Each number means what ?
Each number means what ?
-邮件原件-
发件人: dunl...@gmail.com [mailto:dunl...@gmail.com] 代表 George Dunlap
发送时间: 2015年5月11日 18:05
收件人: 蒋雄伟(蒋冲)
抄送: xen-de...@lists.xensource.com; xen-devel@lists.xen.org
主题: Re: [Xen-devel] 答复: 答复: about xenalyze
On Fri,
On 05/11/2015 07:50 PM, Ian Campbell wrote:
On Fri, 2015-05-08 at 17:33 +0800, Yang Hongyang wrote:
Move the memory allocation before the concrete live/nolive save
in order to avoid the free/alloc memory loop when using Remus.
Rename the to_send bitmap to dirty_bitmap.
Doing the rename in a
On Tue, 2015-05-12 at 00:47 +, Hao, Xudong wrote:
Xen unstable tree doesn't update for about one month, I saw staging
tree has many new commit these days, how often does it merging? Does
unstable tree still work?
There have been some teething troubles with the push gates since osstest
On 05/11/2015 07:49 PM, Ian Campbell wrote:
On Mon, 2015-05-11 at 12:47 +0100, Ian Campbell wrote:
Actually this is another trick that I need to deal with those
hypercall macros.
DECLARE_HYPERCALL_BUFFER_SHADOW will define a user pointer dirty_bitmap
and a shadow buffer, although
On Mon, 2015-05-11 at 15:33 +0100, George Dunlap wrote:
On 05/11/2015 03:23 PM, Ian Campbell wrote:
On Mon, 2015-05-11 at 16:15 +0200, Olaf Hering wrote:
On Thu, May 07, George Dunlap wrote:
On Thu, May 7, 2015 at 11:35 AM, Olaf Hering o...@aepfle.de wrote:
On 05/11/2015 07:01 PM, Andrew Cooper wrote:
On 11/05/15 11:48, Hongyang Yang wrote:
On 05/11/2015 05:00 PM, Andrew Cooper wrote:
On 11/05/15 07:28, Hongyang Yang wrote:
On 05/09/2015 02:12 AM, Andrew Cooper wrote:
On 08/05/15 10:33, Yang Hongyang wrote:
This patchset implement the Remus
Il 11/05/2015 16:13, Wei Liu ha scritto:
On Mon, May 11, 2015 at 03:53:44PM +0200, Fabio Fantoni wrote:
[...]
Can someone help me to complete/fix it please?
Can someone help me about virtio disks support in libxl please?
After this reply about seabios seems can be useful try virtio:
On Tue, 2015-05-12 at 04:42 +, osstest service user wrote:
flight 55270 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55270/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-i386-freebsd10-amd64 13
These are both firmware which is used only with qemuu tests. There is
no point testing either PV guets or HVM with qemut (which always uses
rombios).
Comparing before and after of:
./standalone-generate-dump-flight-runvars | \
awk '// {printf %-32s %s\n, $1, $2}' | \
sort -u
Results in:
On Tue, 2015-05-12 at 15:18 +0800, Yang Hongyang wrote:
On 05/11/2015 07:53 PM, Ian Campbell wrote:
On Fri, 2015-05-08 at 17:33 +0800, Yang Hongyang wrote:
Define a user pointer that can access the hypercall buffer data
Useful when you only need to access the hypercall buffer data
Can
flight 55274 xen-unstable real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55274/
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. 50405
Ian Campbell writes ([Xen-devel] [PATCH OSSTEST v5 00/24] add distro domU
testing flight):
As last time there are some patches in here which I think will be useful
to the Intel folks doing the nested virt testing, specifically the
refactoring of how overlays and ssh host keys are done will be
Fix LINUX_URL to be the canonical url.
Change LIBVIRT_* to point to the tree and branch tested by osstest.
Add comments for the QEMU_TRADITIONAL_REVISION and SEABIOS_REVISION
tags.
Signed-off-by: Stefano Stabellini stefano.stabell...@eu.citrix.com
---
defconfig | 10 +++---
1 file changed,
Refactor xenvif_rx_action by dividing it into build_gops and
submit, similar to what xenvif_tx_action looks like.
