Commit Graph

67 Commits

Author SHA1 Message Date
chris 0e7661f023 Update pmap_update to now take the updated pmap as an argument.
This will allow improvements to the pmaps so that they can more easily defer expensive operations, eg tlb/cache flush, til the last possible moment.

Currently this is a no-op on most platforms, so they should see no difference.

Reviewed by Jason.
2001-09-10 21:19:08 +00:00
simonb a41b7a380e Clean up and standardise across MIPS ports. 2001-09-09 04:20:25 +00:00
thorpej 81bcece4d4 Implement bus_space_mmap(). 2001-09-04 16:32:42 +00:00
simonb c91e08563f Clean up. 2001-09-04 06:26:18 +00:00
simonb 62fb390c64 May as well include <mips/cpuregs.h> in <mips/cpu.h> once rather than
in every MIPS port's <machine/cpu.h>.
2001-09-04 06:23:15 +00:00
simonb 214f5366ea Centralise struct cpu_info declaration and related info to <mips/cpu.h>. 2001-09-04 06:19:21 +00:00
simonb a6b8c86af0 Remove an unneeded comment; ``sync'' with other "just include <mips/foo.h>"
files.
2001-08-31 03:53:22 +00:00
simonb 4ac4da1c52 G/C the unused kernel-only CLK_TCK #define.
XXX: does include/time.h still need <machine/limits.h>?
2001-08-31 03:46:03 +00:00
simonb bcdac55add Fix the display of the CPU clock speed. 2001-08-24 02:11:01 +00:00
simonb 8375df5b6a Whitespace nits. 2001-08-24 02:07:34 +00:00
thorpej 15dcb4a98e Link the kernel at 1MB. 2001-08-22 19:44:47 +00:00
thorpej 9f019df56f Allow a keypress on the console to transition from "halted" to "returning
to monitor".
2001-08-22 19:13:09 +00:00
thorpej f0ea5cb1e9 Don't need to do the Alpha boot block checksum here :-) 2001-08-22 17:53:44 +00:00
thorpej 17c22a634f Set PCI_NETBSD_ENABLE_IDE to 0x01, so that the first IDE channel
will be enabled.
2001-08-22 00:42:43 +00:00
thorpej f5e6f05c1a If PCI_NETBSD_ENABLE_IDE is set, enable the PIIX{3,4} southbridge
IDE channels on the P-5064 and P-6032, since PMON doesn't do it
for us (thus causing the `pciide' driver to ignore them).
2001-08-22 00:40:38 +00:00
wiz ef8d715888 bzero -> memset 2001-07-22 15:11:01 +00:00
thorpej babefc5331 Add BUS_DMA_READ and BUS_DMA_WRITE flags, that hint the back-end
at dmamap load time that the mapping will be used for a unidirectional
transfer of the specified direction.
2001-07-19 15:32:10 +00:00
tsutsui df016928f8 Remove (commented out) ncr* at pci? lines. 2001-07-07 17:09:47 +00:00
thorpej 7a1408c38b Deal with different `memsize' convention on new PMON revs. 2001-06-25 21:25:07 +00:00
thorpej 498641fc7e Print more BONITO rev. info. 2001-06-25 20:15:57 +00:00
thorpej 2a171eff3a Correct a typo. 2001-06-22 14:29:12 +00:00
thorpej 3ae8717568 Add commented-out examples for how to use MEMSIZE and ETHADDR options
in the event that you need them (really old PMON versions).  Add FDESC.
Comment out SCSI-related stuff until I work out why the system goes into
outer orbit when its enabled.
2001-06-22 14:15:56 +00:00
thorpej 04986e1e9a Add FDESC. 2001-06-22 14:14:16 +00:00
thorpej 306b1b09bc Kernel config file for a P-6032. 2001-06-22 14:12:19 +00:00
thorpej 36a4b627d0 Sigh, Jason needs to go to sleep. 2001-06-22 07:00:25 +00:00
thorpej 32e780695d Add a couple of missing p6032 bits. 2001-06-22 06:50:21 +00:00
thorpej 0c37c9e860 Check in work-in-progress of P-6032 support. This is not tested,
but is meant for back-up purposes.
2001-06-22 06:02:54 +00:00
thorpej 7f047ac2b2 Use the other DMA window (with prefetch enabled) on the P-4032
on-board Ethernet.  Suggested by Nigel Stephens <nigel@algor.co.uk>.
2001-06-22 05:57:26 +00:00
thorpej b440db94e8 Fill in one P-6032 snippet. 2001-06-22 04:33:26 +00:00
thorpej e4ce5f4268 A P-6032 will never have a V3 PBC -- don't include its option header. 2001-06-22 03:45:24 +00:00
thorpej b5443137f0 Only call through the function pointer if it's not NULL. 2001-06-22 01:42:20 +00:00
thorpej 1ce7119f17 More ISA interrupt rototilling. I get *some* interrupts on a:
we1 at isa0 port 0x300-0x31f iomem 0xcc000-0xcffff irq 10
we1: WD8013WC Ethernet (16-bit)
we1: Ethernet address 00:00:c0:94:f6:72

