2cf5fc0ece
Add options to allow overriding of PCI0 & PCI1 MEM and I/O start and sizes.
2003-03-27 07:19:11 +00:00
d38e1fa853
Use "rxintr" for the name of the receive interrupts evcnt instead
...
of "txintr". Much less confusing that way...
2003-03-27 01:21:52 +00:00
0c23a8613a
Fix multiple bugs in the way we do the v4 MMU disable -- it was blasting way
...
too many bits (including some reserved ones) and was writing the wrong value
for the TLB flush.
Also, if the flag is off, don't write the control register!
2003-03-26 17:36:56 +00:00
e6b13790fa
SH5 needs __HAVE_FUNCTION_DESCRIPTORS defined in order for ld.elf_so
...
to resolve SHmedia function pointers correctly.
2003-03-26 14:46:32 +00:00
eefd7219fd
Sign-extend the return value, as required by the ABI.
...
While I'm here, simplify the asm operands/constraints.
2003-03-26 12:00:41 +00:00
eb7195a019
Gzboot support for the ADI BRH eval board. This flash address allows
...
the kernel to be decompressed and started at board reset if the BRH's
rotary switch is set to position 5.
2003-03-25 23:38:00 +00:00
0a6cf80b25
Standalone board support routines for the ADI BRH.
2003-03-25 23:34:26 +00:00
78c15d17f8
Memory routines for boards using the ADI BECC companion chip.
2003-03-25 23:33:49 +00:00
9c484bd88d
Allow the frequency to be set by the Makefile.
2003-03-25 23:32:41 +00:00
0abb67bb3b
Bump copyright date for last.
2003-03-25 19:47:30 +00:00
891be168b5
Add support for attaching on-chip peripherals to the BECC using
...
indirect configuration (because the BECC is a soft-core, it could
have a variety of peripherals in the FPGA). Also add support for
local untranslated DMA.
2003-03-25 19:45:52 +00:00
68fc25182e
Implement the beep on halt feature. When enabled, it makes the kernel beep
...
the computer speaker when it's completely safe to power down the machine
(after the "The operating system has halted." message). This is useful for
headless machines.
The feature is only enabled if the BEEP_ONHALT option is defined (disabled by default). It can be tuned through the BEEP_ONHALT_{COUNT,PERIOD,PITCH} options.
Closes my own PR kern/18792.
2003-03-25 19:37:12 +00:00
a63321f624
pmap_kprotect: use supervisor permissions; don't bother checking for `no
...
change' case, it's not worth the effort.
2003-03-25 11:33:46 +00:00
42d1c259a4
duplicate the changes from arm/arm32/stubs.c's dumpsys to remove pmap_map
...
usage.
Seems this file should really be split into the hpcarm specific bits
so it shares stubs.c with the other arm32 ports.
Pointed out by Håvard
2003-03-25 10:41:39 +00:00
9e3ba8326d
Add __KERNEL_RCSID tags
2003-03-25 06:53:15 +00:00
4691b478a0
Add __KERNEL_RCSID tags
2003-03-25 06:12:46 +00:00
3658055d88
Resurrect the REALBASEMEM and REALEXTMEM option for this port; they're
...
sometimes useful.
2003-03-25 00:09:27 +00:00
1ebbd9b463
Make PCI interrupts -> GPP mapping configured via config file.
...
Add bus_spaces for all possile obios. Add gt_halt to make
sure the discovery is not doing dma as soon as possible.
Read the SDRAM register to figure out how much physical RAM
is available.
2003-03-24 17:07:15 +00:00
955a468334
Add PIC support.
2003-03-24 14:26:16 +00:00
2e5c2cd2ae
Define just one mcount for both 32 and 64-bit ABIs; there's really no
...
advantage in having two slightly different versions.
Also save/restore FP register parameters.
2003-03-24 14:24:38 +00:00
28db85d8f6
Relax the initial user stack alignment from 16 to 8 bytes.
...
Add a workaround for what appears to be a bug in binutils whereby
the entry point of an SHmedia binary can sometimes specify
SHcompact code (LSB is clear).
2003-03-24 14:21:27 +00:00
f73433c6f6
Divide `set' instruction in delay slot (from OpenBSD change to rev 1.30).
