KVM: ia64: define kvm_lapic_enabled() to fix a compile error

PATCH Applied Takuya Yoshikawa 4 participants 9 messages Sep 2, 2010 – Sep 29, 2010 KVM

lore

Share & tools

lore

Takuya Yoshikawa

author

2010-09-02 08:54 UTC #0 lore raw

PATCH 1/1 KVM: ia64: define kvm_lapic_enabled() to fix a compile error

The following patch

commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's

ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.

We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch for ia64.

Signed-off-by Takuya Yoshikawa <[email protected]>

---
 arch/ia64/kvm/lapic.h |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

arch/ia64/kvm/lapic.h +1 -0

index ee541ce..c5f92a9 100644
@@ -25,5 +25,6 @@ int kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source,
 int kvm_apic_compare_prio(struct kvm_vcpu *vcpu1, struct kvm_vcpu *vcpu2);
 int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq);
 #define kvm_apic_present(x) (true)
+#define kvm_lapic_enabled(x) (true)
 
 #endif

Signature · 1 line

1.7.0.4

Avi Kivity

Applied

2010-09-02 08:57 UTC #1 lore raw

Quoted 9 lines · Takuya Yoshikawa

The following patch
commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's
ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.
We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch for ia64.

Applied, thanks.

Signature · 1 line

error compiling committee.c: too many arguments to function

Takuya Yoshikawa

author

2010-09-27 06:50 UTC #2 lore raw

Quoted 6 lines · Avi Kivity

Quoted 5 lines · Takuya Yoshikawa
The following patch
commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's

Quoted 7 lines

Quoted 4 lines
ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.
We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch for ia64.
Applied, thanks.

Hi,

Isn't this a candidate for 2.6.36?

Though ia64 kvm is in experimental stage and may not be working now, compile errors may be bad for other people who just want to compile ia64 linux.

Thanks,
Takuya

Avi Kivity

2010-09-27 09:49 UTC #3 lore raw

Quoted 17 lines · Takuya Yoshikawa

Quoted 7 lines · Avi Kivity
Quoted 6 lines · Takuya Yoshikawa
The following patch
commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's
Quoted 8 lines
Quoted 5 lines
ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.
We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch for ia64.
Applied, thanks.

Quoted 8 lines

Hi,
Isn't this a candidate for 2.6.36?
Though ia64 kvm is in experimental stage and may not be working now, compile
errors may be bad for other people who just want to compile ia64 linux.

Maybe we should make ia64 kvm depend on CONFIG_BROKEN.

It has been "experimental" for quite a while.

Signature · 1 line

error compiling committee.c: too many arguments to function

Alexander Graf

2010-09-27 09:52 UTC #4 lore raw

Quoted 20 lines · Avi Kivity

Quoted 19 lines · Takuya Yoshikawa
Quoted 6 lines · Avi Kivity
Quoted 5 lines · Takuya Yoshikawa
The following patch
commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's
Quoted 7 lines
Quoted 4 lines
ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.
We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch for ia64.
Applied, thanks.
Hi,
Isn't this a candidate for 2.6.36?

Quoted 7 lines

Quoted 3 lines
Though ia64 kvm is in experimental stage and may not be working now, compile errors may be bad for other people who just want to compile ia64 linux.
Maybe we should make ia64 kvm depend on CONFIG_BROKEN.
It has been "experimental" for quite a while.

Ack. We should give people on the IA64 Linux ML a chance to step up on it though. CC'ing them.

Alex

Zhang, Xiantao

2010-09-27 10:01 UTC #5 lore raw

In reply to Avi Kivity

Quoted 35 lines · Avi Kivity

Quoted 19 lines · Takuya Yoshikawa
Quoted 10 lines · Avi Kivity
Quoted 9 lines · Takuya Yoshikawa
The following patch
commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's
ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.
Quoted 7 lines
Quoted 4 lines
We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch
for ia64.
Applied, thanks.
Quoted 8 lines
Hi,
Isn't this a candidate for 2.6.36?
Though ia64 kvm is in experimental stage and may not be working now, compile errors may be bad for other people who just want to compile ia64 linux.
Maybe we should make ia64 kvm depend on CONFIG_BROKEN.
It has been "experimental" for quite a while.
--
error compiling committee.c: too many arguments to function

