Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
121,412 Commits 606 Branches 0 Tags 3.1 GiB
13f8d2ce5f
Commit Graph

1117 Commits

Author SHA1 Message Date
nathanw
4d59420344 Move __cpu_simple_lock_t and __SIMPLELOCK_{UN,}LOCKED to machine/types.h 
so that they can be used in a namespace-friendly way.
 2003-09-26 22:45:41 +00:00
matt
6986092cfe Add a machine-dependent SIGTRAMP_VALID macro which is used to test whether 
a trampoline version is valid or not.
 2003-09-26 22:14:19 +00:00
matt
fb54d529b8 Copy out the entire ksiginfo_t, not just the first word. 2003-09-26 18:00:47 +00:00
eeh
a6192488f7 Define ksi. 2003-09-26 00:00:17 +00:00
matt
2deb8b2e7b Deal with the constification of ksiginfo_t and sigset_t in signalling. 2003-09-25 22:22:36 +00:00
christos
ecfb034cb3 constify siginfo/trapsignal 2003-09-25 22:04:17 +00:00
matt
9a305fca59 Use direct entry for siginfo trampoline. Deal with missing 
legacy trampolines in a cleaner manner.
 2003-09-25 21:59:55 +00:00
matt
83003b0798 Fill in ksi_code. 2003-09-25 21:58:46 +00:00
matt
85a5f3065d Add siginfo support for PowerPC. 2003-09-25 18:42:18 +00:00
matt
49317aa149 Remove db_onpanic check. 2003-09-25 17:36:33 +00:00
shige
95fa35eb8e Delete this file because filename includes uppercase letters. 2003-09-24 11:51:19 +00:00
shige
0a36d23137 Copy IBM4xxGPx specific autoconf code from evbppc/walnut/autoconf.c. 2003-09-24 11:47:02 +00:00
matt
9c8a5009b3 Define va_list as __builtin_va_list for GCC 3.x. Change stdarg macros 
appropriately.  (this is committed from a system run a kernel and userland
built with these changes).
 2003-09-24 02:39:56 +00:00
shige
dc98452dce Copy PCI codes for IBM405GPx from evbppc/walnut/pci/{pchb.c,pci_machdep.c}. 2003-09-23 15:30:22 +00:00
shige
03467d74b5 Add devices info for IBM405GPR. 2003-09-23 15:26:46 +00:00
shige
91be153592 Add global cpu_model variable. 2003-09-23 15:25:26 +00:00
shige
e20e959b37 Copy IBM4xxGPx specific autoconf code from evbppc/walnut/autoconf.c. 2003-09-23 15:24:15 +00:00
shige
a8dd07e005 Copy IBM4XX OpenBIOS locore.S from evbppc/walnut. 2003-09-23 15:21:58 +00:00
shige
fc29eb23bf Add some PCI definitions listed in evbppc/include/walnut.h. 2003-09-23 15:19:05 +00:00
shige
2977d5a89a Add IBM40x specific machdep functions. 2003-09-23 15:14:02 +00:00
shige
87f629a7b8 Add IBM405GPr PVR. 2003-09-23 15:10:05 +00:00
shige
94959351e8 Add IBM405GPx specific kernel configuration file. 2003-09-23 15:06:40 +00:00
shige
ee990c494d Move todclock driver from evbppc/walnut/dev. 2003-09-23 15:01:50 +00:00
shige
ff6df75f63 Add on-chip IIC driver. 
Add some definitions for IIC driver.
 2003-09-23 14:56:08 +00:00
cl
b826d7b216 add MD part of SA/pthread pagefault handling on powerpc 2003-09-19 00:16:34 +00:00
kleink
8d633905ac Bring /* LINTSTUBs */ closer to reality. 2003-09-08 22:37:01 +00:00
kleink
a4909cd01e g/c byte-reversing bus_space_set_region_1() entries. 2003-09-07 22:12:05 +00:00
manu
d13828a7dd forgotten commit for KERN_PROCARGS sysctl in COMPAT_DARWIN 2003-09-07 07:50:31 +00:00
matt
cec9cb32df Move CLKF_BASEPRI to machine specific <intr.h> file since it depends on 
the encoding of the spl for the port.
 2003-09-03 21:33:31 +00:00
simonb
05c33c1c05 Need to set BUS_DMA_COHERENT to BUS_DMA_NOCACHE on IBM ppc4xx CPUs. 
The "emac" interface now works properly again on my Walnut.  Much
thanks to Steve Woodford for tracking this down.
 2003-09-03 13:30:05 +00:00
matt
31cc6ab700 Move SFRAMELEN to frame.h and use it in vm_machdep.c. In setfunc, setup 
callframe linkages correctly.  Restore use of ldptr to locore_subr.S
[pthreads and gdb no longer crash/hang the system]
 2003-08-27 20:20:07 +00:00
matt
56f115e896 Switch back to adjusting stack by addi instead of ldptr until I can locate 
who's setting up the stack wrong.
 2003-08-27 19:45:28 +00:00
chs
939df36e55 add support for non-executable mappings (where the hardware allows this) 
and make the stack and heap non-executable by default.  the changes
fall into two basic catagories:

