Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
219,053 Commits 606 Branches 0 Tags 3.1 GiB
e3971f83fc
Commit Graph

2222 Commits

Author SHA1 Message Date
kiyohara
e3971f83fc Call kmem_intr_free() with kmem_intr_alloc-ed memory instead of free(). 2013-06-09 12:52:33 +00:00
macallan
b00e7f24a9 more G5 stuff: 
- call OF_quiesce()
- properly map the message buffer
 2013-05-13 00:12:01 +00:00
macallan
ca1d0e8b1b add OF_quiesce() to shut down OF background tasks, needed on G5 
from Phileas Fogg
 2013-05-12 13:50:11 +00:00
macallan
092e2fd7b4 cleanup mem_regions: 
- use different variables for / and /memory nodes
- remove redundant OF_finddevice("/")
- assume 32bit addresses/sizes in "available" on Apple hardware
 2013-05-12 13:42:39 +00:00
macallan
bdcb52e9e3 move console_node and console_instance to ofw_consinit.c where they're 
actually used
 2013-05-12 13:23:08 +00:00
macallan
f6a41c7ccb remove PMAC_G5 kludge - fixed this elsewhere 2013-04-28 00:45:45 +00:00
macallan
76c99e815c some macppc boxes have onboard devices where parts or the whole PCI interrupt 
register is hardwired to zero - fix this up here
 2013-04-28 00:44:27 +00:00
macallan
9ab0014b38 make delay() work with _ARCH_PPC64 2013-04-25 00:11:35 +00:00
macallan
9c39c70523 we can read the timebase register as a 64bit chunk when _ARCH_PPC64 is defined, 
not necessarily LP64
 2013-04-25 00:08:48 +00:00
macallan
9902bdf86c fix off-by-one 
noticed by Phileas Fogg
 2013-04-11 19:55:10 +00:00
macallan
b321ec3e3e KNF nits 2013-04-11 19:53:51 +00:00
macallan
1a39b65774 in memio_map() don't fail without trying to mapiodev() if we don't find a BAT 
mapping or don't have BATs
now this works on G5
 2013-04-11 18:20:33 +00:00
macallan
b2ee8ed2e2 temporary kludge for G5 - on mine some devices are configured with interrupt 
pin 0 but a valid IRQ line, so don't treat that as an error if PMAC_G5 is set
 2013-04-11 18:14:40 +00:00
macallan
6b3498dcea for some reason we can't use the ROM font on G5, so disable it with 
options OFWOEA_WSCONS_NO_ROM_FONT
from Phileas Fogg
 2013-04-11 18:04:20 +00:00
macallan
8a9318ebf5 make PPC_OEA64_BRIDGE mode work 
from Phileas Fogg
 2013-04-11 17:13:15 +00:00
christos
caafbf4e1b add missing SSIZE_MIN 2013-04-11 00:57:34 +00:00
macallan
0ee8b59e60 deal with /memory "reg" property which may contain 64bit addresses on G5 2013-04-01 20:14:42 +00:00
christos
70c0b4d9e7 Fix loop bug (Phileas Fogg) 2013-03-28 00:22:21 +00:00
christos
3f9b5ba889 Explicitly set KERN_AS (this could be done in /usr/src/sys/conf/Makefile.*) 
so that modules work, on config files that did not set it explicitly. All
the files now use the standard logic, except the ones that set KERN_AS=obj
and Makefile.usermode which sets KERN_AS=library.
 2013-03-10 07:18:19 +00:00
kiyohara
ce02c3a687 Fix KASSERT(l == curlwp)ed in mi_switch(). Don't use INTSTK of cpu_info in 
INTR_PROLOG when user-mode.  Interrupted user context switches to newlwp, if
sched tick.  This context must save to USPACE.
When use our INTSTK?
 2013-02-09 11:22:51 +00:00
kiyohara
b73bbcbf2e More stack for ibm4xx slow machines. 2013-01-29 15:47:16 +00:00
kiyohara
f870b0a58c Support 16bits over stack size. 2013-01-29 15:45:43 +00:00
kiyohara
b38b935659 Set SRR1 to r31 in intr_exit. Not cpuinfo. 2013-01-29 15:33:44 +00:00
matt
7f6d9a714f Use structure initializers. 2013-01-18 06:21:12 +00:00
matt
2b8030185c Update to new pcu_state_{load,state,release} definitions 2012-12-26 19:05:03 +00:00
