after the other CPUs have already return into the PROM through prom_cpustop().
So don't do it; try to halt the other CPUs only when an actual `halt'
is requested.
My guess in this case is that the PROM does not expect its original boot
CPU to call prom_cpustop(0), or possibly treats it as the equivalent of
a prom_halt() which is then called again by some other CPU. In any case,
the result is garbage.
While the pmap functions take care of properly ordering cache and TLB flushes,
not all TLB flush requests are sent to all CPUs (due to per CPU pmap tracking)
while cache flushes are always broadcast.
XXX: need to investigate pmap tracking for cache flush operation as well.
cpu's who have done so. implement pmap_deactivate() for MULTIPROCESSOR and
call it from cpu_switch() when we are about to switch proces and when we
enter idle().
with this change, i see significantly reduced tlb IPI traffic and fork/exec
bound processes -- such as "configure" -- run significantly faster, upto
15%. i also obvserved a small (0-2%) benefit to CPU bound tasks as well.
rather than the level13 software intr xpmsg area. now DDB IPI's don't lock
the xpmsg_lock and we avoid recursion and more.
- don't actually use cpuinfo.msg.lock yet, xpmsg_lock suffices.
- reread the pending register on mbus hypersparc cpus to avoid bugs in the
h/w that cause IPI's to be missed.
cd ${KERNSRCDIR}/${KERNARCHDIR}/compile && ${PRINTOBJDIR}
This is far simpler than the previous system, and more robust with
objdirs built via BSDOBJDIR.
The previous method of finding KERNOBJDIR when using BSDOBJDIR by
referencing _SRC_TOP_OBJ_ from another directory was extremely
fragile due to the depth first tree walk by <bsd.subdir.mk>, and
the caching of _SRC_TOP_OBJ_ (with MAKEOVERRIDES) which would be
empty on the *first* pass to create fresh objdirs.
This change requires adding sys/arch/*/compile/Makefile to create
the objdir in that directory, and descending into arch/*/compile
from arch/*/Makefile. Remove the now-unnecessary .keep_me files
whilst here.
Per lengthy discussion with Andrew Brown.
bus_space_{read,write}_[1248]() functions, which will allow 16-bit
PCMCIA support to work without additional hacks in MI drivers.
this option is not enabled yet.
fact that cache flushes are also passed the context number), most
"long-term" context switches can be eliminated from the SRMMU versions
of the pmap functions.
Setup a DDB context for paused CPUs by defining a soft trap (T_DBPAUSE)
which uses the generic trap handler code to get the trapframe constructed
and then calls on a debugger-defined `suspend' routine.
well as reporting the actual machine model & cpu, rather than first configured
CPU. changes for two machines are:
old:
hw.model = TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU
hw.model = SUNW,UltraSPARC @ 143.002 MHz, version 0 FPU
new:
hw.model = SUNW,SPARCstation-20 (TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU)
hw.model = SUNW,Ultra-1 (SUNW,UltraSPARC @ 143.002 MHz, version 0 FPU)
as per discussion on port-sparc & port-sparc64.
possible to use alternate system call tables. This is usefull for
displaying correctly the arguments in Mach binaries traces.
If NULL is given, then the regular systam call table for the process is used.
cache flush ops.
Also a standard soft interrupt handler for standard cross-call notification
reserving the NMI level 15 softint for urgent cross calls.
must be shared. This requires drivers that request the installation of
a `fast trap' handler to supply a regular interrupt handler as well to fall
back on.
The `fast trap' routine (if present) is now passed as an additional
argument to intr_establish().
To make this work, we now have to use separate handler lists for hardware
and software interrupts as the soft interrupt handlers do not return
an `interrupt handled' status.
Thanks to Matt Fredette for providing an initial set of patches on port-sparc.
given that PROM maps just 4 or 16 this is not going to be a bottle
neck). Doesn't really affect normal kernels, need it for the changed
kernel base address (uncommitted) hack for broken javastation OFW.
