On 6/22/2011 3:21 AM, Chris Wright wrote:
* Nai Xia (nai@gmail.com) wrote:
Introduced kvm_mmu_notifier_test_and_clear_dirty(),
kvm_mmu_notifier_dirty_update()
and their mmu_notifier interfaces to support KSM dirty bit tracking, which
brings
significant performance gain in volatile pages
On Wednesday 22 June 2011 14:15:51 Izik Eidus wrote:
On 6/22/2011 3:21 AM, Chris Wright wrote:
* Nai Xia (nai@gmail.com) wrote:
Introduced kvm_mmu_notifier_test_and_clear_dirty(),
kvm_mmu_notifier_dirty_update()
and their mmu_notifier interfaces to support KSM dirty bit tracking,
On 06/22/2011 12:45 AM, Iordan Iordanov wrote:
It's a job for the management layer; I think it should be easy to script
libvirt to do this.
I read the documentation of libvirt, and out of the box, I don't see
how this can be configured. So, I understand your reply as meaning
that we need
On 06/22/2011 06:17 AM, Chaitra Gorantla wrote:
Hi all,
We are working on Fedora 15 Host. And the KVM is used to create Fedora 14 guest.
The clock-source details are as below.
Our Host supports constant_tsc.
on HOST OS
cat /sys/devices/system/clocksource/clocksource0/current_clocksource
tsc
On Sun, 2011-06-19 at 15:44 +0300, Avi Kivity wrote:
On 06/16/2011 06:50 AM, Zachary Amsden wrote:
The 128-bit multiply in pvclock.h was missing an output constraint for
EDX which caused a register corruption to appear.
Was there any particular symptom associated with corruption at that
On 06/22/2011 12:35 PM, Ian Campbell wrote:
On Sun, 2011-06-19 at 15:44 +0300, Avi Kivity wrote:
On 06/16/2011 06:50 AM, Zachary Amsden wrote:
The 128-bit multiply in pvclock.h was missing an output constraint for
EDX which caused a register corruption to appear.
Was there any
On 06/22/2011 12:35 PM, Ian Campbell wrote:
On Sun, 2011-06-19 at 15:44 +0300, Avi Kivity wrote:
On 06/16/2011 06:50 AM, Zachary Amsden wrote:
The 128-bit multiply in pvclock.h was missing an output constraint for
EDX which caused a register corruption to appear.
Was there any
On Wed, 2011-06-22 at 10:40 +0100, Avi Kivity wrote:
On 06/22/2011 12:35 PM, Ian Campbell wrote:
On Sun, 2011-06-19 at 15:44 +0300, Avi Kivity wrote:
On 06/16/2011 06:50 AM, Zachary Amsden wrote:
The 128-bit multiply in pvclock.h was missing an output constraint for
EDX which
On 06/21/2011 04:32 PM, Nai Xia wrote:
Introduced kvm_mmu_notifier_test_and_clear_dirty(),
kvm_mmu_notifier_dirty_update()
and their mmu_notifier interfaces to support KSM dirty bit tracking, which
brings
significant performance gain in volatile pages scanning in KSM.
Currently,
On 6/22/2011 1:43 PM, Avi Kivity wrote:
On 06/21/2011 04:32 PM, Nai Xia wrote:
Introduced kvm_mmu_notifier_test_and_clear_dirty(),
kvm_mmu_notifier_dirty_update()
and their mmu_notifier interfaces to support KSM dirty bit tracking,
which brings
significant performance gain in volatile pages
On 06/22/2011 02:05 PM, Izik Eidus wrote:
+spte = rmap_next(kvm, rmapp, NULL);
+while (spte) {
+int _dirty;
+u64 _spte = *spte;
+BUG_ON(!(_spte PT_PRESENT_MASK));
+_dirty = _spte PT_DIRTY_MASK;
+if (_dirty) {
+dirty = 1;
+
On 6/22/2011 2:10 PM, Avi Kivity wrote:
On 06/22/2011 02:05 PM, Izik Eidus wrote:
+spte = rmap_next(kvm, rmapp, NULL);
+while (spte) {
+int _dirty;
+u64 _spte = *spte;
+BUG_ON(!(_spte PT_PRESENT_MASK));
+_dirty = _spte PT_DIRTY_MASK;
+if
On 06/22/2011 02:19 PM, Izik Eidus wrote:
On 6/22/2011 2:10 PM, Avi Kivity wrote:
On 06/22/2011 02:05 PM, Izik Eidus wrote:
+spte = rmap_next(kvm, rmapp, NULL);
+while (spte) {
+int _dirty;
+u64 _spte = *spte;
+BUG_ON(!(_spte PT_PRESENT_MASK));
+_dirty
On 06/22/2011 02:24 PM, Avi Kivity wrote:
On 06/22/2011 02:19 PM, Izik Eidus wrote:
On 6/22/2011 2:10 PM, Avi Kivity wrote:
On 06/22/2011 02:05 PM, Izik Eidus wrote:
+spte = rmap_next(kvm, rmapp, NULL);
+while (spte) {
+int _dirty;
+u64 _spte = *spte;
+
On 06/22/2011 02:28 PM, Avi Kivity wrote:
Actually, this is dangerous. If we use the dirty bit for other
things, we will get data corruption.
