On 06/20/2013 07:48 PM, Michael S. Tsirkin wrote:
> vhost_net_ubuf_put_and_wait has a confusing name:
> it will actually also free it's argument.
> Thus since commit 1280c27f8e29acf4af2da914e80ec27c3dbd5c01
> vhost_net_flush tries to use the argument after passing it
> to vhost_net_ubuf_put_and_wai
Signed-off-by: Alexey Kardashevskiy
---
Documentation/vfio.txt |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/Documentation/vfio.txt b/Documentation/vfio.txt
index c55533c..d7993dc 100644
--- a/Documentation/vfio.txt
+++ b/Documentation/vfio.txt
@@ -172,12 +172,12 @@
On Thu, 20 Jun 2013 23:29:22 +0200
Paolo Bonzini wrote:
> > @@ -4385,8 +4385,10 @@ void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm)
> > * The max value is MMIO_MAX_GEN - 1 since it is not called
> > * when mark memslot invalid.
> > */
> > - if (unlikely(kvm_current_mmio_genera
Hey All,
I'm just wondering whether this is what caused my server to crash.
Started last night in NZ land.
Jun 20 19:22:11 elm dovecot: imap-login: Disconnected (tried to use
disallowed plaintext auth): user=<>, rip=attackerip, lip=10.0.0.3,
session=<0C8LzpDfZQDINsQC>
occasionally get
Jun
Il 20/06/2013 17:21, Cornelia Huck ha scritto:
> Hi,
>
> some more fixes for kvm-next, bringing our handling of intercepted
> instructions more into line with what is actually architectured.
>
> Please apply.
>
> Thomas Huth (7):
> KVM: s390: Renamed PGM_PRIVILEGED_OPERATION
> KVM: s390: Pri
Il 20/06/2013 18:34, Takuya Yoshikawa ha scritto:
> From: Takuya Yoshikawa
>
> Without this information, users will just see unexpected performance
> problems and there is little chance we will get good reports from them:
> note that mmio generation is increased even when we just start, or stop,
On 2013-06-20 22:29, Gleb Natapov wrote:
> On Thu, Jun 20, 2013 at 10:10:18PM +0200, Jan Kiszka wrote:
>> On 2013-06-20 13:47, Gleb Natapov wrote:
>>> Jan ping, are you OK with what I proposed below?
>>>
>>> On Thu, Jun 06, 2013 at 11:53:52AM +0300, Gleb Natapov wrote:
Hi Jan,
I bise
On Thu, Jun 20, 2013 at 10:10:18PM +0200, Jan Kiszka wrote:
> On 2013-06-20 13:47, Gleb Natapov wrote:
> > Jan ping, are you OK with what I proposed below?
> >
> > On Thu, Jun 06, 2013 at 11:53:52AM +0300, Gleb Natapov wrote:
> >> Hi Jan,
> >>
> >> I bisected [1] to f1ed0450a5fac7067590317cbf027f5
On 2013-06-20 13:47, Gleb Natapov wrote:
> Jan ping, are you OK with what I proposed below?
>
> On Thu, Jun 06, 2013 at 11:53:52AM +0300, Gleb Natapov wrote:
>> Hi Jan,
>>
>> I bisected [1] to f1ed0450a5fac7067590317cbf027f566b6ccbca. Fortunately
>> further investigation showed that it is not real
On 19.06.2013 16:38, Gleb Natapov wrote:
> On Wed, Jun 19, 2013 at 04:12:04PM +0200, Stefan Pietsch wrote:
>> On 19.06.2013 15:41, Gleb Natapov wrote:
>>> On Sun, Jun 16, 2013 at 02:25:04AM +0200, Stefan Pietsch wrote:
first bad commit: [25391454e73e3156202264eb3c473825afe4bc94]
KVM:
From: Takuya Yoshikawa
Without this information, users will just see unexpected performance
problems and there is little chance we will get good reports from them:
note that mmio generation is increased even when we just start, or stop,
dirty logging for some memory slot, in which case users cann
From: Thomas Huth
We need more fine-grained control about the point in time when we check
for privileged instructions, since the exceptions that can happen during
an instruction have a well-defined priority. For example, for the PFMF
instruction, the check for PGM_PRIVILEGED_OP must happen after
From: Thomas Huth
DIAGNOSE is a privileged instruction and thus we must make sure that we are
in supervisor mode before taking any other actions.
