[Trimming the Cc-list a bit]
On 9/14/17 7:37 AM, Juergen Gross wrote:
On 12/09/17 02:45, anshulmakkar wrote:
Introduces scheduler specific parameter at libxl level which are
passed on to libxc. eg runqueue for credit2
Signed-off-by: Anshul Makkar
int libxl_cpupool_destroy(libxl_ctx *ctx
On 9/15/17 6:35 PM, Dario Faggioli wrote:
static unsigned int __read_mostly opt_migrate_resist = 500;
integer_param("sched_credit2_migrate_resist", opt_migrate_resist);
@@ -1453,6 +1459,26 @@ runq_tickle(const struct scheduler *ops, struct
csched2_vcpu *new, s_time_t now)
On 9/15/17 6:35 PM, Dario Faggioli wrote:
Hello,
Dario Faggioli (4):
xen: sched: introduce 'adjust_affinity' hook.
xen: sched: optimize exclusive pinning case (Credit1 & 2)
xen: sched: improve checking soft-affinity
xen: sched: simplify (and speedup) checking soft-a
On 8/18/17 4:50 PM, Dario Faggioli wrote:
@@ -474,6 +586,12 @@ static inline struct csched2_runqueue_data *c2rqd(const struct scheduler *ops,
return &csched2_priv(ops)->rqd[c2r(cpu)];
}
+/* Does the domain of this vCPU have a cap? */
+static inline bool has_cap(const struct csche
oft-affinity,
- try to preempt someone that is running outside
of its own soft-affinity.
The value of the score is added to a trace record,
so xenalyze's code and tools/xentrace/formats are
updated accordingly.
Signed-off-by: Dario Faggioli
---
Cc: George Dunlap
Cc: Anshul Makkar
Cc: Ian J
: Justin T. Weaver
Reviewed-by: George Dunlap
---
Cc: Anshul Makkar
---
xen/common/sched_credit.c | 97 +++-
xen/include/xen/sched-if.h | 64 +
2 files changed, 79 insertions(+), 82 deletions(-)
diff --git a/xen/common/sche
On 13/06/2017 22:13, Dario Faggioli wrote:
On Tue, 2017-06-13 at 17:07 +0100, Anshul Makkar wrote:
On 12/06/2017 14:19, Dario Faggioli wrote:
@@ -92,6 +92,82 @@
*/
what I want to ask is that if the budget of the domain is
replenished,
but credit for the vcpus of that domain is not
On 12/06/2017 14:19, Dario Faggioli wrote:
Hey,
Budget is burned by the domain's vCPUs, in a similar way to how
credits are.
When a domain runs out of budget, its vCPUs can't run any longer.
if the vcpus of a domain have credit and if budget has run out, will
the
vcpus won't be scheduled.
eep them updated with one modified event.
Signed-off-by: Dario Faggioli
---
Cc: George Dunlap
Cc: Anshul Makkar
Cc: Andrew Cooper
Cc: Jan Beulich
Cc: Ian Jackson
Cc: Wei Liu
---
tools/xentrace/formats |2
tools/xentrace/xenalyze.c | 10 +
xen/common/sched_credit2.c | 470 +
On 08/06/2017 14:12, Andrew Cooper wrote:
Presented herewith is the a plan for the final part of CPUID work, which
primarily covers better Xen/Toolstack interaction for configuring the guests
CPUID policy.
A PDF version of this document is available from:
http://xenbits.xen.org/people/andrewcoo
: check if the ring is full and then break from
the loop to consume the items from the ring.
eg. case: prod = 1272, cons = 248.
Signed-off-by: Anshul Makkar
---
v2:
* resolved the coding style issue.
* modified the "if" condition statement to make it simpler.
tools/firmware/hvmloade
situation, check if the buffer is full and then
break from the loop.
e.g case: prod = 1272, cons = 248.
Signed-off-by: Anshul Makkar
Please avoid indenting the entire commit message.
Oh. sorry about that. I saw this format in few previous commits and
adopted it. Will correct it.
--- a
.
e.g case: prod = 1272, cons = 248.
