On 31.10.2011, at 06:30, Avi Kivity wrote:
> On 10/31/2011 09:53 AM, Alexander Graf wrote:
>> This reverts commit 11d7596e18a712dc3bc29d45662ec111fd65946b. It exceeded
>> the padding on the SREGS struct, rendering the ABI backwards-incompatible.
>
> Can't find the commit hash. Please use hashe
On 10/31/2011 08:24 AM, Avi Kivity wrote:
> On 10/31/2011 09:53 AM, Alexander Graf wrote:
>> From: Scott Wood
>>
>> This implements a shared-memory API for giving host userspace access to
>> the guest's TLB.
>>
>>
>> diff --git a/Documentation/virtual/kvm/api.txt
>> b/Documentation/virtual/kvm/ap
On 10/31/2011 07:50 AM, Avi Kivity wrote:
> On 10/31/2011 09:53 AM, Alexander Graf wrote:
>> +/* sesel is index into the set, not the whole array */
>> +static void write_stlbe(struct kvmppc_vcpu_e500 *vcpu_e500,
>> +struct tlbe *gtlbe,
>> +struct tlbe *stlbe
On 2011-10-31 14:36, Avi Kivity wrote:
> On 10/31/2011 09:53 AM, Alexander Graf wrote:
>> Right now we transfer a static struct every time we want to get or set
>> registers. Unfortunately, over time we realize that there are more of
>> these than we thought of before and the extensibility and flex
On 10/31/2011 09:53 AM, Alexander Graf wrote:
> With hugetlbfs support emerging on e500, we should also support KVM
> backing its guest memory by it.
>
> This patch adds support for hugetlbfs into the e500 shadow mmu code.
>
>
> @@ -673,12 +674,31 @@ static inline void kvmppc_e500_shadow_map(struct
On 10/31/2011 09:53 AM, Alexander Graf wrote:
> Right now we transfer a static struct every time we want to get or set
> registers. Unfortunately, over time we realize that there are more of
> these than we thought of before and the extensibility and flexibility of
> transferring a full struct ever
On 10/31/2011 09:53 AM, Alexander Graf wrote:
> This reverts commit 11d7596e18a712dc3bc29d45662ec111fd65946b. It exceeded
> the padding on the SREGS struct, rendering the ABI backwards-incompatible.
Can't find the commit hash. Please use hashes from the Linus tree when
possible.
This needs to be
On 10/31/2011 09:53 AM, Alexander Graf wrote:
> From: Scott Wood
>
> The hardcoded behavior prevents proper SMP support.
>
> QEMU shall specify the vcpu's PIR as the vcpu id.
>
>
Could also be kvm tool - we generally use the code name 'userspace' to
refer to qemu (but don't rewrite the patch on t
On 10/31/2011 09:53 AM, Alexander Graf wrote:
> From: Scott Wood
>
> This implements a shared-memory API for giving host userspace access to
> the guest's TLB.
>
>
> diff --git a/Documentation/virtual/kvm/api.txt
> b/Documentation/virtual/kvm/api.txt
> index 7945b0b..ab1136f 100644
> --- a/Docume
On 10/31/2011 09:53 AM, Alexander Graf wrote:
> From: Scott Wood
>
> Delay allocation of the shadow pid until we're ready to disable
> preemption and write the entry.
>
> @@ -507,21 +507,16 @@ static inline void kvmppc_e500_deliver_tlb_miss(struct
> kvm_vcpu *vcpu,
> vcpu_e500->mas7 = 0;
>
Hi Alexander,
Alexander Graf hat am Mon 10. Oct, 10:09 (+0200) geschrieben:
> On 07.10.2011, at 23:10, Jörg Sommer wrote:
>
> > Hi,
> >
> > I've got this backtrace:
> >
> > [130902.709711] [ cut here ]
> > [130902.709747] kernel BUG at include/linux/kvm_host.h:603!
>
>
As per specification the decrementer interrupt not happen when DEC is written
with 0. Also when DEC is zero, no decrementer running. So we should not start
hrtimer for decrementer when DEC = 0.
Signed-off-by: Bharat Bhushan
---
arch/powerpc/kvm/emulate.c |3 ++-
1 files changed, 2 insertio
From: Scott Wood
Delay allocation of the shadow pid until we're ready to disable
preemption and write the entry.
