On 2015/5/5 17:24, Jan Beulich wrote:
On 05.05.15 at 04:45, wrote:
Does this work for everyone?
Please first of all explain why the interfaces in asm/flushtlb.h can't
be used here (at least when flushing entire pages). Because - as
I also don't understand any reason we didn't use this previ
-- Forwarded message --
From: "Xuehan Xu"
Date: May 4, 2015 2:35 PM
Subject: How to probe io related operation in dom0 kernel?
To:
Cc:
Hi, everyone
Recently, I tested the disk io latency on dom0 by running the following
instruction:
dd if=/dev/zero of=test_file oflag=direct
Just make this readable while debugging.
Signed-off-by: Tiejun Chen
---
v2:
* Tried to refine code according to comments.
xen/arch/x86/apic.c | 30 ++
1 file changed, 18 insertions(+), 12 deletions(-)
diff --git a/xen/arch/x86/apic.c b/xen/arch/x86/apic.c
index 321
On Tue, May 05, 2015 at 03:30:10PM +0100, Julien Grall wrote:
> On 05/05/15 14:59, Ian Campbell wrote:
> > On Tue, 2015-05-05 at 14:48 +0100, Julien Grall wrote:
> >>> Do we need to also check that we are configuring the same number of
> >>> levels of PT etc, or is that already handled?
> >>
> >> T
From: "Edgar E. Iglesias"
The Stage2 input-size must match what the CPU uses because
the SMMU and the CPU share page-tables.
Test that the SMMU supports the P2M IPA bit size, use it if
supported or bail out if not.
Reviewed-by: Julien Grall
Signed-off-by: Edgar E. Iglesias
---
xen/drivers/pa
From: "Edgar E. Iglesias"
Export p2m_ipa_bits holding the bit size of IPAs used in p2m tables.
Signed-off-by: Edgar E. Iglesias
---
xen/arch/arm/p2m.c| 3 +++
xen/include/asm-arm/p2m.h | 3 +++
2 files changed, 6 insertions(+)
diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c
index
From: "Edgar E. Iglesias"
This is needed to allow the paging setup to probe for
IPA bit sizes to be used in p2m tables prior to iommu setup.
Acked-by: Ian Campbell
Reviewed-by: Julien Grall
Signed-off-by: Edgar E. Iglesias
---
xen/arch/arm/setup.c | 4 ++--
1 file changed, 2 insertions(+), 2
From: "Edgar E. Iglesias"
Hi,
This is a fix for the issue I'm seeing on ZynqMP with missmatched
setup of the SMMU and the shared p2m page-tables with the CPU.
This implements a global p2m_ipa_bits cap for S2 input-size as
discussed in the previous RFC.
Best regards,
Edgar
Changelog:
v4 -> v5:
... so that we can run ./configure in each subsystem.
Make inclusion of Paths-*.mk mandatory. Skip generating a global
Paths.mk since it's of no use anymore.
Signed-off-by: Wei Liu
Cc: Ian Campbell
Cc: Ian Jackson
Acked-by: Ian Campbell
---
v2: update .gitignore, delete Paths-*.mk in distclea
On Tue, May 05, 2015 at 02:23:23PM +0100, Ian Campbell wrote:
> On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
> > From: "Edgar E. Iglesias"
> >
> > Export p2m_ipa_bits holding the bit size of IPAs used in p2m tables.
> >
> > Signed-off-by: Edgar E. Iglesias
> > ---
> > xen/arch/a
On Tue, May 05, 2015 at 02:17:47PM +0100, Ian Campbell wrote:
> On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
> > From: "Edgar E. Iglesias"
> >
> > Hi,
> >
> > This is a fix for the issue I'm seeing on ZynqMP with missmatched
> > setup of the SMMU and the shared p2m page-tables wit
> -Original Message-
> From: Eric Blake [mailto:ebl...@redhat.com]
> Sent: Tuesday, May 05, 2015 10:29 PM
> To: Xu, Quan; stefano.stabell...@eu.citrix.com; stef...@linux.vnet.ibm.com
> Cc: qemu-de...@nongnu.org; wei.l...@citrix.com; dgde...@tycho.nsa.gov;
> xen-devel@lists.xen.org
> Subje
On Tue, 2015-05-05 at 17:38 +0100, Lars Kurth wrote:
>
> The current proposal is to Rename the Xen Project Hackathons to either
> a) Xen Project Architecture & Design Summit,
> b) or Xen Project Design Summit
>
>
> A number of people voted already for b)
> on http://lists.xenproject.org/archive
On Tue, 2015-05-05 at 18:05 +0100, Andrew Cooper wrote:
> On 05/05/15 16:54, Jan Beulich wrote:
> > All,
> >
> > on the hackathon we also discussed possibly changing the versioning
> > of Xen. The main rationale for the proposal is that (just like in many
> > other software projects) version number
On Thu, 2015-04-23 at 17:55 +0800, Chao Peng wrote:
> This is the xc/xl changes to support Intel Cache Allocation
> Technology(CAT).
