summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Expand)Author
2010-07-01Synch instead of flush FPU state on APs before suspending.Mike Larkin
2010-07-01Ensure that IPIs have priority and are run first to avoid deadlock. ThisJoel Sing
2010-07-01Typo in comment.Mike Larkin
2010-07-01I accidentally whacked something important. Sorry.Ted Unangst
2010-07-01another day, another compat gets removed. today is ibcs2's turnTed Unangst
2010-07-01- don't do curupte checks if the current miss address is user spaceJasper Lievisse Adriaanse
2010-07-01Backout recent AML changes, breaks on R210 and othersJordan Hargrave
2010-07-01Create a struct to store FP state and include a pointer to the CPU thatJoel Sing
2010-07-01Provide uvm constraints.Joel Sing
2010-07-01Provide safepri.Joel Sing
2010-07-01Fix incorrect function prototype.Joel Sing
2010-07-01Add missing function prototypes.Joel Sing
2010-07-01Provide mtsp() via cpufunc.h.Joel Sing
2010-07-01Pull in atomic.h and intr.h.Joel Sing
2010-07-01Add missing function prototypes and fix order of includes.Joel Sing
2010-07-01Shuffle function prototypes, add missing include and add missingJoel Sing
2010-07-01Add stubs for soft interrupts.Joel Sing
2010-07-01Include atomic.h via lock.h like other architectures do.Joel Sing
2010-07-01Partially sync atomic.h with hppa so that we get real atomic ops and someJoel Sing
2010-07-01mark acpi_savecpu with the gcc returns_twice attribute as suggestedJonathan Gray
2010-07-01Change scsibus(4)'s scsi_link array to an SLIST to save memory onMatthew Dempsky
2010-07-01Add a look-up table for machines that have special vga cards. This table willPaul Irofti
2010-07-01More accurate comment as to what happens with acpi_savecpuMike Larkin
2010-07-01Add things to enable aesni either ifdef'ed or commented out to easeThordur I. Bjornsson
2010-06-30remove a compat_osf1 fileTed Unangst
2010-06-30remove compat_osf1 support from the kernelTed Unangst
2010-06-30Remove the ``never cache virtual addresses in the DVMA range'' inMiod Vallat
2010-06-30amd64_pa_used is weird.Owain Ainsworth
2010-06-30Make this compile after TTYHOG definition change (but the logic should be fixedMiod Vallat
2010-06-30Remove #ifdef CRYPTO. config will do this for us.Thordur I. Bjornsson
2010-06-30get rid of a few c++ comments and a few spacing nitsThordur I. Bjornsson
2010-06-30remove two useless defines.Thordur I. Bjornsson
2010-06-30change a pair of timeout_add calls to timeout_add_{,m}secBret Lambert
2010-06-29make the xcrypt match the jsg code in opensslTheo de Raadt
2010-06-29Add support for mapping ACPI to PCI devicesJordan Hargrave
2010-06-29Add missing include.Joel Sing
2010-06-29Don't use SUBDIR to traverse tags builds, but punt them straight toTheo de Raadt
2010-06-29aesni, a driver for the crypto framework, similar to theThordur I. Bjornsson
2010-06-29In the boot blocks, stop assuming we have a 1:1 mapping of low physicalMiod Vallat
2010-06-29During kernel bootstrap, stop assuming the kernel image has been loaded inMiod Vallat
2010-06-29There is absolutely no need to double map DVMA addresses into the kernel addressMiod Vallat
2010-06-29Rework cypress cpu cache setup, and enable writeback mode. Adapted fromMiod Vallat
2010-06-29fpu_kernel_{enter,exit}; Functions to allow the use ofThordur I. Bjornsson
2010-06-29Shrink struct scsi_sense_data and thus the sense member of scsi_xfer,Kenneth R Westerback
2010-06-29Shave off some instructions:Jasper Lievisse Adriaanse
2010-06-29Remove COMPAT_HPUX. No one wanted to support it and its fewmets werePhilip Guenthe
2010-06-29COMPAT_ULTRIX is not used, time to retire it. ok deraadt miodTed Unangst
2010-06-29Ansify function declarations.Joel Sing
2010-06-29Factor out code used to save and flushe process FPU context.Joel Sing
2010-06-29Store pointer to process FPU state in struct cpu_info.Joel Sing