Signed-off-by: Scott Wood
Signed-off-by: Alexander Graf
---
arch/powerpc/kvm/e500_tlb.c | 36 +++-
1 files changed, 23 insertions(+), 13 deletions
From: Scott Wood
The hardcoded behavior prevents proper SMP support.
QEMU shall specify the vcpu's PIR as the vcpu id.
Signed-off-by: Scott Wood
Signed-off-by: Alexander Graf
---
arch/powerpc/kvm/booke.c |4 ++--
arch/powerpc/kvm/e500.c |3 ---
2 files changed, 2 insertions(+), 5 de
Until now, we always set HIOR based on the PVR, but this is just wrong.
Instead, we should be setting HIOR explicitly, so user space can decide
what the initial HIOR value is - just like on real hardware.
We keep the old PVR based way around for backwards compatibility, but
once user space uses th
Right now we transfer a static struct every time we want to get or set
registers. Unfortunately, over time we realize that there are more of
these than we thought of before and the extensibility and flexibility of
transferring a full struct every time is limited.
So this is a new approach to the p
From: Scott Wood
The only place it makes sense to call this function already needs
to have preemption disabled.
Signed-off-by: Scott Wood
Signed-off-by: Alexander Graf
---
arch/powerpc/kvm/e500_tlb.c |4 +---
1 files changed, 1 insertions(+), 3 deletions(-)
diff --git a/arch/powerpc/kvm/
From: Scott Wood
Split out the portions of tlbe_priv that should be associated with host
entries into tlbe_ref. Base victim selection on the number of hardware
entries, not guest entries.
For TLB1, where one guest entry can be mapped by multiple host entries,
we use the host tlbe_ref for tracki
This reverts commit 11d7596e18a712dc3bc29d45662ec111fd65946b. It exceeded
the padding on the SREGS struct, rendering the ABI backwards-incompatible.
Signed-off-by: Alexander Graf
---
arch/powerpc/include/asm/kvm.h|8
arch/powerpc/include/asm/kvm_book3s.h |2 --
arch/powe
kvm.h had sparse whitespace at the end of the line. Clean it
up so syncing with QEMU gets easier.
Signed-off-by: Alexander Graf
---
arch/powerpc/include/asm/kvm.h |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm.h b/arch/powerpc/include/asm
From: Scott Wood
It should contain the way, not the absolute TLB0 index.
Signed-off-by: Scott Wood
Signed-off-by: Alexander Graf
---
arch/powerpc/kvm/e500_tlb.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/arch/powerpc/kvm/e500_tlb.c b/arch/powerpc/kvm/e500_tlb.c
Hi Avi / Marcelo,
This is my current patch queue for ppc. Please pull.
Alex
The following changes since commit b796a09c5d808f4013f27ad45953db604dac18fd:
Marcelo Tosatti (1):
Merge remote-tracking branch 'upstream/master' into kvm-devel
are available in the git repository at:
git:/
With hugetlbfs support emerging on e500, we should also support KVM
backing its guest memory by it.
This patch adds support for hugetlbfs into the e500 shadow mmu code.
Signed-off-by: Alexander Graf
Acked-by: Scott Wood
---
v1 -> v2:
- address scott's comments
---
arch/powerpc/kvm/e500_tl
From: Scott Wood
This implements a shared-memory API for giving host userspace access to
the guest's TLB.
Signed-off-by: Scott Wood
Signed-off-by: Alexander Graf
---
Documentation/virtual/kvm/api.txt | 74 +++
arch/powerpc/include/asm/kvm.h | 35 +++
arch/powerpc/include/asm/kv
When syncing KVM headers with QEMU I (or whoever applies the
diff) end up automatically fixing whitespaces. One of them
is in kvm_para.h.
It's a lot more consistent for people who don't do the whitespace
fixups automatically to already have fixed headers in Linux. So
remove the sparse empty line a
From: Bharat Bhushan
The current implementation of mtmsr and mtmsrd are racy in that it does:
* check (int_pending == 0)
---> host sets int_pending = 1 <---
* write shared page
* done
while instead we should check for int_pending after the shared page is written.
Signed-off-by: Bharat
When running with HV KVM and CBE config options enabled, I get
build failures like the following:
arch/powerpc/kernel/head_64.o: In function `cbe_system_error_hv':
(.text+0x1228): undefined reference to `do_kvm_0x1202'
arch/powerpc/kernel/head_64.o: In function `cbe_maintenance_hv':
(.text
27 matches
Mail list logo
