Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
120,285 Commits 606 Branches 0 Tags 3.1 GiB
c5670263e0
Commit Graph

37209 Commits

Author SHA1 Message Date
martin
43b448a711 Redo previous so that it actually works - the OF node still is called "wdc" 
but it is not the parent but the grandparent of the disk device now.
 2003-11-02 10:13:23 +00:00
he
2cd9d70a98 Make the clock device aldo depend on mc146818, to make kernels without 
the 'rtc at ebus' driver configured (e.g. for sbus machines) link.
 2003-11-02 09:33:56 +00:00
simonb
20db9285ba Kill trailing blank lines. 2003-11-02 08:29:06 +00:00
christos
38dd4ae2dc Initialize another fp instance 2003-11-02 08:27:41 +00:00
christos
6dba5c3bc3 only assign to fp when we have a valid lwp. Thanks simon 2003-11-02 08:20:48 +00:00
christos
df7d5f4716 Use siginfo_t not ksiginfo_t in the frame. Doh! 2003-11-02 08:20:09 +00:00
thorpej
dec320ebc1 Squish some recent compiler warnings. 2003-11-02 00:23:38 +00:00
reinoud
4e923603f1 Fix for PR # 23264 thanks to Mike Pumford; it is only attatching clock once 
now.
 2003-11-01 23:37:54 +00:00
tsutsui
bd0f8ee224 - Adapt to MI mk48txx(4) changes. 
- Switch rtc (M5819) routines to use MI mc146818(4) driver.

Tested on Ultra5 (mkclock_ebus), but mkclock_sbus and rtc_ebus are not tested.

XXX This file should be split into each device and each bus attachment,
XXX and moved into sys/arch/sparc64/dev or sys/dev/sun.
 2003-11-01 23:04:32 +00:00
tsutsui
deb5638bdc Adapt to MI mk48txx(4) changes. 
Tested on SS1+ (mainbus), but not tested on sun4/sun4m (obio).

XXX maybe this file should be split into mainbus and obio attachments
XXX and moved into sys/arch/sparc/dev or sys/dev/sun.
 2003-11-01 22:59:24 +00:00
tsutsui
4e8b58dd02 - Adapt mkclock_isa.c to MI mk48txx(4) changes. 
- Switch mcclock_isa.c to use MI mc146818 driver.
- Use todr(9) interface with both todclock drivers.
 2003-11-01 22:54:46 +00:00
tsutsui
2ddfa78de4 Adapt MI mk48txx(4) changes. Tested on NWS-3470D (hb) and NWS-5000X (apbus). 2003-11-01 22:50:45 +00:00
tsutsui
164022f59e Adapt MI mk48txx(4) changes. Tested on NWS-1750. 2003-11-01 22:49:38 +00:00
tsutsui
97717cc9f4 Adapt MI mk48txx(4) changes. (compiles but untested) 2003-11-01 22:48:01 +00:00
scw
f557964d0c Quell an uninitialised variable warning. 2003-11-01 20:04:57 +00:00
tsutsui
c747a93798 Count necpb interrupts with evcnt(9). 2003-11-01 19:23:52 +00:00
matt
1dc67abbb1 Compile at -O2. Fix various GCC3-isms 
XXX mem.c still fails to assemble.
 2003-11-01 18:23:37 +00:00
jdolecek
3abecdb88d avoid stong words in comments 2003-11-01 17:35:42 +00:00
he
d800494086 Based on the historic hp300 pmap, initialize firstpage to TRUE, and 
negate that after processing the first page.
 2003-11-01 17:16:30 +00:00
jdolecek
467746e275 avoid strong words; use 'screw' instead 2003-11-01 17:00:49 +00:00
he
22c1eedd0c Initialize needcflush to FALSE in pmap_do_remove(). 2003-11-01 15:32:47 +00:00
yamt
fe3e5aa0b8 as we know PTPs are zero-filled when being freed, 
mark them PG_ZERO so that they'll be used to serve UVM_PGA_ZERO requests.
 2003-11-01 15:21:02 +00:00
yamt
e36739df67 use explicit "l" suffixes. 
(xor -> xorl)
 2003-11-01 15:14:27 +00:00
tsutsui
fe1d71458c - Flush cache only if mips_sdcache_line_size == 0 in pmap_copy_page() when 
options MIPS3_L2CACHE_ABSENT is defined.
- Fix comments following #endif for MIPS3_L2CACHE_ABSENT.
 2003-11-01 14:48:16 +00:00
jdolecek
d944bca949 rename local index() to strchr() 2003-11-01 13:02:04 +00:00
jdolecek
917bc10805 g/c local index() routine and switch to (libkern's) strchr() 2003-11-01 12:56:32 +00:00
jdolecek
0ce8029b4c shutup one bogus -Wunitialized warning, and fix one correct 
XXX parrw() routine looks like not actually correct for transfers
XXX larger than sc_burst - someone with the hardware should check & fix this
 2003-11-01 12:53:33 +00:00
jdolecek
ca2d6f9331 drop register qualifiers 2003-11-01 12:43:30 +00:00
jdolecek
8064ffcedd fix several places where variables might be used uninitialized 2003-11-01 12:41:59 +00:00
jdolecek
aae2faadc5 the storagesize parameter of extent_create() is a size_t, not a pointer 2003-11-01 12:21:14 +00:00
jdolecek
5a088df3e8 mem_exists(): pacify one apparently bogus -Wuninitialized warning 
setmemrange(): add default case for cputype (same as CPU_68030), to fix
a potential uninitialized use
 2003-11-01 12:18:55 +00:00
jdolecek
b1686553ae drop register qualifiers 2003-11-01 11:50:53 +00:00
jdolecek
a51175c327 fix potential use of uninitialized local variable 'error' 2003-11-01 11:44:46 +00:00
tsutsui
ac30bfe8d4 Remove nonexistent arch/pmax/pmax/mcclock.c. (pmax uses dev/dec/mcclock.c) 2003-11-01 09:52:19 +00:00
yamt
f41e60e6a0 as there shouldn't be any mappings when pmap_destroy() is called, 
assert there is no remaining PTPs at that point.
 2003-11-01 09:07:11 +00:00
sekiya
4f800421dd Use be*toh() and htobe*() to permit use in cross-build environments. Add -h flag to override default volume header size. 2003-11-01 06:30:44 +00:00
shin
70f5a0a5b2 cache_r10k.c rev. 1.1 is broken. Because, 
1) R10k uses VA0 to select cache ways, but in rev. 1.1, VA14
	   is used instead.
	2) R10k does not support HitWriteBack and should map HitWriteBack
	   to HitWriteBackInvalidate, but in rev. 1.1, HitWriteBack is not
	   handled properly.