For example we might want to map clean host pages as writeable-clean
in the spte on a read fault so that we don't get a page fault when
they get
On Wednesday 22 June 2011 19:28:08 Avi Kivity wrote:
On 06/22/2011 02:24 PM, Avi Kivity wrote:
On 06/22/2011 02:19 PM, Izik Eidus wrote:
On 6/22/2011 2:10 PM, Avi Kivity wrote:
On 06/22/2011 02:05 PM, Izik Eidus wrote:
+spte = rmap_next(kvm, rmapp, NULL);
+while (spte) {
+
On 6/22/2011 2:33 PM, Nai Xia wrote:
On Wednesday 22 June 2011 19:28:08 Avi Kivity wrote:
On 06/22/2011 02:24 PM, Avi Kivity wrote:
On 06/22/2011 02:19 PM, Izik Eidus wrote:
On 6/22/2011 2:10 PM, Avi Kivity wrote:
On 06/22/2011 02:05 PM, Izik Eidus wrote:
+spte = rmap_next(kvm, rmapp,
I find qemu-kvm only output a little information when abnormally exit.
For example, if qemu-kvm exit by segmentation fault, there are no
information in /var/log/libvirt/qemu/xx.log.
so i want to solve this by collect some information when qemu-kvm exit.
my idea is register some signal handler,
On 06/21/2011 12:07 PM, Marcelo Tosatti wrote:
The following changes since commit eb47d7c5d96060040931c42773ee07e61e547af9:
hw/9118.c: Implement active-low interrupt support (2011-06-15 13:23:37 +0200)
are available in the git repository at:
Properly check the last mapping, and do not walk to the next level if last spte
is met
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c |9 +
1 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index
Set slot bitmap only if the spte is present
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 15 +++
1 files changed, 7 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index f474e93..8316c2d 100644
---
If the range spans a boundary, the mmio access can be broke, fix it as
write emulation.
And we already get the guest physical address, so use it to read guest data
directly to avoid walking guest page table again
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/x86.c
Introduce vcpu_gva_to_gpa to translate the gva to gpa, we can use it
to cleanup the code between read emulation and write emulation
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/x86.c | 38 +-
1 files changed, 29 insertions(+),
The operations of read emulation and write emulation are very similar, so we
can abstract the operation of them, in larter patch, it is used to cleanup the
same code
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/x86.c | 72
Using the read/write operation to remove the same code
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/x86.c | 149 ---
1 files changed, 47 insertions(+), 102 deletions(-)
diff --git a/arch/x86/kvm/x86.c
If the page fault is caused by mmio, we can cache the mmio info, later, we do
not need to walk guest page table and quickly know it is a mmio fault while we
emulate the mmio instruction
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/include/asm/kvm_host.h |5 +
If dirty bit is not set, we can make the pte access read-only to avoid handing
dirty bit everywhere
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 13 +--
arch/x86/kvm/paging_tmpl.h | 46 ++-
2 files
gw-pte_access is the final access permission, since it is unified with
gw-pt_access when we walked guest page table:
FNAME(walk_addr_generic):
pte_access = pt_access FNAME(gpte_access)(vcpu, pte, true);
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
If 'pt_write' is true, we need to emulate the fault. And in later patch, we
need to emulate the fault even though it is not a pt_write event, so rename
it to better fit the meaning
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 10 +-
Move counting used shadow pages from commiting path to preparing path to
reduce tlb flush on some paths
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/arch/x86/kvm/mmu.c
The idea is from Avi:
| Maybe it's time to kill off bypass_guest_pf=1. It's not as effective as
| it used to be, since unsync pages always use shadow_trap_nonpresent_pte,
| and since we convert between the two nonpresent_ptes during sync and unsync.