 - pmap and trap-handler changes.  these are all MD:
   = alpha: we already track per-page execute permission with the (software)
	PG_EXEC bit, so just have the trap handler pay attention to it.
   = i386: use a new GDT segment for %cs for processes that have no
	executable mappings above a certain threshold (currently the
	bottom of the stack).  track per-page execute permission with
	the last unused PTE bit.
   = powerpc/ibm4xx: just use the hardware exec bit.
   = powerpc/oea: we already track per-page exec bits, but the hardware only
	implements non-exec mappings at the segment level.  so track the
	number of executable mappings in each segment and turn on the no-exec
	segment bit iff the count is 0.  adjust the trap handler to deal.
   = sparc (sun4m): fix our use of the hardware protection bits.
	fix the trap handler to recognize text faults.
   = sparc64: split the existing unified TSB into data and instruction TSBs,
	and only load TTEs into the appropriate TSB(s) for the permissions.
	fix the trap handler to check for execute permission.
   = not yet implemented: amd64, hppa, sh5

 - changes in all the emulations that put a signal trampoline on the stack.
   instead, we now put the trampoline into a uvm_aobj and map that into
   the process separately.

originally from openbsd, adapted for netbsd by me.
 2003-08-24 17:52:28 +00:00
chs
3cf764618d remove bogus code that terminates stack trace too early on OEA. 2003-08-24 16:33:41 +00:00
matt
121c8942a6 Not all PPC ports (e.g. macppc) equate IPL_NONE with 0. 2003-08-18 22:10:33 +00:00
chs
e02e4ecc2b lots-o-stuff: 
- fix opcodes_base[10] to allow us to find "cmpli".
 - fix "cmp" definition.
 - add various missing SPRs.
 - use the right bits for the "BI" operand.
 - fix operand calculation for "rlwinm" and friends, and display
   the operands in the same order as the manuals do.
 - make the buffer bigger to avoid overflow.
 2003-08-17 18:23:17 +00:00
chs
4ba6255f14 add some 745x-specific MSSCR0 bits. 2003-08-17 18:08:17 +00:00
chs
f19809f0fa remove an unused proto. 2003-08-17 18:07:11 +00:00
matt
1d1029332d Remove this. The architecture specific ones are now used. 2003-08-12 18:35:51 +00:00
matt
47483092bd Cleanup/rework cpu_switch*, switch_exit, Idle routine. Remove pcb_psl 
since it was write-only.  When setting up a process, make sure the fake
callframes are properly linked together.