Ok by pk.
raise the ipl in the interrupt handlers to the appropriate level. This avoids
interrupt handler interference if one of the devices actually interrupts at
a lower hardware level than the maximum level assined to a device class.
Based on code from Art Grabowski in openbsd.
original system call number, which can be negative for a Mach trap.
We cannot just replace code by realcode, because ktrsyscall uses it as
an index in the system call table, thus crashing the kernel when the
value is negative.
- disk_unbusy() gets a new parameter to tell the IO direction.
- struct disk_sysctl gets 4 new members for read/write bytes/transfers.
when processing hw.diskstats, add the read&write bytes/transfers for
the old combined stats to attempt to keep backwards compatibility.
unfortunately, due to multiple bugs, this will cause new kernels and old
vmstat/iostat/systat programs to fail. however, the next time this is
change it will not fail again.
this is just the kernel portion.
kqueue provides a stateful and efficient event notification framework
currently supported events include socket, file, directory, fifo,
pipe, tty and device changes, and monitoring of processes and signals
kqueue is supported by all writable filesystems in NetBSD tree
(with exception of Coda) and all device drivers supporting poll(2)
based on work done by Jonathan Lemon for FreeBSD
initial NetBSD port done by Luke Mewburn and Jason Thorpe
layers. Common middle layer shared by kbd_zs and sunkbd is moved into
the new file. Move shared config directives to files.sun and adjust
ports' files.* accordingly.
Need this to support console/Xsun on Mr.Coffee JavaStation.
Tested on sparc, sparc64 (by martin) and sun3 (by jdc).
This merge changes the device switch tables from static array to
dynamically generated by config(8).
- All device switches is defined as a constant structure in device drivers.
- The new grammer ``device-major'' is introduced to ``files''.
device-major <prefix> char <num> [block <num>] [<rules>]
- All device major numbers must be listed up in port dependent majors.<arch>
by using this grammer.
- Added the new naming convention.
The name of the device switch must be <prefix>_[bc]devsw for auto-generation
of device switch tables.
- The backward compatibility of loading block/character device
switch by LKM framework is broken. This is necessary to convert
from block/character device major to device name in runtime and vice versa.
- The restriction to assign device major by LKM is completely removed.
We don't need to reserve LKM entries for dynamic loading of device switch.
- In compile time, device major numbers list is packed into the kernel and
the LKM framework will refer it to assign device major number dynamically.
perform the translation and recursion if t->ranges != NULL. Make
sbus, cpuunit, and bootbus inherit the parent's map/mmap routines,
and delete the now-unused mapping functions. Update all places where
bus space tags are statically allocated.
board has 2 CPU units (one for each processor), one of which also has
a "bootbus", to which the serial ports, etc. are attached.
We can now probe/attach CPUs on a SPARCserver 1000.
counters. These counters do not exist on all CPUs, but where they
do exist, can be used for counting events such as dcache misses that
would otherwise be difficult or impossible to instrument by code
inspection or hardware simulation.
pmc(9) is meant to be a general interface. Initially, the Intel XScale
counters are the only ones supported.
nor software interrupt pending bit is set for the current ipl. Report
this as a "bogus" interrupt (better name anyone?). This is a symptom
of a bug in interrupt handling in one of device drivers interrupting
at this ipl. Reviewed by pk.
be properly used by any misc. cloning device. While here, correct
a comment to indicate that "open" is the only entry point and that
everything else is handled with fileops.
machines always use the Viking CPU, and cache info is not available
from the PROM, so we have to hard-code it in that case.
Rearrange some stuff to make it easier to share.
how many CPU types are configured into the kernel. Then, use this
information to define the CPU predicate macros according to the
following rules:
1. If support for a CPU type is not configured into the kernel,
then the test is always false.
2. Otherwise, if only one CPU type is configured into the kernel,
then the test is always true.
3. Otherwise, we have to reference the cputyp variable.
Use a similar strategy for short-cutting the page size related
definitions.
as (CPU_ISSUN4 || CPU_ISSUN4C) and (CPU_ISSUN4C || CPU_ISSUN4M),
respectively. The compiler can still optimize as desired by expressing
them this way, and it simplifies adding new tests.
