Commit Graph

2448 Commits

Author SHA1 Message Date
tsutsui
99221a563b Use TAILQ_FIRST(3) and TAILQ_NEXT(3) macro. 2007-12-05 12:31:25 +00:00
ad
4b293a84e1 Interrupt handling changes, in discussion since February:
- Reduce available SPL levels for hardware devices to none, vm, sched, high.
- Acquire kernel_lock only for interrupts at IPL_VM.
- Implement threaded soft interrupts.
2007-12-03 15:33:00 +00:00
rafal
7b4886ba60 Clear interrupts as they are established, rearrange interrupt-mapping code
to make sure PCI slot 0 interrupt doesn't accidentally match the (unused)
PCI error interrupt.

Fixes kern/25462 at least on my Netra T1 ac200.
From OpenBSD; ok martin@
2007-12-03 14:13:18 +00:00
kiyohara
a1297d6f3f Add bcsp(4). 2007-12-02 05:19:11 +00:00
he
ddc4103ec7 Allow this to build both for sparc64 and for the GENERIC_SUN4U sparc kernel. 2007-11-28 22:54:46 +00:00
jnemeth
f3ce6c9275 don't include sys/intr.h twice, why aren't the includes sorted? 2007-11-28 20:41:35 +00:00
ad
1a82318781 Use the softint API. 2007-11-28 18:03:56 +00:00
jnemeth
fd5cef8508 Sync with src/sys/sparc/dev/fd.c change to use softint(9) API.
XXX Please keep these two in sync as this one is a drop-in replacement
for the sparc version and the idea is to have both sparc and sparc64
use the same file one day.
2007-11-28 12:44:13 +00:00
yamt
38d5e34116 make kmutex_t and krwlock_t smaller by killing lock id.
ok'ed by Andrew Doran.
2007-11-21 10:19:06 +00:00
macallan
bc546cac16 don't bother looking for console-specific properties on devices that aren't
PCI_CLASS_DISPLAY
2007-11-20 17:06:18 +00:00
ad
dc26833bb6 - Factor out too many copies of the same bit of tty code.
- Fix another tty signalling/wakeup problem.
2007-11-19 18:51:36 +00:00
macallan
09f5c97a83 add genfb at pci
This should work with all firmware-supported PCI framebuffers but so far it's
only been tested with a PGX
2007-11-19 03:43:28 +00:00
macallan
56fd80e18a pull in ofw_subr.c for some convenience OpenFirmware routines 2007-11-19 03:26:04 +00:00
macallan
3cb71681aa do as macppc does and pass a bunch of parameters to PCI console framebuffers
With this we no longer need machine-dependent code in framebuffer console
drivers and we can use genfb at pci for otherwise unsupported graphics
chips as long as there's firmware support.
2007-11-19 02:52:56 +00:00
macallan
f8e81531fa put the physical address of each psycho's memory range into a device
property so we can look it up in device_register() later on
ok'ed by martin
2007-11-19 02:45:50 +00:00
xtraeme
31962fc6e5 Extend the envsys2 API (one more time, sorry) as defined in:
http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.
2007-11-16 08:00:11 +00:00
dsl
c26c7cc378 Constants for system call number now start NETBSD32_SYS_
netbsd32 for sprac64 builds again.
2007-11-10 19:01:28 +00:00
ad
f2c57d8566 Call zs_lock_init() to set up the chanstate's lock. 2007-11-09 00:05:04 +00:00
ad
d37935697b Merge tty changes from the vmlocking branch. 2007-11-07 15:56:11 +00:00
ad
bd6663fc4d Don't set l_usrpri / spc_curpriority here. mi_userret() does it. 2007-11-05 20:37:48 +00:00
martin
b050fccbb2 Extend the sparc64 cpu kcore segment so that it records all kernel mappings
done via locked 4 MB pages. The old format could only record physically
continous kernel text mappings.
For compatibility reasons, just extend the structure and fill in all the
old fields as well.
2007-11-05 00:43:41 +00:00
martin
68aa3a64f4 cosmetics: use the "extern char ..[]" idiom for linker supplied addresses. 2007-11-04 14:32:42 +00:00
ad
bccf777b72 Make ras_lookup() lockless. 2007-10-24 14:50:38 +00:00
martin
18a5eb2b2a Remove an unused .globl declaration 2007-10-22 01:43:39 +00:00
dogcow
4f2396661b It's always nice to prototype functions before using them. 2007-10-19 08:50:07 +00:00
joerg
d238692c3e Initialise the callbacks for tty.t_rstrt_ch in ttymalloc
as all drivers but Sun/SPARC's kd.c use the same arguments.
Separate callout_reset into callout_schedule and the initial