...but still nothing on the PCIC.
2001-06-21 19:00:18 +00:00
thorpej e0901dc354 Fix whitespace botch. 2001-06-21 18:58:18 +00:00
thorpej cc6c89106a Correct the ISA bridge irqmap. 2001-06-21 06:24:23 +00:00
thorpej f341043437 Rototill ISA interrupt code. In particular, I have remembered most
of the horror that is the 8259 PIC.

PCMCIA interrupts still not working yet, but getting closer.
2001-06-21 05:20:54 +00:00
thorpej e51a043945 Yet more interrupt cleanup -- the platform mater interrupt establish
function now just takes an "irq", which is an index into the irqmap
table for that platform.
2001-06-15 04:01:39 +00:00
thorpej 1d161cd563 Correct a comment. 2001-06-14 19:03:43 +00:00
thorpej fe87c9aade When the old PCI address map is in use, disable I/O space. Also disable
I/O space on PBC revs < B2.
2001-06-14 18:52:26 +00:00
thorpej 0ddf0d259b Don't use BUS_END to compute SYS_END. 2001-06-14 18:48:23 +00:00
thorpej 80dfaa3e5a Read the PCI memory space base and the PCI DMA window base from the
V3 PBC, and use them in the bus mem tag and the bus dma tag on the
P-4032.  This allows us to get much further when PMON has configured
the PBC using the old-style PCI address map.
2001-06-14 17:57:26 +00:00
thorpej 4c73c770ce Add MEMSIZE and ETHADDR options, so that they can be set in
the kernel config file, in case you have a buggy PMON which
doesn't provide the environment variables to the kernel.
2001-06-14 16:14:37 +00:00
thorpej defcaa8bcb Initlialize PMON after configuring the console, so we can see
debugging info.
2001-06-14 15:29:23 +00:00
thorpej 587d3ec2b4 Insert some debugging code, conditional on PMON_DEBUG. 2001-06-14 15:28:56 +00:00
thorpej 3bad583046 Oops, use the correct space tag when unmapping the RTC on the
P-4032 after calibrating the clock.
2001-06-14 05:58:19 +00:00
thorpej 7e59a1be7a No longer need MIPS3_5200 here. 2001-06-12 22:32:50 +00:00
thorpej be7629b7e0 Hand off intr evcnt responsiblity in a reasonable way on the 5064. 2001-06-10 09:28:26 +00:00
thorpej 7c074dc806 Check in work-in-progress of generic ISA interrupt support. The
goal here is to get the P-5064 PCMCIA slots working, and serve as
the basis for P-6032 interrupt support.

PCMCIA interrupt auto-detection not working -- more work to be
done here.
2001-06-10 09:13:06 +00:00
thorpej 82418a77b0 Add PCMCIA devices. 2001-06-10 08:45:09 +00:00
thorpej 1930cfd7dc Correct for a data structure change. 2001-06-10 06:17:15 +00:00
thorpej ce66bf0803 Rewrite the interrupt handling code:
- Compute the number of CPU pipeline cycles per second using the
  mc146818.
- Use the COMPARE interrupt for the hardclock interrupt.
- Collapse all interrupt priorities into a single priority, and use
  the CPU interrupt inputs to determine the interrupt source (local
  device, PCI device, ISA device, etc.)

This allows us to have interrupt sharing.
2001-06-10 05:26:58 +00:00