Signed-off-by: Xiao Guangrong
If the page fault is caused by mmio, the gfn can not be found in memslots, and
'bad_pfn' is returned on gfn_to_hva path, so we can use 'bad_pfn' to identify
the mmio page fault.
And, to clarify the meaning of mmio pfn, we return fault page instead of bad
page when the gfn is not allowd to prefetch
Introduce handle_abnormal_pfn to handle fault pfn on page fault path,
introduce mmu_invalid_pfn to handle fault pfn on prefetch path
It is the preparing work for mmio page fault support
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 47
Introduce some interfaces to modify spte as linux kernel does:
- mmu_spte_clear_track_bits, it set the spte from present to nonpresent, and
track the stat bits(accessed/dirty) of spte
- mmu_spte_clear_no_track, the same as mmu_spte_clear_track_bits except
tracking the stat bits
- mmu_spte_set,
Clean up the code between mmu_spte_clear_* and mmu_spte_update
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 75 +++-
1 files changed, 39 insertions(+), 36 deletions(-)
diff --git a/arch/x86/kvm/mmu.c
Now, the spte is just from nonprsent to present or present to nonprsent, so
we can use some trick to set/clear spte non-atomicly as linux kernel does
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c | 82 +++
1
Use rcu to protect shadow pages table to be freed, so we can safely walk it,
it should run fastly and is needed by mmio page fault
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/include/asm/kvm_host.h |8 +++
arch/x86/kvm/mmu.c | 132
Reorganize it for good using the cache
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index dad7ad9..1319050 100644
--- a/arch/x86/kvm/mmu.c
+++
The idea is from Avi:
| We could cache the result of a miss in an spte by using a reserved bit, and
| checking the page fault error code (or seeing if we get an ept violation or
| ept misconfiguration), so if we get repeated mmio on a page, we don't need to
| search the slot list/tree.
|
Add tracepoints to trace mmio page fault
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c |5
arch/x86/kvm/mmutrace.h | 48 +++
arch/x86/kvm/trace.h| 23 ++
arch/x86/kvm/x86.c
On Tue, Jun 21, 2011 at 09:32:39PM +0800, Nai Xia wrote:
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index d48ec60..b407a69 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -4674,6 +4674,7 @@ static int __init vmx_init(void)
kvm_mmu_set_mask_ptes(0ull, 0ull,
If we don't flush the smp tlb don't we risk that we'll insert pages in
the unstable tree that are volatile just because the dirty bit didn't
get set again on the spte?
Yes, this is the trade off we take, the unstable tree will be flushed
anyway -
so this is nothing that won`t be recovered
On 06/22/2011 07:19 AM, Izik Eidus wrote:
So what we say here is: it is better to have little junk in the unstable
tree that get flushed eventualy anyway, instead of make the guest
slower
this race is something that does not reflect accurate of ksm anyway due
to the full memcmp that we will
* Izik Eidus (izik.ei...@ravellosystems.com) wrote:
On 6/22/2011 3:21 AM, Chris Wright wrote:
* Nai Xia (nai@gmail.com) wrote:
+ if (!shadow_dirty_mask) {
+ WARN(1, KVM: do NOT try to test dirty bit in EPT\n);
+ goto out;
+ }
This should never fire with the
On Mon, Jun 20, 2011 at 11:29:47PM +0900, Takuya Yoshikawa wrote:
From: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp
Avoid two step jump to the error handling part. This eliminates the use
of the variables present and rsvd_fault.