Signed-off-by: Joao Martins joao.mart...@neclab.eu
---
drivers/net/xen-netback/netback.c | 180 --
1 file changed, 96 insertions(+), 84
Checks for feature-persistent that indicates persistent grants
support. Adds max_persistent_grants module param that specifies the max
number of persistent grants, which if set to zero disables persistent
grants.
Signed-off-by: Joao Martins joao.mart...@neclab.eu
---
Ian Campbell writes (Re: [OSSTEST v6 10/24] distros: add support for
installing Debian PV guests via d-i, flight and jobs):
On Tue, 2015-05-12 at 16:25 +0100, Ian Jackson wrote:
Shouldn't GUEST_dist be called GUEST_suite (which is probably
already set in various places) ?
Hrm, yes I think
On xenvif_start_xmit() we have an additional queue to the netback RX
kthread that will sends the packet. When using burst1 pktgen sets
skb-xmit_more to tell the driver that there more skbs in the queue.
However, pktgen transmits the same skb burst times, which leads to
the BUG below. Long story
Refactors a little bit how grants are stored by moving
grant_rx_ref/grant_tx_ref and grant_tx_page to its
own structure, namely struct grant.
Signed-off-by: Joao Martins joao.mart...@neclab.eu
---
drivers/net/xen-netfront.c | 56 ++
1 file changed, 32
On Tue, 2015-05-12 at 12:01 -0400, Meng Xu wrote:
Hi Dario,
Hi,
2015-05-12 10:06 GMT-04:00 Dario Faggioli dario.faggi...@citrix.com:
--- a/xen/common/sched_rt.c
+++ b/xen/common/sched_rt.c
@@ -124,6 +124,24 @@
#define TRC_RTDS_BUDGET_REPLENISH TRC_SCHED_CLASS_EVT(RTDS, 4)
#define
Hi Dario,
2015-05-12 10:06 GMT-04:00 Dario Faggioli dario.faggi...@citrix.com:
In fact, printing the cpupool's CPU online mask
for each vCPU is just redundant, as that is the
same for all the vCPUs of all the domains in the
same cpupool, while hard affinity is already part
of the output of
feature-persistent check on xenbus for persistent grants
support on the backend.
Signed-off-by: Joao Martins joao.mart...@neclab.eu
---
drivers/net/xen-netfront.c | 16
1 file changed, 16 insertions(+)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index
Refactors how grants are claimed/released/revoked by moving that code
into claim_grant and release_grant helpers routines that can be shared
in both TX/RX path.
Signed-off-by: Joao Martins joao.mart...@neclab.eu
---
drivers/net/xen-netfront.c | 87 ++
Instead of grant/revoking the buffer related to the skb, it will use
an already granted page and memcpy to it. The grants will be mapped
by xen-netback and reused overtime, but only unmapped when the vif
disconnects, as opposed to every packet.
This only happens if the backend supports
Ian Jackson writes (Re: [OSSTEST v6 13/24] distros: support PV guest install
from Debian netinst media.):
Ian Campbell writes ([OSSTEST v6 13/24] distros: support PV guest install
from Debian netinst media.):
+my $baseurl = $cd eq current ?
+
flight 55348 rumpuserxen real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55348/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-amd64-rumpuserxen 5 rumpuserxen-build fail REGR. vs. 33866
On 12/05/15 17:52, Razvan Cojocaru wrote:
On 05/12/2015 06:53 PM, Andrew Cooper wrote:
On 12/05/15 16:50, Andrew Cooper wrote:
On 12/05/15 16:49, Razvan Cojocaru wrote:
On 05/12/2015 06:35 PM, Jan Beulich wrote:
On 12.05.15 at 16:58, rcojoc...@bitdefender.com wrote:
+/* Supported values for
On Wednesday 07 January 2015 19:20:38 Andrew Cooper wrote:
On 16/12/14 19:33, Mihai Donțu wrote:
Implemented xmem_pool_check(), xmem_pool_check_locked() and
xmem_pool_check_unlocked() to verity the integrity of the TLSF matrix.
Signed-off-by: Mihai Donțu mdo...@bitdefender.com
This
Ian Campbell writes ([OSSTEST v6 15/24] distros: add branch infrastructure):
Since the distro nightlies are not version controlled we cannot use
the usual mechanisms for detecting regressions. Special case things
appropriately. We use an OLD_REVISION of flight-NNN to signify that
the old
Implement the necessary routines for managing the grant tree. These
routines are ported from blkback driver and slightly modified to be
more generic. This patch is separated because it relates to code that
could be shared with other drivers, in case persistent grants are adopted.
