| Age | Commit message (Collapse) | Author |
|---|
|
not necessary, but consistent with other platforms. ok deraadt
|
|
KNF while I'm there. Also remove some #if 0'ed code that I'm not planning
to use in the near future.
Apologies to deraadt@ for ignoring his diff and doing it my own way.
|
|
|
|
the stack region handling in uvm.
|
|
in all MD apmvar.h.
|
|
ok kettenis
|
|
and if so, force soft carrier mode.
|
|
when workqs were introduced, we provided a second argument so you
could pass a thing and some context to work on it in. there were
very few things that took advantage of the second argument, so when
i introduced pools i suggested removing it. since tasks were meant
to replace workqs, it was requested that we keep the second argument
to make porting from workqs to tasks easier.
now that workqs are gone, i had a look at the use of the second
argument again and found only one good use of it (vdsp(4) on sparc64
if you're interested) and a tiny handful of questionable uses. the
vast majority of tasks only used a single argument. i have since
modified all tasks that used two args to only use one, so now we
can remove the second argument.
so this is a mechanical change. all tasks only passed NULL as their
second argument, so we can just remove it.
ok krw@
|
|
|
|
devino into a pseudo-sysino, directly use the devino as the ihandle. The
devhandle is stored in the cbus softc, and accessed through the bus space
tag. This allows us to have more than 256 interrupts on a single cbus, and
avoids relying on the lower bits of the devhandle being zero.
|
|
of pointers to vd requests, and a single task to notify the taskq
that there's work on the ring.
the ring is modelled on what you usually get with network cards,
which is an array with a producer and consumer index. the interrupt
handler fills an entry in the ring at the producers index and
increments it. the task reads entries at the consumer index and
increments it until it reaches the same value as the producer.
tested and slightly tweaked by kettenis@
ok kettenis@
|
|
|
|
usually invoked after initdisklabel() which proactively changes a zero disk
size to MAXDISKSIZE, causing this test to fail.
Allow for MAXDISKSIZE too in that test. This makes spoofed disklabels of SMD
disks have a proper `c' slice size.
luna88k disklabel_om_to_bsd() is modified accordingly, to keep diffability,
even though luna88k can't - to the best of my knowledge - sport SMD disk
controllers.
ok deraadt@ krw@
|
|
make install
from Simon Nicolussi
ok jsing tedu
|
|
This should catch all the same bad cases, but be a little less aggravating
in circumstances where a prototype isn't necessary
ok deraadt
|
|
|
|
|
|
always load all allocated pages instead of the size specified in the
bus_dmamap_load_raw(9) call. Also fixes the corner case where a specified
boundary is less than the page size, which would always create multiple
segments, even if the specified size was smaller than the boundary.
Fixes xhci(4) on sparc64.
|
|
|
|
|
|
|
|
ok deraadt kettenis
|
|
doesn't have all the values and therefore can't be used everywhere.
ok deraadt@ kettenis@
|
|
eliminating the must-be-kept-in-sync UVM_INH_* macros
ok deraadt@ tedu@
|
|
that we delete ours, which isn't actually useful.
ok mpi sthen
|
|
ok miod@
|
|
|
|
ok deraadt, tests on glxpcib and ok mpi
|
|
tweak and ok millert@ miod@
|
|
ok tedu@ deraadt@
|
|
|
|
support was introduced) may return a device path without a unit number?!?
Open Firmware's open-dev routine will crash if given such a path, rendering
the system unbootable. Repair such paths by adding unit number "0".
Breakage found by sebiasta@ on a Sun Enterprise E450, fix tested by him.
helpful hints from deraadt@
|
|
processor in order to give other strands a chance to run. Use it in
__mp_lock_spin_hook() to avoid wasting CPU cycles if we're waiting for
the kernel or scheduler locks. This is instruction is patched in, just like
we already do for the sleep instruction on SPARC64 VI processors. We look
at the hwcap-list property of the cpu nodes in the machine description to
decide whether the pause instruction is available.
|
|
sparc64 machines should now be able to boot from softraid raid1 and crypto.
Tested on sun blade100, sunfire v240, and sun t1000 ldom guests.
Lots of encouragement from deraadt@
|
|
This can be used to pass boot parameters to the kernel which can't be passed
safely via the Open Firmware interface, such as softraid volume IDs and keys.
The kernel already reads the arguments if available but ofwboot won't provide
them until further changes are committed there.
With support from deraadt, kettenis and matthew.
ok deraadt@
|
|
Written with use of the forth by jedi jsing@
For now this only supports 'a' RAID partitions, no other partition letters.
|
|
Not hooked to the build yet.
ok kettenis@
|
|
Will be needed soon.
ok kettenis@
|
|
|
|
|
|
we change the order of the xxx_setvalidand xxx_setstate hypervisor calls
things actually work.
|
|
on the SPARC T5-2. Make error messages a bit more consistent while there.
|
|
queue.
|
|
shouldn't link it into the table.
|
|
|
|
pointer to "struct intrhand" if it is larger than MAXINTNUM. To be used
with the cookie based "vintr" scheme.
|
|
|
|
need this soon.
install tested by jsg
|
|
on newer sun4v implementations (possibly SPARC T3 and later, definitely not
there on SPARC T5) and it should not be necessary on older systems.
Tested by dlg@
|
|
The situation was: memcpy swaps registers and drops into bcopy, which does
the overlap check, before getting around to business. But memcpy is not
supposed to handle overlapped arguments special, so we don't need all this
complex register swapping and tests up front -- when memcpy is the most
common operation. Refactor all these upside downupside down.
Survived multiple make builds, so probably good, and faster.
ok dlg
|
