kernel one as the non-faulting flush address in the loader so
we can can change KERNBASE and VM_MIN_KERNEL_ADDRESS if we
ever want to without needing to worry about using a compatible
loader.
- Correctly check for LOADER_DEBUG.
- Add a missing const for page_sizes[].
whole KVA space using one locked 4MB dTLB entry per GB of physical
memory. On Cheetah-class machines only the dt16 can hold locked
entries though, which would be completely consumed for the kernel
TSB on machines with >= 16GB. Therefore limit the KVA space to use
no more than half of the lockable dTLB slots, given that we need
them also for other things.
- Add sanity checks which ensure that we don't exhaust the (lockable)
TLB slots.
firmware versions which wedge when using the OFW test service,
so given that we don't really depend on SUNW,stop-self just nuke
it altogether instead of risking problems.
- At least Fire V880 have a small hardware glitch which causes the
reception of IDR_NACKs for CPUs we actually haven't tried to send
an IPI to, even not as part of the initial try. According to tests
this apparently can be safely ignored though, so just return if
checking for the individual IDR_NACKs indicates no outstanding
dispatch. Serializing the sending of IPIs between MD and MI code
by the combined usage of smp_ipi_mtx makes no difference to this
phenomenon. [1]
- Provide relevant debugging bits already with the initial panic
in case of problems with the IPI dispatch, which would have
allowed to diagnose the above problem without a specially built
kernel.
- In case of cheetah_ipi_selected() base the delay we wait for
other CPUs which also might want to dispatch IPIs on the total
amount of CPUs instead of just the number of CPUs we let this
CPU send IPIs to because in the worst case all CPUs also want
to IPI us at the same time.
Reported and access for extensive tests provided by: Beat Gaetzi [1]
of OFW access semantics, in order to allow future support for real-mode
OF access and flattened device frees. OF client interface modules are
implemented using KOBJ, in a similar way to the PPC PMAP modules.
Because we need Open Firmware to be available before mutexes can be used on
sparc64, changes are also included to allow KOBJ to be used very early in
the boot process by only using the mutex once we know it has been initialized.
Reviewed by: marius, grehan
be fatal so just inform about this instead of panicing.
- Ensure we use the right softc in case the interrupt of a child is
is routed to the companion PBM instead. This hasn't been seen in the
wild so far but given that it's the case for the Schizo interrupts,
handling this situation also for child interrupts as a precaution
seemed a good idea.
- Deal with broken firmware versions which miss child entries in the
ino-bitmap as seen on V880 by belatedly registering as interrupt
controller in schizo_setup_intr(). [1]
- Add missing '\n' when printing the warning regarding Schizo Errata
I-13.
Reported and tested by: Beat Gaetzi [1]
the code for parsing interrupt maps) to PowerPC and reflect their new MI
status by moving them to the shared dev/ofw directory.
This commit also modifies the OFW PCI enumeration procedure on PowerPC to
allow the bus to find non-firmware-enumerated devices that Apple likes to add,
and adds some useful Open Firmware properties (compat and name) to the pnpinfo
string of children on OFW SBus, EBus, PCI, and MacIO links. Because of the
change to PCI enumeration on PowerPC, X has started working again on PPC
machines with Grackle hostbridges.
Reviewed by: marius
Obtained from: sparc64
Sgtty is a programming interface that has been replaced by termios over
the years. In June we already removed <sgtty.h>, which exposes the
ioctl()'s that are implemented by this interface. The importance of this
flag is overrated right now.
module; the ath module now brings in the hal support. Kernel
config files are almost backwards compatible; supplying
device ath_hal
gives you the same chip support that the binary hal did but you
must also include
options AH_SUPPORT_AR5416
to enable the extended format descriptors used by 11n parts.
It is now possible to control the chip support included in a
build by specifying exactly which chips are to be supported
in the config file; consult ath_hal(4) for information.
of the ABI of the currently executing image. Change some places to test
the flags instead of explicit comparing with address of known sysentvec
structures to determine ABI features.
Discussed with: dchagin, imp, jhb, peter
and ifnet functions
- add memory barriers to <machine/atomic.h>
- update drivers to only conditionally define their own
- add lockless producer / consumer ring buffer
- remove ring buffer implementation from cxgb and update its callers
- add if_transmit(struct ifnet *ifp, struct mbuf *m) to ifnet to
allow drivers to efficiently manage multiple hardware queues
(i.e. not serialize all packets through one ifq)
- expose if_qflush to allow drivers to flush any driver managed queues
This work was supported by Bitgravity Inc. and Chelsio Inc.
and XMITS has to be basically done in the same manner as for
the Sabres, i.e. only for devices behind PCI-PCI-bridges and
after a PIO read on the far side of the farest PCI-PCI-bridge.
