summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2003-10-14Ok, now that the via c3 crypto stuff is in the tree, merge the last bit ofJason Wright
my cleanups before starting "real work".
2003-10-14Oops, unbrek.Miod Vallat
2003-10-14oops, i did not mean to commit this portion (IPSEC mod), sorryJun-ichiro itojun Hagino
2003-10-14IP6F_OFF_MASK is already endian-flipped; sync w/ kameJun-ichiro itojun Hagino
2003-10-14Add htole and bus_dmamap_sync code. These allow if_vr to work on powerpc.Dale Rahn
Code is a collaboration of jason@ and myself. Ok jason@
2003-10-13rx checksum offload support (based on earlier code from aaron@, but withJason Wright
fixes by me). Note: sk v1 chips can sometimes compute an incorrect cksum. To deal with that, we never set the "bad" flags. Ie. if the card says it's good, we assume it's good. If the card says it's bad, we fall back to software. Ugly? You betcha.
2003-10-13rx checksum offload support (based on the freebsd implementation)Jason Wright
2003-10-13Get rid of the "page table group" stuff. "groups" used to be one logical pageMiod Vallat
big anyways. However, getting rid of the various constants (expanding to 0 for logs and shifts, and 1 for size) allows us to do further simplifications in some pmap loops, as there is no risk anymore of address wrapround. While there, fix some typos and comments.
2003-10-13More m68k pmap tweaks:Miod Vallat
- simplify pmap_protect() - simplify reference count managment by using 0 as a base instead of 1 (from NetBSD)
2003-10-13Bring bge and brgphy more up to date with changes from FreeBSD and NetBSD.Kenneth R Westerback
Some bug fixes, support for new hardware like the 5704 and 5705. Testing by deraadt@, danh@, and drahn@ amoung others. ok deraadt@.
2003-10-13Do not define LEDEBUG by default.Miod Vallat
2003-10-13last vtophys (tx data) is now dead. Needs dma sync's and probably moreJason Wright
htole32 now.
2003-10-12Linux shmat allows lookup of segments that are marked as removed soTodd C. Miller
our Linux compat should too. From marius aamodt eriksen
2003-10-12only one vtophys left (tx data)... both the rx and tx descriptors areJason Wright
fully handled by bus_dma now.
2003-10-12rework rx to use bus_dma and sprinkle with a small amount of htole32.Jason Wright
2003-10-11Better sync comment with reality after last commit.Miod Vallat
2003-10-11Get rid of "union cpupid"; makes <machine/cpus.h> disappear as well.Miod Vallat
2003-10-11Nuke trailing whitespace.Miod Vallat
2003-10-11Page tables can be cached again on 187 and 188, as long as they are cachedMiod Vallat
write through and global (for 188).
2003-10-11Simplify DAE debug display.Miod Vallat
2003-10-11Merge sclock and clock pseudo-devices.Miod Vallat
2003-10-11syncThomas Nordin
2003-10-11ATI Radeon 9100. deraadt@ okThomas Nordin
2003-10-11bump minor gap for svnd devices, so more vnds are allowed.Ted Unangst
this change REQUIRES running MAKEDEV again after booting with a new kernel. with a hint from tom cosgrove. ok deraadt@ millert@
2003-10-10unsmoke drugs in vr_encap() (simplify it quite alot)Jason Wright
prepare for bus_dma of tx data
2003-10-10lightly season with a touch of htole32()... a pinch will do.Jason Wright
2003-10-10another easy vtophys removal (tx descriptor setup)... 6 to go.Jason Wright
2003-10-10don't use vtophys on rx lists any moreJason Wright
2003-10-10make sure pd is initialized before use (or byte counters may increaseDaniel Hartmeier
by random values). ok mcbride@, cedric@, henning@
2003-10-10In pmap_expand_kmap(), be sure to initialize new page tables.Miod Vallat
2003-10-10Calibration code from FreeBSD/NetBSD. Some of the auich's are clocked tooJason Wright
high and sound particularly awful on playback. Thanks to millert and marc for testing (and frantzen for giving me a broken machine).
2003-10-09Typo, iquiry->inquiryMiod Vallat
2003-10-09A few 68060 bugfixes with side effects:Miod Vallat
- honor PRM_CFLUSH correctly for 0[46]0 in pmap_remove_mapping(). - be sure to flush caches in pmap_changebit for 060 too when necessary. - make this compile if option DEBUG is defined.
2003-10-09More m68k common include files factorization.Miod Vallat
2003-10-09Support for VIA VT8237 ATA and SATA from NetBSD.Alexander Yurchenko
Thanks to Anton Komarov for providing access to the VIA board.
2003-10-09remove duplicate ifdef __hppa__ that somehow has gotten in thereMichael Shalayeff
2003-10-09go back to non-raw busdma load as some archs still do not have load_rawMichael Shalayeff
2003-10-09Minor cpp changes to let all kernels in conf/ compile.Miod Vallat
2003-10-09- from freebsd: 795[56] AES supportJason Wright
- don't bother with the master/mmio enable stuff (it's done in pci.c for us) based on suggestion from mickey
2003-10-08Sync VMEChip2 timer usage with mvme68k (i.e. make it slightly more sane).Miod Vallat
Also, use it even when the 1x7 is not a system controller - this does not matter.
2003-10-08randomize return from uvm_map_hint. the random increment is limitedTed Unangst
to prevent fragmentation. this has the effect of randomizing unhinted mmap()s, sysV mem, and position of ld.so. tested on many archs by many developers for quite some time. use of MIN to allow m68k to play from miod@. vax is not included. ok deraadt@ miod@
2003-10-08Fix endian bug in macppc bus_space_set_region_N(). ok miod@Dale Rahn
2003-10-08sync; intel proxgbJason Wright
2003-10-08intel PRO/XGBJason Wright
2003-10-08On 167 and 177 boards, use a VMEChip2 timer for accurate delay() wheneverMiod Vallat
possible, rather than a tight loop.
2003-10-08delay() is actually a void function.Miod Vallat
2003-10-08originally from cb@netbsd.org, adapted by provosNikolay Sturm
itojun@ ok fix a race condition between path resolution in userland and the subsequent namei(): inform the kernel portion of valid filenames and then disallow symlink lookups for those filenames by means of a hook in namei(). with suggestions from provos@ also, add (currently unused) seqnr field to struct systrace_replace, from provos@
2003-10-08obviously i'm on drugs, revertHenning Brauer
2003-10-08missing DIOCX* in the securelevel > 1 caseHenning Brauer
2003-10-08Nivas had identified an 88200 specific bug, where flushes of a page or aMiod Vallat
segment would sometimes not work correctly, and he added a brutal workaround leading to non-optimal, but safe, cache handling. The problem affects 88200 rev 7 and 9, and is believed to affect all 88200 revisions (although 88204 seems to be unaffected). Unfortunately, this problem not only affects cache flushes, but also tlb invalidations. So add a similar workaround in cmmu_flush_remote_tlb().