>
> 'xl psr-hwinfo' is updated to show CAT info and two new commands
> for CAT are introduced:
> - xl psr-cat-cbm-set [-s socket]
> Set cache capacity bitmasks(C
On Tue, 2015-05-05 at 13:52 +0100, Ian Campbell wrote:
> On Thu, 2015-04-23 at 17:55 +0800, Chao Peng wrote:
> > @@ -1495,10 +1495,20 @@ for any of these monitoring types.
> >
> > =over 4
> >
> > -=item B
> > +=item B [I]
> >
> > Show CMT hardware information.
> >
> > +B
> > +
> > +=over
[OSSTEST PATCH 1/2] make-flight: Drop unspecified-qemu tests
[OSSTEST PATCH 2/2] make-flight: Drop qemut tests in qemuu branches
I intend to force push these along with the cs-bisection-step
performance fixes etc., as soon as the opportunity arises.
The default since 4.3 onwards is qemuu only, and we do not expect
it to change.
Signed-off-by: Ian Jackson
---
make-flight |7 +++
1 file changed, 7 insertions(+)
diff --git a/make-flight b/make-flight
index 70c1900..cfe96dd 100755
--- a/make-flight
+++ b/make-flight
@@ -388,6 +388,13 @
Signed-off-by: Ian Jackson
---
make-flight |4
1 file changed, 4 insertions(+)
diff --git a/make-flight b/make-flight
index cfe96dd..6ac902e 100755
--- a/make-flight
+++ b/make-flight
@@ -399,6 +399,10 @@ test_matrix_do_one () {
;;
-qemut)
qemuu_runvar=device_
On Tue, 5 May 2015, Ian Campbell wrote:
> On Fri, 2015-04-24 at 16:46 +0100, Stefano Stabellini wrote:
> > Signed-off-by: Stefano Stabellini
>
> This looks like a good start, a few comments below.
>
> > diff --git a/ap-common b/ap-common
> > index 64749e3..985eeec 100644
> > --- a/ap-common
> >
On Tue, 5 May 2015, Ian Jackson wrote:
> Stefano Stabellini writes ("Re: [PATCH] OSSTEST: introduce a raisin build
> test"):
> > I agree that revisions should be passed to ts-raisin-build by osstest,
> > as Ian suggested. AFAICT that should be enough to meet all your
> > criteria.
>
> You also ne
On 05/05/2015 01:25 PM, David Vrabel wrote:
On 04/05/15 16:02, Boris Ostrovsky wrote:
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1760,6 +1760,9 @@ static struct notifier_block xen_hvm_cpu_notifier = {
static void __init xen_hvm_guest_init(void)
{
+ if (xen_pv
On 05/05/15 16:54, Jan Beulich wrote:
> All,
>
> on the hackathon we also discussed possibly changing the versioning
> of Xen. The main rationale for the proposal is that (just like in many
> other software projects) version numbers (in particular the major
> one) currently don't really convey much
On 04/05/15 16:02, Boris Ostrovsky wrote:
>
> --- a/arch/x86/xen/enlighten.c
> +++ b/arch/x86/xen/enlighten.c
> @@ -1760,6 +1760,9 @@ static struct notifier_block xen_hvm_cpu_notifier = {
>
> static void __init xen_hvm_guest_init(void)
> {
> + if (xen_pv_domain())
> + return;
>
On 05/03/2015 10:55 PM, Juergen Gross wrote:
> I did a small measurement of the pure locking functions on bare metal
> without and with my patches.