kiyohara
8a11b9f2d3 Change name of boot image(mount point?) to 'NetBSD' from 'fd'. 2012-12-24 14:01:24 +00:00
kiyohara
fd862c9a85 Obsolete options '-s'(standalone) on bebox. bebox can boot no-standalone image 
from FD or HDD.
 2012-12-19 13:47:20 +00:00
matt
f82647e665 Make the 85xx get closer to spinning up the secondary CPUs. 
Don't assume TLB1[0] has the mapping for VA/PA 0.
Make sure the TLB1 entries that map physical memory have the M (memory
coherent) bit set.
 2012-11-27 19:24:45 +00:00
christos
72708a99b3 provide _ENTRY(x) because some code needs it. 2012-11-25 01:10:37 +00:00
chs
5a4166a003 fix compile problems (for POWERMAC_G5). 2012-10-29 12:52:43 +00:00
matt
2aa18ae8d4 Change a KASSERT to a KASSERTMSG. 2012-10-29 05:25:19 +00:00
matt
57d76ca064 Make bus_dmamap_sync a noop for BookE. 2012-10-29 05:23:44 +00:00
chs
cbab9cadce split device_t/softc for all remaining drivers. 
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.
 2012-10-27 17:17:22 +00:00
kiyohara
515e584834 Support Cache Protocol 'MEI' with MULTIPROCESSOR. 2012-10-20 14:42:15 +00:00
kiyohara
10a93b3b38 Fix broken cpu_hatch_stack. 
And define macro HATCH_STACK_SIZE.
 2012-10-20 13:18:45 +00:00
kiyohara
e08d10659c Remove white-spaces. And remove unused 'extern void tlbia()'. 2012-10-20 13:10:44 +00:00
jdc
4be6ce2eeb Adapt to the changed signature of pckbc_cnattach(). 2012-10-13 17:58:53 +00:00
christos
e62c0b0fdf move common tlb stuff to uvm 2012-10-02 23:51:39 +00:00
mrg
d87a39703d increase powerpc NKMEMPAGES_MAX_DEFAULT to 256MB. remove the macppc 
overrides that are now the same as the powerpc default.
 2012-09-23 22:31:38 +00:00
mrg
172bbb2df1 correct some comments on #endif to match their #if. 2012-09-23 06:05:32 +00:00
matt
17674df3f2 Add support for _UC_TLSBASE. Make sure to preserve backwards compat for 
programs built before TLS support was added.
 2012-09-11 00:15:19 +00:00
matt
56d64289ed Fool gcc into thinking we aren't using r31. We do use it but we don't 
clobber it since it save and restore it from a temporary register.
 2012-09-07 18:05:11 +00:00
macallan
eb30b37d82 when calculating per-IPL virq masks, take into account that shared IRQs may 
request different IPLs
fixes vr(4)-related problems seen on ofppc
 2012-08-31 13:12:52 +00:00
he
c7664459b0 Make this build again by adding a couple of right-parens. 2012-08-14 13:02:19 +00:00
matt
2b43b3296b Deal with PCIe/PCI hoses with more than one window so the bus_space 
encompasses all of it and reserves those portions not assigned so
no one can allocate from them.
 2012-08-13 00:52:45 +00:00
matt
834f7f1974 Also supprese printing fatal traps if the process is being debugged. 2012-08-02 14:07:47 +00:00
matt
5565c9eb97 Also supprese printing fatal traps is the process is being debugged. 2012-08-02 14:06:34 +00:00
matt
331e204f59 Only print fatal trap info if the resulting signal would be uncaught. 
(this avoid the messages from libcrypto when it probes for capabilities).
 2012-08-02 13:50:14 +00:00
matt
143ca138fb Add a machine splhist command to give (a incomplete) spl history. 
(only the most recent are going to be accurate).

splraise(6) from 0 at 549214603
splraise(7) from 6 at 549214643 (+40)
splx(6) from 7 at 549214691 (+48)
splx(0) from 6 at 549214730 (+39)
 2012-08-01 21:30:21 +00:00
matt
73fa065a77 Add diagnostic code emptyidlespin verifying that the current IPL 
is IPL_NONE and EE is set in the MSR.
 2012-08-01 20:35:52 +00:00