The changes
This patch implements persistent grants for xen-net{back,front}. There has
been work on persistent grants in the past[1], but the one here
described is a bit different: 1) using zerocopy skbs for RX path in
xen-netfront as opposed to memcpy every packet; 2) using a tree to store
the grants (and
flight 55349 libvirt real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55349/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-amd64-libvirt 11 guest-start fail REGR. vs. 53854
Tests which did not
2015-05-12 10:06 GMT-04:00 Dario Faggioli dario.faggi...@citrix.com:
--- a/xen/common/sched_rt.c
+++ b/xen/common/sched_rt.c
@@ -124,6 +124,24 @@
#define TRC_RTDS_BUDGET_REPLENISH TRC_SCHED_CLASS_EVT(RTDS, 4)
#define TRC_RTDS_SCHED_TASKLETTRC_SCHED_CLASS_EVT(RTDS, 5)
+ /*
On 05/12/2015 08:54 PM, Andrew Cooper wrote:
On 12/05/15 17:52, Razvan Cojocaru wrote:
On 05/12/2015 06:53 PM, Andrew Cooper wrote:
On 12/05/15 16:50, Andrew Cooper wrote:
On 12/05/15 16:49, Razvan Cojocaru wrote:
On 05/12/2015 06:35 PM, Jan Beulich wrote:
On 12.05.15 at 16:58,
On 05/07/15 09:43, Tim Deegan wrote:
At 14:18 +0200 on 07 May (1431008321), Dario Faggioli wrote:
On Thu, 2015-05-07 at 12:00 +0100, Tim Deegan wrote:
At 17:38 +0100 on 05 May (1430847486), Lars Kurth wrote:
The current proposal is to Rename the Xen Project Hackathons to either
a) Xen
flight 55331 qemu-mainline real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55331/
Failures :-/ but no regressions.
Regressions which are regarded as allowable (not blocking):
test-amd64-i386-libvirt 11 guest-start fail REGR. vs. 54832
Ian Jackson wrote:
Andrew Cooper writes (Tools backports for 4.5):
From the XenServer patch queue,
(Andrew forgot to CC xen-devel, doing that now:)
3380f5b6270e ocaml/xenctrl: Check return values from hypercalls
c8945d516134 ocaml/xenctrl: Make failwith_xc() thread safe
On 05/12/2015 06:53 PM, Andrew Cooper wrote:
On 12/05/15 16:50, Andrew Cooper wrote:
On 12/05/15 16:49, Razvan Cojocaru wrote:
On 05/12/2015 06:35 PM, Jan Beulich wrote:
On 12.05.15 at 16:58, rcojoc...@bitdefender.com wrote:
+/* Supported values for the vm_event_write_ctrlreg index. */
Signed-off-by: Stefano Stabellini stefano.stabell...@eu.citrix.com
---
Changes in v5:
- remove setting DEFAULT_REVISION_RAISIN to master
- disable raisin when branch is not xen*
- do not overwrite raisin default variables unless both revision_* and
tree_* are set of a given component
Changes in
It starts by doing a lookup in the tree for a gref. If no persistent
grant is found on the tree, it will do grant copy and prepare
the grant maps. Finally valides the grant map and adds it to the tree.
After mapped these grants can be pulled from the tree in the subsequent
requests. If it's out of
By introducing persistent grants we speed up the RX thread with the
decreased copy cost, that leads to a throughput decrease of 20%.
It is observed that the rx_queue stays mostly at 10% of its capacity,
as opposed to full capacity when using grant copy. And a finer measure
with lock_stat (below
Prints the total/max number of persistent grants and how many of
them are in use.
Signed-off-by: Joao Martins joao.mart...@neclab.eu
---
drivers/net/xen-netback/xenbus.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/drivers/net/xen-netback/xenbus.c
Stefano Stabellini writes ([PATCH v5] OSSTEST: introduce a raisin build test):
Signed-off-by: Stefano Stabellini stefano.stabell...@eu.citrix.com
...