>
> spin_lock() for the first time (lock and code not in cache) dropped from
> about 600 to 500 cycles.
>
> spin_unlock() for first time dropped from 1
Stefano Stabellini writes ("Re: [PATCH] OSSTEST: introduce a raisin build
test"):
> I agree that revisions should be passed to ts-raisin-build by osstest,
> as Ian suggested. AFAICT that should be enough to meet all your
> criteria.
You also need to call store_revision for all the trees involved,
On Tue, 5 May 2015, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH] OSSTEST: introduce a raisin build test"):
> > On Fri, 2015-04-24 at 16:46 +0100, Stefano Stabellini wrote:
> > > Signed-off-by: Stefano Stabellini
> >
> > This looks like a good start, a few comments below.
> ...
> > You s
On 05/05/15 16:47, Paul Durrant wrote:
> There are actually very few HVM parameters that a guest needs to read
> and even fewer that a guest needs to write. Use white-lists to specify
> those parameters and also ensre that, by default, newly introduced
> parameters are not accessible.
>
> Signed-of
Hi Konrad,
On 05/05/15 16:50, Konrad Rzeszutek Wilk wrote:
>> ---
>> xen/arch/arm/p2m.c | 4 +++-
>> 1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c
>> index 65efa94..59dd23a 100644
>> --- a/xen/arch/arm/p2m.c
>> +++ b/xen/arch/arm/p2m.c
>
On 05/05/15 17:09, Vijay Kilari wrote:
> On Tue, May 5, 2015 at 7:39 PM, Julien Grall wrote:
>> On 05/05/15 13:14, Vijay Kilari wrote:
>>> Proposal 2:
>>>
>>> Here when guest writes command to vITS queue and updates CWRITER registers,
>>> it is trapped in XEN and below steps are f
Hi all,
one piece of feedback we got at the last and this Hackathon, was that the name
for the event is really misleading. Although some Hacking goes on, in reality
the event is really more of an Architecture, Design and Deployment Summit.
There was some discussion about this topic last year, (
The following 3 patches re-structure the code implementing HVMOP_set_param
and HVMOP_get_param.
Patch #1 gives each operation its own function
Patch #2 splits out checks for getting/setting non-reflexive params and
setting params with change-once semantics, as well as the XSM check into
separate
On 05/05/2015 11:58 AM, Jan Beulich wrote:
On 05.05.15 at 17:46, wrote:
On 05/04/2015 05:14 AM, Andrew Cooper wrote:
On 04/05/2015 09:52, Jan Beulich wrote:
On 04.05.15 at 04:16, wrote:
--- a/xen/drivers/passthrough/vtd/x86/vtd.c
+++ b/xen/drivers/passthrough/vtd/x86/vtd.c
@@ -56,7 +56,9 @@
On Fri, 2015-04-17 at 19:39 +0800, Chen Baozi wrote:
> >
> > Acked-by: Ian Campbell
> >
> > If it is confirmed that I got the correct meaning in the first one I can
> > fix both on commit.
>
> Yes, please. Thanks.
Done and pushed, thanks.
___
Xen-
The level of switch nesting in those ops is getting unreadable. Giving
them their own functions does introduce some code duplication in the
the pre-op checks but the overall result is easier to follow.
This patch is code movement (including style fixes). There is no
functional change.
Signed-off-
There are actually very few HVM parameters that a guest needs to read
and even fewer that a guest needs to write. Use white-lists to specify
those parameters and also ensre that, by default, newly introduced
parameters are not accessible.
Signed-off-by: Paul Durrant
Cc: Keir Fraser
Cc: Jan Beuli
Some parameters can only (validly) be set once. Some should not be set
by a guest for its own domain, and others must not be set since they
require the domain to be paused. Consolidate these checks, along with
the XSM check, in a new hvm_allow_set_param() function for clarity.
Also, introduce hvm_
On Tue, May 5, 2015 at 7:39 PM, Julien Grall wrote:
> On 05/05/15 13:14, Vijay Kilari wrote:
>> Hi,
>>
>
> Hi Vijay,
>
>>As discussed, here is the design doc/txt.
>
> I will comment on the proposal 2 as it seems to be the preferred one
> assuming you are able to find why it's slow.
