Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
103,338 Commits 606 Branches 0 Tags 3.1 GiB
Commit Graph

33196 Commits

Author SHA1 Message Date
joda 337993c988 use the same mnemonics as linux for cpu features (this is -o linux 
after all)
 2002-10-04 19:39:54 +00:00
junyoung 8c56af5cb9 Don't init IDT twice. 2002-10-04 19:27:05 +00:00
junyoung 857da21d88 Correctly identify model 6, 7, and 8 of AMD Athlon and Duron processors. 
Before:
cpu0: AMD Athlon Model 6 (Palomino) (686-class), 1532.11 MHz

After:
cpu0: AMD Athlon XP 1800+ (686-class), 1532.11 MHz
 2002-10-04 18:42:34 +00:00
elric d19d268a95 assign majors for raw and cooked cgd's. 2002-10-04 18:28:24 +00:00
rearnsha 16930e9cb8 Hmm, fix some bogosity in my previous attempt to fix some bogosity:-) 
Keep SYMTAB_SPACE (but clean it up).
Restore iophy PHYs.
 2002-10-04 17:14:48 +00:00
matt 2f1192b327 CFATTACH mem_sbi, not sbi 2002-10-04 17:09:51 +00:00
tsutsui 714e6adf24 Add le* at isapnp? 2002-10-04 16:24:47 +00:00
scw 4c6f867793 s/COPTS/DEFCOPTS/ 2002-10-04 10:24:05 +00:00
scw 26ebb442bf - Hook the PCI arbiter and error interrupts, and print something 
useful should one occur.
- Manually poke some config values into the sh5pci host bridge's
  config registers since it doesn't appear in config. space.
- Reserve the first 256 bytes of i/o space to avoid assigning i/o
  address 0 to any cards.
- Slight tweak to the initialisation code after consultation with
  SuperH and the linux driver.
 2002-10-04 10:22:24 +00:00
chris 5dded94793 Fixup IPL_LEVELS to be correct. This matches the change I did to footbridge 
based systems.  Untested on shark, but is the right thing to do.  I suspect
the original arm32 intr.h had the bug, and when the ports split we just took
the bug.
 2002-10-04 10:21:33 +00:00
scw a520b3b2c8 Convert to new devsw world order. 2002-10-04 10:16:56 +00:00
scw db80ebe781 Add needs-flag to sysfpga, and check it before calling sysfpga_nmi_clear(). 2002-10-04 09:24:52 +00:00
scw d4c3160197 pmap_bootstrap() needs to know the physical address of KSEG0. 2002-10-04 09:23:27 +00:00
scw f7f3db86ff When loading a DMA map, make sure the BUS_DMA_COHERENT flag in the 
map accurately tracks the same flag in the segments belonging to it.
The map's copy can be set only if all the segments are coherent.

This finally gets NFS writes fully working on my PCI ex(4) card.
 2002-10-04 09:20:20 +00:00
scw 99fec1b346 Minor pmap rototil: 
- Track unmanaged mappings of RAM more closely by allocating a pvo
   for them. This allows us to check more accurately for multiple
   cache-mode-incompatible mappings.