Signed-off-by: Anshul Makkar
---
tools/firmware/hvmloader/xenbus.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/tools/firmware/hvmloader/xenbus.c
b/tools/firmware/hvmloader/xenbus.c
index 448157d..f8fd730 100644
--- a/tools/firmware
On 02/03/17 10:38, Dario Faggioli wrote:
Chacking whether or not a vCPU can be 'stolen'
from a peer pCPU's runqueue is relatively cheap.
Therefore, let's do that as early as possible,
avoiding potentially useless complex checks, and
cpumask manipulations.
Signed-off-by: Dario Faggioli
---
C
On 02/03/17 10:38, Dario Faggioli wrote:
Since we're holding the lock on the pCPU from which we
are trying to steal, it can't have disappeared, so we
can drop the check for that (and convert it in an
ASSERT()).
And since we try to steal only from busy pCPUs, it's
unlikely for such pCPU to be i
On 18/01/17 13:21, bharat gohil wrote:
Hello
I am new to GPU and GPU virtualization and found that xen support
intel GPU virtualization using XenGT.
I want to know,
1) What are the critical GPU command pass from xen to Dom0?
2) How the Dom0 mediator or xen validate the GPU command which is
Hi Jan,
On 09/02/17 16:22, Jan Beulich wrote:
On 08.02.17 at 16:32, wrote:
On 07.02.17 at 18:26, wrote:
Facing a issue where bootstorm of guests leads to host crash. I debugged
and found that that enabling PML introduces a race condition during
guest teardown stage while disabling PML on
May be "causes". But not sure, as disabling this feature solves the
issue (no hypervisor crash) and we have reports that the same code base
works fine on Haswell (again no hardware support for PML in Haswell, but
code base is same. But inherently it leads to non-execution of PML code
path, so n
me vcpus .
With a single domain reboot, no race is observed.
Commit e18d4274772e52ac81fda1acb246d11ef666e5fe causes this race condition.
Anshul
On 07/02/17 17:58, anshul makkar wrote:
Hi, Sorry, forgot to include you in cc list.
Anshul
On 07/02/17 17:26, anshul makkar wrote:
Hi,
Facing a iss
are disabled during
context switch and execution of vm_write in vmx_fpu_leave.. This has got
me confused.
Also, I am not sure if I understand the handling of foreign_vmcs
correctly, which can also be the cause of the race.
Please if you can share some suggestions here.
Th
On 11/01/17 05:37, Tian, Kevin wrote:
From: Anshul Makkar [mailto:anshul.mak...@citrix.com]
Sent: Friday, January 06, 2017 2:42 AM
Current codebase doesn't implement and use vmptrst. Implementing it as it may
be required in future.
Signed-off-by: Anshul Makkar
Then let's do it
Current codebase doesn't implement and use vmptrst. Implementing it as it may
be required in future.
Signed-off-by: Anshul Makkar
---
xen/include/asm-x86/hvm/vmx/vmx.h | 22 ++
1 file changed, 22 insertions(+)
diff --git a/xen/include/asm-x86/hvm/vmx/vmx.h
b/xen/in
On 05/01/2017 08:39, Dario Faggioli wrote:
> On Thu, 2017-01-05 at 02:05 +, Andrew Cooper wrote:
>> On 05/01/2017 01:52, Konrad Rzeszutek Wilk wrote:
>>> It works just fine with credit1 (see further down the log)
>>> but if I try credit2 it ends up hanging during bootup.
>>>
>>> I am a going to
On 03/01/17 18:20, Daniel De Graaf wrote:
On 12/19/2016 11:03 PM, Doug Goldstein wrote:
On 12/19/16 10:02 AM, Doug Goldstein wrote:
On 12/14/16 3:09 PM, Daniel De Graaf wrote:
On 12/12/2016 09:00 AM, Anshul Makkar wrote:
During guest migrate allow permission to prevent
spurious page faults
On 08/12/16 23:40, Boris Ostrovsky wrote:
On 12/08/2016 05:21 PM, Andrew Cooper wrote:
On 08/12/2016 19:18, Stefano Stabellini wrote:
Of course even the largest virtual machine today (2TB on Amazon AFAIK)
is not close to reaching the current memory limit, but it's just a
matter of time.