>
>> Propos
>>> On 05.05.15 at 17:46, wrote:
> On 05/04/2015 05:14 AM, Andrew Cooper wrote:
>> On 04/05/2015 09:52, Jan Beulich wrote:
>> On 04.05.15 at 04:16, wrote:
--- a/xen/drivers/passthrough/vtd/x86/vtd.c
+++ b/xen/drivers/passthrough/vtd/x86/vtd.c
@@ -56,7 +56,9 @@ unsigned int get_
On Tue, May 5, 2015 at 7:21 PM, Stefano Stabellini
wrote:
> On Tue, 5 May 2015, Vijay Kilari wrote:
>> Proposal 2:
>>
>> Here when guest writes command to vITS queue and updates CWRITER registers,
>> it is trapped in XEN and below steps are followed to process ITS command
>>
>> -
All,
on the hackathon we also discussed possibly changing the versioning
of Xen. The main rationale for the proposal is that (just like in many
other software projects) version numbers (in particular the major
one) currently don't really convey much information. The proposal is
to take gcc's new v
On Tue, May 05, 2015 at 04:02:09PM +0100, Julien Grall wrote:
> The commit 569fb6c "xen/arm: Data abort exception (R/W) mem_access
> events" makes apply_p2m_changes to call hypercall_preempt_check for any
> operation rather than for relinquish.
>
> The function hypercall_preempt_check call local_e
On 05/04/2015 05:14 AM, Andrew Cooper wrote:
On 04/05/2015 09:52, Jan Beulich wrote:
On 04.05.15 at 04:16, wrote:
--- a/xen/drivers/passthrough/vtd/x86/vtd.c
+++ b/xen/drivers/passthrough/vtd/x86/vtd.c
@@ -56,7 +56,9 @@ unsigned int get_cache_line_size(void)
void cacheline_flush(char * ad
> On 5 May 2015, at 16:44, Jan Beulich wrote:
>
> All,
>
> on the hackathon we discussed this topic. With the rate of changes
> going into the stable trees not being as high anymore as a few
> years ago, we settled on trying to stick to a 4 month release
> cycle going forward.
+1
Lars
_
All,
on the hackathon we discussed this topic. With the rate of changes
going into the stable trees not being as high anymore as a few
years ago, we settled on trying to stick to a 4 month release
cycle going forward. Of course, issues like the currently severely
lagging osstest can always lead to
On Tue, 2015-05-05 at 16:02 +0100, Julien Grall wrote:
> The commit 569fb6c "xen/arm: Data abort exception (R/W) mem_access
> events" makes apply_p2m_changes to call hypercall_preempt_check for any
> operation rather than for relinquish.
>
> The function hypercall_preempt_check call local_events_n
On Tue, 2015-05-05 at 10:57 +0100, Ian Campbell wrote:
> I did some testing of v6 over the weekend and fixed a couple of minor
> regressions which I caused to test-*-debianhvm-*. A final test is
> running now so I hope to post v6 either this afternoon or tomorrow AM.
Unfortunately that test showed
>>> On 05.05.15 at 17:17, wrote:
> At 16:10 +0100 on 05 May (1430842206), Jan Beulich wrote:
>> From what I
>> can tell (and assuming other code works correctly) the fact that
>> arch_iommu_populate_page_table() sets d->need_iommu to -1
>> first thing should make sure that any subsequent changes t
Michal Privoznik wrote:
> On 04.05.2015 22:35, Jim Fehlig wrote:
>
>> The xend driver and the parsing/formating code in src/xenconfig
>> have long supported soundhw. Add support in the libxl driver too.
>>
>> Signed-off-by: Jim Fehlig
>> ---
>>
>> This patch continues the tradition of silently
Hi Ian,
On 07/04/15 12:24, Chen Baozi wrote:
> We have already had the boot pagetable when reaching the point
> of early_printk fixmap setup. Thus there is no longer necessary
> to calculate physical address of xen_fixmap.