Given that the Tomatillo documentation mentions no difference
to the Schizo bridges in this regard and this is also still
part of the procedure described Schizo documentation this
seems about right so adjust accordingly (the unconditional
CDMA flushing/syncing previously done was based on how Linux
behaves).
- Implement CDMA flushing/syncing for Schizo version >= 5,
which requires the workaround described in Schizo Errata I-23.
According to Schizo Errata I-13 it's just unusable with
version < 5 though. [1]
- Don't register the Schizo streaming buffer for now until it's
usage is sorted out according to the erratas.
- Register our interrupt filters with the revived INTR_FAST so
they these interrupts can even interrupt filters of device
drivers as necessary.
- Remove the comment regarding lack of newbus'ified bus_dma(9)
as being able to associate a DMA tag with a device would
allow to implement CDMA flushing/syncing in bus_dmamap_sync(9)
but that would totally kill performance. Given that for devices
not behind a PCI-PCI bridge the host-to-PCI bridges also only
do CDMA flushing/syncing based on interrupts there's no
additional disadvantage for polling(4) callbacks in the case
schizo(4) has to do the CDMA flushing/syncing but rather a
general problem.
Reported by: Michael Moll [1]
filters instead of PIL_FAST and allow special filters and handlers
for interrupts which need to be able to interrupt even filters, f.e.
bus error interrupts, to be registered with the revived INTR_FAST
at PIL_FAST.
rerun of the streaming cache for silicon bug workarounds.
- Announce the presence of a streaming cache on attach for
informational purposes.
- For performance reasons don't do unnecessary flushes of the
streaming cache when coherent mappings are synced.
- Fix some minor style issues.
class CPUs. In theory one could also use versions additionally
taking advantage of the prefetch cache with cheetah-class CPUs,
in my worldstone runs these either didn't provide extra speedup
(USIII+) in comparison to the existing spitfire versions or were
even slightly slower (USIIIi) though, so they aren't committed
for now.
The basic problem leading to the VIS-based copy/zero functions
being initially disabled for cheetah-class CPUs was solved by
letting cheetah_init() clear DCR_IFPOE.
while doing the block store workaround so we restore the correct
floating-point registers state in case of nested floating-point
operations resulting from nested interrupts. This allows the
VIS-based block copy/zero functions to be used on machines
requiring this workaround. Alternatively, we could take care of
saving the floating-point registers here, which would be more
inefficiently though and also involves turning off interrupts.
- It turns out that the SCZ_PCI_DMA_SYNC register doesn't work
like the TOMXMS_PCI_DMA_SYNC_PEND one (but more like the
corresponding register in of Hummingbird and Sabre bridges)
and writing the INO of the respective device to it causes a
Safari bus error. However, due to the Schizo errata I-23,
SCZ_PCI_DMA_SYNC can't be used as intended either, so remove
consistent DMA syncing for Schzio bridges for now, which means
that add-on cards with non-"sun4u compliant" (whatever that
means exactly) PCI-PCI-bridges should be avoided until the
proper workaround is implemented. [1]
Reported by: Michael Moll [1]
V9 stack bias so we no longer need to add it in db_backtrace()
and stack_capture() respectively. This also reverts r182018,
which kludged around the resulting unaligned access.
- Sync the sun4v versions of db_trace.c and stack_machdep.c with
the sparc64 ones and fix some style bugs.
MFC after: 3 days
controllers and it seems to work just fine with at least an
add-on SAS3080X. While at it, remove the commented out ncr(4)
as it doesn't even use bus_dma(9), which isn't worth fixing
though as sym(4) already supports a superset of the controllers
driven by ncr(4).
JBus to PCI 2.2 bridges. In theory, this driver should also handle
`XMITS' Fireplane/Safari to PCI-X bridges but due to lack of access
to such hardware, support for these hasn't be fleshed out, yet.
After I removed all the unit2minor()/minor2unit() calls from the kernel
yesterday, I realised calling minor() everywhere is quite confusing.
Character devices now only have the ability to store a unit number, not
a minor number. Remove the confusion by using dev2unit() everywhere.
This commit could also be considered as a bug fix. A lot of drivers call
minor(), while they should actually be calling dev2unit(). In -CURRENT
this isn't a problem, but it turns out we never had any problem reports
related to that issue in the past. I suspect not many people connect
more than 256 pieces of the same hardware.
Reviewed by: kib
to the C99 style. At least, it is easier to read sysent definitions
that way, and search for the actual instances of sigcode etc.
Explicitely initialize sysentvec.sv_maxssiz that was missed in most
sysvecs.