2003-03-24 13:27:44 +00:00
3034a15d1f
+ fix a crash when write-back cache is used, by calling PTE_SYNC()
...
after tweaking page table entry.
+ 4th argument of bus_space_map() is not only for BUS_SPACE_MAP_CACHEABLE.
2003-03-24 04:15:49 +00:00
aa6bfd42a7
It appears that we now need locators.h. Other files in this directory
...
have it, and adding it makes this compile. If this is not the right
way to fix this problem, please let me know the right way...
2003-03-24 03:32:08 +00:00
c9033077aa
Garbage collect pmap_map, the last (and only?) use has been removed.
2003-03-23 15:59:23 +00:00
a97b660835
When doing a kernel dump use the pmap_k* funcs. Also make sure that all
...
data is written to ram. This avoids issues with tlb's not being flushed
etc.
As discussed a long time ago on port-arm
2003-03-23 15:49:25 +00:00
9fd86b683f
Add __KERNEL_RCSID tags to footbridge files.
2003-03-23 14:12:25 +00:00
0a11ce171f
Increase the INSTALL kernel's memory disk by 200k, in advance of some
...
additions to the ramdisk.
2003-03-23 12:44:25 +00:00
ea5200bf17
Add rf (RX01/02 floppy device driver)
2003-03-23 11:02:29 +00:00
f8f0fcb3af
don't make kernels with options DEBUG print too much debug messages.
2003-03-23 08:59:02 +00:00
5e9e4bd811
Make it compilable after something is modified..
2003-03-23 05:01:22 +00:00
0a30e5fb17
Fix a grammatical nit.
2003-03-22 14:26:41 +00:00
281c2b075c
GEOM_DEBUG was left set :-(
2003-03-22 13:05:26 +00:00
7bb8250096
PCI-CardBus bridge support for sparc64:
...
- add CardBus and PCMCIA devices configrations.
- add workaround for internal compiler error on dev/ic/elink3.c.
2003-03-22 06:39:43 +00:00
b1a9940dae
PCI-CardBus bridge support for sparc64:
...
- add MD part of cardbus support.
2003-03-22 06:36:18 +00:00
11529a0609
PCI-CardBus bridge support for sparc64:
...
- add splsoftserial.
2003-03-22 06:34:28 +00:00
aad45dd3ad
PCI-CardBus bridge support for sparc64:
...
- handle devices which has no OBP node.
- move PCI latency-timer initialization from pci_intr_map to
pci_enumerate_bus.
- make PCI bus free space extents for cardbus devices.
- fix PCI config space map size.
- some code integrations.
2003-03-22 06:33:09 +00:00
bd99e3429d
Use 'void *' instead of 'caddr_t' in prototypes of VOP_IOCTL, VOP_FCNTL
...
and VOP_ADVLOCK, delete casts from callers (and some to copyin/out).
2003-03-21 23:11:19 +00:00
05bdb0b97d
make these compile after struct consdev change
2003-03-21 17:23:22 +00:00
aab55006b2
Follow a struct consdev changes.
...
Pointed out by Tsuyoshi AKIHO in the port-hpcmips-ja mailing list.
2003-03-21 14:44:28 +00:00
1654f7e057
Yet another leftover of struct consdev changes..
2003-03-21 04:35:02 +00:00
6f16ef3a25
Strip some stuff that's not needed.
2003-03-20 22:30:53 +00:00
b52610cf4a
Limit the KVA used by buffers.
2003-03-20 22:30:28 +00:00
2e8a9de36c
memcmp()->bcmp(), this gives us 32 bytes and makes bootxx_ffs
...
build again
2003-03-20 18:41:46 +00:00
8d3aeb9abc
Rewrite user space access functions which use gcc's "labels as values"
...
extension in asm to avoid unexpected code omission during optimization
which causes wrong return value on faults.
2003-03-20 17:45:14 +00:00
11d8ce0e4e
catch up with constification in sys/lib/libsa/nfs.c
2003-03-19 17:27:43 +00:00
7349f5c301
"bootparam" specific variables are defined in libsa/bootparam.c now.