>
> Signed-off-by: Chen Baozi
This patch has been fully ack and Chen agr
At 16:10 +0100 on 05 May (1430842206), Jan Beulich wrote:
> >>> On 16.04.15 at 11:28, wrote:
> > At 22:35 +0100 on 11 Apr (1428791713), Andrew Cooper wrote:
> >> I am not certain that it is the correct way to fix the issue, nor that
> >> the ioreq server code is the only way to trigger it. There
>>> On 05.05.15 at 16:43, wrote:
> On 05/05/15 15:29, Jan Beulich wrote:
> On 05.05.15 at 16:09, wrote:
From: Andrew Cooper [mailto:andrew.coop...@citrix.com]
Sent: 05 May 2015 11:54
On 05/05/15 11:25, Paul Durrant wrote:
> @@ -5657,21 +5664,11 @@ static int hvm_allow_set_p
>>> On 16.04.15 at 11:28, wrote:
> At 22:35 +0100 on 11 Apr (1428791713), Andrew Cooper wrote:
>> I am not certain that it is the correct way to fix the issue, nor that
>> the ioreq server code is the only way to trigger it. There are several
>> ways to shoot a gfn mapping from the guests physmap
The commit 569fb6c "xen/arm: Data abort exception (R/W) mem_access
events" makes apply_p2m_changes to call hypercall_preempt_check for any
operation rather than for relinquish.
The function hypercall_preempt_check call local_events_need_delivery
which rely on the current VCPU is not an idle VCPU.
On Thu, 2015-04-16 at 12:24 +0100, Stefano Stabellini wrote:
> Introduce a configure option to disable the in-tree qemu-traditional
> build and use an externally provided qemu-dm binary. The option is very
> similar to the existing --with-system-qemu-xen.
This overlaps a bit with the existing --wi
Ian Campbell writes ("Re: [PATCH] OSSTEST: introduce a raisin build test"):
> On Fri, 2015-04-24 at 16:46 +0100, Stefano Stabellini wrote:
> > Signed-off-by: Stefano Stabellini
>
> This looks like a good start, a few comments below.
...
> You should also call store_revision() for each git repo wh
On 05/05/15 15:29, Jan Beulich wrote:
On 05.05.15 at 16:09, wrote:
>>> From: Andrew Cooper [mailto:andrew.coop...@citrix.com]
>>> Sent: 05 May 2015 11:54
>>> On 05/05/15 11:25, Paul Durrant wrote:
@@ -5657,21 +5664,11 @@ static int hvm_allow_set_param(struct domain
>>> *d,
case
On 05/05/15 15:09, Paul Durrant wrote:
>> -Original Message-
>> From: Andrew Cooper [mailto:andrew.coop...@citrix.com]
>> Sent: 05 May 2015 11:54
>> To: Paul Durrant; xen-de...@lists.xenproject.org
>> Cc: Keir (Xen.org); Jan Beulich
>> Subject: Re: [PATCH v3 3/3] x86/hvm: Use white-lists fo
On 05/05/15 13:29, Ian Campbell wrote:
> On Tue, 2015-05-05 at 13:00 +0100, Julien Grall wrote:
>>> I'm inclined to just go with this patch for now, unless Stefano is
>>> nacking it.
>>
>> This patch seem to turn into a workaround, would it be better to move
>> check idle_check in apply_p2m_check?
On 05/05/15 14:59, Ian Campbell wrote:
> On Tue, 2015-05-05 at 14:48 +0100, Julien Grall wrote:
>>> Do we need to also check that we are configuring the same number of
>>> levels of PT etc, or is that already handled?
>>
>> The SMMU only care about the number of IPA bits.
>
> What ensures that the
On 05/04/2015 01:22 AM, Quan Xu wrote:
> Signed-off-by: Quan Xu
>
> --Changes in v6:
> -Remove stray insertion.
> ---
> configure| 14 ++
> hmp.c| 2 ++
> qapi-schema.json | 16 ++--
> qemu-options.hx | 13 +++--
> tpm.c| 7
On 05/05/15 13:14, Vijay Kilari wrote:
> Hi,
>
Hi Vijay,
>As discussed, here is the design doc/txt.
I will comment on the proposal 2 as it seems to be the preferred one
assuming you are able to find why it's slow.
