summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2003-03-31Protect tdb access w/ spltdb; Patrick LatifiTodd C. Miller
Since bridgeintr_frame() is called at splsoftnet() this isn't a big deal but should still be fixed. jason@ OK.
2003-03-31Avoid using FREEd data when we get a crypto error; Patrick LatifiTodd C. Miller
Also move the session ID reset into the crp_etype == EAGAIN case (noticed by angelos@). OK jason@ and angelos@
2003-03-31Add missing splx; Patrick LatifiTodd C. Miller
jason@ OK
2003-03-31Only delete rule structure when no state refer to it.Cedric Berger
Fix a bunch of issues. Removal of unneeded (r != null) tests coming soon... ok dhartmei@
2003-03-31remove queue ID assigning code. it's dead since some time as we assign queueHenning Brauer
IDs in userland already.
2003-03-30Ensure that all files in /kern end with a newline (machine, model,Bruno Rohee
osrelease and ostype were not). ok tedu@
2003-03-30let this compile with TCPDEBUGTed Unangst
ok millert@ deraadt@
2003-03-30Fix probing of dual channel 7899 with some fixes from current FreeBSDKenneth R Westerback
code, plus an extra ahc_flush_device_writes(). ok deraadt@ tdeval@
2003-03-30Use snprintf() to simplify device name construction, removing nowKenneth R Westerback
superfluous variables and the function 'number()'. ok deraadt@ and millert@
2003-03-30Delete unused local version of strncat.Kenneth R Westerback
ok mjacob@.
2003-03-30As long as I am here pull in some useful changes from NetBSD.Kenneth R Westerback
1) Eliminate some magic numbers. 2) Eliminate some double semi-colons. 3) Rename iha_scsi_req_q to iha_scb.
2003-03-30better MAXSSIZMichael Shalayeff
2003-03-29A few final (I hope) tweaks to the dma fixes.Kenneth R Westerback
1) Return XS_DRIVER_STUFFUP if request sense cannot be loaded. 2) Try to avoid bus_dmamap_unload()'ing maps that have already been unloaded. 3) Try to avoid bus_dmamap_load()'ing maps that are loaded. 4) Be extra paranoid and ensure bus_dmamap_sync()'s are done before a bus_dmamap_unload().
2003-03-29Another chunk of bus_dma fixes. May actually approach correctness now.Kenneth R Westerback
1) Actually map and sync scatter gather list that is also DMA'd. 2) Actually map and sync request sense buffer. 3) Actually POSTREAD/WRITE sync data buffer after I/O, rather than skipping it because the I/O was successful and no chars were left to xfer. 4) Eliminate extra fields, clarify some names, reorder fields to clarify their use. 5) Put common scatter gather preparation logic into separate function. 6) If insertion of request sense command fails, complete active command with XS_SENSE, but zero'd sense data. 7) Eliminate physical address arithmetic as much as possible, centralizing remaining manipulations in code actually sending addresses to adapter.
2003-03-29Very limited support for Promise PDC20376 SATA. Only parallel ATAAlexander Yurchenko
compatibility mode works for now. Based on tests by Jedi/Sector One <j@pureftpd.org>.
2003-03-29Use sc_print_addr() where possible to improve messages.Kenneth R Westerback
Put debug messages in #ifdef sections. Eliminate a less than useful error message.
2003-03-29ubchist is not a fully cooked kadaver and though use the other well formed ↵Michael Shalayeff
pdhist one until ubc gaets back. art@ ok
2003-03-29try to blinkMichael Shalayeff
2003-03-29fetch the device address list for where supported; say 16 addresses is ought ↵Michael Shalayeff
to be enough for every case. dim off the disk led amongst the other lights
2003-03-29add some author refs from the older versions of these from much older projectsMichael Shalayeff
2003-03-29nothing uses UADDRMichael Shalayeff
2003-03-29bus_space_vaddr and related definitions (needs actual implementations)Michael Shalayeff
2003-03-29allow attaching onto phantomas. check graphics id to seed out the medusa ↵Michael Shalayeff
fddi. fix some debugging printfs
2003-03-29fix fpu printout and remove dead code pieceMichael Shalayeff
2003-03-29some mount stinger onto phantomas and there is not need to flag stingers tooMichael Shalayeff
2003-03-29device/vendor id is always < 7 chars. store it in the array and save on the ↵Michael Shalayeff
extra pointer average of 50% space savings. regen the eisadevs
2003-03-29add some more eisa devicesMichael Shalayeff
2003-03-28unbreak after SiS pcidevs changeTodd C. Miller
2003-03-28add support for the i830 that has 8 pci irq lines configure by the extra ↵Michael Shalayeff
0x68 pci config reg. does not change the behaviour for the other chipsets; ho@ testing
2003-03-28a small boot for a cdMichael Shalayeff
2003-03-28s/SADIR/STANDIR/ to avoid conflict w/ the libsa; install raw boot as well as ↵Michael Shalayeff
cooked
2003-03-28regenAlexander Yurchenko
2003-03-28Little consistency in SiS sectionAlexander Yurchenko
ok deraadt@
2003-03-28another const char * from David HillHenning Brauer
2003-03-28Improve debug output a bitAlexander Yurchenko
2003-03-28Fix the color inversion of the cursor (now it's not a solid block of black =)Jason Wright
2003-03-28const char *Henning Brauer
from David Hill <david at phobia.ms>
2003-03-28unbreakHenning Brauer
STRIOCCLONE and SYSTR_CLONE have the same value now ok mickey@
2003-03-28fake off the SYSTR_CLONE through a normally numbered other ioctl; as of ↵Michael Shalayeff
previous rename; after discussion w/ niels
2003-03-28OCHIO* dies horrible death by dd; as per renaming done beforeMichael Shalayeff
2003-03-28zero'd -> zeroed;Jason McIntyre
ok art@, miod@
2003-03-28Ip -> IP;Jason McIntyre
ok miod@
2003-03-28mu-law and A-law spelling;Jason McIntyre
mu-law and u-law are the same, so used mu-law for consistency; ok jason@
2003-03-28more register defns (from linux)Jason Wright
2003-03-28Use the brand new types instead of WSDISPLAY_TYPE_UNKNOWNJason Wright
2003-03-28add a whole bunch of display types (will be used by the accelerated XJason Wright
stuff coming soon to a tree near you).
2003-03-28In ccdsize(), check for CCDF_INITED before doing the ccdopen().Todd C. Miller
Adapted from NetBSD.
2003-03-28Add a sysctl to option USER_LDT, to control its behaviour, which will beMiod Vallat
disabled by default. Enable with sysctl -w machdep.userldt=1 This will allow people to use the few ports that require this functionality to work without requiring the user to recompile a kernel. The option USER_LDT remains in order to not increase size on the installation media. It is now enabled in GENERIC. Per espie@'s idea, but my diff was much simpler than his; ok deraadt@
2003-03-28Fix VIA C3 comment.Tobias Weingartner
Ok deraadt@
2003-03-28If console acceleration was not enabled, do not reinitialize the blitterMiod Vallat
when leaving X11.