2003-03-19 17:21:41 +00:00
5d74e4ddb2
Now that we can deal with debug exceptions, use a BRK instruction for
...
ddb breakpoints instead of TRAPA in order to simplify the trapa (syscall)
handler.
2003-03-19 11:37:57 +00:00
2bce28091c
Fix some comments.
...
Set up stack frames in copyin/copyout and friends (if DDB is defined) so
that ddb stack traces can trace through them.
2003-03-19 11:34:28 +00:00
7c18481cf2
G/C an obsolete comment.
2003-03-19 11:31:55 +00:00
a06ca06a19
Spell 'strategy' correctly.
2003-03-19 03:05:43 +00:00
93d2b9ee12
Add some underscores in curcpu() to avoid a -Wshadow.
2003-03-18 22:29:58 +00:00
13db562ca9
Remove unneeded code in pci_bus_configure (was pci_config_bus). Fix
...
setting of bus_spaces in gt_attach.
2003-03-18 19:35:01 +00:00
0200088d83
Enable iobats for the PCI memory space(s).
2003-03-18 19:33:50 +00:00
8a5a3a480a
Switch/adapt to new bus space infrastructure.
2003-03-18 16:40:18 +00:00
0a2b7ed22b
Switch to gtmpsccnattach. Fix a few bugs in bus_space initialization.
2003-03-18 14:59:12 +00:00
0b9b6810a5
Removed unused variable.
2003-03-18 14:56:48 +00:00
3ec00b9a8f
work around a cache problem. (flush both caches and disable the D-cache).
2003-03-18 14:56:21 +00:00
ae4f6e5092
add interrupt numbers for built-in peripherals.
...
add register definitions for DMA, AC97, and USB.
2003-03-18 11:23:03 +00:00
f47a5ffeea
simplify CPU ID test for Intel PXA2xx by masking core revision part.
...
This also allows the kernel to run on pxa255.
2003-03-18 11:20:56 +00:00
40eb5e9921
fix XScale core revision mask, and add masks for core generation and
...
product number.
2003-03-18 11:17:31 +00:00
a935178cad
remove "kludge" com console.
...
Simplify timebase initialization.
2003-03-17 23:28:09 +00:00
0fb35cae8c
Get CLOCKBASE from opt_ppcparam.h.
2003-03-17 23:24:41 +00:00
ecc8926154
Add CLOCKBASE=100000000
2003-03-17 23:23:47 +00:00
bff46769b9
fix typo in comment.
2003-03-17 23:15:33 +00:00
afb54fa917
Add hardware accelerated copy/erase ops.
2003-03-17 18:39:23 +00:00
582f976e54
Add __HAVE_GENERIC_SOFT_INTERRUPT support.
2003-03-17 16:54:16 +00:00
c7b0df67b4
Add CLOCKBASE to opt_ppcparam.h.
2003-03-17 16:53:52 +00:00
090fac3d8b
Update for devopen() change.
2003-03-17 03:07:14 +00:00
5bd093cfbd
If the boot device starts with a PCI bus spec, skip over that when looking
...
for a boot device type; this gets the boot.ip32 booting a kernel off disk.
While here remove some unrolled string compares in favor of strncmp() to
make code more readable.
2003-03-17 03:04:51 +00:00
3306d06ea7
Don't use bitfields, make the fxsave structure packed instead. Makes
...
life a bit easier (for i.e. taking the address of a field).
2003-03-16 20:35:28 +00:00
de21c5e413
kernal -> kernel
2003-03-16 19:42:09 +00:00
4b261fc76f
allocate majors for vmmon/vmnet - VMware pseudo devices
...
we have the sources in othersrc tree now, so it's useful to have some
other bits in tree as well to simplify module use
2003-03-16 18:04:08 +00:00
cc8c490f0c
Adapt for changed struct reg
2003-03-16 16:26:58 +00:00
f6215de412
Sync with other reg state structures.
2003-03-16 16:26:11 +00:00
cea7b7a329
Add the newly separated-out ether_sprintf.c here, so that the booters
...
can be allowed to link with no unresolved symbols again.
2003-03-16 08:39:58 +00:00
77533df068
Remove some debugging code.