Signed-off-by: Thomas Huth
Acked-by: Cornelia Huck
Signed-off-by: Cornelia Huck
---
arch/s390/kvm/diag.c | 3 +++
1 file changed, 3 insertions(+)
From: Thomas Huth
TPROT is a privileged instruction and thus should generate a privileged
operation exception when the problem state bit is not cleared in the PSW.
Signed-off-by: Thomas Huth
Acked-by: Cornelia Huck
Signed-off-by: Cornelia Huck
---
arch/s390/kvm/priv.c | 3 +++
1 file changed
From: Thomas Huth
When a guest calls the TPI instruction, the second operand address could
point to an invalid location. In this case the problem should be signaled
to the guest by throwing an access exception.
Signed-off-by: Thomas Huth
Acked-by: Cornelia Huck
Signed-off-by: Cornelia Huck
--
From: Thomas Huth
Added some missing validity checks for the operands and fixed the
priority of exceptions for some function codes according to the
"Principles of Operation" document.
Signed-off-by: Thomas Huth
Acked-by: Cornelia Huck
Signed-off-by: Cornelia Huck
---
arch/s390/kvm/priv.c | 2
From: Thomas Huth
Renamed the PGM_PRIVILEGED_OPERATION define to PGM_PRIVILEGED_OP since this
define was way longer than the other PGM_* defines and caused the code often
to exceed the 80 columns limit when not split to multiple lines.
Signed-off-by: Thomas Huth
Acked-by: Cornelia Huck
Signed-
Hi,
some more fixes for kvm-next, bringing our handling of intercepted
instructions more into line with what is actually architectured.
Please apply.
Thomas Huth (7):
KVM: s390: Renamed PGM_PRIVILEGED_OPERATION
KVM: s390: Privileged operation check for TPROT
KVM: s390: Privileged operation
From: Thomas Huth
LCTL and LCTLG are also privileged instructions, thus there is no need for
treating them separately from the other instructions in priv.c. So this
patch moves these two instructions to priv.c, adds a check for supervisor
state and simplifies the "handle_eb" instruction decoding
On Thu, 2013-06-20 at 18:48 +1000, Alexey Kardashevskiy wrote:
> On 06/20/2013 05:47 PM, Benjamin Herrenschmidt wrote:
> > On Thu, 2013-06-20 at 15:28 +1000, David Gibson wrote:
> >>> Just out of curiosity - would not get_file() and fput_atomic() on a
> >> group's
> >>> file* do the right job inste
Change two functions (test_mmx_movq_mf and test_movabs) using
unified trap_emulator.
Signed-off-by: Arthur Chunqi Li
---
x86/emulator.c | 70
1 file changed, 15 insertions(+), 55 deletions(-)
diff --git a/x86/emulator.c b/x86/emulator.c
Add memcpy(void *dest, const void *src, size_t n) to lib/string.c.
This function acts the same as memcpy in libc.
Signed-off-by: Arthur Chunqi Li
---
lib/libcflat.h |1 +
lib/string.c | 12
2 files changed, 13 insertions(+)
diff --git a/lib/libcflat.h b/lib/libcflat.h
index
Add a function trap_emulator to run an instruction in emulator.
Set inregs first (%rax is invalid because it is used as return
address), put instruction codec in alt_insn and call func with
alt_insn_length. Get results in outregs.
Signed-off-by: Arthur Chunqi Li
---
x86/emulator.c | 95 +++
On Thu, 20 Jun 2013 15:14:42 +0200
Paolo Bonzini wrote:
> Il 20/06/2013 14:54, Gleb Natapov ha scritto:
> >> If they see mysterious peformance problems induced by this wraparound, the
> >> only
> >> way to know the cause later is by this kind of information in the syslog.
