| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Stop lying about supporting cpu_est_clockrate() when TSC is invariant. This | jkim | 2010-12-14 | 2 | -3/+18 |
| * | In fpudna()/npxdna(), mark FPU context initialized and optionally | kib | 2010-12-12 | 1 | -1/+3 |
| * | Reduce the Xen timecounter from 1GHz to 2^-9 GHz, thereby increasing the | cperciva | 2010-12-11 | 1 | -2/+3 |
| * | Make the machdep.independent_wallclock sysctl do what it says on the box. | cperciva | 2010-12-11 | 1 | -1/+2 |
| * | When r207410 eliminated the acquisition and release of the page queues | alc | 2010-12-09 | 1 | -61/+40 |
| * | Replace i386/i386/busdma_machdep.c and amd64/amd64/busdma_machdep.c | cperciva | 2010-12-09 | 1 | -1222/+0 |
| * | Do not subtract 0.5% from estimated frequency if DELAY(9) is driven by TSC. | jkim | 2010-12-08 | 1 | -3/+7 |
| * | On amd64, we have (since r1.72, in December 2005) MAX_BPAGES=8192, | cperciva | 2010-12-08 | 1 | -0/+4 |
| * | Merge sys/amd64/amd64/tsc.c and sys/i386/i386/tsc.c and move to sys/x86/x86. | jkim | 2010-12-08 | 1 | -281/+0 |
| * | Use int for 'tsc_present' instead of u_int. It is just a boolean. | jkim | 2010-12-07 | 2 | -2/+2 |
| * | Remove stale comments about P-state invariant TSC and fix style(9) nits. | jkim | 2010-12-07 | 2 | -8/+4 |
| * | Do not register a event handler for CPU freqency changes when it is found | jkim | 2010-12-07 | 1 | -4/+16 |
| * | Now the P-state invariant TSC is probed early enough, do not register event | jkim | 2010-12-07 | 1 | -10/+8 |
| * | Probe P-state invariant TSC from rightful place. | jkim | 2010-12-07 | 2 | -22/+23 |
| * | MFamd64 r204214: Enforce stronger alignment semantics (require that the | cperciva | 2010-12-05 | 1 | -12/+18 |
| * | Remove gratuitous i386/amd64 inconsistency in favour of the less verbose | cperciva | 2010-12-04 | 1 | -2/+1 |
| * | Remove unnecessary #includes which seem to have been accidentally added | cperciva | 2010-12-04 | 1 | -3/+0 |
| * | Revert r216161. It is not necessary because we zero-fill BSS anyway. | jkim | 2010-12-03 | 1 | -1/+1 |
| * | Explicitly initialize TSC frequency. To calibrate TSC frequency, we use | jkim | 2010-12-03 | 1 | -1/+1 |
| * | Do not change CPU ticker frequency if TSC is P-state invariant. Note this | jkim | 2010-12-03 | 1 | -0/+3 |
| * | Revert r216134. This checkin broke platforms where bus_space are macros: | brucec | 2010-12-03 | 1 | -24/+12 |
| * | Disallow passing in a count of zero bytes to the bus_space(9) functions. | brucec | 2010-12-02 | 1 | -12/+24 |
| * | Fix bug introduced by r194784: Under XEN, the page(s) allocated to dpcpu | cperciva | 2010-11-29 | 1 | -0/+2 |
| * | Calling fill_fpregs() for curthread is legitimate, and ELF coredump | kib | 2010-11-28 | 1 | -1/+2 |
| * | Remove npxgetregs(), npxsetregs(), fpugetregs() and fpusetregs() | kib | 2010-11-26 | 3 | -150/+48 |
| * | Merge amd64/i386 _align.h by aligning on the size of register_t (copied | tijl | 2010-11-26 | 1 | -44/+3 |
| * | Remove kernel support for BB profiling, now that kernbb(8) is gone, too. | uqs | 2010-11-26 | 1 | -20/+0 |
| * | Revert r215819 and fix the bug properly. In pmap_qremove, paging table | cperciva | 2010-11-25 | 1 | -12/+1 |
| * | Use unambiguous inline assembly to load a float variable. GNU as | dim | 2010-11-25 | 1 | -1/+1 |
| * | Add device IDs for two more ServerWorks Host-PCI bridges so that we can | jhb | 2010-11-25 | 1 | -0/+5 |
| * | Work around paging bug. Somehow we seem to be ending up with entries in | cperciva | 2010-11-25 | 1 | -0/+12 |
| * | Rename HYPERVISOR_multicall (which performs the multicall hypercall) to | cperciva | 2010-11-25 | 2 | -1/+23 |
| * | Remove vestigal debugging code which, in fork-heavy workloads, can cause | cperciva | 2010-11-25 | 1 | -1/+0 |
| * | Remove a stale tunable introduced in r215703. | jkim | 2010-11-23 | 1 | -2/+1 |
| * | specialreg.h: add definitions for some useful bits found in CPUID.6 EAX and ECX | avg | 2010-11-23 | 1 | -0/+9 |
| * | - Disable caches and flush caches/TLBs when we update PAT as we do for MTRR. | jkim | 2010-11-22 | 1 | -92/+70 |
| * | In xen_get_timecount, return the full ns-precision time rather than | cperciva | 2010-11-22 | 1 | -1/+1 |
| * | Unifdef XEN. This file is only compiled with the XEN kernel option set, | cperciva | 2010-11-20 | 1 | -103/+2 |
| * | Add VTOM(va) macro as xpmap_ptom(VTOP(va)) to convert to machine addresses. | cperciva | 2010-11-20 | 4 | -35/+34 |
| * | Make pmap_release consistent with pmap_pinit with respect to unpinning | cperciva | 2010-11-19 | 1 | -0/+5 |
| * | specialreg.h: add definitions for MPERF/APERF pair of MSRs | avg | 2010-11-19 | 1 | -0/+2 |
| * | specialreg.h: add AMD-specific "Hardware Configuration Register" MSR | avg | 2010-11-19 | 2 | -2/+3 |
| * | specialreg.h: add definition for AMD Core Performance Boost bit | avg | 2010-11-19 | 1 | -0/+1 |
| * | Make pmap_release match pmap_pinit by invoking pmap_qremove(pmap->pm_pdpt) | cperciva | 2010-11-18 | 1 | -0/+3 |
| * | Don't KASSERT in pmap_release that | cperciva | 2010-11-18 | 1 | -2/+3 |
| * | Restore CR0 after MTRR initialization for correctness sakes. There will be | jkim | 2010-11-16 | 1 | -10/+8 |
| * | Invalidate TLBs explicitly. r1.4 of sys/i386/i386/i686_mem.c removed this | jkim | 2010-11-16 | 1 | -1/+3 |
| * | Do not use __FreeBSD_version prefix for the special osrel version. | kib | 2010-11-14 | 1 | -2/+1 |
| * | Use symbolic names instead of hardcoding values for magic p_osrel constants. | kib | 2010-11-14 | 1 | -3/+3 |
| * | MFamd64: (based on) r209957 | jkim | 2010-11-12 | 4 | -43/+1 |