2003-03-16 08:12:26 +00:00
f5110f7988
EV64260 now gets to single user prompt using COM1 (2nd uart) as console
...
with PPCBOOT as firmware. XXX PCI devices are not found.
2003-03-16 07:07:17 +00:00
5504cc0c26
Use "b" constraint so r0 won't be used.
2003-03-16 06:57:31 +00:00
1d04c635f2
Make the result an early clobber so gcc won't get clever.
2003-03-16 06:56:47 +00:00
8524a72241
Honor PMAP_NC for pmap_kenter_pa. Fix goof in pmap_pte_to_va.
2003-03-16 06:54:46 +00:00
ea542f761a
addi is not the same as add. :)
2003-03-16 06:52:39 +00:00
8a37a3ec5d
Fix bus_dmamap_sync (add offset to ds->ds_addr when needed).
...
Add a set of parenthesis to PMAP_NC so it evaluates properly.
2003-03-16 05:37:37 +00:00
3735c66934
Add snprintf, divdi3, moddi3, and qdivrem so that ufsboot builds.
2003-03-16 01:29:04 +00:00
c5eb2581e1
Make struct sigcontext consistent with other regstate-holding structures.
2003-03-15 23:41:25 +00:00
640bee3010
Reduce visible globals. (gt_handle is now gt_memh and is only known to
...
machdep.c, gt_mainbus.c, and extintr.c)
2003-03-15 19:51:48 +00:00
101a152db1
This contains no marvell specific code now. It now completely hides the
...
underlying PCI MD implementation. XXX want to move this to
powerpc/include/ someday.
2003-03-15 19:50:31 +00:00
d69e6401db
Note the DDB_ONPANIC option (commented out). Closes my own PR port-i386/17649.
2003-03-15 19:34:16 +00:00
a0b2076415
New generic powerpc bus_space framework.
2003-03-15 08:03:19 +00:00
0f1794e44a
Make lint happy and use __asm && __volatile.
...
manipulate netisr via lwarx/stwcx. to get atomicity.
2003-03-15 07:50:28 +00:00
7ad3fe6e03
When mapping a page with BUS_SPACE_NOCACHE, make sure to flush the
...
physical page from the data cache.
2003-03-15 07:25:20 +00:00
35962f72c9
s;backside;; and report L2CR_L2DO & L2CR_L2IO
2003-03-15 07:22:46 +00:00
2b8417e488
Add LINTSTUBs and make oea_init() agree with them.
2003-03-15 07:21:02 +00:00
df24bda908
Make lint happy.
2003-03-15 07:19:20 +00:00
866b37e61b
pmap_kremove the pages before uvm_km_free'ing them. Thanks jason!
2003-03-14 18:47:53 +00:00
3b3cb97b5b
Attach esiop to a scsibus.
2003-03-14 17:29:05 +00:00
d03db36449
Use aprint_normal. Print 2MB L2 sizes with 7410.
2003-03-14 06:27:40 +00:00
e0242aaebc
Remove Debugger call.
2003-03-14 06:25:58 +00:00
d50c91e021
Make lint happy by not assigning to casted lvalue.
2003-03-14 06:23:48 +00:00
12dca1407b
Condition ({ ... }) by __GNUC__. Remove redundant SPR_IBAT0U definiton.
2003-03-14 06:22:51 +00:00
d7fc76efe1
Add _LOCORE protection.
2003-03-14 06:21:19 +00:00
61920c743e
Make ALI trap print DSISR.
2003-03-14 05:38:53 +00:00
a7b613e469
Use __asm and __volatile to make lint happy.
2003-03-14 05:37:51 +00:00
ce05df7bc7
Quiet lint warning.
2003-03-14 05:37:14 +00:00
8ceb32c0af
make LINTSTUB work with this.
2003-03-14 05:36:39 +00:00
c894ddaec1
Print more useful messages on kernel ALI or PGM traps.
2003-03-14 05:32:27 +00:00
0c29e154e3
Use __asm & __volatile to make lint(1) happy.
2003-03-13 17:30:38 +00:00
8344279675
Synchronize macppc and evbppc majors when possible. Add a lot of missing
...
devices to evbppc.
