Hi, This series is a revamp of the series John Cooper sent at Message-ID: <4ddad592.3020...@redhat.com> (http://marc.info/?l=qemu-devel&m=130618852625770).
The results of applying this series is, bit-by-bit exactly the same of applying the previous series from John. All I did was to rewrite Subject lines and split one large patch changing CPU flags into small steps, so the changes can be more easily reviewed/discussed. The proof can be checked by looking at the 'cpudefs-v1-john' and 'cpudefs-v2-ehabkost' branches at <git://github.com/ehabkost/qemu-hacks.git>. Eduardo Habkost (5): cpu defs: use Intel flag names for Intel models cpu defs: remove replicated flags from Intel cpu defs: uncomment empty extfeatures_ecx definition for Opteron_G1 reorder cpuid feature bits on target-x86_64.conf cpu defs: add pse36, mca, mtrr to AMD CPU definitions john cooper (6): correct archaic CPU model "model" field for Intel CPUs. Allow an optional qemu_early_init_vcpu() Add kvm emulated x2apic flag to config defined cpu models (v2) Support -readconfig "?" to debug config file loading add Westmere as a qemu cpu model add "default" pseudo CPU model name cpus.c | 8 ++++ hw/pc.c | 41 +++++++++++++++++++--- kvm-all.c | 36 +++++++++++++++---- kvm.h | 1 + qemu-common.h | 2 + qemu-config.c | 30 ++++++++++------ qemu-config.h | 2 +- sysconfigs/target/target-x86_64.conf | 58 +++++++++++++++++++----------- target-i386/cpuid.c | 65 ++++++++++++++++++++++++---------- target-i386/helper.c | 2 + vl.c | 20 +++++++---- 11 files changed, 193 insertions(+), 72 deletions(-) -- 1.7.3.2
