Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
82,301 Commits 606 Branches 0 Tags 3.1 GiB
Commit Graph

82301 Commits

Author SHA1 Message Date
fredette 7a083fffc2 Don't set up the g0 and g4 handlers here, now prom_abort() 
does this.  Include the common m68k 16-bit aligned copy routines.
 2001-06-27 03:25:31 +00:00
fredette 4d85da3f79 Moved much softint and bus_space/bus_dma code 
to the common sun68k code.
 2001-06-27 03:16:02 +00:00
fredette ee7503cf71 Converted to use the new bus-specific attach args. 2001-06-27 03:00:45 +00:00
fredette a50332b3e7 Converted to use the new bus-specific attach args, 
and the bus_space_peek_N functions.
 2001-06-27 02:59:26 +00:00
fredette 19f9c211b0 Added a kernel files/devices config file covering 
the common sun68k code.
 2001-06-27 02:55:30 +00:00
fredette c0cf0f4a11 Added a Makefile to install sun68k common headers. 2001-06-27 02:53:39 +00:00
fredette fef1479cf8 Now give at least the appearance of having bus-specific 
attach args.  In reality everything is still one structure,
now called mainbus_attach_args, and a series of macros
creates the illusion.
 2001-06-27 02:53:07 +00:00
fredette 1a29a16e5a Now define splzs() here. 2001-06-27 02:51:45 +00:00
fredette 5d81b8d269 Added common sun68k autoconfiguration and bus_space 
related code.  Currently used by sun2.
 2001-06-27 02:48:32 +00:00
simonb 2de3233da6 Fixes to compile on alpha. Reported by Caffeinate The World on 
current-users.
 2001-06-27 02:46:25 +00:00
thorpej 25f00d4c18 In getnewvnode(), allocate a vnode from the pool with NOWAIT. If 
that fails, just try to recycle a vnode.  If we can't allocate or
recycle, issue a warning, sleep a bit, and try the whole thing
again.

This prevents us from blocking forever if we want to use a very large
number of vnodes, but don't have {memory,kva} resources from which to
allocate them.
 2001-06-26 22:52:03 +00:00
chris e2d29421a9 Add checks (make non DIAGNOSTIC) to make sure that we never call uvm_fault from an interrupt. 2001-06-26 22:09:19 +00:00
bjh21 25d9d85b42 seeq8005-based drivers (ea and eb) now support full-sized frames. 2001-06-26 22:01:45 +00:00
bjh21 770912bb85 Add support for handling full-size 802.1Q packets. 2001-06-26 22:00:44 +00:00
uwe 6a127dddd8 KNF: put parens around returned values. 2001-06-26 20:35:54 +00:00
uwe b4ded8ad61 Some OpenFirmware client services return -1 on error, while the rest 
of the code expects zero.  Add some wrappers so that opf promops are
consistent with obp promops.  Now microSPARC-IIep correctly panics ;)
While here, propagate 'static' to opf_* definitions.
 2001-06-26 20:21:59 +00:00
uwe 67a5f8fc0a OF_finddevice uses -1 to signal error. Make it zero to match what 
the rest of the code expects the error value to be.
 2001-06-26 19:37:54 +00:00
bjh21 e0e53bb8fe Change the DUBUG check on interrupt depth in do_fault() into a KASSERT, since 
we're never allowed to call uvm_fault from an interrupt.
 2001-06-26 19:37:18 +00:00
lukem afa0abb87f s/tise/tize/ in docco, add "advertize" as synonym for "advertise" 2001-06-26 19:30:44 +00:00
bjh21 ead93c630b Chomp out the old, unused undefined_handler(). 2001-06-26 19:22:02 +00:00
jdolecek e65c47a67f vfs_rootmountalloc: take advantage of LIST_FOREACH() 2001-06-26 19:14:25 +00:00
thorpej a279b0973b Reduce some complexity in the fault path -- Rather than maintaining 
an spl-protected "interrupt safe map" list, simply require that callers
of uvm_fault() never call us in interrupt context (MD code must make
the assertion), and check for interrupt-safe maps in uvmfault_lookup()
before we lock the map.
 2001-06-26 17:55:14 +00:00
thorpej 0910b96106 Panic if we get an MM trans/prot fault in interrupt context. 2001-06-26 17:29:28 +00:00
thorpej 78ae4127bb Note that uvm_fault() must NEVER EVER EVER be called in interrupt 
context.
 2001-06-26 17:27:31 +00:00
itojun b12a7f19d9 drop setgid. suggested by deraadt 2001-06-26 17:10:33 +00:00
thorpej a160a0b77c Make the kmemstats limits unsigned, and use unsigned arithmetic 
to compute them.  Prevents overflows when using a very large number
of kmem pages.