2003-03-13 17:02:52 +00:00
6b8fef6400
cleanup, delete declarations now in libsa/iodesc.h
2003-03-13 16:02:39 +00:00
66a5580c6e
Remove remaining uses of libsa/netif -- the "netif" structure was only
...
used to lookup a "struct of_dev", everything else was ballast.
Do it straightforward now and assign the ofdev directly to io_netif.
2003-03-13 15:36:06 +00:00
42cc9f8839
Add wscons to MAKEDEV. Add wsfont major number.
2003-03-13 14:56:56 +00:00
6e14bfebea
Remove remaining uses of libsa/netif -- the "netif" structure was only
...
used to lookup a "struct romdev", everything else was ballast.
Do it straightforward now and assign the romdev directly to io_netif.
2003-03-13 14:49:12 +00:00
bfca0494bd
Try to configure all children of cprc0, not just clock.
2003-03-13 14:37:36 +00:00
7d2a35f773
kill the last use of NENTS
...
(We shouldn't use libsa/netif here at all.)
2003-03-13 14:15:58 +00:00
a15dcfdce7
Avoid conflict with sockets[] definition in net.h
...
and add some typecasts.
Thanks to Juergen Hannken-Illjes for compile-checking.
2003-03-13 14:05:53 +00:00
20b8952600
cope with removal of this sick NENTS macro from libsa/netif.h
2003-03-13 13:59:11 +00:00
e537123c66
cope with changes to libsa/netif.h
...
(This shouldn't use libsa/netif at all.)
2003-03-13 13:55:24 +00:00
9053a4f127
NEC RISCstation 2200 (NEC-R94) doesn't have the secondary SCSI,
...
so prepare struct pica_dev for R94 in p_nec_r94.c to handle its quirk.
The problem was reported by David Hopper on port-arc.
XXX We should use ARC BIOS to get info of these devices.
2003-03-13 13:52:09 +00:00
d4f824014d
A bunch of things:
...
- Clean up the way cpu-specific tlb/cache functions are configured
and used.
- Add a workaround for a problem whereby cpu* at superhyway? fails
to probe.
- Print more info about the cpu/cache.
- Move the RESVEC handlers back into generic sh5 code and ditch
the panic stack hack.
- Make the on-chip SCIF device the default console on Cayman.
- Add experimental support for booting via a standalone bootstrap
program (not yet committed) and using the boot parameters passed
in by it.
- Add a few more SH elf constants.
- Tick a couple of items off the TODO list.
2003-03-13 13:44:16 +00:00
caa65123a8
cope with removal of this sick NENTS macro from libsa/netif.h
2003-03-13 13:35:55 +00:00
31b3a7c37f
In scif_getc(), check SCIF_SCFSR2_RDF in addition to SCIF_SCFSR2_DR to
...
detect waiting Rx data. This fixes an occasional console lockup when
when pasting text at the ddb prompt.
2003-03-13 13:19:01 +00:00
e7bde1e9c7
adapt for void *io_netif
2003-03-13 12:44:48 +00:00
47fe380b4c
make netif_close() return value void
...
fix prototype
2003-03-13 12:15:31 +00:00
f2746edc5b
Explicit cast for netif.
...
Approved by: Matthias Drochner <drochner@netbsd.org>
2003-03-13 12:02:54 +00:00
e5cc00b939
Remove remaining uses of libsa/netif -- the "netif" structure was only
...
used to lookup a "struct of_dev", everything else was ballast.
Do it straightforward now and assign the ofdev directly to io_netif.
2003-03-13 11:35:54 +00:00
3c7ff59b5b
Sprinkle some "volatile"; fixes problems with the {read,write}_{1,2}
...
functions big-endian Au1xxx CPUs.
2003-03-13 03:04:13 +00:00
9b72b507ef
-use our own netif layer
...
-call the pxe unload function only on close(), not on open() errors,
so we have a chance to retry
-remove some remaining references to NFS
2003-03-12 17:35:57 +00:00
168322ea77
kick out mi libsa/netif.c references
2003-03-12 17:33:55 +00:00
542ccfb0d5
Kick out the netif layer from mi libsa and implement an own, smaller, one
...
