* [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM
@ 2021-09-21 22:22 Sean Christopherson
2021-09-21 22:22 ` [PATCH 1/2] KVM: arm64: Unconditionally include generic KVM's Kconfig Sean Christopherson
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: Sean Christopherson @ 2021-09-21 22:22 UTC (permalink / raw)
To: Catalin Marinas, Will Deacon, Marc Zyngier
Cc: Sean Christopherson, linux-kernel, kvmarm, linux-arm-kernel
Fix an oddity where arm64 can end up with KVM=y and HAVE_KVM=n. This
doesn't cause problems today as no generic code that supports arm64 uses
HAVE_KVM, but that may not always be the case, e.g. I stumbled across this
when exploring hiding perf's guest callbacks behind HAVE_KVM=y.
Sean Christopherson (2):
KVM: arm64: Unconditionally include generic KVM's Kconfig
KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF
arch/arm64/Kconfig | 1 +
arch/arm64/kvm/Kconfig | 9 ++-------
2 files changed, 3 insertions(+), 7 deletions(-)
--
2.33.0.464.g1972c5931b-goog
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH 1/2] KVM: arm64: Unconditionally include generic KVM's Kconfig
2021-09-21 22:22 [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Sean Christopherson
@ 2021-09-21 22:22 ` Sean Christopherson
2021-09-21 22:22 ` [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF Sean Christopherson
2021-10-11 9:18 ` [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Marc Zyngier
2 siblings, 0 replies; 6+ messages in thread
From: Sean Christopherson @ 2021-09-21 22:22 UTC (permalink / raw)
To: Catalin Marinas, Will Deacon, Marc Zyngier
Cc: Sean Christopherson, linux-kernel, kvmarm, linux-arm-kernel
Unconditionally "source" the generic KVM Kconfig instead of wrapping it
with KVM=y. A future patch will select HAVE_KVM so that referencing
HAVE_KVM in common kernel code doesn't break, and because KVM=y and
HAVE_KVM=n is weird. Source the generic KVM Kconfig unconditionally so
that HAVE_KVM and KVM don't end up with a circular dependency.
Note, all but one of generic KVM's "configs" are of the HAVE_XYZ nature,
and the one outlier correctly takes a dependency on CONFIG_KVM, i.e. the
generic Kconfig is intended to be included unconditionally.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
arch/arm64/kvm/Kconfig | 7 +------
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/arch/arm64/kvm/Kconfig b/arch/arm64/kvm/Kconfig
index a4eba0908bfa..c50f75cf76fe 100644
--- a/arch/arm64/kvm/Kconfig
+++ b/arch/arm64/kvm/Kconfig
@@ -4,6 +4,7 @@
#
source "virt/lib/Kconfig"
+source "virt/kvm/Kconfig"
menuconfig VIRTUALIZATION
bool "Virtualization"
@@ -42,10 +43,4 @@ menuconfig KVM
If unsure, say N.
-if KVM
-
-source "virt/kvm/Kconfig"
-
-endif # KVM
-
endif # VIRTUALIZATION
--
2.33.0.464.g1972c5931b-goog
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF
2021-09-21 22:22 [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Sean Christopherson
2021-09-21 22:22 ` [PATCH 1/2] KVM: arm64: Unconditionally include generic KVM's Kconfig Sean Christopherson
@ 2021-09-21 22:22 ` Sean Christopherson
2021-09-29 15:53 ` Will Deacon
2021-10-11 9:18 ` [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Marc Zyngier
2 siblings, 1 reply; 6+ messages in thread
From: Sean Christopherson @ 2021-09-21 22:22 UTC (permalink / raw)
To: Catalin Marinas, Will Deacon, Marc Zyngier
Cc: Sean Christopherson, linux-kernel, kvmarm, linux-arm-kernel
Select HAVE_KVM if the KVM dependency is met (OF / Open Firmware), and
make KVM depend on HAVE_KVM instead of directly on OF. This fixes the
oddity where arm64 configs can end up with KVM=y and HAVE_KVM=n, and will
hopefully prevent breakage if there are future users of HAVE_KVM.
