summaryrefslogtreecommitdiff
path: root/sys/arch/sparc
AgeCommit message (Collapse)Author
2000-02-21move fpproc into the cpuinfo structure.Artur Grabowski
2000-02-21The last pieces of hypersparc support.Artur Grabowski
- Split get_faultstatus into get_syncflt and get_asyncflt. - Get the syncflt status before setting up the trap frame and put the results in cpuinfo.syncfltdump (related to the next change). - unlock the sfsr/sfva after reading from ASI_SRMMUFP (it gets locked if the translation fails). - + other interface changes and cleanups. (most work from NetBSD).
2000-02-21In cpumatch_hypersparc - zap the instruction cache if we are the main cpu.Artur Grabowski
2000-02-21remove the annoying "hypersparc support still under construction" warning.Artur Grabowski
2000-02-21hypersparc has a writeback cacheArtur Grabowski
2000-02-21When mapping something into iommu space hypersparc requires us to align itArtur Grabowski
so that cache_alias_bits match in the kernel mapping and the iommu mapping. (see code for better explaination).
2000-02-21Take the hypersparc cache size from prom, enable instruction cache andArtur Grabowski
disable "Unimplemented Flush Traps". From NetBSD.
2000-02-21add char 124 to be SES devicemjacob
2000-02-20Add support for CDBs > 12 bytes.mjacob
2000-02-19flush the caches that need flushing on context switch.Artur Grabowski
2000-02-19replacemul has been broken all the time. comment it out for now.Artur Grabowski
2000-02-19get a symbol for pure_vcache_flushArtur Grabowski
2000-02-19Add two new cache operations.Artur Grabowski
cache_flush_all - (pretty obvious). pure_vcache_flush - flush all VIVT caches (needed on context switch). (From NetBSD).
2000-02-19link boot at a starting address 0x10000 higher than bootxx; avoids ↵Theo de Raadt
hypersparc cache aliasing
2000-02-18No need to call memerr4_4c through a pointer. We know when we are a 4/4c.Artur Grabowski
2000-02-18In vmapbuf expand uvm_km_valloc_wait into a direct call to uvm_mapArtur Grabowski
so that we can specify an offset. This allows uvm_map to use PMAP_PREFER and removes a big source of bad cache aliases. With this change I have not seen any bad cache aliases during normal use. (it is still possible to force them).
2000-02-18setpte4m is now only legal for kernel mappings.Artur Grabowski
2000-02-18Zap an unnecessary and confusing prototype for setpte4m.Artur Grabowski
2000-02-18Add a bunch of hypersparc registers. (from NetBSD).Artur Grabowski
2000-02-18Rename some registers to match NetBSD and litterature.Artur Grabowski
(SFADDR -> SFAR, SFSTAT -> SFSR, AFADDR -> AFAR, AFSTAT -> AFSR).
2000-02-17add wrasr (the name says it all)Artur Grabowski
2000-02-17In dvma_mapout don't try to flush unmapped memory from cache.Artur Grabowski
2000-02-15Use extents instead of rmaps to handle dvma space.Artur Grabowski
2000-02-09don't include netinet6/ip6.hJun-ichiro itojun Hagino
2000-02-04In cases where we flush a page from the cache and tlb we want toArtur Grabowski
flush it from the tlb _after_ the cache because some cache flushes can reload the tlb. (fixes random coredumps on some cpus).
2000-02-04Some cpus might have the cache running even before we enable it.Artur Grabowski
Uncache the pagetables in pmap_bootstrap4m and recache them again in pmap_cache_enable if we can. This fixes the problems seen on SM30.
2000-02-01Add a bunch of XXX on incorrect handling of pv_flags.Artur Grabowski
None of the noted problems are critical (the code still works), but they can have a severe impact on the performance and they are all really hard to fix in an elegant way.
2000-02-01Opps. Missed two cases of ps_alias_recache.Artur Grabowski
2000-02-01Keep stats of pages we uncache because of bad cache aliases.Artur Grabowski
2000-01-31Clean up memerr*. (from NetBSD)Artur Grabowski
2000-01-27remove bogus commentArtur Grabowski
2000-01-27reduce global symbol pollutionArtur Grabowski
2000-01-27use {round,trunc}_page instead of manually doing the same thingArtur Grabowski
2000-01-27pmap_writetext - use getptep4m/setpgt and avoid one unnecessary table walkArtur Grabowski
2000-01-27It's amazing what you can find with some code reading.Artur Grabowski
In pv_unlink4m we check if the page that was uncached due to bad aliases can be cached again. The check was correct but instead of clearing this flag we clear all other flags except PV_ANC.
2000-01-27zap setptesw4m, not used anymoreArtur Grabowski
2000-01-27zap getptesw4m, not used anymore (and the comment was bogus)Artur Grabowski
2000-01-27update comment to reflect reality (remove it).Artur Grabowski
2000-01-27pmap_rmk4m: There is no need to flush the segment from the tlb orArtur Grabowski
to zero the table when we have 0 mappings in a segment.
2000-01-26cleanup and simplify pmap_page_protect4m:Artur Grabowski
- Use a normal while-loop when walking the pv lists instead of an ugly hack. - always clean the pte even if we know that it will be freed. - No need to flush the segment from the tlb in kernel. - add some ifdef DIAGNOSTIC. - clear the PV_ANC flag on the pv.
2000-01-26Replace more table-walks and getpte4m/setpte4m with getptep4m/setpgt4m.Artur Grabowski
2000-01-22changes to GENERIC must be matched in GENERIC_SCSI3Theo de Raadt
2000-01-17pmap_{zero,copy}_page4m: use getptep4m to get the ptes for vpages.Artur Grabowski
This way we only walk the tables once every boot and we don't have to switch to context 0 on every call.
2000-01-14Cleanup the handling of ptes on 4m.Artur Grabowski
Implement a generic getptep4m that returns the pointer to the pte for (pmap, va). reimplement (and make static __inline) {set,get}ptesw4m and setpte4m using getptep4m. (more on this topic coming soon)
2000-01-14make setpgt4m static __inline, remove dead code.Artur Grabowski
2000-01-12ifdef SUN4M, ok art@David Leonard
2000-01-09Do the bootpath_store dance so we can boot off of a disk attachedmjacob
to an isp controller. This involves rewriting the matched bootpath component for the HBA with the generic 'isp' name so that matching can be done correctly elsewhere. Also use nulling the pointer to f/w to indicate not to load the current f/w set (for older PTI cards) instead of f/w length.
2000-01-09add INET6 into conf/GENERIC.Jun-ichiro itojun Hagino
add pseudo-devices for IPv6 into arch/*/conf/GENERIC. nuke arch/*/conf/GENERIC.v6.
2000-01-03.Sh "SEE ALSO" -> .Sh SEE ALSOPaul Janzen
2000-01-01check db_panic in kdb_trap() as well; similar to i386 fix from assarTheo de Raadt
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-19 18:00:05 +0000