+echo config XEN_URL=\\$r{tree_xen}\\
+echo config XEN_REVISION=\\$r{revision_xen}\\
This is very repetitive. In ts-xen-build, the
Hi Ian,
On 12/05/15 16:02, Ian Campbell wrote:
On Tue, 2015-05-05 at 17:44 +0530, Vijay Kilari wrote:
Hi,
As discussed, here is the design doc/txt.
There seems to be no consideration of multiple guests or VCPUs all
accessing one or more vITS in parallel and the associated issues around
On 05/12/2015 08:10 PM, Andrew Cooper wrote:
On 12/05/15 12:25, Yang Hongyang wrote:
When we use a DECLARE_HYPERCALL_BUFFER_SHADOW define a user
pointer '_name' and a shadow xc_hypercall_buffer_t.
then call xc_hypercall_buffer_free_pages(_xch, _name, _nr),
the complier will report '_name'
On 05/12/2015 08:11 PM, Andrew Cooper wrote:
On 12/05/15 12:25, Yang Hongyang wrote:
introduce setup() and cleanup() which subsume the
ctx-save.ops.{setup,cleanup}() calls.
The SHADOW_OP_OFF hypercall also included in the cleanup().
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
CC: Ian
Split read/handle qemu info. The receiving of qemu info
should be done while we receive the migration stream,
handle_qemu will be called when the stream complete.
Otherwise, it will break Remus because read_record()
won't read qemu info and stream_complete will be called
at failover.
Move record handle codes into a function process_record().
It will be used multiple times by Remus.
No functional change.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson
In last patch we added dirty bitmap to the save context,
we no longer need to pass this param to send_some_pages.
We can get dirty bitmap from the save context.
'entries' should stay as it is a useful sanity check.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper
From: Andrew Cooper andrew.coop...@citrix.com
There are some records which should only be sent once in the stream, and not
repeated for each checkpoint. {start,end}_of_stream() become per-checkpoint,
and a new start_of_stream() is introduced.
There is no resulting change record order, but the
From: Andrew Cooper andrew.coop...@citrix.com
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC: Yang Hongyang yan...@cn.fujitsu.com
---
tools/libxc/include/xenguest.h | 1 +
There are cases that we only need to use the hypercall buffer data,
and do not use the xc_hypercall_buffer_t struct.
DECLARE_HYPERCALL_BUFFER_SHADOW define a user pointer that can allow
us to access the hypercall buffer data but it also define a
xc_hypercall_buffer_t that we don't use, the
On 05/12/2015 08:05 PM, Andrew Cooper wrote:
On 12/05/15 12:25, Yang Hongyang wrote:
From: Andrew Cooper andrew.coop...@citrix.com
Checkpointed streams need to signal the end of a consistent view of VM state,
and the start of the libxl data.
Signed-off-by: Andrew Cooper
flight 55353 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55353/
Failures :-/ but no regressions.
Regressions which are regarded as allowable (not blocking):
test-amd64-i386-freebsd10-amd64 13 guest-localmigrate fail REGR. vs. 53940
test-amd64-i386-freebsd10-i386 13
2015-05-12 18:59 GMT-04:00 Dario Faggioli dario.faggi...@citrix.com:
On Sun, 2015-05-10 at 22:36 -0400, Meng Xu wrote:
Hi Dario and George,
Hi Meng,
Hi Dario,
I gave a quick look at the slides. Nice work.
Thanks for your encouragement! :-)
Although I don't have much time, I also wanted
From: Andrew Cooper andrew.coop...@citrix.com
Checkpointed streams need to signal the end of a consistent view of VM state,
and the start of the libxl data.
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
Signed-off-by: David Vrabel david.vra...@citrix.com
Signed-off-by: Yang Hongyang
rename send_some_pages to send_dirty_pages, no functional change.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC: Andrew Cooper andrew.coop...@citrix.com
---
From: Andrew Cooper andrew.coop...@citrix.com
The migration code itself should be able to validly assume all mandatory
callbacks are set up.
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu
Move the memory allocation before the concrete live/nolive save
in order to avoid the free/alloc memory loop when using Remus.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson
introduce bitmap_set() to set the entire bitmap.
in send_all_pages(), set the entire bitmap and call send_dirty_pages().
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson
introduce setup() and cleanup() which subsume the
ctx-save.ops.{setup,cleanup}() calls.