Note, arm64 unconditionally selects OF, and has always done so (see
commit 8c2c3df31e3b ("arm64: Build infrastructure"). Keep the somewhat
pointless HAVE_KVM dependency on OF to document that KVM requires Open
Firmware support.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
arch/arm64/Kconfig | 1 +
arch/arm64/kvm/Kconfig | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index b5b13a932561..38c0f36a5ed4 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -187,6 +187,7 @@ config ARM64
select HAVE_GCC_PLUGINS
select HAVE_HW_BREAKPOINT if PERF_EVENTS
select HAVE_IRQ_TIME_ACCOUNTING
+ select HAVE_KVM if OF
select HAVE_NMI
select HAVE_PATA_PLATFORM
select HAVE_PERF_EVENTS
diff --git a/arch/arm64/kvm/Kconfig b/arch/arm64/kvm/Kconfig
index c50f75cf76fe..e1e400bd8de5 100644
--- a/arch/arm64/kvm/Kconfig
+++ b/arch/arm64/kvm/Kconfig
@@ -20,7 +20,7 @@ if VIRTUALIZATION
menuconfig KVM
bool "Kernel-based Virtual Machine (KVM) support"
- depends on OF
+ depends on HAVE_KVM
select MMU_NOTIFIER
select PREEMPT_NOTIFIERS
select HAVE_KVM_CPU_RELAX_INTERCEPT
--
2.33.0.464.g1972c5931b-goog
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF
2021-09-21 22:22 ` [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF Sean Christopherson
@ 2021-09-29 15:53 ` Will Deacon
2021-09-29 15:56 ` Sean Christopherson
0 siblings, 1 reply; 6+ messages in thread
From: Will Deacon @ 2021-09-29 15:53 UTC (permalink / raw)
To: Sean Christopherson
Cc: Catalin Marinas, linux-kernel, Marc Zyngier, kvmarm, linux-arm-kernel
On Tue, Sep 21, 2021 at 03:22:31PM -0700, Sean Christopherson wrote:
> Select HAVE_KVM if the KVM dependency is met (OF / Open Firmware), and
> make KVM depend on HAVE_KVM instead of directly on OF. This fixes the
> oddity where arm64 configs can end up with KVM=y and HAVE_KVM=n, and will
> hopefully prevent breakage if there are future users of HAVE_KVM.
>
> Note, arm64 unconditionally selects OF, and has always done so (see
> commit 8c2c3df31e3b ("arm64: Build infrastructure"). Keep the somewhat
> pointless HAVE_KVM dependency on OF to document that KVM requires Open
> Firmware support.
>
> No functional change intended.
>
> Signed-off-by: Sean Christopherson <seanjc@google.com>
> ---
> arch/arm64/Kconfig | 1 +
> arch/arm64/kvm/Kconfig | 2 +-
> 2 files changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index b5b13a932561..38c0f36a5ed4 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -187,6 +187,7 @@ config ARM64
> select HAVE_GCC_PLUGINS
> select HAVE_HW_BREAKPOINT if PERF_EVENTS
> select HAVE_IRQ_TIME_ACCOUNTING
> + select HAVE_KVM if OF
Honestly, I'd just drop the 'if OF' here. We select it unconditionally a
few lines below and so I think it's more confusing to have the check.