> Proposal 2:
>
> Here when guest writes command to vITS qu
>>> On 05.05.15 at 16:09, wrote:
>> From: Andrew Cooper [mailto:andrew.coop...@citrix.com]
>> Sent: 05 May 2015 11:54
>> On 05/05/15 11:25, Paul Durrant wrote:
>> > @@ -5657,21 +5664,11 @@ static int hvm_allow_set_param(struct domain
>> *d,
>> > case HVM_PARAM_IDENT_PT:
>> > case HVM_PAR
Reminder: Our Document Day is tomorrow!
Theme for this month: "No-Can-Do Without a HowTo"
We've got a lot of good information in the documentation, but it isn't
always in a form that will help someone use a particular feature for
the first time. Or, the information is present, but it might need t
On 05/05/2015 09:51 AM, Andy Lutomirski wrote:
On Mon, May 4, 2015 at 8:02 AM, Boris Ostrovsky
wrote:
Commit 61f01dd941ba ("x86_64, asm: Work around AMD SYSRET SS descriptor
attribute issue") makes AMD processors set SS to __KERNEL_DS in
__switch_to() to deal with cases when SS is NULL.
This b
> -Original Message-
> From: Andrew Cooper [mailto:andrew.coop...@citrix.com]
> Sent: 05 May 2015 11:54
> To: Paul Durrant; xen-de...@lists.xenproject.org
> Cc: Keir (Xen.org); Jan Beulich
> Subject: Re: [PATCH v3 3/3] x86/hvm: Use white-lists for HVM param guest
> accessibility checks
>
>
On Thu, 2015-04-23 at 12:48 +0100, Andrew Cooper wrote:
> Presented here is v10 of the Migration v2 series (libxc subset), which is able
> to function when transparently inserted under an unmodified xl/libxl.
I've applied the v11 branch with my minor comments addressed which you
provided privately
On Fri, Apr 17, 2015 at 08:30:59AM +, Olaf Hering wrote:
> Signed-off-by: Olaf Hering
> Cc: Ian Campbell
> Cc: Ian Jackson
> Cc: Jan Beulich
> Cc: Keir Fraser
> Cc: Tim Deegan
> ---
> xen/include/public/io/vscsiif.h | 68
> +
> 1 file changed, 68
On Tue, 2015-05-05 at 14:48 +0100, Julien Grall wrote:
> > Do we need to also check that we are configuring the same number of
> > levels of PT etc, or is that already handled?
>
> The SMMU only care about the number of IPA bits.
What ensures that the starting level of the SMMU matches the starti
On 05/05/15 14:56, Ian Campbell wrote:
> On Thu, 2015-04-30 at 16:33 +0100, David Vrabel wrote:
>>
>> void _spin_lock_irq(spinlock_t *lock)
>> {
>> -LOCK_PROFILE_VAR;
>> -
>> ASSERT(local_irq_is_enabled());
>> local_irq_disable();
>> -check_lock(&lock->debug);
>> -while (
On Thu, 2015-04-30 at 16:33 +0100, David Vrabel wrote:
>
> void _spin_lock_irq(spinlock_t *lock)
> {
> -LOCK_PROFILE_VAR;
> -
> ASSERT(local_irq_is_enabled());
> local_irq_disable();
> -check_lock(&lock->debug);
> -while ( unlikely(!_raw_spin_trylock(&lock->raw)) )
> -
On Tue, 5 May 2015, Ian Campbell wrote:
> On Fri, 2015-05-01 at 16:48 +0100, Stefano Stabellini wrote:
> > Hi all,
> >
> > this patch series introduces a framework to execute simple unit and
> > functional tests in raisin. It can be used by developers to validate
> > their changes before submittin
On Fri, Apr 17, 2015 at 08:30:58AM +, Olaf Hering wrote:
> Port pvscsi support from xend to libxl:
>
> vscsi=['pdev,vdev{,options}']
> xl scsi-attach
> xl scsi-detach
> xl scsi-list
>
> Signed-off-by: Olaf Hering
> Cc: Ian Jackson
> Cc: Stefano Stabellini
> Cc: Ian Campbell
> Cc: Wei
On 05/05/15 14:51, Stefano Stabellini wrote:
>> - With this irq/LPI descriptor we can identify the domain/vITS.
>> - Info of all the ongoing ITS requests(put in pITS Queue) of this domain is
>> stored in ITS command status array (called its_requests). This is
>> managed per vITS.