Only lower SPL when in Idle loop.  Raise spl to previous level (which would
be IPL_SCHED) when exiting Idle loop.  Never lower SPL anyplace else.
 2003-08-12 18:34:47 +00:00
matt
b986a8c215 D'oh! The PCB can't be accessed with the MMU off. So get the pmap pointer 
*before* disabling the MMU.
 2003-08-12 15:40:02 +00:00
matt
a3a18840be Remove SPILLSTK leftovers. 2003-08-12 05:15:41 +00:00
matt
c0d6cb285d Nuke ci_curpm and curpm. Nuke pcb_pmreal. Those were use for spill stacks 
and those no longer exist.  for few uses that need CURPM, use CURPCB/PCB_PM
 2003-08-12 05:06:53 +00:00
matt
3527dc3950 When getting the lwp to switch to, test to see if the scheduling queue is 
empty, and if so panic.  Change references to "proc/process" to lwp.
 2003-08-11 15:39:24 +00:00
chs
b90614b54e catch up with changes elsewhere. 2003-08-11 05:13:20 +00:00
matt
341742a7e9 Add PTEGCOUNT defparam for opt_pmap.h 2003-08-11 01:33:30 +00:00
matt
3d5b7190ad Nuke ci_spillstk/CI_SPILLSTK. No longer needed. 2003-08-08 07:14:26 +00:00
matt
add426be7e Add a type for trapstart so ddb will use it. 2003-08-08 06:11:48 +00:00
matt
938aca290c Don't try to spill entries from the kernel's pmap. As of now, they can no 
longer be evicted, only user-mappings can be evicted.
 2003-08-08 06:10:43 +00:00
matt
330dc2e202 Allow only user-mappings to be evicted (spillage). This prevents the 
dreaded eviction of a kernel stack page.
 2003-08-08 06:06:48 +00:00
agc
aad01611e7 Move UCB-licensed code from 4-clause to 3-clause licence. 
Patches provided by Joel Baker in PR 22364, verified by myself.
 2003-08-07 16:26:28 +00:00
matt
c60f393ec5 Don't overwrite the value of DAR in r30. Use r31 instead. 2003-08-06 07:18:56 +00:00
simonb
20a37f9627 Note that the used interrupt bits are 405GP specific. 
Fix a tyop.
 2003-08-05 02:10:31 +00:00
matt
9254350d6a Change bc x,y,z to their simplified mnemonics. During a kernel DSI fault, 
if the exception address is < 1 page away from the KSP, switch to the that
CPU's spill stack to handle the trap.  Otherwise you can get in a infinite
DSI fault loop.
 2003-08-04 22:29:59 +00:00
matt
0e50e47bb9 Make that OEA based kernels can properly deal with kernel ISI faults. Now 
that LKMs are supported, it is possible for a LKM page to be "outspilled"
resulting in a possible ISI fault.  Try to spill the page back in.
 2003-08-04 22:26:59 +00:00
matt
9b7d071bab Eliminate stmw/lmw substituting the individual load/store instructions. 
Use more symbolic constants.  These are now safe for use on PPC64.
 2003-08-04 00:32:49 +00:00
matt
a4a468e215 Add SFRAME_USER_SR (even though the 4XX doesn't have one). 2003-08-04 00:30:51 +00:00
eeh
f77f1feee5 Use %r<n> for register names. 2003-08-03 23:26:55 +00:00
matt
777589be9a Add each register in trapframe, switchframe, and faultbuf. 2003-08-03 21:40:13 +00:00
matt
e3e80aa5d7 Add PPC_OEA64 and netbsd32_sigcode.S 2003-08-03 21:38:08 +00:00
matt
618c78b771 Don't use stmw. Do each store individually. 2003-08-03 21:27:49 +00:00
matt
42382e46c4 Add CPUSAVE_* and each register in the saveframe and faultbuf. 2003-08-03 21:26:27 +00:00
matt
dd1c661661 Nuke stmreg/ldmreg. PPC64 doesn't have a lmd/stmd so make sure lmw/stmw 
don't invoke valid instructions on PPC64.
 2003-08-02 19:40:39 +00:00
matt
c9d56ac39d Add symbolic offsets for what's in cpu save locations. 
Add a PPC64 variant of mftb
 2003-08-02 19:35:26 +00:00
matt
d89b685a6e Switch to regiser prefixes. Also change numeric constants to symbolic ones. 2003-07-31 15:30:41 +00:00
matt
f9c46681fd Change switchframe to use register_t (anything that uses stmX/lmX needs 
to be defined as register_t).
 2003-07-31 15:29:29 +00:00
matt
18a1f8d4c7 Switch to m[tf]sprg[0-3] 2003-07-31 14:02:22 +00:00
matt
52bfbfda86 Use register prefixes and load/store pseudo-instructions. 2003-07-31 13:59:54 +00:00
matt
478364a5c5 Convert to the load/store pseudo-instructions. 2003-07-31 08:04:21 +00:00
matt
e51e13713a Use symbolic register names (%rN and %crN) throughout. Change a few more 
numeric constants to their symbolic counterparts.
 2003-07-31 07:51:16 +00:00
matt
82ebd07f24 Convert most numeric constants to their symbolic equivalents. (step 2 of 
cleaning up this file).
 2003-07-31 06:49:32 +00:00
matt
f5444cea2f Define SZREG {4,8} appropriately. Add pseudo-instructions (via #define) 
to load/store int, long, pointer, register, multiple registers.  This is so
assembly files can be support IPL32 and LP64 PowerPC implementations.
 2003-07-31 06:23:55 +00:00
matt
0039d6a257 Replace 'm[tf]sprg x,y' with the appropriate 'm[tf]sprg[0-3] r' (this makes the 
SPRG used unambiguous).  This causes no change in the generated object.
 2003-07-31 06:21:09 +00:00
matt
18eb53cd22 add PSL_TGPR (for MPC603) 2003-07-31 01:25:38 +00:00
simonb
14fc7f3334 Whitespace alignment nits. 2003-07-27 23:45:44 +00:00
scw
f0f724e255 Fix the default bus dma tag. 2003-07-25 10:50:13 +00:00
scw
728102e66c Switch ibm4xx over to using the more flexible powerpc bus_space/bus_dma code. 2003-07-25 10:12:42 +00:00
hannken
879ba21504 Typo: __HAVE_BITENDIAN_BITOPS -> __HAVE_BIGENDIAN_BITOPS 2003-07-21 16:10:50 +00:00
simonb
9853da3d35 Remove an unused variable. 2003-07-19 08:20:22 +00:00
matt
6783111f9f Elimindate MD setrunqueue/remrunqueue (which were out-of-date compared to 
the canonical versions in kern_synch.c).  Define __HAVE_BITENDIAN_BITOPS
so the canonical versions will be used but will store priorities in the
desired (MSB) order for PowerPC (which allows the use of the cntlzw (count
leading zeroes, word) instruction in locore_subr.S to find the proper
priority).
 2003-07-18 01:08:11 +00:00
matt
5819fb160d Elimindate MD setrunqueue/remrunqueue (which were out-of-date compared to 
the canonical versions in kern_synch.c).  Define __HAVE_BIGENDIAN_BITOPS
so the canonical versions will be used but will store priorities in the
desired (MSB) order for PowerPC (which allows the use of the cntlzw (count
leading zeroes, word) instruction in locore_subr.S to find the proper
priority).
 2003-07-18 01:08:10 +00:00
simonb
d854b5c002 SZ_BOARD_CFG_DATA is not used anywhere, remove it. 2003-07-16 03:52:31 +00:00
lukem
ed51729135 __KERNEL_RCSID() 2003-07-15 02:54:31 +00:00
simonb
649e4ff2d5 Check return value of prop_get() correctly; serial ports now attach on 
a walnut, but don't seem to accept input(?!).
 2003-07-14 05:21:25 +00:00
matt
2a6c2aa546 Back out rev 1.19. It's wrong. Add comments so it doesn't happen again. 2003-07-10 04:18:01 +00:00
matt
18a43bbe99 Don't do vtophys on instruction addresses. This would cause problems for 
user addresses (think trap from user mode) or lkms).
 2003-07-10 02:06:11 +00:00
matt
554b4aa05f Enhance db_trace to understand syscalls and print the syscall number 
that the user requested.  For example:

