summaryrefslogtreecommitdiffstats
path: root/sys/i386
Commit message (Expand)AuthorAgeFilesLines
* Use npxdrop in cpu_thread_exit to save some cycles.davidxu2003-10-191-1/+3
* Add the cpu_idle_hook() function pointer so that other idlers can benjl2003-10-181-9/+17
* Don't forget to load %es with the kernel data segment selector inbde2003-10-162-0/+6
* Add a workaround for the fact that the priv field was removed fromdfr2003-10-161-4/+18
* Get some more data if we hit the pmap_enter() thing.peter2003-10-151-1/+8
* Fix just about as many bugs in my last commit here as there were lines thatpeter2003-10-141-2/+3
* - Modify pmap_is_current() to return FALSE when a pmap's page table is inalc2003-10-131-4/+12
* Initialize CMAP3 to 0phk2003-10-121-0/+1
* Fix a multitude of security bugs in the iBCS2 emulator:tjr2003-10-126-12/+42
* Set page zero read/write right from the start rather than trying topeter2003-10-102-8/+7
* Move the pmap_kenter(KERNBASE, 0) a bit earlier so that it works forpeter2003-10-101-15/+24
* Implement preliminary support for the PT_SYSCALL command to ptrace(2).robert2003-10-091-0/+5
* Move pmap_resident_count() from the MD pmap.h to the MI pmap.h.bms2003-10-061-1/+0
* Don't bother setting a page table page's valid field. It is unused andalc2003-10-051-2/+0
* Fix the apm problem for real. We leave the first 4K page for the bios topeter2003-10-042-5/+6
* - The proper test is CPU_ENABLE_SSE and not CPU_ENABLED_SSE. Thisjeff2003-10-041-1/+1
* Emulate bugs in the old PSE code so that apm works again.peter2003-10-041-0/+5
* Migrate pmap_prefault() into the machine-independent virtual memory layer.alc2003-10-031-98/+19
* Add #include "opt_pmap.h" so locore picks up DISABLE_PSE etc options.peter2003-10-031-0/+1
* Commit Bosko's patch to clean up the PSE/PG_G initialization to andpeter2003-10-0111-133/+145
* - Add a memory barrier before the sse2_pagezero() function returns. Thisjeff2003-10-011-0/+1
* - Hide more #ifdef logic in a new invlcaddr inline. This function flushesjeff2003-10-011-15/+13
* - Define an inline pagezero() to select the appropriate full-page zeroingjeff2003-10-011-16/+20
* - Add ss2_pagezero() for zeroing pages using the movnti instruction. Thisjeff2003-10-012-0/+15
* - Correct a problem with the last commit. The CMAP ptes need to be zeroedjeff2003-10-011-5/+5
* - On my Pentium4-M laptop, invalpg takes ~1100 cycles if the page is found injeff2003-09-304-36/+42
* - Correct a typo in a comment.jeff2003-09-301-1/+1
* MFsio (sio.c 1.413: cleaned up and fixed setting of speeds in comparam()).bde2003-09-271-23/+12
* The present defaults for the open and close for device drivers whichphk2003-09-272-4/+0
* Quick fix for bitrot in locking in the SMP case. cd_getreg() andbde2003-09-271-4/+24
* Addendum to the previous revision: If vm_page_alloc() for the pagealc2003-09-271-5/+10
* Initialize cn_name, ignore cn_dev.phk2003-09-261-1/+1
* - Eliminate the pte object.alc2003-09-253-28/+18
* Add sysentvec->sv_fixlimits() hook so that we can catch cases on 64 bitpeter2003-09-254-16/+8
* Per TRB vote: restore the aquire_timer0 and associated goo. This willimp2003-09-242-2/+150
* Implement the bus_space_map() function to allocate resources and initializenyan2003-09-232-4/+25
* Microoptimization to allow the compiler to evaluate ntohl() etc onpeter2003-09-221-4/+49
* Allocate the page table directory page(s) as "no object" pages. (Thisalc2003-09-211-10/+14
* Reimplement pmap_release() such that it uses the page table rather than thealc2003-09-201-10/+5
* Mention the puc(4) glue driver in a commented-out example so the userjoerg2003-09-191-0/+4
* add Atheros driversam2003-09-191-0/+5
* Fix a busted constant related to PCI configuration acess method #1. Thejhb2003-09-181-1/+1
* Don't forget to reenable interrupts after a breakpoint and trace traps frombde2003-09-181-0/+1
* Add final adjustment code of battery status based on the battery capacityiwasaki2003-09-171-0/+9
* Simplify (and micro-optimize) pmap_unuse_pt(): Only one caller,alc2003-09-131-12/+3
* Add a new parameter to pmap_extract_and_hold() that is needed to eliminatealc2003-09-121-2/+3
* Sort 'bge' correctly.obrien2003-09-101-1/+1
* Whitespace.jhb2003-09-101-2/+2
* We represent PCI intpin's two different ways. One is the way that thejhb2003-09-102-2/+2
* Finish an earlier commit:jhb2003-09-101-3/+11
OpenPOWER on IntegriCloud