The SHADOW_OP_OFF hypercall also included in the cleanup().
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC:
rename to_send to dirty_bitmap.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
---
tools/libxc/xc_sr_save.c | 22
When we use a DECLARE_HYPERCALL_BUFFER_SHADOW define a user
pointer '_name' and a shadow xc_hypercall_buffer_t.
then call xc_hypercall_buffer_free_pages(_xch, _name, _nr),
the complier will report '_name' unused error, it's because
xc_hypercall_buffer_free_pages() is a MACRO and '_name'
This is the combination of Andrew Cooper's misc patches and mine
to aid migration v2 Remus support.
See individual patches for details.
Git tree available at:
https://github.com/macrosheep/xen/tree/misc-remus-v5
v4-v5
Most of the changes are trival, like drop brackets in
Il 11/05/2015 17:04, Fabio Fantoni ha scritto:
Il 21/04/2015 14:53, Stefano Stabellini ha scritto:
On Tue, 21 Apr 2015, Fabio Fantoni wrote:
Il 21/04/2015 12:49, Stefano Stabellini ha scritto:
On Mon, 20 Apr 2015, Fabio Fantoni wrote:
I updated xen and qemu from xen 4.5.0 with its upstream
On 05/12/2015 04:19 PM, Ian Campbell wrote:
On Tue, 2015-05-12 at 15:18 +0800, Yang Hongyang wrote:
On 05/11/2015 07:53 PM, Ian Campbell wrote:
On Fri, 2015-05-08 at 17:33 +0800, Yang Hongyang wrote:
Define a user pointer that can access the hypercall buffer data
Useful when you only need
On 11.05.15 at 17:15, konrad.w...@oracle.com wrote:
On Fri, May 08, 2015 at 01:08:05PM +0100, Jan Beulich wrote:
1- or 2-byte operations never alter the high halves of registers.
That is impressive. How possibly did you spot this?
Was it due to reviewing some other code that got you
On 05/12/2015 05:43 PM, Ian Campbell wrote:
On Tue, 2015-05-12 at 17:24 +0800, Yang Hongyang wrote:
On 05/12/2015 04:19 PM, Ian Campbell wrote:
On Tue, 2015-05-12 at 15:18 +0800, Yang Hongyang wrote:
On 05/11/2015 07:53 PM, Ian Campbell wrote:
On Fri, 2015-05-08 at 17:33 +0800, Yang
On 05/05/15 13:34, Jan Beulich wrote:
On 30.04.15 at 15:28, david.vra...@citrix.com wrote:
From: Malcolm Crossley malcolm.cross...@citrix.com
Performance analysis of aggregate network throughput with many VMs
shows that performance is signficantly limited by contention on the
maptrack lock
The right place for talk_to_blkback() to query backend features and transport
parameters is after backend entered XenbusStateInitWait. There is no problem
with this yet, but it is an violation of the design and furthermore it would not
allow frontend/backend to negotiate 'multi-page' and
See individual patches for details.
Git tree available at:
git://xenbits.xen.org/people/andrewcoop/xen.git remus-migv2-v3
~Andrew
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC: Yang Hongyang yan...@cn.fujitsu.com
---
tools/libxc/include/xenguest.h |1 +
tools/libxc/xc_sr_common.h |3 +++
There are some records which should only be sent once in the stream, and not
repeated for each checkpoint. {start,end}_of_stream() become per-checkpoint,
and a new start_of_stream() is introduced.
There is no resulting change record order, but the X86_PV_INFO record is
identified as once per
Checkpointed streams need to signal the end of a consistent view of VM state,
and the start of the libxl data.