We also use the const type qualifier to show that
On Wed, Jun 22, 2011 at 11:39:40AM -0400, Rik van Riel wrote:
On 06/22/2011 07:19 AM, Izik Eidus wrote:
So what we say here is: it is better to have little junk in the unstable
tree that get flushed eventualy anyway, instead of make the guest
slower
this race is something that does
On 06/22/2011 07:46 PM, Marcelo Tosatti wrote:
if (unlikely(!is_present_gpte(pte))) {
- present = false;
- break;
+ errcode |= PFERR_PRESENT_MASK;
+ goto error;
}
Assignment of PFERR_PRESENT_MASK is
On Wed, Jun 22, 2011 at 10:28:04PM +0800, Xiao Guangrong wrote:
Properly check the last mapping, and do not walk to the next level if last
spte
is met
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c |9 +
1 files changed, 5 insertions(+), 4
Am 08.06.2011 16:10, schrieb Jan Kiszka:
This helps reducing our build-time checks for feature support in the
available Linux kernel headers. And it helps users that do not have
sufficiently recent headers installed on their build machine.
Consequently, the patch removes and build-time checks
On 2011-06-22 22:51, Stefan Weil wrote:
Am 08.06.2011 16:10, schrieb Jan Kiszka:
This helps reducing our build-time checks for feature support in the
available Linux kernel headers. And it helps users that do not have
sufficiently recent headers installed on their build machine.
After receiving some feedback on get_started.py, resolved
to remove a question regarding NFS shares. Since the script
prints the directories clearly, if one wants to setup NFS or
symlinks, he/she will do it.
Signed-off-by: Lucas Meneghel Rodrigues l...@redhat.com
---
Xiao,
On Wed, Jun 22, 2011 at 10:36:16PM +0800, Xiao Guangrong wrote:
The idea is from Avi:
| We could cache the result of a miss in an spte by using a reserved bit, and
| checking the page fault error code (or seeing if we get an ept violation or
| ept misconfiguration), so if we get
On Wed, Jun 22, 2011 at 11:39 PM, Rik van Riel r...@redhat.com wrote:
On 06/22/2011 07:19 AM, Izik Eidus wrote:
So what we say here is: it is better to have little junk in the unstable
tree that get flushed eventualy anyway, instead of make the guest
slower
this race is something that
On Wed, Jun 22, 2011 at 11:03 PM, Andrea Arcangeli aarca...@redhat.com wrote:
On Tue, Jun 21, 2011 at 09:32:39PM +0800, Nai Xia wrote:
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index d48ec60..b407a69 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -4674,6 +4674,7 @@
On Thu, Jun 23, 2011 at 07:13:54AM +0800, Nai Xia wrote:
I agree on this point. Dirty bit , young bit, is by no means accurate. Even
on 4kB pages, there is always a chance that the pte are dirty but the contents
are actually the same. Yeah, the whole optimization contains trade-offs and
Just a
On 06/22/2011 07:13 PM, Nai Xia wrote:
On Wed, Jun 22, 2011 at 11:39 PM, Rik van Rielr...@redhat.com wrote:
On 06/22/2011 07:19 AM, Izik Eidus wrote:
So what we say here is: it is better to have little junk in the unstable
tree that get flushed eventualy anyway, instead of make the guest
On Thu, Jun 23, 2011 at 12:55 AM, Andrea Arcangeli aarca...@redhat.com wrote:
On Wed, Jun 22, 2011 at 11:39:40AM -0400, Rik van Riel wrote:
On 06/22/2011 07:19 AM, Izik Eidus wrote:
So what we say here is: it is better to have little junk in the unstable
tree that get flushed eventualy
On Wed, Jun 22, 2011 at 11:03 PM, Andrea Arcangeli aarca...@redhat.com wrote:
On Tue, Jun 21, 2011 at 09:32:39PM +0800, Nai Xia wrote:
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index d48ec60..b407a69 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -4674,6 +4674,7 @@
On Thu, Jun 23, 2011 at 07:19:06AM +0800, Nai Xia wrote:
OK, I'll have a try over other workarounds.
I am not feeling good about need_pte_unmap myself. :-)
The usual way is to check VM_HUGETLB in the caller and to call another
function that doesn't kmap. Casting pmd_t to pte_t isn't really nice
On Thu, Jun 23, 2011 at 07:37:47AM +0800, Nai Xia wrote:
On 2MB pages, I'd like to remind you and Rik that ksmd currently splits
huge pages before their sub pages gets really merged to stable tree.
So when there are many 2MB pages each having a 4kB subpage
changed for all time, this is already
On 06/22/2011 07:37 PM, Nai Xia wrote:
On 2MB pages, I'd like to remind you and Rik that ksmd currently splits
huge pages before their sub pages gets really merged to stable tree.
