summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2001-08-11GC some unused malloc types.Artur Grabowski
2001-08-11remove some old cruft.Artur Grabowski
2001-08-11swapmap is gone forever.Artur Grabowski
2001-08-11Make m18x_cmmu.c only compiled if support for 187 or 188 is required.Miod Vallat
2001-08-11Use more sensible malloc type.Artur Grabowski
2001-08-11Include files serve a purpose. Move lots of extern statements fromMiod Vallat
foo_cmmu.c to cmmu.h ; also don't check for 197 in m18x_cmmu.c, as this code is 18x specific.
2001-08-11Make iomap_mapout() a void function, since its return value is always 1Miod Vallat
and never used anyway. While there, beautify the extent_alloc() call.
2001-08-10allocate tx dma maps in attach instead of creating/destroying them for each txJason Wright
2001-08-10convert V->P table memory allocations to pool.Dale Rahn
2001-08-10- make transmit use linked lists not indexed arrays (more like it was when I ↵Jason Wright
started) - also, only bus_dmamap_unload() loaded maps in fxp_stop()
2001-08-09- scb_wait() is now time based (not cpu speed based)Jason Wright
- rearrange multicast setup to be part of init - convert cu operations to bus_dma(9) (still have to do RFA ops).
2001-08-09Use a startup hook for the rest of the chip initialization.Hakan Olsson
2001-08-09Don't check the source address on the packet vs. the one on the SA, asAngelos D. Keromytis
this prevents use of ESP in mobility; pointed out on the IETF mailing list by Francis Dupont.
2001-08-09typo: ALTQ_CQB -> ALTQ_CBQKenjiro Cho
2001-08-09Force alignment of blocks so that we don't crash on strict alignmentHeikki Korpela
archs. This is a stopgap until we get a rijndael implementation that doesn't assume 4 byte alignment. ok deraadt@
2001-08-09Unshare signal actions on exec, from NetBSD. While I am here a little KNF ↵Niklas Hallqvist
nit too
2001-08-09change a maze of altq options into just "altq" for the base+red+cbq, andTheo de Raadt
then altq_* for each of the other * experimental options. and.. enable it by default in GENERIC.
2001-08-09Correct linux-compat sigaltstack implementation.Niklas Hallqvist
Helps HotSpot Java VMs a bit, but not totally. I can at least run javac on J2SE 1.4beta with good results, but anything that uses swing seems to fail on something else.
2001-08-09syncTheo de Raadt
2001-08-09yet another damn winmodemTheo de Raadt
2001-08-08New timeouts.Miod Vallat
2001-08-08New timeouts, drahn@ sez go ahead.Miod Vallat
2001-08-08New timeouts. I've been running this since almost a whole year.Miod Vallat
2001-08-08New timeouts.Miod Vallat
if_gm tested by matthieu@, the whole ok'd by drahn@
2001-08-08New timeouts.Miod Vallat
2001-08-08move the com_raisedtr proto for timeout_setMichael Shalayeff
2001-08-08vga_{isa,pci}.h -> vga.h since the former no longer exists.Todd C. Miller
2001-08-08syncTheo de Raadt
2001-08-08IPCOMP removal. deraadt@ ok.Jean-Jacques Bernard-Gundol
2001-08-08remove IPCOMP. derradt@ ok.Jean-Jacques Bernard-Gundol
2001-08-08new stuffTheo de Raadt
2001-08-08initialize timeouts, since it's a copy of comattach; reported by Larry Moore ↵Michael Shalayeff
<lmoore@starwon.com.au>
2001-08-08remove IPCOMP option. deraadt@ ok.Jean-Jacques Bernard-Gundol
2001-08-08Remove IPCOMP option, it's now part of IPSEC option. You still need toJean-Jacques Bernard-Gundol
enable ipcomp via sysctl to use it. deraadt@ ok.
2001-08-08__dead doesn't work on function pointers.Artur Grabowski
2001-08-08leave the dma engines polling for a bit after the last descriptor has beenJason Wright
passed through. After a few seconds (5 currently) of no activity, stop them completely.
2001-08-08Initial support for RC4 operations (This probably won't work yet asJason Wright
the caller of *_process() doesn't seem to initialize crd_klen correctly; it has a zero value... still debugging...).
2001-08-08__dead not __noreturn__Todd C. Miller
2001-08-08nuke __attribute__((__noreturn__)) when used in conjunction with __deadTodd C. Miller
2001-08-08nuke __attribute__((__noreturn__)) when used in conjunction with __deadTodd C. Miller
2001-08-08Fix broken logic in wait flags passed to extent_alloc.Artur Grabowski
2001-08-07Not used.Miod Vallat
2001-08-07One more typo.Miod Vallat
There are probably lots of free commits left for spelling wizards out there.
2001-08-07Change tsleep into ltsleep.Artur Grabowski
ltsleep takes an additional argument - a simplelock and unlocks it when it's safe to do so. tsleep now becomes a wrapper around ltsleep. From NetBSD
2001-08-07Sprinkle a few Holy Volatiles (tm).Miod Vallat
2001-08-07Make config(8) choose which source files to exclude, from the kernelMiod Vallat
configuration file options, rather than putting ugly preprocessor statements and compile empty files.
2001-08-07For gcc >= 2.5 #define __dead and __pure using the __noreturn__Todd C. Miller
and __const__ attributes.
2001-08-07We now have predict_{true,false}Artur Grabowski
2001-08-07tag fixes; Denis Afonin <dfa@solo.ee>Jason Wright
2001-08-07Add a bunch of macros from NetBSD.Artur Grabowski
__GNUC_PREREQ__(minor,major) evaluates to true if use gcc that has the version major.minor or newer. __predict_{true,false} are there to make use of the new branch prediction hint mechanisms in gcc 3.0.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-25 17:45:43 +0000