While here, just remove CPU_ISSUN4MOR4U; it's not used by anything.
MALLOC_NOINLINE, and VNODE_OP_NOINLINE. The exceptions are when they
include another config files that already defines the options, or if
they are for an embedded board, just define a few extra options, and
do not already define PIPE_SOCKETPAIR.
* struct sigacts gets a new sigact_sigdesc structure, which has the
sigaction and the trampoline/version. Version 0 means "legacy kernel
provided trampoline". Other versions are coordinated with machine-
dependent code in libc.
* sigaction1() grows two more arguments -- the trampoline pointer and
the trampoline version.
* A new __sigaction_sigtramp() system call is provided to register a
trampoline along with a signal handler.
* The handler is no longer passed to sensig() functions. Instead,
sendsig() looks up the handler by peeking in the sigacts for the
process getting the signal (since it has to look in there for the
trampoline anyway).
* Native sendsig() functions now select the appropriate trampoline and
its arguments based on the trampoline version in the sigacts.
Changes to libc to use the new facility will be checked in later. Kernel
version not bumped; we will ride the 1.6C bump made recently.
to be double-word aligned and it is not when we are called from bpf.
Remove "should be able to std here" comment in generic microtime.
While here, simplify some asm mnemonics and fix a typo in comment.
NULL for root PCI busses. For busses behind a bridge, it points to
a persistent copy of the bridge's pcitag_t. This can be very useful
for machine-dependent PCI bus enumeration code.
* Implement a machine-dependent pci_enumerate_bus() for sparc64 which
uses OFW device nodes to enumerate the bus. When a PCI bus that is
behind a bridge is attached, pci_attach_hook() allocates a new PCI
chipset tag for the new bus and sets it's "curnode" to the OFW node
of the bridge. This is used as a starting point when enumerating
that bus. Root busses get the OFW node of the host bridge (psycho).
* Garbage-collect "ofpci" and "ofppb" from the sparc64 port.
- unify sparc_bbinfo (1064 bytes, with 256 block entries)
and sun68k_bbinfo (296 byte, with 64 block entries)
into shared_bbinfo (512 bytes, with 118 block entries),
which will be also shared by future bbinfo-using platforms
(including macppc)
- add datestamp to *_BBINFO_MAGIC strings, to prevent installboot vs
bootxx version skew.
- add macppc support
*/bootxx.c:
- migrate to new shared_bbinfo structure
installboot:
- add macppc support (still needs applepartmap support and testing)
- improve and add some more warnings & errors to installboot
- implement shared_bbinfo_clearboot() and shared_bbinfo_setboot(), which
perform the majority of the work for bbinfo-using back-ends
(rather than replicating that across multiple back-ends).
DHCP/Bootp is tried before bootparams in the MI nfs mountroot code.
- Change the order in boot.net from "bootparams then bootp" to
"bootp then bootparams", to be consistent with the MI nfs mountroot code.
(Consistency with other NetBSD ports is good, and things still work
for sites that run bootparams but not dhcp/bootp, although I'd argue that
dhcp/bootp is much easier to setup and manage and is easier to debug
than our rpc.bootparamd...)
Per discussion with Paul Kranenburg and Matt Green.
files from sys/param.h. The problem is that lint stubs are generated
with -C (keep comments) and gcc complains about macro redefinition,
since "assym.h" doesn't have the comments that "sys/param.h" has.
* Pull in dev/mii/files.mii from conf/files, rather than playing
the magic "files include order" dance in N machine-dependent
configuration definitions.
as a stub bus that knows how to attach drivers for various functions
of PCIC.
This change is a follow up to timer0 at msiiep0 change, since all
children must share same attach arguments and so we no longer can
attach pci0 directly under msiiep0.