On 20/12/2016 04:03, Doug Goldstein wrote:
On 12/19/16 10:02 AM, Doug Goldstein wrote:
On 12/14/16 3:09 PM, Daniel De Graaf wrote:
On 12/12/2016 09:00 AM, Anshul Makkar wrote:
During guest migrate allow permission to prevent
spurious page faults.
Prevents these errors:
d73: Non-privileged (73
passthrough for hvm guest:
avc: denied { send_irq } for domid=0 target=10
scontext=system_u:system_r:dom0_t
tcontext=system_u:system_r:domU_t tclass=hvm
Signed-off-by: Anshul Makkar
---
tools/flask/policy/modules/xen.if |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a
Hi,
In Xen-Summit, Toronto, I did a presentation on XSM where two
shortcomings with the present implementation were highlighted:
1) Allow more finer grain control. e.g all introspection domain doesn't
need to have pci-passthrough capability while few need to have and all
should have introspe
gioli
---
Cc: George Dunlap
Cc: Anshul Makkar
---
/*
* If a vcpu is meant to be picked up by another processor, and such
@@ -2282,6 +2288,7 @@ runq_candidate(struct csched2_runqueue_data *rqd,
if ( svc->tickled_cpu != -1 && svc->
On 05/09/16 15:55, Dario Faggioli wrote:
On Thu, 2016-09-01 at 11:52 +0100, anshul makkar wrote:
On 17/08/16 18:19, Dario Faggioli wrote:
+/*
+ * We're doing soft-affinity, and we know that the current
vcpu on cpu
+ * has a soft affinity. We now want to know whether cpu itse
On 05/09/16 14:26, Dario Faggioli wrote:
On Thu, 2016-09-01 at 12:08 +0100, anshul makkar wrote:
On 17/08/16 18:19, Dario Faggioli wrote:
Can't we
just read their workload or we can change the locktype to allow
reading ?
Reading without taking the lock would race against the load
On 05/09/16 14:47, Dario Faggioli wrote:
On Wed, 2016-08-31 at 18:10 +0100, anshul makkar wrote:
On 17/08/16 18:18, Dario Faggioli wrote:
@@ -1266,6 +1272,7 @@ csched2_alloc_vdata(const struct scheduler
*ops, struct vcpu *vc, void *dd)
ASSERT(svc->sdom != NULL);
orge Dunlap
Cc: Anshul Makkar
---
xen/common/sched_credit2.c | 43 +++
1 file changed, 39 insertions(+), 4 deletions(-)
diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c
index 6963872..f03ecce 100644
--- a/xen/common/sched_credit2.c
+
Dunlap
Cc: Anshul Makkar
---
xen/common/sched_credit2.c | 359
1 file changed, 326 insertions(+), 33 deletions(-)
diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c
index 2d7228a..3722f46 100644
--- a/xen/common/sched_credit2.c
+++ b
d runqueue among the soft
affinity "friendly" ones, and the overall least loaded
one, in the same pass.
(Also, kill a spurious ';' when defining MAX_LOAD.)
Signed-off-by: Dario Faggioli
Signed-off-by: Justin T. Weaver
---
Cc: George Dunlap
Cc: Anshul Makkar
---
xen
s soft affinity. In the second step,
we fall back to use all the CPUs from its hard
affinity (as it is doing now, without this patch).
Signed-off-by: Dario Faggioli
Signed-off-by: Justin T. Weaver
---
Cc: George Dunlap
Cc: Anshul Makkar
---
xen/common/sched_c
least, we consider that), as running somewhere non-ideal
is better than not running at all.
The commit also adds performance counters for each of
the possible situations.