Your proposal appears to add a condition that causes ksmd to skip
doing that, which can cause the system to start
On Thu, Jun 23, 2011 at 7:44 AM, Andrea Arcangeli aarca...@redhat.com wrote:
On Thu, Jun 23, 2011 at 07:19:06AM +0800, Nai Xia wrote:
OK, I'll have a try over other workarounds.
I am not feeling good about need_pte_unmap myself. :-)
The usual way is to check VM_HUGETLB in the caller and to
On Thu, Jun 23, 2011 at 7:59 AM, Andrea Arcangeli aarca...@redhat.com wrote:
On Thu, Jun 23, 2011 at 07:37:47AM +0800, Nai Xia wrote:
On 2MB pages, I'd like to remind you and Rik that ksmd currently splits
huge pages before their sub pages gets really merged to stable tree.
So when there are
On Thu, Jun 23, 2011 at 8:00 AM, Rik van Riel r...@redhat.com wrote:
On 06/22/2011 07:37 PM, Nai Xia wrote:
On 2MB pages, I'd like to remind you and Rik that ksmd currently splits
huge pages before their sub pages gets really merged to stable tree.
Your proposal appears to add a condition
On Thu, Jun 23, 2011 at 08:31:56AM +0800, Nai Xia wrote:
On Thu, Jun 23, 2011 at 7:59 AM, Andrea Arcangeli aarca...@redhat.com wrote:
On Thu, Jun 23, 2011 at 07:37:47AM +0800, Nai Xia wrote:
On 2MB pages, I'd like to remind you and Rik that ksmd currently splits
huge pages before their sub
On Thu, Jun 23, 2011 at 7:28 AM, Rik van Riel r...@redhat.com wrote:
On 06/22/2011 07:13 PM, Nai Xia wrote:
On Wed, Jun 22, 2011 at 11:39 PM, Rik van Rielr...@redhat.com wrote:
On 06/22/2011 07:19 AM, Izik Eidus wrote:
So what we say here is: it is better to have little junk in the
On Thu, Jun 23, 2011 at 7:25 AM, Andrea Arcangeli aarca...@redhat.com wrote:
On Thu, Jun 23, 2011 at 07:13:54AM +0800, Nai Xia wrote:
I agree on this point. Dirty bit , young bit, is by no means accurate. Even
on 4kB pages, there is always a chance that the pte are dirty but the
contents
are
On Thu, Jun 23, 2011 at 8:44 AM, Andrea Arcangeli aarca...@redhat.com wrote:
On Thu, Jun 23, 2011 at 08:31:56AM +0800, Nai Xia wrote:
On Thu, Jun 23, 2011 at 7:59 AM, Andrea Arcangeli aarca...@redhat.com
wrote:
On Thu, Jun 23, 2011 at 07:37:47AM +0800, Nai Xia wrote:
On 2MB pages, I'd like
On 06/23/2011 01:13 AM, Marcelo Tosatti wrote:
On Wed, Jun 22, 2011 at 10:28:04PM +0800, Xiao Guangrong wrote:
Properly check the last mapping, and do not walk to the next level if last
spte
is met
Signed-off-by: Xiao Guangrong xiaoguangr...@cn.fujitsu.com
---
arch/x86/kvm/mmu.c |9
Marcelo,
Thanks for your review!
On 06/23/2011 05:59 AM, Marcelo Tosatti wrote:
static int is_large_pte(u64 pte)
@@ -2123,6 +2158,9 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
u64 spte, entry = *sptep;
int ret = 0;
+if (set_mmio_spte(sptep, gfn, pfn,
Am 22.06.2011 23:37, schrieb Jan Kiszka:
On 2011-06-22 22:51, Stefan Weil wrote:
If not, kvm=yes should be restricted to platforms with kvm support.
Otherwise, QEMU builds will fail very early:
ERROR: Host kernel lacks signalfd() support,
but KVM depends on it when the IO thread is disabled.
I find this problem have already solved by this patch:)
Subject: [PATCH] [virtio] Replace virtio-net with native gPXE driver
http://git.etherboot.org/gpxe.git/commitdiff/180dcf4363bf1d8889a2398a4944e94ca150b311
2011/2/16 Michael S. Tsirkin m...@redhat.com:
On Wed, Feb 16, 2011 at 04:00:15PM
72 matches
Mail list logo