So, cache_r10k.c rev. 1.1 was replaced by new implementation.
 2003-11-01 04:42:56 +00:00
uwe
72b8dfe22a In hd64461video_update_videochip_status init depth to supress bogus 
-Wuninitialized.
 2003-11-01 03:45:52 +00:00
uwe
e9dc4000d1 XXX: Init dummy variable that we only use in asm() to pacify 
XXX: -Wuninitialized.  There must be a better way.
 2003-11-01 03:24:12 +00:00
uwe
289bc4802a #include "debug_kloader.h" 2003-11-01 02:29:08 +00:00
uwe
818209017c In intc_intr_priority init iprreg to pacify gcc. 2003-11-01 02:24:49 +00:00
cl
b936d40f39 Avoid sneaking past signal delivery in sa_upcall_userret(): 
generate unblocked upcalls in sa_unblock_userret(), before signal
delivery/p_userret handling in userret().

Also defer getting state for preempted upcalls because on some ports
preemption can happen between sa_unblock_userret() and sa_upcall_userret().
 2003-11-01 01:38:46 +00:00
mycroft
a4061d6dec Update for structure changes. 2003-10-31 21:49:39 +00:00
fvdl
ab3d5bdcea Catch up with the new acpica code. 2003-10-31 20:56:55 +00:00
petrov
9cc6a54d18 Don't attach zs at mainbus. 2003-10-31 20:10:06 +00:00
petrov
5a0a3beffd Attach zs to sbus only, current driver cannot attach to mainbus. 2003-10-31 20:06:54 +00:00
tsutsui
d135aa4588 In swap_endian_disklabel(), don't forget copy members which are not 
needed to byteswap.
 2003-10-31 17:04:12 +00:00
scw
b4c2d3dbb0 Insert some nops around the load instructions in badaddr_read_N(). 
That way, we can be sure any imprecise data abort actually happens
before clearing pcb_onfault.
 2003-10-31 16:54:05 +00:00
cl
ef56cc40ab Reduce code duplication by adding mi_userret() in sys/userret.h 
containing signal posting, kernel-exit handling and sa_upcall processing.

XXX the pc532, sparc, sparc64 and vax ports should have their
XXX userret() code rearranged to use this.
 2003-10-31 16:44:34 +00:00
scw
336806eb55 Overhaul arm32's abort handlers: 
- Assume a permission fault is always the result of an attempted
   write, so no need to disassemble the opcode.
   (as discussed with Richard Earnshaw/Jason Thorpe a week or two ago)

 - Split out non-MMU data aborts into separate functions, and deal
   correctly with XScale imprecise aborts. Specifically, the old code
   made no attempt to handle the double abort faults which can occur
   as a result of two consecutive external (imprecise) aborts. This
   was easy to provoke by read(2)ing from a /dev/mem offset which caused
   an external abort. With the old code, this would bring the system
   down instantly, with little clue as to why. (hint: tf_spsr held
   PSR_ABT32_MODE...)

 - Re-write badaddr_read() to use pcb_onfault instead of adding extra
   overhead to data_abort_handler(). A side effect of this is that it
   now benefits from the XScale double abort recovery.

 - Invoke the cpu-specific prefetch/data abort fixup routines only if
   the host cpu actually needs it. On other cpus, the code is optimised
   away.

 - Sprinkle __predict_{false,true} in all the right places.

 - G/C some excess debugging baggage.
 2003-10-31 16:30:15 +00:00