>>
>> 1) Trap of
At 14:49 +0100 on 05 May (1430837355), Ian Campbell wrote:
> On Thu, 2015-04-30 at 16:33 +0100, David Vrabel wrote:
> > add_sized(ptr, inc) adds inc to the value at ptr using only the correct
> > size of loads and stores for the type of *ptr. The add is /not/ atomic.
>
> atomic.h is an odd place
Hi Pranav,
On 29/04/15 10:38, Pranavkumar Sawargaonkar wrote:
> In old X-Gene Storm firmware and DT, secure mode addresses have been
> mentioned in GICv2 node. In this case maintenance interrupt is used
> instead of EOI HW method.
>
> This patch checks the GIC Distributor Base Address to enable E
On Tue, 5 May 2015, Vijay Kilari wrote:
> Proposal 2:
>
> Here when guest writes command to vITS queue and updates CWRITER registers,
> it is trapped in XEN and below steps are followed to process ITS command
>
> - Dom0 creates a ITS completion device with device id (00:00.1) and
On Mon, May 4, 2015 at 8:02 AM, Boris Ostrovsky
wrote:
> Commit 61f01dd941ba ("x86_64, asm: Work around AMD SYSRET SS descriptor
> attribute issue") makes AMD processors set SS to __KERNEL_DS in
> __switch_to() to deal with cases when SS is NULL.
>
> This breaks Xen PV guests who do not want to lo
On Thu, 2015-04-30 at 16:33 +0100, David Vrabel wrote:
> Now that all architecture use a common ticket lock implementation for
> spinlocks, remove the architecture specific byte lock implementations.
>
> Signed-off-by: David Vrabel
> Reviewed-by: Tim Deegan
> Acked-by: Jan Beulich
Acked-by: Ia
On Thu, 2015-04-30 at 16:33 +0100, David Vrabel wrote:
> add_sized(ptr, inc) adds inc to the value at ptr using only the correct
> size of loads and stores for the type of *ptr. The add is /not/ atomic.
atomic.h is an odd place for it then ;-) But given you use the
infrastructure I can't suggest
Hi Ian,
On 05/05/15 14:24, Ian Campbell wrote:
> On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
>> From: "Edgar E. Iglesias"
>>
>> The Stage2 input-size must match what the CPU uses because
>> the SMMU and the CPU share page-tables.
>>
>> Test that the SMMU supports the P2M IPA bit s
Hi Ian,
On 05/05/15 14:17, Ian Campbell wrote:
> On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
>> From: "Edgar E. Iglesias"
>>
>> Hi,
>>
>> This is a fix for the issue I'm seeing on ZynqMP with missmatched
>> setup of the SMMU and the shared p2m page-tables with the CPU.
>
> Lookin
On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias"
>
> The Stage2 input-size must match what the CPU uses because
> the SMMU and the CPU share page-tables.
>
> Test that the SMMU supports the P2M IPA bit size, use it if
> supported or bail out if not.
>
> Si
On Tue, 2015-05-05 at 14:04 +0100, Andrew Cooper wrote:
> On 05/05/15 14:03, Ian Campbell wrote:
> > On Thu, 2015-04-23 at 12:48 +0100, Andrew Cooper wrote:
> >> From: David Vrabel
> >>
> >> Add the specification for a new migration stream format. The document
> >> includes all the details but to
On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias"
>
> Export p2m_ipa_bits holding the bit size of IPAs used in p2m tables.
>
> Signed-off-by: Edgar E. Iglesias
> ---
> xen/arch/arm/p2m.c| 5 +
> xen/include/asm-arm/p2m.h | 3 +++
> 2 files chan
On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias"
>
> This is needed to allow the paging setup to probe for
> IPA bit sizes to be used in p2m tables prior to iommu setup.
>
> Reviewed-by: Julien Grall
> Signed-off-by: Edgar E. Iglesias
This seems like a m
On Fri, 2015-05-01 at 11:40 +1000, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias"
>
> Hi,
>
> This is a fix for the issue I'm seeing on ZynqMP with missmatched
> setup of the SMMU and the shared p2m page-tables with the CPU.
Looking back at previous conversations it seems like your SMMU h
Thank you, Wei.
