summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2006-07-12Shorten the dmesg one line.Miod Vallat
2006-07-12Remove commented out defopt goo.Miod Vallat
2006-07-12Fix dependenciesMiod Vallat
2006-07-12Turn GetCPSR() into an inline get_cpsr(), and remove more dead code.Miod Vallat
2006-07-12Orphaned stuff.Miod Vallat
2006-07-12include uvm_extern.h for atop()Martin Reindl
2006-07-12do not include uvm_extern.h twiceMartin Reindl
2006-07-11the scsi_link structure contained a copy of the inquiry flags and the wholeDavid Gwynne
inquiry. this removes the flags member and makes all its users refer to the whole inquiry now. ok miod@ krw@
2006-07-11syncMichael Shalayeff
2006-07-11Remove duplicate prototypes already in <sys/systm.h>Miod Vallat
2006-07-11Hook up the hw.product sysctl on macppc,Gordon Willem Klok
requested by deraadt@, ok drahn@
2006-07-11Remove old debug goo; drahn@ okMiod Vallat
2006-07-11Remove DISKLABEL_ALL, DISKLABEL_ALPHA, leave DISKLABEL_I386 inKenneth R Westerback
GENERIC. Remove now superfluous check for DISKLABEL_ALL. 'whack whack whack!' deraadt@
2006-07-10Do not blindly enable pci io/mem space.Mark Kettenis
ok deraadt@
2006-07-10The setperf mechanism cannot handle multiprocessor systems. Ensure thatGordon Willem Klok
p4tcc and speedstep are not initialized in multiprocessor kernels. ok kettenis@
2006-07-10Accept non-page-aligned addresses and sizes in mapiodev() and unmapiodev().Miod Vallat
2006-07-10Remove the error_cnt cacluation, this only makes sense when the suppliedDale Rahn
clock speed is extrememly accurate and not evenly divisible. Both of which is not true on armish.
2006-07-10As pointed out by miod, cpu_initclocks() can use setstatclockrate(), theDale Rahn
spl() is unnecessary. Also cleanup the extra print and dont bother counting 'tick' interrupts since clock and stat counters exist.
2006-07-10Adding missing $OpenBSD$ tags, pointed out by miod.Dale Rahn
2006-07-09I don't think we're talking about animals in here, soMichael Knudsen
s/wether/whether/g.
2006-07-09DISKLABEL_ALL was never used in hppa, hppa64, mips64 config files soKenneth R Westerback
just eliminate the check to set DISKLABEL_I386 with it. If you want DISKLABEL_I386, use that option.
2006-07-09Remove DLT_* logic from Alpha as well, being careful to keep theKenneth R Westerback
checksumming of the native disklabel. No functional change.
2006-07-0968020-based hp300 systems are not supported anymore now. If you still useMiod Vallat
any, I'll suggest you buy a boat so that they can be recycled as boat anchors.
2006-07-09If running on a 68030 or 68040 system, use the transparent translationMiod Vallat
registers to program an 1GB untranslated window over the physical memory, and use it to define __HAVE_PMAP_DIRECT for faster pool operation.
2006-07-09Make this compile on platforms which define __HAVE_PMAP_DIRECT.Miod Vallat
2006-07-09*enty->*ently where applicableMiod Vallat
2006-07-07Unroll DLT_* logic into simple 'look for native disklabel, then tryKenneth R Westerback
DOS MBR' logic. Leave DOS MBR logic inside #if defined(DISKLABEL_I386) blocks until impact on install media can be evaluated. Eliminates most remnants of cross-endian, cross-architecture disklabel processing attempts. Fix some mips64 issues with last commit, found by kettenis@. No functional change.
2006-07-07Get rid of the fixed iomap for device mappings, and allocated va rangesMiod Vallat
off kernel_map whenever necessary.
2006-07-07ANSIfyGordon Willem Klok
ok deraadt@, brad@, toby@
2006-07-06Make sure we alter ddb variables as ints, not longs; only matters on 64 bitMiod Vallat
platforms.
2006-07-06Make dummy $g0 in ddb the right size.Miod Vallat
2006-07-06PMAP_MD_RWLOW kluge is no longer necessary now that mac68k does TRT.Miod Vallat
2006-07-06Insert an empty page at the beginning of the kernel, so that we can map itMiod Vallat
invalid and have NULL pointer dereferences in the kernel fault now.
2006-07-06Adapt rtclock_intr() to the current interrupt scheme; from NetBSDMiod Vallat
2006-07-05machine is hppa64, not hppa; comment out orphaned wsdisplay at sti;Martin Reindl
2006-07-04use empty INSTALL_STRIP instead of STRIPFLAG to indicate we don't want toMartin Reindl
strip xxboot on install
2006-07-04disable alipm until someone can find this spurious bug, which happens mostTheo de Raadt
often on the x1, and is really bizzare: data error type 32 sfsr=0 sfva=43caa000 afsr=84000000 afva=1fe02004000 tf=0x18e999c0 data fault: pc=1314578 addr=43caa000 sfsr=0 alipm_smb_exec(2389200, 1, 18, 0, 1, 18e99d4e) at alipm_smb_exec+0x258
2006-07-03Repair initialization value of m68060_pcr_init, per 68060 errata I14.Miod Vallat
2006-07-03Eliminate most of the #if/#endif maze in alpha, hppa, hppa64, and mips64Kenneth R Westerback
disklabel code that was intended to support cross-endian and cross-architecture use of disklabels. Never actually used except for Alpha<->Amiga moves that now can't happen anyway. Leave native and DOS MBR disklabel code. Tested on alpha and hppa. No functional change.
2006-07-03Backout premature incomplete 060 stuff.Miod Vallat
2006-07-02get rid of an unused variable.David Gwynne
ok deraadt@
2006-07-02count interrupts (uvmexp).Dale Rahn
2006-07-02Oh what the heck. Enable zyd* even though it does not work yet :)Theo de Raadt
2006-07-01Cleanup dmesg printing.Mark Kettenis
2006-07-01one more dmesg tweakTheo de Raadt
2006-07-01Eliminate redundant get_le() in favour of letoh32(). Add letoh32()Kenneth R Westerback
calls in alpha, hppa, hppa64 and mips63 when accessing the two DOS MBR u_int32_t fields dp_size and dp_start. No functional change.
2006-07-01clean dmesg output for iommu more; tested miod dlgTheo de Raadt
2006-07-01Make probeget() and related code in trap.c #ifdef DDBMiod Vallat
2006-07-01Add PMAP_LITTLE to the list of forbidden pa bits in pmap_kenter_pa().Miod Vallat
2006-07-01Provide our own alloc_attr() routine for sun frame buffer, since the computationMiod Vallat
of highlighted color indexes is slightly different due to different black and white color code (and the fact that the default white is the highlighted white).