summaryrefslogtreecommitdiffstats
path: root/sys/amd64
Commit message (Expand)AuthorAgeFilesLines
* Go back to the old (icu.s rev.1.7 1993) way of keeping the AST-pendingbde1999-07-1011-126/+32
* Fixed a longstanding scheduling bug. ASTs and softclock interrupts werebde1999-07-102-24/+26
* Implement support for hardware debug registers on the i386.jlemon1999-07-098-8/+198
* These changes appear to give us benefits with both small (32MB) andmckusick1999-07-082-5/+6
* Rename bpfilter to bpf.des1999-07-061-3/+3
* Quieten gcc paranoia.peter1999-07-061-2/+2
* Typo: s/0ff0/0xff0/peter1999-07-061-2/+2
* Implement SA_SIGINFO for i386. Thanks to Bruce Evans for much morecracauer1999-07-063-44/+82
* Add Centaur/IDT WinChip support.green1999-07-061-1/+14
* I made some cleanups, rearranged things a bit, and made AMD Features defaultgreen1999-07-061-47/+52
* Move the initialisation/tuning of nmbclusters from param.c/machdep.cmsmith1999-07-051-7/+1
* Add an extra space to " AMD Features=" to make it line up well.green1999-07-051-2/+2
* K6-III CPUs are now case:d in the appropriate switch; also, ingreen1999-07-051-4/+54
* Some cleanup and rearrangement. hw.physmem is now an absolute quantity;jlemon1999-07-041-121/+98
* The buffer queue mechanism has been reformulated. Instead of havingmckusick1999-07-041-2/+4
* printf int/dev_t (pointer) warningpeter1999-07-031-3/+3
* Improve compatibility with other systems by changing the defaultimp1999-07-034-4/+40
* Fix warnings in last commit (dev_t is not an int, and not even intpeter1999-07-031-4/+5
* Be more informative and try to ask the user in some instances if we can'tphk1999-07-031-8/+18
* An SMP-specific change: Add the lock prefix to RMW operationsalc1999-07-033-5/+13
* Lightly overhaul the memory sizing code again.msmith1999-07-021-80/+127
* Correct some ugly formatting. Remember to initialize the alignment tag.mjacob1999-07-021-40/+35
* Zap totally the npx0 memory size override. It only worked if staticallypeter1999-07-021-14/+2
* Look up the kernel environment for MAXMEM as a final override for thepeter1999-07-011-3/+9
* Move kern_envp and preload initialization a little earlier so that wepeter1999-07-011-9/+8
* Slight reorganization of kernel thread/process creation. Instead of usingpeter1999-07-011-3/+3
* Put on my asbestos suit and attempt to tidy up and add some simple docspeter1999-06-291-118/+108
* Save common_tssd before it's loaded and the busy bit set.luoqi1999-06-281-3/+3
* Use the same -UKERNEL strategy as the alpha to avoid the inlines etc.peter1999-06-281-1/+4
* An SMP-specific change: Remove an unnecessary lock acquire and releasealc1999-06-272-10/+26
* Shut up gcc.peter1999-06-272-2/+6
* Fix warning message; that was 4GB, not 2GB. I apparently can't dojlemon1999-06-241-2/+2
* Explicitly ignore any memory > 2GB, we don't support it yet.jlemon1999-06-241-1/+7
* Only include AMD wt_alloc routines if I586_CPU is defined. Fixesjlemon1999-06-241-2/+2
* This commit gives support for the Rise mP6 CPU. It has two changes:green1999-06-243-12/+26
* Changes in the way that the APs are started appears to have removed themsmith1999-06-233-33/+9
* Do not setup 4M pdir until all APs are up.luoqi1999-06-235-64/+44
* Remove an unnecessary panic when sparse PCI bus numbering is encountered.msmith1999-06-223-9/+3
* Clean up some of the documentation at the top.jkh1999-06-201-8/+12
* Harmless change to prevent possible problems in the future. I madegreen1999-06-181-2/+4
* Changed the global `idt' from an array to a pointer so that npx.cbde1999-06-185-21/+18
* Clean up bitrot in interrupt tracing code.tegge1999-06-161-3/+3
* Kill option FAILSAFE.des1999-06-152-8/+2
* Use pmap_kenter instead of pmap_enter to map the message buffer.alc1999-06-131-3/+2
* Change variable used for calculating ending address of physical memoryjlemon1999-06-101-2/+2
* Use kmem_alloc_nofault() rather than kmem_alloc_pageable() to allocatedt1999-06-081-2/+2
* Fix an accounting problem when prefaulting 4M pages.luoqi1999-06-051-2/+2
* remove references to isa_device, it's no longer associated with interrupts.peter1999-06-031-5/+1
* Unbreak memory sizing for SMP.jlemon1999-06-011-11/+15
* Introduce the makebdev() function, it does the same as the makedev()phk1999-06-011-7/+7
OpenPOWER on IntegriCloud