summaryrefslogtreecommitdiff
path: root/sys/arch/mips64
AgeCommit message (Collapse)Author
2004-12-06Use uvm_grow() to account for stack growth, rather than home-grown codeMiod Vallat
or nothing. Inspired by a similar recent change in NetBSD.
2004-11-28MAXSLP is not really an MD-configurable define so move it to param.h; miod@ ↵Michael Shalayeff
testing
2004-11-26Kill __HAVE_EVCOUNT, now that all architectures provide them.Miod Vallat
2004-11-11Somewhat improve single-stepping.Mark Kettenis
ok pefo@, miod@
2004-11-11say hello to XKSEG0 and XKSEG1!Per Fogelstrom
2004-11-11up default data size so things doesn't blow up to easyPer Fogelstrom
2004-11-09Fix single-stepping.Mark Kettenis
ok miod@
2004-11-08TypoMiod Vallat
2004-11-02Fix problem with floating point restore at sigreturn.Per Fogelstrom
Regress lib/libpthread/preemption_float now passes.
2004-11-02missed reg index calc fixPer Fogelstrom
2004-10-30remove debugPer Fogelstrom
2004-10-30Make the default boot partition p.Per Fogelstrom
Migrate float exception code to 64 bit. Add a few new "uniplemented".
2004-10-20Fix some 64 bit address problems.Per Fogelstrom
Some function names made more unique. Other changes for the upcoming Origin 200 support.
2004-10-18Move exec.h out from sgi to share it with other mips64 ports; withAlexander Yurchenko
a tip from miod@. ok pefo@ miod@
2004-10-08Fix DDB/DEBUG ifdefs dance.Alexander Yurchenko
Help and ok from miod, ok pefo@
2004-10-08Compile netisr code only for network-capable kernels.Alexander Yurchenko
ok pefo@ miod@
2004-10-08trapdebug_enter() really takes 2 args, not 1.Alexander Yurchenko
ok pefo@ miod@
2004-10-08disksubr.c should depend on disk.Alexander Yurchenko
ok pefo@ miod@
2004-10-05some more typesPer Fogelstrom
2004-09-30put back ifdef PMAP_DEBUG on VAC warn. don't confuse ppl.Per Fogelstrom
2004-09-29move cache syncing for now. delayed sync needs more debugPer Fogelstrom
2004-09-29fix typoPer Fogelstrom
2004-09-29Better RM7K cache init dealing with lazy firmwarePer Fogelstrom
2004-09-27R5K seems to have bugs/hazardsPer Fogelstrom
2004-09-27Rewrite parts of the interrupt system to achive:Per Fogelstrom
o Remove do_pending code and take a real int instead. The performance impact seems to be very low and it simplifies the code considerably. o Allow interrupt nesting at first level. Run softints with HW ints enabled.
2004-09-27move ITLBNOPFIX to asm.h and make it conditionalPer Fogelstrom
2004-09-27reorder slightly to avoid asm warningsPer Fogelstrom
2004-09-27cleanPer Fogelstrom
2004-09-27rearrange code to avoid asm warningsPer Fogelstrom
2004-09-27Change busdma to map coherent dmamem memory as uncached if the hostPer Fogelstrom
system has non-coherent caches. This will help some drivers to work better.
2004-09-24new style interrupt counters; pefo okTheo de Raadt
2004-09-23Clean clean clean... should be OK nowPer Fogelstrom
2004-09-23More pmap bugs cleaned out. Some old, some new.Per Fogelstrom
Better structured pmap.c. Evil still in there, more work to do.
2004-09-22Add systrace support.Miod Vallat
2004-09-22Partially revert 1.9, the new mod/ref accounting has issues with mmap().Miod Vallat
2004-09-21trap() should return a 64 bit value now.Miod Vallat
2004-09-21Make RM7K get L3 cache from config reg. cleanupPer Fogelstrom
2004-09-21Previous commit removed proper caching when growing segmaps by mistake;Miod Vallat
spotted by pefo@
2004-09-21Nuke commons.Miod Vallat
2004-09-20Move <machine/pte.h> to mips64 land.Miod Vallat
2004-09-20There was one CF_5_xx constant left in the 52xx code, change it to CF_52_xxMiod Vallat
(purely cosmetic, as these constant share the same values).
2004-09-20Some cleanups for RM52x0 cpus.Per Fogelstrom
2004-09-20Add support for R10K cpu classPer Fogelstrom
2004-09-20R10K cpu class cache supportPer Fogelstrom
2004-09-19nop missing in bd slotPer Fogelstrom
2004-09-17In vmapbuf(), pass VM_PROT_READ | VM_PROT_WRITE as flags hints too.Miod Vallat
2004-09-17Now that we can check for pmap_extract() failure properly, do so.Miod Vallat
2004-09-17Long-due mips pmap potpourri:Miod Vallat
- un-__P()ee. - remove splimp() protection. - allow pmap_extract() to return FALSE, dammit! XXX some code under arch/mips64 considers pa == 0 as failure, instead of checking the return value of pmap_extract(). Free commits. - do not peek at uvm vm_page flags for ref/mod accounting, but use real MD flags in pv lists, and implement pmap_is_* and pmap_clear_*, so uvm (which knows better) does the right thing. - debug code #ifdef PMAPDEBUG, not #ifdef DIAGNOSTIC. - merge pmap_init() and pmap_pinit(), pmap_destroy() and pmap_release(), and update comments. - better pmap_enter() logic, from NetBSD. - do not use IS_VM_PHYSADDR() to know if we can use PHYS_TO_VM_PAGE() unchecked. Instead, compute PHYS_TO_VM_PAGE(), and check it against NULL. Saves a vm_physseg_find() call each time. - pass vm_page_t to internal function whenever possible, rather than paddr_t, so that PHYS_TO_VM_PAGE() needs not be computed again. - in the same way, do not compute PHYS_TO_KSEG0(foo) several times in a row (though PHYS_TO_KSEG0 is trivial). - allow userspace address to cross 0x80000000 (may be useful in the future). - implement PMAP_CANFAIL. As a bonus: - switch from HAVE_PMAP_PHYSSEG to HAVE_VM_PAGE_MD. - remove dead (pre-pmap_pv_pool) code. - KNF fixes.
2004-09-17Crank MSIZE and NMBCLUSTERS, per other 64bit arches.Miod Vallat
2004-09-17Properly setup register ra in sendsig() in 64bit world.Miod Vallat
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-27 21:15:29 +0000