 - As part of the above, implement pmap_steal_memory(). This has the
   beneficial side-effect of moving a fair chunk of kernel data
   structures into KSEG0.
 2002-10-04 09:17:57 +00:00
simonb a66f0d74ef This file hasn't been used in a long time. 2002-10-04 08:48:35 +00:00
simonb dcd65411b2 Only declare and set the "step" variable in identifycpu() ifdef CPUDEBUG. 2002-10-04 08:44:08 +00:00
junyoung acd50bf487 Back out pushl/popl -> movl change. copy_fault is also used by kcopy(9). 2002-10-04 06:43:40 +00:00
explorer 3a856ad198 use config_match() 2002-10-04 06:02:38 +00:00
junyoung f9939562a1 Fix a mistake in previous commit. 2002-10-04 04:40:12 +00:00
junyoung 8ab9e2ef57 Use single movl instruction rather than pushl/popl pair. 2002-10-04 03:20:59 +00:00
itojun 8f0eabf697 whitespace 2002-10-04 03:17:00 +00:00
thorpej d90d300b7b Overhaul the way cfattach structures are looked up. The cfdata entry 
now carries the name of the attachment (e.g. "tlp_pci" or "audio"),
and cfattach structures are registered at boot time on a per-driver
basis.  The cfdriver and cfattach pointers are cached in the device
structure when attached.
 2002-10-04 01:50:53 +00:00
thorpej 1a5a7fe7b7 Use indirect configuration for on-board devices (i.e. those things 
attached to "obio") on the IQ80310 and IQ80321.  It makes more sense
to do it this way for this type of system (the goal being to encapsulate
as much information about the board as possible into one file).
 2002-10-03 20:14:58 +00:00
thorpej cea0a147b4 Add macros to convert an "IRQ" number to its respective XINT3/XINT0 bit. 2002-10-03 20:11:42 +00:00
thorpej 072eedb728 Add a symbolic constant for where external interrupts start. 2002-10-03 20:10:40 +00:00
fvdl 547337a4ce FIx off by one error for brand id array index. 2002-10-03 19:39:51 +00:00
elric 70bf1b763b prepend syscall_fancy and syscall_plain with mach_ to get this to compile 
again.
 2002-10-03 19:17:01 +00:00
uwe fbcf88f171 Add wscons. Misc catch-up with GENERIC. 2002-10-03 17:45:20 +00:00
uwe fa76f154c6 Add majors for wscons devices. 
And, no!  This does not mean that sparc switches to wscons.
Only Krups has experimental wscons support so far.
 2002-10-03 17:06:18 +00:00
uwe 7235b5d479 First cut at PS/2 keyboard and mouse drivers that talk Sun firm events. 
Console and Xsun for Mr.Coffee.
 2002-10-03 16:27:04 +00:00
uwe fb2195ae8f Search children of the "8042" node to check if keyboard is the stdin. 2002-10-03 16:22:46 +00:00
uwe 191626f788 Refactor kbd and mouse drivers so that they can use different middle 
layers.  Common middle layer shared by kbd_zs and sunkbd is moved into
the new file.  Move shared config directives to files.sun and adjust
ports' files.* accordingly.

Need this to support console/Xsun on Mr.Coffee JavaStation.

Tested on sparc, sparc64 (by martin) and sun3 (by jdc).
 2002-10-03 16:13:24 +00:00
fvdl 4eb09bb2aa Make shared IRQs for the normal PIC case work again as they did before 
the MP merge (i.e. treat them as having the lowest level of all registered
handlers in the chain for the IRQ).
 2002-10-03 15:58:56 +00:00
uwe 50de7c9595 If we don't get record drq from pnpbios, set it to play drq. 
Otherwise attaching half-duplex device will trigger panic.
NB: ad1848_isa_get_props should probably check for recdrq == -1 as well.
 2002-10-03 03:02:11 +00:00
thorpej 0c1efadb23 Include <sys/param.h> so that we get the defn of NULL. 2002-10-03 01:57:24 +00:00
thorpej dd14964387 Fix script-o. 2002-10-03 01:29:30 +00:00
jdolecek 47ae4d5c59 remove local declaration of iwm_cd and fd_cd for LKM case; it's superfluous 
and doesn't even compile after latest autoconf changes
 2002-10-02 20:35:28 +00:00
rearnsha 16c4b46282 Revert previous change. GENERIC already has an sb* at isapnp? 2002-10-02 20:33:44 +00:00
jdolecek 2bbadb1fa6 make ha_name const 2002-10-02 20:21:32 +00:00
jdolecek c802e99987 make ka_name const 2002-10-02 20:21:01 +00:00
jdolecek a38b5eed0e adapt to latest dev/ic/wi* changes 2002-10-02 20:13:48 +00:00
scw c859106ab8 Gee, this file is showing its origins... cpu_swapout() may invoke 
panic() on MIPS, but really shouldn't do the same on SH5.
 2002-10-02 16:19:59 +00:00
thorpej 4bf871a755 Add trailing ; to CFATTACH_DECL. 2002-10-02 16:02:08 +00:00
thorpej 89bf5a8f8e Add trailing ; to CFATTACH_DECL. 2002-10-02 15:52:22 +00:00
thorpej bd5bb4652b Add trailing ; to CFATTACH_DECL 2002-10-02 15:45:10 +00:00
drochner 59ec26b0d3 _C_LABEL(curproc) disappeared - have cpu_info now 
(still uniprocessor only)
 2002-10-02 15:23:31 +00:00
drochner 5a62e785ca add kvm86_incall again which got lost in the SMP merge 2002-10-02 14:54:43 +00:00
scw b13817324e Fix the initial cacheline alignment case. This gets my PCI ex(4) card 
pretty much working, at least for non-NFS use.

With NFS, it fails under pressure probably due to operand cache aliases
between KSEG0 and regular 4KB mappings elsewhere. Sigh.
 2002-10-02 14:40:27 +00:00
scw 9e1133a710 In pmap_page_is_cacheable(), add an explicit check for KSEG0 addresses. 
They don't show up in the page tables, so the default "not cacheable"
status is wrong.

This finally gets my ex(4) working on the Cayman's PCIbus.
 2002-10-02 12:19:38 +00:00