With that:
Acked-by: Will Deacon <will@kernel.org>
Will
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF
2021-09-29 15:53 ` Will Deacon
@ 2021-09-29 15:56 ` Sean Christopherson
0 siblings, 0 replies; 6+ messages in thread
From: Sean Christopherson @ 2021-09-29 15:56 UTC (permalink / raw)
To: Will Deacon
Cc: Catalin Marinas, linux-kernel, Marc Zyngier, kvmarm, linux-arm-kernel
On Wed, Sep 29, 2021, Will Deacon wrote:
> On Tue, Sep 21, 2021 at 03:22:31PM -0700, Sean Christopherson wrote:
> > Select HAVE_KVM if the KVM dependency is met (OF / Open Firmware), and
> > make KVM depend on HAVE_KVM instead of directly on OF. This fixes the
> > oddity where arm64 configs can end up with KVM=y and HAVE_KVM=n, and will
> > hopefully prevent breakage if there are future users of HAVE_KVM.
> >
> > Note, arm64 unconditionally selects OF, and has always done so (see
> > commit 8c2c3df31e3b ("arm64: Build infrastructure"). Keep the somewhat
> > pointless HAVE_KVM dependency on OF to document that KVM requires Open
> > Firmware support.
> >
> > No functional change intended.
> >
> > Signed-off-by: Sean Christopherson <seanjc@google.com>
> > ---
> > arch/arm64/Kconfig | 1 +
> > arch/arm64/kvm/Kconfig | 2 +-
> > 2 files changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > index b5b13a932561..38c0f36a5ed4 100644
> > --- a/arch/arm64/Kconfig
> > +++ b/arch/arm64/Kconfig
> > @@ -187,6 +187,7 @@ config ARM64
> > select HAVE_GCC_PLUGINS
> > select HAVE_HW_BREAKPOINT if PERF_EVENTS
> > select HAVE_IRQ_TIME_ACCOUNTING
> > + select HAVE_KVM if OF
>
> Honestly, I'd just drop the 'if OF' here. We select it unconditionally a
> few lines below and so I think it's more confusing to have the check.
Work for me. I all but flipped a coin when deciding whether or not to keep the
OF dependency.
Thanks!
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM
2021-09-21 22:22 [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Sean Christopherson
2021-09-21 22:22 ` [PATCH 1/2] KVM: arm64: Unconditionally include generic KVM's Kconfig Sean Christopherson
2021-09-21 22:22 ` [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF Sean Christopherson
@ 2021-10-11 9:18 ` Marc Zyngier
2 siblings, 0 replies; 6+ messages in thread
From: Marc Zyngier @ 2021-10-11 9:18 UTC (permalink / raw)
To: Will Deacon, Catalin Marinas, Sean Christopherson
Cc: linux-kernel, kvmarm, linux-arm-kernel
On Tue, 21 Sep 2021 15:22:29 -0700, Sean Christopherson wrote:
> Fix an oddity where arm64 can end up with KVM=y and HAVE_KVM=n. This
> doesn't cause problems today as no generic code that supports arm64 uses
> HAVE_KVM, but that may not always be the case, e.g. I stumbled across this
> when exploring hiding perf's guest callbacks behind HAVE_KVM=y.
>
> Sean Christopherson (2):
> KVM: arm64: Unconditionally include generic KVM's Kconfig
> KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF
>
> [...]
Applied to next, thanks!
[1/2] KVM: arm64: Unconditionally include generic KVM's Kconfig
commit: c8f1e96734069c788b10545f4fd82bcbb6b55dfa
[2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF
commit: e26bb75aa2f17fc079e6a24dff653b098e1f5d37
Cheers,
M.
--
Without deviation from the norm, progress is not possible.
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2021-10-11 9:18 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-09-21 22:22 [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Sean Christopherson
2021-09-21 22:22 ` [PATCH 1/2] KVM: arm64: Unconditionally include generic KVM's Kconfig Sean Christopherson
2021-09-21 22:22 ` [PATCH 2/2] KVM: arm64: Depend on HAVE_KVM => OF instead of directly on OF Sean Christopherson
2021-09-29 15:53 ` Will Deacon
2021-09-29 15:56 ` Sean Christopherson
2021-10-11 9:18 ` [PATCH 0/2] KVM: arm64: Clean up CONFIG_KVM vs CONFIG_HAVE_KVM Marc Zyngier
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).