Re: [Qemu-devel] [PULL 43/50] update-linux-headers: prepare for hyperv.h removal

[Roman Kagan]

On Tue, Sep 19, 2017 at 02:29:32PM +0200, Paolo Bonzini wrote:
> From: Roman Kagan <address@hidden>
> 
> All definitions related to Hyper-V emulation are now taken from the QEMU
> own header, so the one imported from the kernel is no longer needed.
> 
> Unfortunately it's included by kvm_para.h.
> 
> So, until this is fixed in the kernel, teach the header harvesting
> script to substitute kernel's hyperv.h with a dummy.
> 
> Signed-off-by: Roman Kagan <address@hidden>
> Message-Id: <address@hidden>
> Signed-off-by: Paolo Bonzini <address@hidden>
> ---
>  scripts/update-linux-headers.sh |  4 +++-
>  target/i386/hyperv-proto.h      |  3 +++
>  target/i386/kvm.c               | 17 +++++------------
>  3 files changed, 11 insertions(+), 13 deletions(-)
> 
> diff --git a/scripts/update-linux-headers.sh b/scripts/update-linux-headers.sh
> index 2f906c4..ad80fe3 100755
> --- a/scripts/update-linux-headers.sh
> +++ b/scripts/update-linux-headers.sh
> @@ -104,7 +104,9 @@ for arch in $ARCHLIST; do
>          cp "$tmpdir/include/asm/unistd-common.h" 
> "$output/linux-headers/asm-arm/"
>      fi
>      if [ $arch = x86 ]; then
> -        cp_portable "$tmpdir/include/asm/hyperv.h" 
> "$output/include/standard-headers/asm-x86/"
> +        cat <<-EOF >"$output/include/standard-headers/asm-x86/hyperv.h"
> +        /* this is a temporary placeholder until kvm_para.h stops including 
> it */
> +        EOF
>          cp "$tmpdir/include/asm/unistd_32.h" "$output/linux-headers/asm-x86/"
>          cp "$tmpdir/include/asm/unistd_x32.h" 
> "$output/linux-headers/asm-x86/"
>          cp "$tmpdir/include/asm/unistd_64.h" "$output/linux-headers/asm-x86/"
> diff --git a/target/i386/hyperv-proto.h b/target/i386/hyperv-proto.h
> index 1b3fa6e..cb4d7f2 100644
> --- a/target/i386/hyperv-proto.h
> +++ b/target/i386/hyperv-proto.h
> @@ -34,6 +34,8 @@
>  #define HV_VP_INDEX_AVAILABLE        (1u << 6)
>  #define HV_RESET_AVAILABLE           (1u << 7)
>  #define HV_REFERENCE_TSC_AVAILABLE   (1u << 9)
> +#define HV_ACCESS_FREQUENCY_MSRS     (1u << 11)
> +
>  
>  /*
>   * HV_CPUID_FEATURES.EDX bits
> @@ -44,6 +46,7 @@
>  #define HV_CPU_DYNAMIC_PARTITIONING_AVAILABLE   (1u << 3)
>  #define HV_HYPERCALL_PARAMS_XMM_AVAILABLE       (1u << 4)
>  #define HV_GUEST_IDLE_STATE_AVAILABLE           (1u << 5)
> +#define HV_FREQUENCY_MSRS_AVAILABLE             (1u << 8)
>  #define HV_GUEST_CRASH_MSR_AVAILABLE            (1u << 10)
>  
>  /*
> diff --git a/target/i386/kvm.c b/target/i386/kvm.c
> index 0498ab7..b1e32e9 100644
> --- a/target/i386/kvm.c
> +++ b/target/i386/kvm.c
> @@ -639,21 +639,14 @@ static int hyperv_handle_properties(CPUState *cs)
>          env->features[FEAT_HYPERV_EAX] |= HV_APIC_ACCESS_AVAILABLE;
>      }
>      if (cpu->hyperv_time) {
> -<<<<<<< HEAD
> -        env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_HYPERCALL_AVAILABLE;
> -        env->features[FEAT_HYPERV_EAX] |= 
> HV_X64_MSR_TIME_REF_COUNT_AVAILABLE;
> -        env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_REFERENCE_TSC_AVAILABLE;
> -
> -        if (has_msr_hv_frequencies && tsc_is_stable_and_known(env)) {
> -            env->features[FEAT_HYPERV_EAX] |= HV_X64_ACCESS_FREQUENCY_MSRS;
> -            env->features[FEAT_HYPERV_EDX] |=
> -                HV_FEATURE_FREQUENCY_MSRS_AVAILABLE;
> -        }
> -=======
>          env->features[FEAT_HYPERV_EAX] |= HV_HYPERCALL_AVAILABLE;
>          env->features[FEAT_HYPERV_EAX] |= HV_TIME_REF_COUNT_AVAILABLE;
>          env->features[FEAT_HYPERV_EAX] |= HV_REFERENCE_TSC_AVAILABLE;
> ->>>>>>> hyperv: add header with protocol definitions
> +
> +        if (has_msr_hv_frequencies && tsc_is_stable_and_known(env)) {
> +            env->features[FEAT_HYPERV_EAX] |= HV_ACCESS_FREQUENCY_MSRS;
> +            env->features[FEAT_HYPERV_EDX] |= HV_FREQUENCY_MSRS_AVAILABLE;
> +        }
>      }
>      if (cpu->hyperv_crash && has_msr_hv_crash) {
>          env->features[FEAT_HYPERV_EDX] |= HV_GUEST_CRASH_MSR_AVAILABLE;

I guess the conflicts were not supposed to be here.

Nor the FREQUENCY_MSRS stuff which would better be merged in the
previous patch.

Roman.