summaryrefslogtreecommitdiffstats
path: root/sys/amd64
Commit message (Expand)AuthorAgeFilesLines
* This mega-commit is meant to fix numerous interrelated problems. Theredyson1998-03-072-6/+4
* The APs now reload the interrupt descriptor table pointer aftertegge1998-03-074-4/+21
* Remove special handling for resuming clock interrupt when using APIC_IO.tegge1998-03-055-37/+18
* Use t_idt instead of idt inside setidt() if f00f_hack() has relocated the IDT.tegge1998-03-051-2/+7
* Defined CCR6 and CCR7 (configuration registers of M2 CPU.)kato1998-03-041-1/+5
* Reviewed by: msmith, bde long agodufault1998-03-042-6/+26
* When entering the apic version of slow interrupt handler, leveltegge1998-03-038-69/+544
* Forward the signal if the process runs on a different CPU. This reducestegge1998-03-036-17/+174
* Reduce timeout before assuming that forwarding of hardclock or softclocktegge1998-03-033-21/+33
* forward_statclock and forward_hardclock are located in mp_machdep.c.tegge1998-03-031-3/+5
* Update the ELF image activator to use some of the exec resources ratherpeter1998-03-021-2/+2
* 1) Use a more consistent page wait methodology.dyson1998-03-014-39/+17
* Prevent the TSC from being used on APM machines, we have no idea ifphk1998-02-282-4/+12
* Fix page prezeroing for SMP, and fix some potential paging-in-progressdyson1998-02-252-32/+38
* Removed vestiges of previous microtime() implementation.bde1998-02-251-6/+2
* Try to dynamically size the VM_KMEM_SIZE (but is still able to be overriddendyson1998-02-231-2/+18
* Quick fix for the i8254 timecounter often gaining 10 msec.bde1998-02-232-2/+4
* Add missing CLOCK_UNLOCK() before write_eflags().jkh1998-02-212-2/+4
* Replace TOD clock code with more systematic approach.phk1998-02-203-201/+231
* Removed unused #includes.bde1998-02-201-2/+1
* Remove DISABLE_PSE option which was masking (but not fixing) the problem.msmith1998-02-161-3/+1
* TEMPORARILY disable support for the 4MB kernel page, as it appears to bemsmith1998-02-161-1/+3
* Removed a superstitious fnop() that broke the usefulness of the FPU'sbde1998-02-152-4/+2
* Use RDMSR instruction instead of WRMSR.kato1998-02-131-16/+16
* Ifdefed SMP-only declarations.bde1998-02-131-3/+8
* Update timer0_prescaler_count before calling hardclock() while timer0bde1998-02-132-30/+44
* Ifdefed some npx code. npx should be optional again.bde1998-02-131-1/+3
* Fixed missing privilege checking and off-by-1 bounds checking inbde1998-02-131-6/+9
* Fixed initialization of the 4MB page. Kernels larger than about 2.75MBbde1998-02-121-2/+2
* Only use the i586-optimized copying and zeroing functions if they arebde1998-02-122-6/+76
* Fix warning after previous staticization.eivind1998-02-101-2/+2
* Staticize.eivind1998-02-0910-44/+50
* Remove warnings from f00f_hack.eivind1998-02-091-5/+5
* Back out DIAGNOSTIC changes.eivind1998-02-068-18/+8
* 1) Start using a cleaner and more consistant page allocator insteaddyson1998-02-052-122/+147
* Turn DIAGNOSTIC into a new-style option.eivind1998-02-048-8/+18
* Make FAILSAFE a new-style option.eivind1998-02-041-1/+2
* Converted DISABLE_PSE to a new-style option.bde1998-02-031-1/+3
* Ifdefed some SMP and VM86 code. Note that although VM86 is not a globalbde1998-02-032-2/+21
* Forward declare a union so that this file is self-sufficient.bde1998-02-031-11/+10
* Ifdefed use of a GNU feature.bde1998-02-031-1/+3
* Fixed disordering of busdma* and swi_vm.bde1998-02-011-4/+4
* Fixed a recently broken comment.bde1998-02-011-4/+4
* Declare printf() instead of including <stdio.h>, so that this doesn'tbde1998-02-011-4/+3
* Change the busy page mgmt, so that when pages are freed, theydyson1998-01-311-7/+21
* Make the debug options new-style.eivind1998-01-312-2/+4
* Make POWERFAIL_NMI, PPS_SYNC and NATM new style options.eivind1998-01-311-1/+2
* Skip probing devices which have already probed true.eivind1998-01-311-1/+13
* Include "opt_nfs.h"eivind1998-01-312-2/+4
* Retire LFS.phk1998-01-301-20/+1
OpenPOWER on IntegriCloud