(we really don't need to support multiple interfaces)
Also arrange things so that we are able to unload the PXE stacks only
when we are sure that we don't need them anymore. (To make this more
useful, a hook in exec() is needed.
2003-03-12 17:33:10 +00:00
1ebc8c2dbd
cleanup:
...
-update comments
-remove remaining references to "bootparam"
-don't try to interpret extra arguments to net_open() - we
get only passed a NULL anyway (see devopen.c)
2003-03-12 13:15:08 +00:00
c693137139
sync with mi libsa/nfs.c (with the intent to replace it eventually):
...
-kill "register" attributes
-obey "LIBSA_NO_TWIDDLE"
-ignore leading "/" also in the NFS_NOSYMLINK case
2003-03-12 12:23:06 +00:00
cc782e8a7e
Make this compile with only SUN4 defined.
2003-03-12 07:51:00 +00:00
cd26de2684
Same code exists in both halves in #ifdef/#else/#endif. move it outside.
2003-03-12 06:00:36 +00:00
7d8f167f16
Split bus_space and bus_dma into separate files.
2003-03-12 00:09:51 +00:00
717651ac94
Band aid - the infamous dcache_flush_page change independently suggested
...
by Andrey Petrov and Jason Thorpje (AFAIK).
This is a hack that covers some symptoms while we have no idea where
the real problem is. Anyway, since this avoids random data corruption
we better be safe and have this in-tree until the problem is solved the
right way.
2003-03-11 21:28:23 +00:00
a82c138b1c
If we get passed a filename (from the command line or default "netbsd"),
...
pass it to the BOOTP server in the "filename" field.
(as the "netboot" bootROMs already did)
So the user can easily switch between different kernels/configurations.
2003-03-11 18:32:59 +00:00
dd15fa8102
Fix bug in readudp(): we have to return the correct length
...
of the received packet to the caller.
TFTP works now.
2003-03-11 18:29:00 +00:00
df62259ae1
kick out RARP/bootparam - PXE uses BOOTP anyway
2003-03-11 16:30:02 +00:00
ff28ac27ec
use our own NFS fs implementation, which is smaller, does a better job
...
in error reporting and coexists with the TFTP fs
2003-03-11 15:01:51 +00:00
8e04272b51
add prototypes for local functions
2003-03-11 14:38:45 +00:00
f8f0dcfd17
Need .code32 after real_to_prot (not .code16)
2003-03-11 12:37:35 +00:00
a6fda7bcfb
create a symlink so that eg <x86/trap.h> is found
2003-03-11 11:03:23 +00:00
249e0067fe
Add a port to the NCD Explora451 ThinClient.
...
- Supported devices: le, com, lpt, pckbd, frame buffer.
- Support missing: phy, audio, pcmcia.
- Boots from network or md.
Approved by: Eduardo Horvath <eeh@netbsd.org>
2003-03-11 10:57:54 +00:00
44b1e07ec9
Add support for the IBM 403GCX cpu. Enabled with "options PPC_IBM403".
...
- different set of device control registers.
- non-standard access to the time base.
- 16 byte cache lines.
Approved by: Eduardo Horvath <eeh@netbsd.org>
2003-03-11 10:40:15 +00:00
f86b0e1741
In tlb_exception(), don't panic immediately on access at va == 0
...
from kernel mode if fault hander is set.
2003-03-09 16:14:29 +00:00
1565f9401d
Remove remnants of support for console on ite, since the ite device
...
driver is no more.
2003-03-09 13:59:56 +00:00
60b9080da3
don't depend on "../../../../../gnu/dist/toolchain/include/coff/rs6000.h".
2003-03-09 00:39:10 +00:00
8a6f95626f
Avoid pulling strncpy() into minimal boot code
2003-03-08 21:30:59 +00:00
ef18027d4d
Don't trust %bp to point to same place in real and protected modes.
...
prot_to_real preserves %ebp but changes %ss and %sp.
(ok if real %ss is offset n*64k from protected %ss - which it has been)
2003-03-08 21:09:37 +00:00
b648593b37
Initialize the two new members of "struct consdev" to NULL, so that
...
this file compiles again.
2003-03-08 10:59:02 +00:00
matt