2006-08-28 17:15:24 +00:00
he
03d728a806 Collapse two more separate instances of MD vuid_event.h header
files to instead use the common <dev/sun/vuid_event.h> header.
2006-08-28 16:52:09 +00:00
he
4493c76313 Use the common <dev/sun/vuid_event.h> header file instead of maintaining
separate copies in port-specific directories.  Add duplicate include
protection.  Makes the alpha and atari ports build usr.bin/kdump/ again.
2006-08-28 16:41:04 +00:00
bjh21
44ef93a9be Be more careful with registers in LDR[T] so that GAS doesn't warn us about
constructs that aren't allowed in ARMv3.
2006-08-28 15:15:50 +00:00
tsutsui
e67dfa30bf Remove unneeded #include <sys/device.h> which also requires <prop/proplib.h>. 2006-08-28 14:31:45 +00:00
bjh21
d5930df82e unifdef -U __APCS_26__: we've been using APCS-32 for years. 2006-08-28 14:31:18 +00:00
bjh21
3612709d56 unifdef __APCS26__: we've been using APCS-32 for years. 2006-08-28 14:26:34 +00:00
tsutsui
73e8e44a34 options SYSTRACE is now disabled in GENERIC,
so remove "no options SYSTRACE".
2006-08-28 13:53:04 +00:00
yamt
e527ebac6f - remove unused bdbtofsb.
- move the following macros from MD headers to sys/param.h.
	ctod
	dtoc
	ctob
	btoc
	dbtob
	btodb
2006-08-28 13:43:35 +00:00
tsutsui
7ecca3ca2e Revert previous, because header search path for src/common/include is
now handled in src/sys/lib/libkern/Makefile.
2006-08-28 13:42:39 +00:00
tsutsui
25b6b97dcd Add src/common/include to include header search path since
<sys/device.h> includes <prop/proplib.h>.
2006-08-28 13:38:43 +00:00
gdamore
5cdb703d36 First pass at cleanup AR5312 WiSoC support to enable better & cleaner
sharing of code with the AR5315, which has many similarities, but many
differences from the AR5312.

No functional change at this time, other than the cpu_model string
(and also sysctl.hw.model node) is changed to reflect the WiSoC cpu
name rather than the identification string in ROM (which tends to not
be very informative.)
2006-08-28 07:21:15 +00:00