> >> So even the first
On 06/11/2013 06:51 AM, Christoffer Dall wrote:
This patch series adds support for user space save/restore of the VGIC
state. Instead of expanding the ONE_REG interface, which works on
VCPUs, we first introduce support for the new KVM device control API and
the VGIC. Now, instead of calling KVM
On Thu, 20 Jun 2013 15:54:38 +0300
Gleb Natapov wrote:
> On Thu, Jun 20, 2013 at 09:28:37PM +0900, Takuya Yoshikawa wrote:
> > On Thu, 20 Jun 2013 14:45:04 +0300
> > Gleb Natapov wrote:
> >
> > > On Thu, Jun 20, 2013 at 12:59:54PM +0200, Paolo Bonzini wrote:
> > > > Il 20/06/2013 10:59, Takuya
Il 20/06/2013 14:54, Gleb Natapov ha scritto:
>> If they see mysterious peformance problems induced by this wraparound, the
>> only
>> way to know the cause later is by this kind of information in the syslog.
>> So even the first wraparound may better be printed out IMO.
> Think about starting hun
Some guests do a large number of mask/unmask
calls which currently trigger expensive route update
system calls.
Detect that route in unchanged and skip the system call.
Reported-by: "Zhanghaoyu (A)"
Signed-off-by: Michael S. Tsirkin
---
kvm-all.c | 4
1 file changed, 4 insertions(+)
diff
kvm_add_routing_entry makes an attempt to
zero-initialize any new routing entry.
However, it fails to initialize padding
within the u field of the structure
kvm_irq_routing_entry.
Other functions like kvm_irqchip_update_msi_route
also fail to initialize the padding field in
kvm_irq_routing_entry.
On Thu, Jun 20, 2013 at 09:28:37PM +0900, Takuya Yoshikawa wrote:
> On Thu, 20 Jun 2013 14:45:04 +0300
> Gleb Natapov wrote:
>
> > On Thu, Jun 20, 2013 at 12:59:54PM +0200, Paolo Bonzini wrote:
> > > Il 20/06/2013 10:59, Takuya Yoshikawa ha scritto:
> > > > Without this information, users will ju
Hello.
On 20-06-2013 15:48, Michael S. Tsirkin wrote:
vhost_net_ubuf_put_and_wait has a confusing name:
it will actually also free it's argument.
Thus since commit 1280c27f8e29acf4af2da914e80ec27c3dbd5c01
Please also specify that commit's summary line in parens.
vhost_net_flush tries to
On Thu, Jun 20, 2013 at 02:35:36PM +0200, Paolo Bonzini wrote:
> Il 20/06/2013 13:45, Gleb Natapov ha scritto:
> > On Thu, Jun 20, 2013 at 12:59:54PM +0200, Paolo Bonzini wrote:
> >> Il 20/06/2013 10:59, Takuya Yoshikawa ha scritto:
> >>> Without this information, users will just see unexpected per
Please, send any topic that you are interested in covering.
Thanks, MST
--
MST
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Il 20/06/2013 13:45, Gleb Natapov ha scritto:
> On Thu, Jun 20, 2013 at 12:59:54PM +0200, Paolo Bonzini wrote:
>> Il 20/06/2013 10:59, Takuya Yoshikawa ha scritto:
>>> Without this information, users will just see unexpected performance
>>> problems and there is little chance we will get good repor
On Thu, Jun 20, 2013 at 06:45:22PM +0800, Arthur Chunqi Li wrote:
> Change two functions (test_mmx_movq_mf and test_movabs) using
> unified trap_emulator.
>
> Signed-off-by: Arthur Chunqi Li
> ---
> x86/emulator.c | 70
>
> 1 file chang
On Thu, Jun 20, 2013 at 06:45:21PM +0800, Arthur Chunqi Li wrote:
> Add a function trap_emulator to run an instruction in emulator.