Quoted 24 lines · Avi Kivity

Quoted 23 lines · Takuya Yoshikawa
Quoted 10 lines · Avi Kivity
Quoted 9 lines · Takuya Yoshikawa
The following patch
commit 57ce1659316f4ca298919649f9b1b55862ac3826
KVM: x86: In DM_LOWEST, only deliver interrupts to vcpus with enabled LAPIC's
ignored the fact that kvm_irq_delivery_to_apic() was also used by ia64.
Quoted 7 lines
Quoted 4 lines
We define kvm_lapic_enabled() to fix a compile error caused by this. This will have the same effect as reverting the problematic patch
for ia64.
Applied, thanks.
Hi,
Isn't this a candidate for 2.6.36?

Quoted 8 lines

Quoted 4 lines
Though ia64 kvm is in experimental stage and may not be working now, compile errors may be bad for other people who just want to compile ia64 linux.
Maybe we should make ia64 kvm depend on CONFIG_BROKEN.
It has been "experimental" for quite a while.

I don't think the kvm/ia64 is broken in the upstream Linux, and it should work according to our last try. The big issue is about userspace support. Latest qemu maybe not work with latest kernel, but if you choose an old qemu, it should work! Xiantao

Quoted 2 lines

--
error compiling committee.c: too many arguments to function

Takuya Yoshikawa

author

2010-09-27 10:29 UTC #6 lore raw

Quoted 9 lines

Quoted 9 lines
Quoted 8 lines
Hi,
Isn't this a candidate for 2.6.36?
Though ia64 kvm is in experimental stage and may not be working now, compile errors may be bad for other people who just want to compile ia64 linux.

Quoted 7 lines

Quoted 3 lines
Maybe we should make ia64 kvm depend on CONFIG_BROKEN.
It has been "experimental" for quite a while.
I don't think the kvm/ia64 is broken in the upstream Linux, and it should work according to our last try. The big issue is about userspace support. Latest qemu maybe not work with latest kernel, but if you choose an old qemu, it should work! Xiantao

But without my patch, I failed to (cross-)compile kvm/ia64.

So for a few months, kvm/ia64 was broken in that sense, I guess.

I hope we would have more chance to know in which version kvm/ia64 works!
From my side, I'm sometimes worried whether my patches made something bad for kvm/ia64, hopefully not of course.

Thanks,
Takuya

Avi Kivity

2010-09-28 09:27 UTC #7 lore raw

In reply to Zhang, Xiantao

Quoted 6 lines · Xiantao

Quoted 4 lines
Maybe we should make ia64 kvm depend on CONFIG_BROKEN.
It has been "experimental" for quite a while.
I don't think the kvm/ia64 is broken in the upstream Linux, and it should work according to our last try.

When was that?

Quoted 1 line

The big issue is about userspace support. Latest qemu maybe not work with latest kernel, but if you choose an old qemu, it should work!

There's always the chance that changes in the host kernel or architecture independent kvm code will cause breakage. We need regular use or testing.

Signature · 1 line

error compiling committee.c: too many arguments to function

Zhang, Xiantao

2010-09-29 14:10 UTC #8 lore raw

Quoted 10 lines · Avi Kivity

Quoted 7 lines · Xiantao
Quoted 4 lines
Maybe we should make ia64 kvm depend on CONFIG_BROKEN.
It has been "experimental" for quite a while.
I don't think the kvm/ia64 is broken in the upstream Linux, and it should work according to our last try.
When was that?

It should be pre-2.6.35. You know, if the API between kernel and userspace is not changed, it should keep workable.

Quoted 7 lines

Quoted 3 lines
The big issue is about userspace support. Latest qemu maybe not
work with latest kernel, but if you choose an old qemu, it should work!
There's always the chance that changes in the host kernel or architecture independent kvm code will cause breakage. We need regular use or testing.

Agree, and we will check it regularly.
Xiantao