FreeBSD source tree https://cgit-dev.freebsd.org/src-test2/atom?h=release%2F10.0.0 2012-10-27T22:54:23Z 2012-10-27T22:54:23Z Neel Natu neel@FreeBSD.org 2012-10-27T22:54:23Z urn:sha1:1ea1e1849014192d01e735715b2d3eecebde8f82 always present. Suggested by: grehan Obtained from: NetApp 2012-01-04T17:01:12Z Sean Bruno sbruno@FreeBSD.org 2012-01-04T17:01:12Z urn:sha1:fe5476a6e7f318b2ac8b0fbceb2808aff76af285 Reviewed by: grehan@ 2011-11-07T19:53:20Z Peter Grehan grehan@FreeBSD.org 2011-11-07T19:53:20Z urn:sha1:f9f4638d7a4cefcd45a5d70e536f00e9b2154520 Obtained from: NetApp 2011-09-15T22:14:35Z Peter Grehan grehan@FreeBSD.org 2011-09-15T22:14:35Z urn:sha1:fab4c373af4f465f03ae217f491d8cf3a3eeb26d sys/dev/bvm/bvm_console.c - move up to the new alt-break order. 2011-05-19T21:53:25Z Peter Grehan grehan@FreeBSD.org 2011-05-19T21:53:25Z urn:sha1:1f3025e13378bfb7be746e15743c2050fae45a93 run as a 1/2 CPU guest on an 8.1 bhyve host. bhyve/inout.c inout.h fbsdrun.c - Rather than exiting on accesses to unhandled i/o ports, emulate hardware by returning -1 on reads and ignoring writes to unhandled ports. Support the previous mode by allowing a 'strict' parameter to be set from the command line. The 8.1 guest kernel was vastly cut down from GENERIC and had no ISA devices. Booting GENERIC exposes a massive amount of random touching of i/o ports (hello syscons/vga/atkbdc). bhyve/consport.c dev/bvm/bvm_console.c - implement a simplistic signature for the bvm console by returning 'bv' for an inw on the port. Also, set the priority of the console to CN_REMOTE if the signature was returned. This works better in an environment where multiple consoles are in the kernel (hello syscons) bhyve/rtc.c - return 0 for the access to RTC_EQUIPMENT (yes, you syscons) amd64/vmm/x86.c x86.h - hide a bunch more CPUID leaf 1 bits from the guest to prevent cpufreq drivers from probing. The next step will be to move CPUID handling completely into user-space. This will allow the full spectrum of changes from presenting a lowest-common-denominator CPU type/feature set, to exposing (almost) everything that the host can support. Reviewed by: neel Obtained from: NetApp 2011-05-14T18:37:24Z Peter Grehan grehan@FreeBSD.org 2011-05-14T18:37:24Z urn:sha1:6c4c7d0f96c632ef104b202a633a06db33f74d6b This branch is now considered frozen: future bhyve development will take place in a branch off -CURRENT. sys/dev/bvm/bvm_console.c sys/dev/bvm/bvm_dbg.c - simple console driver/gdb debug port used for bringup. supported by user-space bhyve executable sys/conf/options.amd64 sys/amd64/amd64/minidump_machdep.c - allow NKPT to be set in the kernel config file sys/amd64/conf/GENERIC - mptable config options; bhyve user-space executable creates an mptable with number of CPUs, and optional vendor extension - add bvm console/debug - set NKPT to 512 to allow loading of large RAM disks from the loader - include kdb/gdb sys/amd64/amd64/local_apic.c sys/amd64/amd64/apic_vector.S sys/amd64/include/specialreg.h - if x2apic mode available, use MSRs to access the local APIC, otherwise fall back to 'classic' MMIO mode sys/amd64/amd64/mp_machdep.c - support AP spinup on CPU models that don't have real-mode support by overwriting the real-mode page with a message that supplies the bhyve user-space executable with enough information to start the AP directly in 64-bit mode. sys/amd64/amd64/vm_machdep.c - insert pause statements into cpu shutdown busy-wait loops sys/dev/blackhole/blackhole.c sys/modules/blackhole/Makefile - boot-time loadable module that claims all PCI bus/slot/funcs specified in an env var that are to be used for PCI passthrough sys/amd64/amd64/intr_machdep.c - allow round-robin assignment of device interrupts to CPUs to be disabled from the loader sys/amd64/include/bus.h - convert string ins/outs instructions to loops of individual in/out since bhyve doesn't support these yet sys/kern/subr_bus.c - if the device was no created with a fixed devclass, then remove it's association with the devclass it was associated with during probe. Otherwise, new drivers do not get a chance to probe/attach since the device will stay married to the first driver that it probed successfully but failed to attach. Sponsored by: NetApp, Inc.