summaryrefslogtreecommitdiffstats
path: root/sys/amd64/amd64/mp_machdep.c
Commit message (Expand)AuthorAgeFilesLines
...
* Small comment nit: "run time" -> "run-time".sobomax2009-03-081-1/+1
* Fix typo in comments in r189023.sobomax2009-02-251-1/+1
* Make machdep.hyperthreading_enabled tunable working with the SCHED_ULE.sobomax2009-02-251-9/+43
* Improve robustness of NMI handling, for NMIs recognized in kerneljkoshy2009-02-031-0/+10
* - Allocate apic vectors on a per-cpu basis. This allows us to allocatejeff2009-01-291-0/+3
* Introduce cpu_vendor_id and replace a lot of strcmp(cpu_vendor, "...").jkim2008-11-261-2/+2
* Remove ipi_all() and ipi_self() as the former hasn't been used atmarius2008-09-281-30/+0
* The pcb_gs32p should be per-cpu, not per-thread pointer. This iskib2008-09-081-0/+1
* Provide private per-CPU GDTs on amd64. This is required at least for thekib2008-09-081-4/+11
* Band-aid a problem with 32 bit selector setup.peter2008-07-091-1/+5
* Eliminate an unused declaration. (In fact, the declaration is bogusalc2008-07-041-1/+0
* - Rather than repeating the same preemption code everywhere call the schedulerjeff2008-03-101-9/+2
* - Remove the old smp cpu topology specification with a new, more flexiblejeff2008-03-021-41/+30
* Move mp_topology() from apic_init(i386) and apic_setup_local(amd64) topeter2007-08-021-14/+6
* Commit 10/14 of sched_lock decomposition.jeff2007-06-041-23/+4
* Remove extern struct pcpu __pcpu[]; from the header file andkan2007-05-191-0/+2
* Handle CPUs with APIC IDs higher than 32 (at least one IBM server usesjhb2007-05-081-30/+57
* Change the x86 interrupt code to use FreeBSD CPU IDs (i.e. PCPU_GET(cpuid))jhb2007-03-061-5/+6
* Acquiring smp_ipi_mtx on every call to pmap_invalidate_*() is wasteful.alc2007-03-051-2/+8
* Don't send interrupts to CPUs disabled via lapic hints.jhb2007-02-081-0/+2
* - Allow the schedulers to IPI_PREEMPT idlethread. This puts the decisionjeff2007-01-231-8/+5
* - Use the correct test in the ipi bitmask handler for IPI_PREEMPT so thatjeff2007-01-111-4/+2
* Change the x86 interrupt code to suspend/resume interrupt controllersjhb2006-10-101-1/+1
* Actually hook up the IPI_INVLCACHE IDT vectors backingjhb2006-09-111-1/+4
* Since DELAY() was moved, most <machine/clock.h> #includes have beenphk2006-05-161-1/+0
* Add various constants for the PAT MSR and the PAT PTE and PDE flags.jhb2006-05-011-0/+3
* Add a new 'pmap_invalidate_cache()' to flush the CPU caches via thejhb2006-05-011-0/+8
* Check if reported HTT cores are physical cores. This commit does notjkim2006-04-251-0/+8
* Adjust dangerous-shared-cache-detection logic from "all shared datacperciva2006-04-241-2/+2
* Enable global pages TLB extension on Application Processors.ups2006-03-181-0/+7
* Flip the switch and don't route interrupts to hyperthreads in a HT system.jhb2006-03-091-2/+2
* Rework how we wire up interrupt sources to CPUs:jhb2006-02-281-22/+18
* Modify the way we account for CPU time spent (step 1)phk2006-02-071-1/+1
* Tweak how the MD code calls the fooclock() methods some. Instead ofjhb2005-12-221-1/+1
* MFi386:jhb2005-12-081-6/+16
* Revert a part of the previous commits to these files that made the NMIjhb2005-11-161-5/+4
* Rename the KDB_STOP_NMI kernel option to STOP_NMI and make it apply to alljhb2005-10-241-12/+47
* - Various small whitespace and style nits.jhb2005-10-241-15/+10
* Avoid code duplication and implement bitcount32() function in systm.h only.pjd2005-08-191-53/+1
* Actually create the double fault stack page for AP cpus so that we have apeter2005-07-211-1/+6
* Move the KDB_STOP_NMI option from opt_global.h to opt_kdb.hpeter2005-06-291-0/+1
* Add IPI support for preempting a thread on another CPU.ups2005-06-091-0/+16
* Sync the style of these two files.obrien2005-05-231-11/+15
* Adjust the start_ap delay to match i386.obrien2005-05-201-1/+1
* Fix mismerge in rev 1.226: wait 5 seconds as the comment documents,obrien2005-05-201-1/+1
* Default hyperthreading on in -CURRENT. No seatbelts in CURRENT (^_^)nectar2005-05-131-1/+1
* Add a knob for disabling/enabling HTT, "machdep.hyperthreading_allowed".nectar2005-05-131-0/+101
* Implement an alternate method to stop CPUs when entering DDB. Normally we usedwhite2005-04-301-0/+74
* Fully initialize the required TSS fields so that the io permissioncperciva2005-04-061-0/+1
* Divorce critical sections from spinlocks. Critical sections as denoted byjhb2005-04-041-0/+16
OpenPOWER on IntegriCloud