target/i386/kvm/kvm.c | 2 ++ 1 file changed, 2 insertions(+)
Commit bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG
enablement") broke !CONFIG_SYNDBG builds as hyperv_syndbg_query_options()
is missing there. The idea probably was that as "hv-syndbg" is now under
'#ifdef CONFIG_SYNDBG', hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) cannot
be true anyway.
It would have been possible to add a stub for hyperv_syndbg_query_options()
instead of resurrecting '#ifdef CONFIG_SYNDBG' but avoiding
HV_X64_MSR_SYNDBG_OPTIONS altogether instead of zeroing it when
!CONFIG_SYNDBG seems preferable.
Reported-by: Michael Tokarev <mjt@tls.msk.ru>
Fixes: bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG enablement")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
---
target/i386/kvm/kvm.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index 8e17942c3ba1..11f9526c8f8c 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -4034,11 +4034,13 @@ static int kvm_put_msrs(X86CPU *cpu, int level)
kvm_msr_entry_add(cpu, HV_X64_MSR_TSC_EMULATION_STATUS,
env->msr_hv_tsc_emulation_status);
}
+#ifdef CONFIG_SYNDBG
if (hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) &&
has_msr_hv_syndbg_options) {
kvm_msr_entry_add(cpu, HV_X64_MSR_SYNDBG_OPTIONS,
hyperv_syndbg_query_options());
}
+#endif
}
if (hyperv_feat_enabled(cpu, HYPERV_FEAT_VAPIC)) {
kvm_msr_entry_add(cpu, HV_X64_MSR_APIC_ASSIST_PAGE,
--
2.47.0
On 14/11/24 13:01, Vitaly Kuznetsov wrote: > Commit bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG > enablement") broke !CONFIG_SYNDBG builds as hyperv_syndbg_query_options() > is missing there. The idea probably was that as "hv-syndbg" is now under > '#ifdef CONFIG_SYNDBG', hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) cannot > be true anyway. > > It would have been possible to add a stub for hyperv_syndbg_query_options() > instead of resurrecting '#ifdef CONFIG_SYNDBG' but avoiding > HV_X64_MSR_SYNDBG_OPTIONS altogether instead of zeroing it when > !CONFIG_SYNDBG seems preferable. > > Reported-by: Michael Tokarev <mjt@tls.msk.ru> > Fixes: bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG enablement") > Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> > --- > target/i386/kvm/kvm.c | 2 ++ > 1 file changed, 2 insertions(+) Patch queued to hw-misc, thanks.
18.11.2024 15:39, Philippe Mathieu-Daudé wrote: > On 14/11/24 13:01, Vitaly Kuznetsov wrote: >> Commit bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG >> enablement") broke !CONFIG_SYNDBG builds as hyperv_syndbg_query_options() >> is missing there. The idea probably was that as "hv-syndbg" is now under >> '#ifdef CONFIG_SYNDBG', hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) cannot >> be true anyway. >> >> It would have been possible to add a stub for hyperv_syndbg_query_options() >> instead of resurrecting '#ifdef CONFIG_SYNDBG' but avoiding >> HV_X64_MSR_SYNDBG_OPTIONS altogether instead of zeroing it when >> !CONFIG_SYNDBG seems preferable. >> >> Reported-by: Michael Tokarev <mjt@tls.msk.ru> >> Fixes: bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG enablement") >> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> >> --- >> target/i386/kvm/kvm.c | 2 ++ >> 1 file changed, 2 insertions(+) > > Patch queued to hw-misc, thanks. Please note once again there's another fix from pbonzini for this. /mjt
On 11/18/24 13:40, Michael Tokarev wrote: > 18.11.2024 15:39, Philippe Mathieu-Daudé wrote: >> On 14/11/24 13:01, Vitaly Kuznetsov wrote: >>> Commit bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG >>> enablement") broke !CONFIG_SYNDBG builds as >>> hyperv_syndbg_query_options() >>> is missing there. The idea probably was that as "hv-syndbg" is now under >>> '#ifdef CONFIG_SYNDBG', hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) >>> cannot >>> be true anyway. >>> >>> It would have been possible to add a stub for >>> hyperv_syndbg_query_options() >>> instead of resurrecting '#ifdef CONFIG_SYNDBG' but avoiding >>> HV_X64_MSR_SYNDBG_OPTIONS altogether instead of zeroing it when >>> !CONFIG_SYNDBG seems preferable. >>> >>> Reported-by: Michael Tokarev <mjt@tls.msk.ru> >>> Fixes: bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG >>> enablement") >>> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> >>> --- >>> target/i386/kvm/kvm.c | 2 ++ >>> 1 file changed, 2 insertions(+) >> >> Patch queued to hw-misc, thanks. > > Please note once again there's another fix from pbonzini for this. I'll send a pull request for this soon; Philippe, can you drop this one? (I'll drop the hda fix from my tree instead) Thanks, Paolo
On 18/11/24 12:40, Michael Tokarev wrote: > 18.11.2024 15:39, Philippe Mathieu-Daudé wrote: >> On 14/11/24 13:01, Vitaly Kuznetsov wrote: >>> Commit bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG >>> enablement") broke !CONFIG_SYNDBG builds as >>> hyperv_syndbg_query_options() >>> is missing there. The idea probably was that as "hv-syndbg" is now under >>> '#ifdef CONFIG_SYNDBG', hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) >>> cannot >>> be true anyway. >>> >>> It would have been possible to add a stub for >>> hyperv_syndbg_query_options() >>> instead of resurrecting '#ifdef CONFIG_SYNDBG' but avoiding >>> HV_X64_MSR_SYNDBG_OPTIONS altogether instead of zeroing it when >>> !CONFIG_SYNDBG seems preferable. >>> >>> Reported-by: Michael Tokarev <mjt@tls.msk.ru> >>> Fixes: bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG >>> enablement") >>> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> >>> --- >>> target/i386/kvm/kvm.c | 2 ++ >>> 1 file changed, 2 insertions(+) >> >> Patch queued to hw-misc, thanks. > > Please note once again there's another fix from pbonzini for this. OK, dropped.
On 14/11/24 13:01, Vitaly Kuznetsov wrote: > Commit bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG > enablement") broke !CONFIG_SYNDBG builds as hyperv_syndbg_query_options() > is missing there. The idea probably was that as "hv-syndbg" is now under > '#ifdef CONFIG_SYNDBG', hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNDBG) cannot > be true anyway. > > It would have been possible to add a stub for hyperv_syndbg_query_options() > instead of resurrecting '#ifdef CONFIG_SYNDBG' but avoiding > HV_X64_MSR_SYNDBG_OPTIONS altogether instead of zeroing it when > !CONFIG_SYNDBG seems preferable. > > Reported-by: Michael Tokarev <mjt@tls.msk.ru> > Fixes: bbf3810f2c4f ("target/i386: Fix conditional CONFIG_SYNDBG enablement") > Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> > --- > target/i386/kvm/kvm.c | 2 ++ > 1 file changed, 2 insertions(+) Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
© 2016 - 2024 Red Hat, Inc.