> Set inregs first (%rax is invalid because it is used as return
> address), put instruction codec in alt_insn and call func with
> alt_insn_length. Get results in out
On Thu, 20 Jun 2013 14:45:04 +0300
Gleb Natapov wrote:
> On Thu, Jun 20, 2013 at 12:59:54PM +0200, Paolo Bonzini wrote:
> > Il 20/06/2013 10:59, Takuya Yoshikawa ha scritto:
> > > Without this information, users will just see unexpected performance
> > > problems and there is little chance we wil
Make two more functions static - they only have local callers.
Signed-off-by: Michael S. Tsirkin
---
Will be merged through the vhost tree - no need to apply
directly.
drivers/vhost/net.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhos
vhost_net_ubuf_put_and_wait has a confusing name:
it will actually also free it's argument.
Thus since commit 1280c27f8e29acf4af2da914e80ec27c3dbd5c01
vhost_net_flush tries to use the argument after passing it
to vhost_net_ubuf_put_and_wait, this results
in use after free.
To fix, don't free the ar
vhost_net_ubuf_put_and_wait has a confusing name:
it will actually also free it's argument.
Thus since commit 1280c27f8e29acf4af2da914e80ec27c3dbd5c01
vhost_net_flush tries to use the argument after passing it
to vhost_net_ubuf_put_and_wait, this results
in use after free.
To fix, don't free the ar
Jan ping, are you OK with what I proposed below?
On Thu, Jun 06, 2013 at 11:53:52AM +0300, Gleb Natapov wrote:
> Hi Jan,
>
> I bisected [1] to f1ed0450a5fac7067590317cbf027f566b6ccbca. Fortunately
> further investigation showed that it is not really related to removing
> APIC timer interrupt rein
On Thu, Jun 20, 2013 at 12:59:54PM +0200, Paolo Bonzini wrote:
> Il 20/06/2013 10:59, Takuya Yoshikawa ha scritto:
> > Without this information, users will just see unexpected performance
> > problems and there is little chance we will get good reports from them:
> > note that mmio generation is in
Il 20/06/2013 10:59, Takuya Yoshikawa ha scritto:
> Without this information, users will just see unexpected performance
> problems and there is little chance we will get good reports from them:
> note that mmio generation is increased even when we just start, or stop,
> dirty logging for some memo
On 2013-06-20 12:45, Arthur Chunqi Li wrote:
> Add a function trap_emulator to run an instruction in emulator.
> Set inregs first (%rax is invalid because it is used as return
> address), put instruction codec in alt_insn and call func with
> alt_insn_length. Get results in outregs.
Small hint: Yo
Add a function trap_emulator to run an instruction in emulator.
Set inregs first (%rax is invalid because it is used as return
address), put instruction codec in alt_insn and call func with
alt_insn_length. Get results in outregs.
Signed-off-by: Arthur Chunqi Li
---
lib/libcflat.h |1 +
lib/
Change two functions (test_mmx_movq_mf and test_movabs) using
unified trap_emulator.
Signed-off-by: Arthur Chunqi Li
---
x86/emulator.c | 70
1 file changed, 15 insertions(+), 55 deletions(-)
diff --git a/x86/emulator.c b/x86/emulator.c
Without this information, users will just see unexpected performance
problems and there is little chance we will get good reports from them:
note that mmio generation is increased even when we just start, or stop,
dirty logging for some memory slot, in which case users should never
expect all shado
ok, I will handle all above in the following commit.
Arthur Chunqi Li
Department of Computer Science
School of EECS
Peking University
Beijing, China
>From my iPhone
在 2013-6-20,16:48,Gleb Natapov 写道:
> On Wed, Jun 19, 2013 at 11:00:56PM +0800, Arthur Chunqi Li wrote:
>> Add a function trap_emu
On Thu, Jun 20, 2013 at 10:34:39AM +0200, Paolo Bonzini wrote:
> Il 20/06/2013 09:31, Gleb Natapov ha scritto:
> >> I agree with you on the potential problems but I think we are completely
> >> ignoring the "non-migration" use case. These users will probably benefit
> >> from a correct value of (v
On 06/20/2013 05:47 PM, Benjamin Herrenschmidt wrote:
> On Thu, 2013-06-20 at 15:28 +1000, David Gibson wrote:
>>> Just out of curiosity - would not get_file() and fput_atomic() on a
>> group's
>>> file* do the right job instead of vfio_group_add_external_user() and
>>> vfio_group_del_external_user
On Wed, Jun 19, 2013 at 11:00:56PM +0800, Arthur Chunqi Li wrote:
> Add a function trap_emulator to run an instruction in emulator.
