CVS log for src/sys/arch/macppc/macppc/locore.S

Up to [cvs.NetBSD.org] / src / sys / arch / macppc / macppc

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

add comment to explain why we zero SPRG0 here
NFC

#ifdef DEBUG the previous change, it seems to cause trouble on qemu which
needs further investigation

Explicitly zero SPRG0 before doing anything else - it's used to store a pointer
to struct cpuinfo, this way we can catch attempts to use it before it's set
up properly

Sync with HEAD.

Declare oeacpufeat once, in powerpc/oea/cpu_subr.c, rather than in
N different locore.S files.

Document the register state on entry per the OpenFirmware PowerPC CPU
bindings.  Ensure we save off %r6 and %r7 before calling any other
functions.

Improve readability of this file by adding register prefixes.

Call cpu_model_init() after clearing the MSR.

remove unused/useless #ifdef block

Mostly merge changes from HEAD upto 20200411

don't zero the MSR when building with options FIRMWORKSBUGS
with this a FIRMWORKSBUGS kernel boots on my TiBook
first step to address PR2231

Sync with HEAD

Sync with HEAD

Retire ipkdb entirely. The option was removed from the config files
yesterday.

ok kamil christos

update from HEAD

Pull up following revision(s) (requested by chs in ticket #291):
	sys/arch/ofppc/ofppc/locore.S: revision 1.49
	sys/arch/macppc/macppc/locore.S: revision 1.72
in the initial asm code, save the interesting caller-saved registers
(containing "arg" and "len") across function calls to C code.
fixes kernels built with gcc 4.8.

in the initial asm code, save the interesting caller-saved registers
(containing "arg" and "len") across function calls to C code.
fixes kernels built with gcc 4.8.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

resync from head

more OEA64_BRIDGE support
from Phileas Fogg

Catchup with rmind-uvmplock merge.

Don't include *pmap.h in assembly files.  Get the needed definitions
from "assym.h".

Make a common genassym.cf (powerpc/genassym.cf) and then
only put specific stuff in foo/genassym.cf

Make __HAVE_FAST_SOFTINTS appear in assym.h so all oeas don't need to include
<powerpc/intr.h>
XXX <powerpc/intr.h> should be <powerpc/oea/intr.h>

we need powerpc/intr.h n order to see __HAVE_FAST_SOFTINTS

Sync with HEAD.

sync with head

Kill _NOREGNAMES.  Everything should be using %rX now.  If it doesn't it
soon will.

Deal with split <spr.h> (and a few other misc fixes).

Sync with HEAD.

sync with head

Adapt to <spr.h> breakup.

Kill proc0paddr.  Use lwp0.l_addr instead.

Sync with HEAD.

Commit is split, to avoid a "too many arguments" protocol error.

sync with head.

Sync with HEAD.

Use "defopt MODULAR" in sys/conf/files, and #include "opt_modular.h"
in all kernel sources that use the MODULAR option.
Proposed in tech-kern on 18 Jan 2009.

Sync with HEAD.

Sync with HEAD.

Update haad-dm branch to haad-dm-base2.

actually call cpu_model_init()
This might be enough to get 601-based boxes going

LKM -> MODULAR

sync with HEAD

Sync with HEAD.

sync with head.

Because macppc shares code with ofppc, it needs oeacpufeat defined.
Other OEA ports do not yet make use of this code, so they don't require it.

sync with HEAD

sync with head.

Sync with HEAD.

Follow the merge of pmap.c on i386 and amd64 and move
pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup
code to restore CR4 before jumping back into kernel space as the large
page option might cover that.

Sync with HEAD.

Sync with head.

sync with head.

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.

Move a bunch of the macppc SMP code out of macppc, and down into the
generic OEA code.  Add a set of md callbacks into these shared routines,
that any oeappc SMP machine needs to provide.  This allows for generally
shared SMP startup code, but still allows for machine-specific
differences in the setup and kicking of the new CPU.

Convert macppc to this new layout.  Tested by Macallan.

Add an initial attempt at SMP to prep.  Tested on UP machine only,
untested on SMP yet.

Sync with HEAD

sync with head.

Sync with HEAD.

Sync with HEAD.

Sync with HEAD.

sync with HEAD

Make MP kernels build again.  Heck, if I'm lucky, maybe these changes will
even make one boot.

Convert macppc to powerpc bus_space.  Lots of work here from Matt Thomas,
as well as the common ofwoea code from myself.  Compile tested only,
still probably needs some fine tuning.

Also in this commit:
Convert macppc to new shared ofwoea routines.
Lots of KNF.

sync with head.

sync with head

sync with head.