No objection from: jhb
MFC after: 1 month
consists of CPUs running at different speeds, for driving hardclock as
these timers in turn are driven at frequencies as low as 5MHz, resulting
in bad granularity compared to the TICK timers. However, don't employ
the workaround for the BlackBird erratum #1 when using the TICK timer
on machines with cheetah-class CPUs for performance reasons.
Reported by: Florian Smeets
could trigger an error interrupt that we can't actually to do anything
against as soon as enabling the error handlers.
While at it don't bother about writing only to the write-one-to-clear
bits when clearing error bits.
disable interrupts and loop forever with these.
- Hide all MP-related bits in <machine/smp.h> underneath #ifdef SMP.
- Inline ipi_all_but_self(9) and ipi_selected(9). We don't expose any
additional bits but save a few cycles by doing so.
- Remove ipi_all(9), which actually only called panic(9). It can't be
implemented natively anyway and having it removed at least causes
MI users to fail already fail when linking.
Safari- and JBus-based machines. Currently the main purpose of
these drivers is debugging of the resource allocation on nexus(4)
and the register content of these devices though.
the locked entry in it16 slot 0, which typically is occupied by the
PROM, and manually entering locked entries in slots != 0.
Thanks to Hubert Feyrer for donating the Blade 2000 this change was
developed on.
for all three contexts and configure the dt512_1 to hold 4MB pages for
them (e.g. for direct mappings).
This might allow for additional optimization by using the faulting
page sizes provided by AA_DMMU_TAG_ACCESS_EXT for bypassing the page
size walker for the dt512 in the superpage support code.
Submitted by: nwhitehorn (initial patch)
to synchronization needed after stores to internal ASIs in order
to make side-effects visible. This mainly requires the MEMBAR #Sync
after such stores to be replaced with a FLUSH. We use KERNBASE as
the address to FLUSH as it is guaranteed to not trap. Actually,
the USII synchronization rules also already require a FLUSH in
pretty much all of the cases changed.
We're also hitting an additional USIII synchronization rule which
requires stores to AA_IMMU_SFSR to be immediately followed by a DONE,
FLUSH or RETRY. Doing so triggers a RED state exception though so
leave the MEMBAR #Sync. Linux apparently also has gotten away with
doing the same for quite some time now, apart from the fact that
it's not clear to me why we need to clear the valid bit from the
SFSR in the first place.
Reviewed by: nwhitehorn
the ncr53c9x.c core where it actually belongs so future front-ends
don't need to add it.
o Use the correct OFW property when looking for the initiator ID of the
SBus device.
o Don't specify an alignment when creating the parent DMA tag for
SUNW,fas; their DMA engine doesn't require an alignment constraint
and it's no inherited by the child DMA tags anyway (which probably
is a bug though).
o Drop the superfluous sc_maxsync and use sc_minsync instead. The
former apparently was added due to a confusion with the maximum
frequency used in cam(4), which basically corresponds to the
inverse of minimum sync period.
o Merge ncr53c9x.c from NetBSD:
1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and
NCRCMD_DMA command in ncr53c9x_select().
1.125: free allocated resources on detach.
o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset()
as these are not required outside of ncr53c9x.c.
o In ncr53c9x_attach() don't leak the device mutex in case attaching
fails.
o Register an asynchronous notification handler so in case cam(4)
reports a lost device we can cancel outstanding commands and
restore the default parameters for the target in question.
o For FAS366 correctly support 16-bit target IDs and let it know
that we use 32-bit transfers.
o Overhaul the negotiation of transfer settings. This includes
distinguishing between current and goal transfer settings of the
target so we can renegotiate their goal settings when necessary
and correcting the order in which tagged, wide and synchronous
transfers are negotiated.
o If we are requesting sense, force a renegotiation if we are
currently using anything different from asynchronous at 8 bit
as the target might have lost our transfer negotiations.
o In case of an XPT_RESET_BUS just directly call ncr53c9x_init()
instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an
interrupt that is treated as an unexpected SCSI bus reset by
ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now
no longer used ncr53c9x_scsi_reset().
o Correct an off-by-one error when setting cpi->max_lun.
o In replace printf(9) with device_printf(9) calls where appropriate
and in ncr53c9x_action() remove some unnecessarily verbose messages.
o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing
it and consolidate two tagging-related target info checks into one.
o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when
appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI
status information.
o In ncr53c9x_dequeue() ensure the tags are cleared.
o Use ulmin() instead of min() where appropriate.
o In ncr53c9x_msgout() consistently use the reset label.
o When we're interrupted during a data phase and the DMA engine is
still active, don't panic but reset the core and the DMA engine as
this should be sufficient. Also, the typical problem for triggering
this was the lack of renegotiation when requesting sense.
o Correctly handle DEVICE RESETs.
o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving
the calls of lsi64854_attach() to the bus front-ends so it can pass
the esp(4) mutex to bus_dma_tag_create(9).
o Change the LSI64854 driver to not create a DMA tag and map for the
Ethernet channel as le(4) will handle these on its own as well as
sync and unload the DMA maps for the SCSI and parallel port channel
after a DMA transfer.
o Cam(4)'ify some NetBSD-centric comments.
o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9)
and take advantage of rman_get_rid(9) in order to save some softc
members.