> Set inregs first (%rax is invalid because it is used as return
> address), put instruction codec in alt_insn and call func with
> alt_insn_length. Get results in out
On Wed, 19 Jun 2013 15:29:31 +0200
Paolo Bonzini wrote:
> Il 19/06/2013 15:20, Batalov Eugene ha scritto:
> >
> > I've missed this detail. It looks like Igor's patch doesn't bring
> > secondary cpus kvm_clocksource behavior back to one before the regression,
> > Before the regression per_cpu var
Il 20/06/2013 10:30, Igor Mammedov ha scritto:
> On Wed, 19 Jun 2013 15:29:31 +0200
> Paolo Bonzini wrote:
>
>> Il 19/06/2013 15:20, Batalov Eugene ha scritto:
>>>
>>> I've missed this detail. It looks like Igor's patch doesn't bring
>>> secondary cpus kvm_clocksource behavior back to one before
Il 20/06/2013 09:31, Gleb Natapov ha scritto:
>> I agree with you on the potential problems but I think we are completely
>> ignoring the "non-migration" use case. These users will probably benefit
>> from a correct value of (virtual) msr_platform_info. And it appears, the
>> easiest way to give
On Thu, Jun 20, 2013 at 10:29:42AM +0200, Paolo Bonzini wrote:
> Il 19/06/2013 18:03, Gleb Natapov ha scritto:
> > On Wed, Jun 19, 2013 at 11:07:18PM +0800, 李春奇 wrote:
> >> Hi Gleb,
> >> This version can set %rsp before trapping into emulator, because
> >> insn_page and alt_insn_page is statically
Il 19/06/2013 18:03, Gleb Natapov ha scritto:
> On Wed, Jun 19, 2013 at 11:07:18PM +0800, 李春奇 wrote:
>> Hi Gleb,
>> This version can set %rsp before trapping into emulator, because
>> insn_page and alt_insn_page is statically defined and their relative
>> position to (save) is fixed during executi
Il 19/06/2013 17:00, Arthur Chunqi Li ha scritto:
> static void test_movabs(uint64_t *mem, uint8_t *insn_page,
> uint8_t *alt_insn_page, void *insn_ram)
> {
> -uint64_t val = 0;
> -ulong *cr3 = (ulong *)read_cr3();
> -
> -// Pad with RET instructions
> -memset
Il 20/06/2013 07:21, Rob Landley ha scritto:
> On 06/19/2013 04:09:25 AM, Xiao Guangrong wrote:
>> Document it to Documentation/virtual/kvm/mmu.txt
>
> Why break a change to a single documentation file into 7 pieces.
>
> Are we going to bisect the documentation?
It is explaining 7 different opti
On Thu, 2013-06-20 at 15:28 +1000, David Gibson wrote:
> > Just out of curiosity - would not get_file() and fput_atomic() on a
> group's
> > file* do the right job instead of vfio_group_add_external_user() and
> > vfio_group_del_external_user()?
>
> I was thinking that too. Grabbing a file refere
On Wed, Jun 19, 2013 at 01:50:45PM -0400, Bandan Das wrote:
> Gleb Natapov writes:
>
> > On Tue, Jun 18, 2013 at 11:29:27AM -0400, Bandan Das wrote:
> >> Gleb Natapov writes:
> >>
> >> > On Tue, Jun 18, 2013 at 04:05:08PM +0200, Paolo Bonzini wrote:
> >> >> Il 05/06/2013 10:42, Gleb Natapov ha
60 matches
Mail list logo