Signed-off-by: Dario Faggioli
---
Cc: George Dunlap
Cc: Anshul Makkar
Cc: Jan Beulich
Cc: Andrew Cooper
---
xen/common/sched
Jackson
CC: Wei Liu
CC: Dario Faggioli
CC: Anshul Makkar
---
tools/xentrace/xenalyze.c | 95 +--
1 file changed, 68 insertions(+), 27 deletions(-)
diff --git a/tools/xentrace/xenalyze.c b/tools/xentrace/xenalyze.c
index 455cbdf..a4d8823 100644
On 03/08/16 11:16, George Dunlap wrote:
On 26/07/16 17:21, Dario Faggioli wrote:
On Mon, 2016-07-25 at 17:12 +0100, Anshul Makkar wrote:
It introduces context-switch rate-limiting. The patch enables the VM
The subject, which will become the "title" of the commit, is way too
long. Th
It introduces context-switch rate-limiting. The patch enables the VM to batch
its work and prevents the system from spending most of its time in context
switches because of a VM that is waking/sleeping at high rate.
ratelimit can be disabled by setting it to 0.
Signed-off-by: Anshul Makkar
On 22/07/16 11:36, Dario Faggioli wrote:
On Mon, 2016-07-18 at 13:22 +0100, Anshul Makkar wrote:
Hey, Anshul.
Thanks, and sorry for the delay in reviewing.
This version is an improvement, but it looks to me that you've missed a
few of the review comments to v1.
Sorry about that. !
On 19/07/16 14:30, Doug Goldstein wrote:
On 7/19/16 4:05 AM, Anshul Makkar wrote:
Signed-off-by: Anshul Makkar
---
* Resending the patch due to incomplete subject in the previous patch.
docs/misc/xsm-flask.txt | 8
1 file changed, 4 insertions(+), 4 deletions(-)
---
diff --git a
Signed-off-by: Anshul Makkar
---
* Resending the patch due to incomplete subject in the previous patch.
docs/misc/xsm-flask.txt | 8
1 file changed, 4 insertions(+), 4 deletions(-)
---
diff --git a/docs/misc/xsm-flask.txt b/docs/misc/xsm-flask.txt
index 62f15dd..bf8bb6e 100644
--- a
Signed-off-by: Anshul Makkar
---
docs/misc/xsm-flask.txt | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/docs/misc/xsm-flask.txt b/docs/misc/xsm-flask.txt
index 62f15dd..bf8bb6e 100644
--- a/docs/misc/xsm-flask.txt
+++ b/docs/misc/xsm-flask.txt
@@ -9,8 +9,8
It introduces a minimum amount of latency to enable a VM to batch its work and
it also ensures that system is not spending most of its time in
VMEXIT/VMENTRY because of VM that is waking/sleeping at high rate.
ratelimit can be disabled by setting it to 0.
Signed-off-by: Anshul Makkar
On 13/07/16 09:53, Dario Faggioli wrote:
On Tue, 2016-07-12 at 17:16 +0100, George Dunlap wrote:
On Wed, Jul 6, 2016 at 6:33 PM, Makkar anshul.mak...@citrix.com
wrote:
--- a/xen/common/sched_credit2.c
+++ b/xen/common/sched_credit2.c
+#define MAX_TSLICE(t1, t2) \
+ ({ typeo
scontext=system_u:system_r:dom0_t
tcontext=system_u:system_r:domU_t tclass=domain
Signed-off-by: Anshul Makkar
Acked-by: Daniel De Graaf
---
Changed Since V1:
* added getpodtarget.
tools/flask/policy/modules/xen.if | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools
=system_u:system_r:dom0_t
tcontext=system_u:system_r:domU_t tclass=domain
Signed-off-by: Anshul Makkar
---
---
tools/flask/policy/modules/xen.if | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/flask/policy/modules/xen.if
b/tools/flask/policy/modules/xen.if
index 8c43c28
On 07/07/16 16:36, Daniel De Graaf wrote:
On 07/06/2016 12:19 PM, anshul makkar wrote:
On 06/07/16 16:59, Daniel De Graaf wrote:
On 07/06/2016 11:34 AM, anshul makkar wrote:
Hi,
It allows the resource to be added and removed by the source domain to
target domain, but its use by target
From: Anshul Makkar
This patch resolves the following permission denied scenarios while creating
new domU :
avc: denied { setdomainhandle } for domid=0 target=1
scontext=system_u:system_r:dom0_t tcontext=system_u:system_r:domU_t
tclass=domain
avc: denied { settime } for domid=0 target=1
From: Anshul Makkar
Rate limit assures that a vcpu will execute for a minimum amount of time before
being put at the back of a queue or being preempted by higher priority thread.