Reviewed by: scottl
MFC after: 1 month
table. This is required in order to set obp-control-relinquished
within the PROM, allowing to safely read the OFW translations node.
Without this, f.e. a `ofwdump -ap` triggers a fatal reset error or
worse things on machines based on USIII and beyond.
In theory this should allow to remove touching %tba in cpu_setregs(),
in practice we seem to currently face a chicken and egg problem when
doing so however.
to 43 bits so update TD_PA_BITS accordingly. For the most part this
increase is transparent to the existing code except for when reading
the physical address from ASI_{D,I}TLB_DATA_ACCESS_REG, which we
only do in the loader and which was already adjusted in r182478, or
from the OFW translations node.
While at it, ensure we are only taking valid OFW mapping entries
into account.
frequencies (and having different cache sizes) so use the STICK
(System TICK) timer, which was introduced due to this and is
driven by the same frequency across all CPUs, instead of the
TICK timer, whose frequency varies with the CPU clock, to drive
hardclock. We try to use the STICK counter with all CPUs that are
USIII or beyond, even when not necessary due to identical CPUs,
as we can can also avoid the workaround for the BlackBird erratum
#1 there. Unfortunately, using the STICK counter currently causes
a hang with USIIIi MP machines for reasons unknown, so we still
use the TICK timer there (which is okay as they can only consist
of identical CPUs).
- Given that we only (try to) synchronize the (S)TICK timers of APs
with the BSP during startup, we could end up spinning forever in
DELAY(9) if that function is migrated to another CPU while we're
spinning due to clock drift afterwards, so pin to the CPU in order
to avoid migration. Unfortunately, pinning doesn't work at the
point DELAY(9) is required by the low-level console drivers, yet,
so switch to a function pointer, which is updated accordingly, for
implementing DELAY(9). For USIII and beyond, this would also allow
to easily use the STICK counter instead of the TICK one here,
there's no benefit in doing so however.
While at it, use cpu_spinwait(9) for spinning in the delay-
functions. This currently is a NOP though.
- Don't set the TICK timer of the BSP to 0 during at startup as
there's no need to do so.
- Implement cpu_est_clockrate().
- Unfortunately, USIIIi-based machines don't provide a timecounter
device besides the STICK and TICK counters (well, in theory the
Tomatillo bridges have a performance counter that can be (ab)used
as timecounter by configuring it to count bus cycles, though unlike
the performance counter of Schizo bridges, the Tomatillo one is
broken and counts Sun knows what in this mode). This means that
we've to use a (S)TICK counter for timecounting, which has the old
problem of not being in sync across CPUs, so provide an additional
timecounter function which binds itself to the BSP but has an
adequate low priority.
sizes (and running at different frequencies) so move the cacheinfo
to the PCPU data. While at it, remove some redundant and/or unused
members from struct cacheinfo.
- In sparc64_init don't assume the first CPU node we find in the OFW
device tree is the BSP.
in order to avoid the invasive probes done by identify-routines of
ISA drivers, which may access unassigned addresses or those of
unrelated devices and thus in turn can trigger master/target aborts
as revealed by r182108 and ahc(4). I think that this is also the
cause of the hang previously seen on B100 blades during boot.
Bypassing isa_probe_children(9) also avoids adding ISA hints, which
just can be wrong for sparc64.
Reported by: gavin
in the kernel and copying it out, causing a panic when faulting on a
nofault entry. Handle this case gracefully by letting the kernel copy
functions return EFAULT instead. As such this change addresses the
same problem as r154721 does for i386.
MFC after: 3 days
In syscall, always make a copy of parameters from trapframe, this
becauses some syscalls using set_mcontext can sneakily change
parameters and later when those syscalls references parameters,
they will wrongly use register values in mcontext_t.
PR: 72998
MFC after: 3 days
the latency based on the Min_Gnt register so use the algorithm found
in OpenSolaris as they probably know how to interpret the value Sun
puts into these registers (previously, the latency calculated for
66MHz was most likely wrong) and for bridges additionally set up the
secondary latency register. Also set up the bridge control register
the way it's done in OpenSolaris. As the latency register don't apply
to PCI-Express and the bridge control setup wasn't tested on sun4v
(besides most likely not being needed), expand the #ifndef SUN4V
accordingly.
MFC after: 3 days