From chopps@netbsd.org.
 2001-06-26 16:02:27 +00:00
wrstuden 716d3ae08f In vcount(), when getting rid of unused aliases, don't vgone one which 
has VXLOCK set - it's already being vgoned, most likely by one of our
callers. If we call vgone, we can end up sleeping against ourself
with VXLOCK set - we'll start the race for root.

Pointed out by Love <lha@stacken.kth.se> on tech-kern. Analysis from
Artur Grabowski <art@openbsd.org> via Love.

Should resolve PR kern/13077
 2001-06-26 15:51:06 +00:00
bouyer 90d0e67408 Add a XS_CTL_SILENT_NODEV flag: if the sense info is "not ready, medium 
not present" don't print any message but still return ENODEV.
Use this in cd driver to allow open of character raw partition even if
the drive is empty (older drives fails at LOAD_UNLOAD time, newer
ones fail at TEST_UNIT_READY time).
 2001-06-26 15:32:02 +00:00
itojun d54eb3a39d change tcpdump dir to refer to. Daniel Carosone <dan@geek.com.au> 2001-06-26 13:43:29 +00:00
matt 6bad4f04a0 Record any stack growth. This fixes a serious problem with core dumps. 
Without this change, coredumps will only contain one page of stack
regardless of how many pages of stack the process actually has.
 2001-06-26 13:00:18 +00:00
bouyer f47d097cf5 Make it compile with SCSIPI_DEBUG 2001-06-26 12:46:56 +00:00
bouyer e27abfaddf Make it compile with SCSIPI_DEBUG 2001-06-26 12:19:49 +00:00
jdolecek 142a3390df two more 2001-06-26 09:36:40 +00:00
jdolecek a91872d1cf ntfs_readdir(): use NTFS_ROOTINO instead of ip->i_number for '..' entry 
XXX it's not correct either way
 2001-06-26 09:36:06 +00:00
jdolecek 61c37aef21 Allocate the struct dirent in ntfs_readdir() via MALLOC(), rather than 
on stack.
 2001-06-26 09:24:36 +00:00
onoe 65f0069f5b restore awi_power() to reenable awi after resume. 
... forgot why it is disabled...
 2001-06-26 08:41:19 +00:00
hubertf 2e9a29abf2 regen for MAKEDEV2manpage.pl change 2001-06-26 02:28:47 +00:00
hubertf d50569f713 Use \&foo instead of f\&oo for better readability. 
By uwe@netbsd.org.
 2001-06-26 02:28:21 +00:00
hubertf 3dcda53fea "make man man.ps" for some 5000 page book. 
Needs more work (intro page, sorting, ...), but better than before.
 2001-06-26 02:26:06 +00:00
hubertf 505430a429 Regen for foo -> f\&oo on .Xr 2001-06-26 02:11:54 +00:00
hubertf b569c48c16 Try cross-referencing our VAX and/or hp300 port's ct(4) manpage: 
.Xr ct 4

^- This does not work, as "ct" is some *roff command. Work around
it by adding a non-printing char:

	.Xr c\&t 4

(for all manpages, just in case and out of laziness)
 2001-06-26 02:03:08 +00:00
hubertf 941ea41c4a Add argument to -Bl -width - groff complains while nroff doesn't 2001-06-26 01:46:26 +00:00
enami 4655e8da67 Some man page here requires processing by tbl. 2001-06-26 01:43:58 +00:00
hubertf d8ec602681 Note: just because our macros/groff/whatever terminates .Bl internall for a 
new .Sh doesn't mean the ending .El should be omitted
 2001-06-26 00:52:59 +00:00
thorpej 93c03c9726 - Implement a real pmap_kenter_pa()/pmap_kremove(). 
- Use pools for pmap structures and pv_entry structures.
- Remove a bunch of splvm()/splx(), no longer needed now that
  pmap_kenter_pa() and pmap_kremove() are as they should be.

Mostly from Chuck Silvers.
 2001-06-26 00:31:32 +00:00
hubertf 04eb606c4b .El a .Bl 2001-06-26 00:14:32 +00:00
hubertf 864848be63 close a .Bl 2001-06-25 23:42:42 +00:00
hubertf d91e447e6b PRevent one of these: 
List open at EOF -- A .Bl directive has no matching .El
 2001-06-25 23:37:27 +00:00
chris b8ceed365f Improve the vac_me_harder function, it is now slightly faster, however pmap_enter_pv is also now slightly slower, so they appear to balance out. 
Note that I've some ideas in the works on how to improve the pv handling, so the slow down is short term only.

Also added non-advertising licence and copyright to myself and richard.
 2001-06-25 23:22:38 +00:00
eeh 55e676fb15 Remove another unused file. 2001-06-25 23:07:03 +00:00