0xd5c56f40: SC trap #240 by 0x15668c60: srr1=0xd032
            r1=0xffffe470 cr=0x44000045 xer=0 ctr=0xeff27ab8
 2003-07-09 22:51:50 +00:00
thorpej
d00b22d0e7 Consult the "mac-addr" property associated with the emac device in 
the dev_propdb rather than referencing board_data.
 2003-07-04 02:34:47 +00:00
thorpej
5b7c5eadea Consult the "frequency" property associated with the device in the 
dev_propdb, rather than using "board_info".
 2003-07-04 02:21:02 +00:00
scw
b8b2b1d895 Fix resident page accounting for the kernel pmap. 2003-07-03 13:18:42 +00:00
fvdl
d5aece61d6 Back out the lwp/ktrace changes. They contained a lot of colateral damage, 
and need to be examined and discussed more.
 2003-06-29 22:28:00 +00:00
darrenr
257443876f 'struct proc *' -> 'struct lwp *' as required to get GENERIC for macppc built 2003-06-29 11:02:21 +00:00
simonb
ccbfec4208 Sprinkle some KNF. 2003-06-28 14:32:02 +00:00
darrenr
960df3c8d1 Pass lwp pointers throughtout the kernel, as required, so that the lwpid can 
be inserted into ktrace records.  The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V
 2003-06-28 14:20:43 +00:00
martin
d505b18964 Make sure to include opt_foo.h if a defflag option FOO is used. 2003-06-23 11:00:59 +00:00
thorpej
452a8fdae2 Rename IPL_IMP -> IPL_VM. 2003-06-16 20:00:56 +00:00
fvdl
7dd7f8baa2 Handle 64bit DMA addresses on PCI for platforms that can (currently only 
enabled on amd64). Add a dmat64 field to various PCI attach structures,
and pass it down where needed. Implement a simple new function called
pci_dma64_available(pa) to test if 64bit DMA addresses may be used.
This returns 1 iff _PCI_HAVE_DMA64 is defined in <machine/pci_machdep.h>,
and there is more than 4G of memory.
 2003-06-15 23:08:53 +00:00
simonb
67417f64a8 Sprinkle some TABs to line up the columns in the cputab models array. 2003-06-13 04:29:39 +00:00
msaitoh
6f197c635e Add support IBM 405GPr 2003-06-13 04:05:26 +00:00
scw
e05ce46e3e Oops, delete an accidentally committed local change. 2003-06-12 08:47:21 +00:00