[PATCH] kvm: iommu: Add cond_resched to legacy device assignment code

[PATCH] kvm: iommu: Add cond_resched to legacy device assignment code

[Joerg Roedel]

From: Joerg Roedel <jroedel@suse.de>

When assigning devices to large memory guests (>=128GB guest
memory in the failure case) the functions to create the
IOMMU page-tables for the whole guest might run for a very
long time. On non-preemptible kernels this might cause
Soft-Lockup warnings. Fix these by adding a cond_resched()
to the mapping and unmapping loops.

Signed-off-by: Joerg Roedel <jroedel@suse.de>
---
 virt/kvm/iommu.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
index c1e6ae9..ac427e8 100644
--- a/virt/kvm/iommu.c
+++ b/virt/kvm/iommu.c
@@ -137,7 +137,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 
 		gfn += page_size >> PAGE_SHIFT;
 
-
+		cond_resched();
 	}
 
 	return 0;
@@ -311,6 +311,8 @@ static void kvm_iommu_put_pages(struct kvm *kvm,
 		kvm_unpin_pages(kvm, pfn, unmap_pages);
 
 		gfn += unmap_pages;
+
+		cond_resched();
 	}
 }
 
-- 
1.9.1

Re: [PATCH] kvm: iommu: Add cond_resched to legacy device assignment code

[Chen, Tiejun]

On 2014/12/16 23:47, Joerg Roedel wrote:
> From: Joerg Roedel <jroedel@suse.de>
>
> When assigning devices to large memory guests (>=128GB guest
> memory in the failure case) the functions to create the
> IOMMU page-tables for the whole guest might run for a very
> long time. On non-preemptible kernels this might cause
> Soft-Lockup warnings. Fix these by adding a cond_resched()
> to the mapping and unmapping loops.
>
> Signed-off-by: Joerg Roedel <jroedel@suse.de>
> ---
>   virt/kvm/iommu.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
> index c1e6ae9..ac427e8 100644
> --- a/virt/kvm/iommu.c
> +++ b/virt/kvm/iommu.c

This file is already gone after one latest commit c274e03af705, "kvm: 
x86: move assigned-dev.c and iommu.c to arch/x86/" is introduced, so you 
need to pull your tree firstly :)

Tiejun


> @@ -137,7 +137,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
>
>   		gfn += page_size >> PAGE_SHIFT;
>
> -
> +		cond_resched();
>   	}
>
>   	return 0;
> @@ -311,6 +311,8 @@ static void kvm_iommu_put_pages(struct kvm *kvm,
>   		kvm_unpin_pages(kvm, pfn, unmap_pages);
>
>   		gfn += unmap_pages;
> +
> +		cond_resched();
>   	}
>   }
>
>

Re: [PATCH] kvm: iommu: Add cond_resched to legacy device assignment code

[Joerg Roedel]

On Wed, Dec 17, 2014 at 09:38:57AM +0800, Chen, Tiejun wrote:
> On 2014/12/16 23:47, Joerg Roedel wrote:
> >diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
> >index c1e6ae9..ac427e8 100644
> >--- a/virt/kvm/iommu.c
> >+++ b/virt/kvm/iommu.c
> 
> This file is already gone after one latest commit c274e03af705,
> "kvm: x86: move assigned-dev.c and iommu.c to arch/x86/" is
> introduced, so you need to pull your tree firstly :)

Hmm, I based the patch on kvm/master, where the file is still present.
What is the branch to base kvm patches on these days?


	Joerg

Re: [PATCH] kvm: iommu: Add cond_resched to legacy device assignment code

[Paolo Bonzini]

On 17/12/2014 11:35, Joerg Roedel wrote:
>> > 
>> > This file is already gone after one latest commit c274e03af705,
>> > "kvm: x86: move assigned-dev.c and iommu.c to arch/x86/" is
>> > introduced, so you need to pull your tree firstly :)
> Hmm, I based the patch on kvm/master, where the file is still present.
> What is the branch to base kvm patches on these days?

kvm/master in the -rc period, kvm/next during the merge window and for
the next release.  But it's not a problem, I can sort it out.

Paolo

Re: [PATCH] kvm: iommu: Add cond_resched to legacy device assignment code

[Joerg Roedel]

On Wed, Dec 17, 2014 at 11:46:24AM +0100, Paolo Bonzini wrote:
> 
> 
> On 17/12/2014 11:35, Joerg Roedel wrote:
> >> > 
> >> > This file is already gone after one latest commit c274e03af705,
> >> > "kvm: x86: move assigned-dev.c and iommu.c to arch/x86/" is
> >> > introduced, so you need to pull your tree firstly :)
> > Hmm, I based the patch on kvm/master, where the file is still present.
> > What is the branch to base kvm patches on these days?
> 
> kvm/master in the -rc period, kvm/next during the merge window and for
> the next release.  But it's not a problem, I can sort it out.

Great, thanks Paolo. So I will base my future patches on kvm/next.


	Joerg