* put various time-of-day clock, timer and eeprom drivers in separate files
* avoid unnecessary overhead in interrupt handlers
* make the MSIIEP timer attach through the regular autoconf mechanisms
indicating an unhandled "command". ERESTART is -1, which can lead to
confusion. ERESTART has been moved to -3 and EPASSTHROUGH has been
placed at -4. No ioctl code should now return -1 anywhere. The
ioctl() system call is now properly restartable.
become ippp (ISDN ppp) and irip (ISDN raw IP). The character device now
are called: /dev/isdn (isdnd <-> kernel communication), /dev/isdnctl (dialing
and other control), /dev/isdntrc* (tracing), /dev/isdnbchan* (raw B channel
access, i.e. for user land PPP) and /dev/isdntel* (telephone devices, i.e.
for answering machines).
Audio-related stuff is left almost intact.
* support audiocs at ebus playback and capture
tested on krups and u5 (thanks, martin)
* make first attempt at supporting audiocs at sbus capture
* nb: full-duplex is not tested
* while here, fix CSAUDIO_MONITOR_MUTE to be of CSAUDIO_MONITOR_CLASS
i.e. outputs.monitor.mute -> monitor.monitor.mute
Ok by pk, eeh.
* Use BUS_ADDR() where appropriate to encode I/O space and physical
address offset into a `bus_addr_t' value.
* Drop obio_bus_map() since it's now completely equivalent to bus_space_map()
* Use bus_space_map2() to map device space at a fixed virtual address.
* Remove the virtual address argument from sbus_sbus_addr()
deal with shortages of the VM maps where the backing pages are mapped
(usually kmem_map). Try to deal with this:
* Group all information about the backend allocator for a pool in a
separate structure. The pool references this structure, rather than
the individual fields.
* Change the pool_init() API accordingly, and adjust all callers.
* Link all pools using the same backend allocator on a list.
* The backend allocator is responsible for waiting for physical memory
to become available, but will still fail if it cannot callocate KVA
space for the pages. If this happens, carefully drain all pools using
the same backend allocator, so that some KVA space can be freed.
* Change pool_reclaim() to indicate if it actually succeeded in freeing
some pages, and use that information to make draining easier and more
efficient.
* Get rid of PR_URGENT. There was only one use of it, and it could be
dealt with by the caller.
From art@openbsd.org.
Be consistant in the way that MSIZE, MCLSHIFT, MCLBYTES and NMBCLUSTERS
are defined.
Remove old VM constants from cesfic port.
Bump MSIZE to 256 on mipsco (the only one that wasn't already 256).
locations. If passed NULL, the old behaviour using bus_space_{read,write}_1()
is used. Otherwise, all access to the chip goes via the MD functions.
This is necessary for mvmeppc boards where the mk48txx NVRAM/RTC is not
directly addressable.
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).
the information provided by the loader if possible (defaulting to `end').
If the DDB symbols aren't needed, `kernel_top' is adjusted in
autoconf:bootstrap() before calling pmap_bootstrap(). It will also
preserve the bootinfo data (if passed by the loader) for non-DDB kernels.
While we're here, enable RAIDframe (and RAID_AUTOCONFIG) by default for
architectures that I'm comfortable can deal with it being on by default.
Also: bump the number of 'raid' devices from 4 to 8, since 4 seems to
be insufficient in practise.
translation, get rid of sbus_bus_addr. Since sbus_bus_addr is used by
framebuffer drivers that are shared with sparc64 we cannot just change
the callers to use BUS_ADDR() w/out coordinating the change with
sparc64 that uses a different calling protocol, so stub our
sbus_bus_addr to just return BUS_ADDR.
which will cause the values of tf_pc and tf_npc set here in cpu_fork()
to be restored and used to return to user space from sigreturn() eventually
(before that the tf_pc and tf_npc set in sendsig() will be used in
proc_trampoline()).
Also, the fork system call never uses the SYSCALL_G2RFLAG shortcut, so drop
the code handling that.
(instead of `tf_npc') contains the user-space return address. This ensures
entering the signal trampoline code correctly in case we take handled
signal during child_return().
