summaryrefslogtreecommitdiff
path: root/sys/arch/mvme88k
AgeCommit message (Collapse)Author
2006-11-28give scsi controllers a real attach args to fill in when attaching scsibus.David Gwynne
ok miod@ marco@ deraadt@
2006-11-28rename scsibus_attach_args to scsi_attach_args. this can help avoidDavid Gwynne
confusing when trying to attach scsibus to a hba, since it is really meant for attaching scsi devices to scsibus. ok deraadt@ marco@
2006-11-20Hold the kernel lock when processing low IPL interrupts; one more step towardsMiod Vallat
stable MULTIPROCESSOR kernels.
2006-11-20Directly invoke m188_setipl() instead of setipl() in the external interruptMiod Vallat
handler.
2006-11-19I completely broke pfsr_188_save_double when introducing per-cpuinfoMiod Vallat
pfsr pointers seven months ago. This would work, but not in all fault cases; now it does the right thing.
2006-11-18On 88100 systems which do not need the xxx.usr instruction workaround,Miod Vallat
override the DAE helper routines with workaroundless code while the kernel text is still writable.
2006-11-18Rework the PFSR register analysis code on mvme88k: split it into fourMiod Vallat
independent subfunctions, turn PFSR_SAVE into a couple of NOP, and replace them early at runtime with a branch to the selected routine, which will return to pfsr_save. This is really better for 188 systems.
2006-11-18A better and shorter wait sequence between writes to the command registerMiod Vallat
(spec wants 3 X1 clock cycles, thus 0.82 microsecond on mvme188)
2006-11-18Do not include <machine/frame.h> (forgotten when the former was removed).Miod Vallat
2006-11-16Ratibibugle struct frame and <machine/frame.h>Miod Vallat
2006-08-17Check d_secpercyl in all readdisklable() functions, and have all ofKenneth R Westerback
them return 'invalid geometry' when d_secpercyl == 0. While there move the check to a consistant location (after the check of d_secperunit) and use a consistant idiom (i.e. some readdisklabel()'s have no 'done' label). prodded by thib@ after a bad macppc experience. ok deraadt@
2006-08-13Remove remains of cut'n'paste from ffs code.Miod Vallat
2006-07-284KB buffer on the stack, and would you like fries with that?Miod Vallat
2006-07-27Compile all kernels with -Wstack-usage-larger-than-2047, now that allMiod Vallat
offending code has been taken out and shot. ok deraadt@
2006-07-07Get rid of the fixed iomap for device mappings, and allocated va rangesMiod Vallat
off kernel_map whenever necessary.
2006-06-19move clock_subr.c to a better place, and now it is always in the kernelTheo de Raadt
so that things can use it; tested on all architectures; ok kettenis
2006-06-11TypoMiod Vallat
2006-05-28Configure network devices at ipl 3, not ipl 1.Miod Vallat
2006-05-18Use standard swap16() and swap32() functions instead of abusing <db.h> to getMiod Vallat
similar M_xx_SWAP().
2006-05-18Move otherwise unused <machine/vid.h> to stand/Miod Vallat
2006-05-17Delay board type retrieval.Miod Vallat
2006-05-16Crank bootblocks version for today's changes.Miod Vallat
2006-05-16Nuke whitespace at EOLMiod Vallat
2006-05-16Make standalone binaries compile with -Wall -WerrorMiod Vallat
2006-05-16Cleanup of mvme88k standalone code, with rewritten startup code withMiod Vallat
fewer assembly statements and much less magic, a few display artefacts removed, and private definitions moved out of the main prom.h.
2006-05-13Oops, intr_findvec() proto is still necessary.Miod Vallat
2006-05-08Replace gazillions of badvaddr() or badwordaddr() calls with badaddr() calls.Miod Vallat
With a few prototype declarations shuffling, this finally allows <machine/locore.h> to die.
2006-05-08Clean the internal m88k trap type codes; while there, simplify andMiod Vallat
factorize the build of the VBR page betweem luna88k and mvme88k. Tested by aoyama@ and I.
2006-05-07No need to mask misaligned address exceptions for global flush or invalidateMiod Vallat
operations (those do not take a page number in the low address bits of the 88410 control register).
2006-05-07Do not panic on MVME188 if the abort switch is pressed while we are at ipl 7Miod Vallat
(fallout from always enabling the abort switch).
2006-05-06Sync VM games on startup with earlier pmap_table[] cleanups.Miod Vallat
2006-05-06Decide whether we are running on 88100 or 88110 upon the processorMiod Vallat
identification register value, rather than upon the MVME board number.
2006-05-06On 88100-based systems, take the two hardwired BATC into account inMiod Vallat
pmap_extract(), and do not bother creating regular page table mappings for obio regions which are covered by these BATC entries.
2006-05-04Now that .S files have proper dependencies, do not hardcode dependenciesMiod Vallat
on assym.h in Makefile, especially since some files have been renamed...
2006-05-02Wave goodbye to bugtty on mvme88k as well.Miod Vallat
2006-04-27<machine/mvme1x7.h> is not really used anymore, remove it.Miod Vallat
2006-04-27Adjust sizes for the mandatory 1:1 mappings created in pmap_bootstrap().Miod Vallat
Saves up to 12KB of no longer necessary page tables.
2006-04-27Manage a local copy of the MVME188 nvram and update the real data whenMiod Vallat
necessary; allow opening /dev/nvram0 on MVME188 again (but no mmap).
2006-04-27typoMiod Vallat
2006-04-26On MVME188, the 2KB nvram is non contiguous; each byte is mapped into aMiod Vallat
32 bit int. So map the right size, and fail open() until the uiomove() wrapper is modified to cope with this.
2006-04-26Define BUS_SPACE_MAP_LINEAR and use it when appropriate. Right nowMiod Vallat
bus_space_map() always maps everything linear, but this could change in the future.
2006-04-26The dual-ported memory of the MVME376 boards is D32 addressable, say theMiod Vallat
documentation, so we can use the MI {zero,copy{to,from}}buf_contig callbacks, which rely upon bcopy() and bzero(), instead of their d16_bcopy() and d16_bzero() equivalents. No functional change, except an unnoticeable speedup.
2006-04-26In nmihand(), do not enter ddb with Debugger(), which causes an exception,Miod Vallat
as we might not be in a state where we can process a nested exception. This would then cause an error exception, which handler calls nmihand. Kaboom. Instead, directly invoke m88k_db_trap(); from the AV tree.
2006-04-26We do not need to reserve the 0x60..0x7f interrupt vector range on MVME188,Miod Vallat
so let VME interrupt numbers start at 0x60 on these machines.
2006-04-19Get rid of the clock device attachment - since the clock is not somethingMiod Vallat
we can live without, move it into the board-dependent code. This even makes the code slightly smaller. clock.c is moved from dev/ to mvme88k/ and only keeps common variables and delay().
2006-04-19Count spurious interrupt as part of the alarming interrupt issues as well.Miod Vallat
2006-04-19Better diagnostic messages in the MVME188 interrupt handler; also no needMiod Vallat
to register an interrupt handler for the SYSCON board interrupts anymore, as we never trigger it.
2006-04-18Bring back a kernel_text symbol, config -e needs it.Miod Vallat
2006-04-17Do not mask the MVME188 abort button interrupt at splhigh(), similar to MVME197.Miod Vallat
2006-04-17On MVME188, clear SYSFAIL and set up a decent VME data transfer timeout,Miod Vallat
instead of relying on what the BUG leaves us with.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-22 04:40:52 +0000