callout_setfunc using that.
2007-10-18 18:54:56 +00:00
garbled
d974db0ada Merge the ppcoea-renovation branch to HEAD.
This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree.  Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches.  The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.
2007-10-17 19:52:51 +00:00
martin
09fda47045 Waste a bit of kernel VA space (who cares?) and speed up crash dumps
significantly by writing larger chunks with one (polled) command.
Based on work and testing done by Chris Ross.
2007-10-09 23:35:18 +00:00
ad
36a1712707 Merge run time accounting changes from the vmlocking branch. These make
the LWP "start time" per-thread instead of per-CPU.
2007-10-08 20:06:17 +00:00
ad
2af68666da Merge disk init changes from the vmlocking branch. These seperate init /
destroy of 'struct disk' from attach / detach.
2007-10-08 16:41:05 +00:00
martin
464d49ce18 printf -> DPRINTF 2007-10-01 14:18:52 +00:00
martin
da0f1debda Turn a few printf into DPRINTF 2007-10-01 14:17:34 +00:00
martin
594ab8c151 No need to cast the size argument of the dump function to "int" - make
it size_t instead. Pointed out by Chris Ross.
2007-10-01 09:15:46 +00:00
martin
7c4a68e67a Fix printf format. 2007-10-01 09:13:51 +00:00
martin
2e57b74770 On second thought: a valid mapping is a valid mapping, period.
No need to special case the kernel pmap here.
2007-10-01 08:53:35 +00:00
martin
ed85693634 Fix printf format string for 32bit builds - pointed out by Kurt S.
on current-users.
2007-09-30 23:24:56 +00:00
martin
898738b834 Cosmetics - improve the countdown while doing a crashdump.
XXX - need to add an ascii version of the game of life here someday.
2007-09-30 13:56:02 +00:00
martin
8841effc73 Add support to pmap_extract() for kernal mappings with PA 0. 2007-09-30 13:54:00 +00:00
martin
7cd933d009 When calculating memory sizes and related values for kernel dumps,
consistently use uint64_t. Fixes crash dumps on machines with > 2GB
memory. Found by Chris Ross.
While there, remove a #if 0'd part of code that prevented the first
physical memory page to be dumped - I'll solve that differently.
2007-09-30 12:06:14 +00:00
martin
0ceb0c0edd Cosmetic changes in ddb online help strings. 2007-09-24 08:20:16 +00:00
mrg
8a76c90ed6 add mskc/msk & skc/sk network drivers.
reported to work by Chris Ross on port-sparc64.
2007-09-24 06:14:03 +00:00
martin
d405af593c add DDB_VERBOSE_HELP 2007-09-24 00:12:26 +00:00
martin
66eb3cf2ac The "mach dtlb" and "mach itlb" had two ways to produce the same output
(depending on an address given or not) - simplify to always use the
locore.s version.
2007-09-23 18:55:12 +00:00
martin
e2240a6d26 Add verbose help for ddb mach commands. 2007-09-23 14:01:15 +00:00
martin
b2e87907f2 Remove the driver specific "mach esp" command from the "mach" table.
We can do that better now.
2007-09-22 23:30:55 +00:00
martin
d9a407bba2 Add a new option DDB_VERBOSE_HELP that adds online help to ddb.
From Adam Hamsik.
Minor modifications by me, all bugs are probably mine.
2007-09-22 18:40:20 +00:00
dsl
8b937dec65 Fix previous, 'long long' is 8 byte aligned on sparc32 2007-09-16 22:44:31 +00:00
dsl
06b0a1bdab Define netbsd32_uint64 for 64bit integers with the alignment requirement
of the corresponding 32bit architecture.
Use it for the 64bit items in netbsd32_statvfs so that the structure
doesn't collect 8byte alignment (and 4 bytes of trailing padding).
This replaces the 'packed' attribute which wasn't architecture specific
and would cause massive overheads accessing every member of sparc64.
Should allow the MIPS64 port do DTRT.
2007-09-16 22:35:01 +00:00
martin
913f43886d Cleanup cpu_info: get rid of ci_number and ci_upaid, use ci_index
and ci_cpuid instead.
2007-09-11 16:00:05 +00:00
martin
e680b6ed1a Make cpufrequency and friends per cpu values.
Prepare a hz tick interrupt on secondary CPUs via %tick, but do not
enable it yet, as it breaks ddb.
2007-09-09 22:37:39 +00:00