It introduces a minimum amount of latency to enable a VM to batch its work and
it also ensures that system is not
On 06/07/16 16:59, Daniel De Graaf wrote:
On 07/06/2016 11:34 AM, anshul makkar wrote:
Hi,
It allows the resource to be added and removed by the source domain to
target domain, but its use by target domain is blocked.
This rule only mandates the use of resource_type for resource types. If
rence
which is expected to be permissive in most of the scenarios so that it
doesn't affect the basic functionalities, is this "neverallow" rule needed ?
Thanks
Anshul Makkar
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
function, we use a new cpumask which indeed needs
to be updated in other places).
In addition to disentangling SMT awareness from load
balancing, this also allows us to support the
sched_smt_power_savings parametar in Credit2 as well.
Signed-off-by: Dario Faggioli
Reviewed-by: Anshul Makkar
On 18/06/16 00:11, Dario Faggioli wrote:
In fact, they always operate on the svc->processor of
the csched2_vcpu passed to them.
No functional change intended.
Signed-off-by: Dario Faggioli
Reviewed-by: Anshul Makkar
---
Cc: George Dunlap
Cc: Anshul Makkar
Cc: David Vrabel
---
ill indeed
be idle. The only thing we can do, therefore, is to
remove the BUG_ON() check.
Signed-off-by: Dario Faggioli
Reviewed-by: Anshul Makkar
---
Cc: George Dunlap
Cc: Anshul Makkar
Cc: David Vrabel
---
xen/common/sched_credit.c | 12 ++--
xen/common/sched_credit
On 15/03/16 17:56, Konrad Rzeszutek Wilk wrote:
All of XENVER_* have now an XSM check for their sub-ops.
The subop for XENVER_commandline is now a priviliged operation.
To not break guests we still return an string - but it is
just '\0'.
The rest: XENVER_[version|extraversion|capabilities|
para
-Original Message-
From: Xen-devel [mailto:xen-devel-boun...@lists.xen.org] On Behalf Of Konrad
Rzeszutek Wilk
Sent: 10 March 2016 19:12
To: Daniel De Graaf ; Ian Jackson
; jbeul...@suse.com
Cc: xen-de...@lists.xenproject.org; car...@cardoe.com; Andrew Cooper
Subject: Re: [Xen-devel]
h the
fact that you can't reconfigure your cpupools)?
-George
Sorry, lost the original thread so replying at the top of mail chain.
+static XSM_INLINE int xsm_schedop_pin_temp(XSM_DEFAULT_VOID)
+{
+ XSM_ASSERT_ACTION(XSM_PRIV);
+ return xsm_default_action(action, current->domain, NULL);
Based on the discussion below, can I assume there is an agreement for using
processor model for filtering or chipset ID will be the preferred candidate.
Thanks
Anshul Makkar
-Original Message-
From: Tian, Kevin [mailto:kevin.t...@intel.com]
Sent: 26 November 2015 07:17
To: Malcolm
Snabbswitch (virtualized switch) also encountered similar problem :
https://groups.google.com/forum/#!topic/snabb-devel/xX0yFzeXylI
Thanks
Anshul Makkar
-Original Message-
From: Andrew Cooper [mailto:andrew.coop...@citrix.com]
Sent: 01 December 2015 10:34
To: Konrad Rzeszutek Wilk
Cc
From: Anshul Makkar
Sandybridge or earlier processors don't have huge page support for
IOTLB which leads to fallback on 4k pages and causes performance issues.
Shared EPT will be disabled only if the user has not provided explicit
choice on the command line.
Signed-off-by: Anshul M
62 matches
Mail list logo