Pull up following revision(s) (requested by sanjayl in ticket #72):
	sys/arch/macppc/macppc/locore.S: revision 1.57
	sys/arch/macppc/macppc/failsafe.c: revision 1.2
the following address a couple of issues with the G5 port.
(1) failsafe.c  allows character input from the failsafe
console device. This is the early console device that is used for
output before
the pmap layer is initialized.
(1) locore.S: make sure that the size of the cacheline for
dcbz is set to the default of 128 bytes. It also sets up other mode
bits on
the 970.  This fix allows the G5 to boot into full multi user mode
using a
NFS mounted root file system.

Pull up following revision(s) (requested by sanjayl in ticket #71):
	sys/arch/macppc/macppc/locore.S: revision 1.56
G5/970 updates:
Setup cache bits in HID5 early on so that dcbz/icbi use the
128 byte cache line size.  The mismatch was causing dynamically linked
apps to randomly coredump.
With this fix, the G5 can boot into full multi-user by using a NFS
mounted root file system.

the following address a couple of issues with the G5 port.

(1) failsafe.c  allows character input from the failsafe
console device. This is the early console device that is used for output before
the pmap layer is initialized.

(1) locore.S: make sure that the size of the cacheline for
dcbz is set to the default of 128 bytes. It also sets up other mode bits on
the 970.  This fix allows the G5 to boot into full multi user mode using a
NFS mounted root file system.

G5/970 updates:
Setup cache bits in HID5 early on so that dcbz/icbi use the
128 byte cache line size.  The mismatch was causing dynamically linked
apps to randomly coredump.

With this fix, the G5 can boot into full multi-user by using a NFS
mounted root file system.

sync with head

1st cut of Powermac G5 support (uses bridge mode).

merge ktrace-lwp.

Sync with HEAD. Here we go again...

remove vestigal {,e}intr{names,cnt} symbols now that vmstat doesn't need them.

Fix the sync with head I botched.

Sync with HEAD.

Sync with HEAD

Preserve kernel symbol table not only #if defined(DDB) but
also #if NKSYMS > 0 || defined(LKM).
Should fix port-macppc/21742.

Add #define _NOREGNAMES (to suppress asm.h from #define r1 1 ...)

Rename PPC_MPC6XX to PPC_OEA (and any mpc6xx reference to oea).

More cleanup for OLDPMAP.

Perform a rototill of the powerpc code.  Mandate use of SPRG0 to store
a pointer to current cpu's cpu_info structure.  Use cpu_info for
intstk,intr_depth,still_stk,idle_pcb,curpcb,curlwp,etal even on
non-MULTIPROCESSOR machines.  Add common macros GET_CPUINFO and
INIT_CPUINFO to get and initialize the cpu_info struct on startup.  Make
ibm4xx use the standard <powerpc/frame.h>.  Use IFRAME_xx in ibm4xx
trap_subr.S instead of explicit magic offsets.  Move INTSTK and SPILLSTK
to std.<platform>.  Change faultbuf to a struct instead of an array.

On MPC6XX cpus, stop using the vector page for temporary space and use
reserved space in cpu_info.

sync kqueue branch with HEAD

catch up with -current.

Catch up to -current.

terminate the MP idle stack for the boot CPU.

catch up with -current.

put the OF buffer page in the BSS rather than allocate it at runtime.
from PR 14497.

Common OFW code has been moved to powerpc/mpc6xx/ofw_subr.S.  Now use it.

Catch up to -current.
(CVS: It's not just a program. It's an adventure!)

Catch up with -current.

switch all mpc6xx powerpc ports to NEWPMAP by default.
the old pmap is still available with the OLDPMAP option.

Catch up to -current.

More catchup.

Sync kqueue branch with -current.

Catch up to -current.

add support for kgdb over zs

save and restore sprg[0-3] when calling openfirmware
this fixes calling OF_finddevice() with invalid devices,
and as a result fixes booting with -d.
Apparently openfirmware expects persistence of sprg2 and/or sprg3
I suspect this is used when handling exceptions

file locore.S was added on branch nathanw_sa on 2002-01-08 00:26:12 +0000

#include <powerpc/spr.h> and SPR_PIR in place of 1023

Catch up to -current.

Make the new pmap optional.  Use the old (stable!) pmap by default.

Changes new pmap, common param.h, vmparam.h, and moved includes.

Catch up with -current.

Sync with HEAD.

Merge my MP tree.  At this point, Daystar dual 604 card (i.e. Power Macintosh
9500/180MP and 9600/200MP) runs at least into single-user mode, possibly
multi-user mode (not stable yet).

Apply patch, revision 1.2 from powerpc/locore_subr.S (requested by tsubai):
  Fix a bug in cpu_switch().