On 5/5/2015 5:12 PM, Wei Liu wrote:
On Mon, May 04, 2015 at 08:51:56PM +0800, Yu, Zhang wrote:
Hi Wei,
Thanks for your reply.
On 5/4/2015 5:44 PM, Wei Liu wrote:
(Thanks for trimming the CC list before hand)
On Mon, May 04, 2015 at 02:05:49PM +0800, Yu, Zhang wrote:
Hi Wei,
On 05/05/15 14:03, Ian Campbell wrote:
> On Thu, 2015-04-23 at 12:48 +0100, Andrew Cooper wrote:
>> From: David Vrabel
>>
>> Add the specification for a new migration stream format. The document
>> includes all the details but to summarize:
>>
>> The existing (legacy) format is dependant on the w
On Thu, 2015-04-23 at 12:48 +0100, Andrew Cooper wrote:
> From: David Vrabel
>
> Add the specification for a new migration stream format. The document
> includes all the details but to summarize:
>
> The existing (legacy) format is dependant on the word size of the
> toolstack. This prevents d
On Thu, 2015-04-23 at 12:48 +0100, Andrew Cooper wrote:
> + * Set a pfn as populated, expanding the tracking structures if needed. To
> + * avoid realloc()ing too excessivly, the size increased to the nearest power
"excessively".
Otherwise: Acked-by: Ian Campbell
On Thu, 2015-04-23 at 12:48 +0100, Andrew Cooper wrote:
> Progress information is logged via a different logger to regular libxc log
> messages, and currently can only express a range. However, not everything
> which needs reporting as progress comes with a range. Extend the interface to
> allow
On Thu, 2015-04-23 at 17:55 +0800, Chao Peng wrote:
> +Maximum RMID and supported monitor types in the system can be obtained by:
> +
> +`xl psr-cmt-hwinfo`
I think this is now `xl psr-hwinfo --cmt` ?
> +For example, assuming a system with 8 portions and 3 domains:
> +
> + * A CBM of 0xff for ev
On Thu, 2015-04-23 at 17:55 +0800, Chao Peng wrote:
> This is the xc/xl changes to support Intel Cache Allocation
> Technology(CAT).
>
> 'xl psr-hwinfo' is updated to show CAT info and two new commands
> for CAT are introduced:
> - xl psr-cat-cbm-set [-s socket]
> Set cache capacity bitmasks(C
On Tue, 2015-05-05 at 13:43 +0100, Ian Campbell wrote:
> On Thu, 2015-04-23 at 17:55 +0800, Chao Peng wrote:
> > Add libxl_socket_bitmap_alloc() to allow allocating a socket specific
> > libxl_bitmap (as it is for cpu/node bitmap).
> >
> > Internal function libxl__count_physical_sockets() is intro
On Thu, 2015-04-23 at 17:55 +0800, Chao Peng wrote:
> Add libxl_socket_bitmap_alloc() to allow allocating a socket specific
> libxl_bitmap (as it is for cpu/node bitmap).
>
> Internal function libxl__count_physical_sockets() is introduced together
> to get the socket count when the size of bitmap
flight 53015 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/53015/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-i386-freebsd10-amd64 13 guest-localmigrate fail REGR. vs. 52776
test-amd64-amd64-xl-qemut-de
>>> On 30.04.15 at 15:28, wrote:
> From: Malcolm Crossley
>
> Performance analysis of aggregate network throughput with many VMs
> shows that performance is signficantly limited by contention on the
> maptrack lock when obtaining/releasing maptrack handles from the free
> list.
>
> Instead of a
On Tue, 2015-05-05 at 13:00 +0100, Julien Grall wrote:
> > I'm inclined to just go with this patch for now, unless Stefano is
> > nacking it.
>
> This patch seem to turn into a workaround, would it be better to move
> check idle_check in apply_p2m_check?
>
> I will prepare a follow-up to avoid pr
On Wed, 2015-04-29 at 15:08 +0530, Pranavkumar Sawargaonkar wrote:
> In old X-Gene Storm firmware and DT, secure mode addresses have been
> mentioned in GICv2 node. In this case maintenance interrupt is used
> instead of EOI HW method.
>
> This patch checks the GIC Distributor Base Address to enab
1 - 100 of 162 matches
Mail list logo