Sync with HEAD.

Include trap_subr_mp.S if MULTIPROCESSOR is defined.

Use common switch / setfault / trampoline code.  (except amigappc because
it doesn't use trap_subr.S either).

Re-sync with HEAD files which were out of sync.

Sync with HEAD

Allocate intstk and spillstk for cpu0.

Switch to the idle stack before releasing the sched_lock (rev 1.31 was
wrong).

Make want_resched and astpending per-CPU.
Add some per-CPU variables to cpu_info.

Work around of "trap type 700 at ff8099b8".
It seems to work for me but I'm not sure...

Switch to the idle stack in the idle loop.

Sync with HEAD.

Move INTSTK and SPILLSTK definitions to locore.[Ss] to allow allocation of
them in early startup.

Add KERNEL_LOCK (et al.) to run MULTIPROCESSOR kernel on a single cpu system.

Sync with HEAD.

Update thorpej_scsipi to -current as of a month ago
A i386 GENERIC kernel compiles without the siop, ahc and bha drivers
(will be updated later). i386 IDE/ATAPI and ncr work, as well as
sparc/esp_sbus. alpha should work as well (untested yet).
siop, ahc and bha will be updated once I've updated the branch to current
-current, as well as machine-dependant code.

Use <powerpc/powerpc/trap_subr.S>

Scheduler locking changes.

Make it compile with "options MULTIPROCESSOR".

First attempt to spin up the secondary processor on 180MP and 200MP.
XXX Need more work.

Sync w/ netbsd-1-5-base.

Add a comment about needing to initialize p_cpu when multiple
processors are supported.

First sweep at scheduler state cleanup.  Collect MI scheduler
state into global and per-CPU scheduler state:

	- Global state: sched_qs (run queues), sched_whichqs (bitmap
	  of non-empty run queues), sched_slpque (sleep queues).
	  NOTE: These may collectively move into a struct schedstate
	  at some point in the future.

	- Per-CPU state, struct schedstate_percpu: spc_runtime
	  (time process on this CPU started running), spc_flags
	  (replaces struct proc's p_schedflags), and
	  spc_curpriority (usrpri of processes on this CPU).

	- Every platform must now supply a struct cpu_info and
	  a curcpu() macro.  Simplify existing cpu_info declarations
	  where appropriate.

	- All references to per-CPU scheduler state now made through
	  curcpu().  NOTE: this will likely be adjusted in the future
	  after further changes to struct proc are made.

Tested on i386 and Alpha.  Changes are mostly mechanical, but apologies
in advance if it doesn't compile on a particular platform.

Introduce a new process state distinct from SRUN called SONPROC
which indicates that the process is actually running on a
processor.  Test against SONPROC as appropriate rather than
combinations of SRUN and curproc.  Update all context switch code
to properly set SONPROC when the process becomes the current
process on the CPU.

Update several ports to at least work again without IPKDB
after the update of that.
Sorry for breaking this with my last changes :-(.
Fixes PR#9671 by Lennart Augustsson.

Enter power saving mode in idle loop on 603, 750, and 7400.
603 uses doze mode, 750 and 7400 use nap mode.

Pull up revision 1.19 (requested by danw):
  Add (the beginnings of) an unaligned access handler for the
  powerpc, to get around the fact that gcc currently generates bad
  floating point copies sometimes, which breaks various things,

Add a special trap handler for EXC_ALI that saves dsisr and dar

Pull up to last week's -current.

synchronize various changes to comments between these files to make it
even more obvious that they're 90% identical and really ought to be
using common code in arch/powerpc for a most of this...

Move signal management code common to all PowerPC platforms (sendsig(),
sigreturn(), sigcode) to a single place in the shared powerpc directory.
(The struct sigcontext etc. definitions were already being shared.)

Use battable to map obio space rather than using bat1 registers statically.

Sync w/ -current.

Add more sanity check.

Now we can use symbols in DDB. :-)

remove opt_uvm.h

Oops, last change is wrong...

Use private (its own) segment registers in Open Firmware call,
instead of BAT register.

ofmsr[IP] = 0.

Add irq32 - irq63 to intrnames.

Correct battable[] usage.

Make signal delivery work again.

Adjust for the new "reaper" kernel thread: do not free the vmspace and
u-area in machine-dependent code.  Instead, call exit2() to schedule
the reaper to free them for us, once it is safe to do so (i.e. we are
no longer running on the dead proc's vmspace and stack).

Preserve OpenFirmware's msr and use it in openfirmware().

defopt DDB.

Stop using "options FIRMWORKSBUGS".

Add support for UVM and MACHINE_NEW_NONCONTIG.

Initial import of macppc port.