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: David Vrabel david.vra...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei
George,
given that this question is coming up occasionally, I was going to create a
wiki page. I can't find xenalyze.html and xenalyse on xenbuts though
I can find
* https://github.com/nikmis/xentrace-analysis - but that does not contain the
html file and is probably a different tool
*
On 12.05.15 at 13:01, david.vra...@citrix.com wrote:
On 05/05/15 13:34, Jan Beulich wrote:
On 30.04.15 at 15:28, david.vra...@citrix.com wrote:
@@ -1430,6 +1456,17 @@ gnttab_setup_table(
gt = d-grant_table;
write_lock(gt-lock);
+/* Tracking of mapped foreign frames table */
On Tue, 2015-05-12 at 10:25 +0100, George Dunlap wrote:
On 05/12/2015 10:16 AM, Stefano Stabellini wrote:
Signed-off-by: Stefano Stabellini stefano.stabell...@eu.citrix.com
What's the purpose of doing this? In particular it seems like having to
manually update the changesets as we go along
[Adjusting the Cc list:
- removing hypervisor only people
- adding more tools maintainers
- adding George]
On Thu, 2015-05-07 at 12:05 -0500, Chong Li wrote:
Change sched_rtds_domain_get/set functions to support per-VCPU settings for
RTDS scheduler.
More on this patch (I had to run
On 05/12/2015 05:40 PM, Andrew Cooper wrote:
On 12/05/15 09:12, Yang Hongyang wrote:
That sounds like COLO wants a should_checkpoint() callback which
separates the decision to make a checkpoint from the logic of
implementing a checkpoint.
We use checkpoint callback to do
Ian Campbell writes (Re: [PATCH v4] OSSTEST: introduce a raisin build test):
More generally this now ties everything together into one build job.
I hadn't spotted that.
I think this is very undesirable.
Currently while bisecting osstest can try and reuse e.g. the xen job
with a different
Move the memory allocation before the concrete live/nolive save
in order to avoid the free/alloc memory loop when using Remus.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC:
There are cases that we only need to use the hypercall buffer data,
and do not use the xc_hypercall_buffer_t struct.
DECLARE_HYPERCALL_BUFFER_SHADOW define a user pointer that can allow
us to access the hypercall buffer data but it also define a
xc_hypercall_buffer_t that we don't use, the
rename to_send to dirty_bitmap.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC: Andrew Cooper andrew.coop...@citrix.com
---
tools/libxc/xc_sr_save.c | 22
introduce setup() and cleanup() which subsume the
ctx-save.ops.{setup,cleanup}() calls.
The SHADOW_OP_OFF hypercall also included in the cleanup().
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu
This is the combination of Andrew Cooper's misc patches and mine
to aid migration v2 Remus support.
See individual patches for details.
Git tree available at:
https://github.com/macrosheep/xen/tree/misc-remus-v4
Andrew Cooper (4):
libxc/migration: Be rather stricter with illformed callers
From: Andrew Cooper andrew.coop...@citrix.com
The migration code itself should be able to validly assume all mandatory
callbacks are set up.
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu
From: Andrew Cooper andrew.coop...@citrix.com
Checkpointed streams need to signal the end of a consistent view of VM state,
and the start of the libxl data.
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
Signed-off-by: David Vrabel david.vra...@citrix.com
Signed-off-by: Yang Hongyang
From: Andrew Cooper andrew.coop...@citrix.com
There are some records which should only be sent once in the stream, and not
repeated for each checkpoint. {start,end}_of_stream() become per-checkpoint,
and a new start_of_stream() is introduced.
There is no resulting change record order, but the
From: Andrew Cooper andrew.coop...@citrix.com
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC: Yang Hongyang yan...@cn.fujitsu.com
---
tools/libxc/include/xenguest.h | 1 +
Move record handle codes into a function process_record().
It will be used multiple times by Remus.
No functional change.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson
On Tue, 2015-05-12 at 12:20 +0100, Stefano Stabellini wrote:
We currently have build jobs for both normal and the XSM case.
Duplicating the raisin jobs is also going to duplicate the build for
everything else it builds, which might be wasteful?
We could disable the raisin build if [
The migration code itself should be able to validly assume all mandatory
callbacks are set up.
Signed-off-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell ian.campb...@citrix.com
CC: Ian Jackson ian.jack...@eu.citrix.com
CC: Wei Liu wei.l...@citrix.com
CC: Yang Hongyang
Some basic infrastructure.
CC: Yang Hongyang yan...@cn.fujitsu.com
---
I started attempting to explain this in text, but concluded that pseudocode
would be easier.
My design for the restore side handling was always based around a single do {
} while ( not end ) loop, and specifically not to
From: Yang Hongyang yan...@cn.fujitsu.com
Move record handle codes into a function process_record().
It will be used multiple times by Remus.
No functional change.
Signed-off-by: Yang Hongyang yan...@cn.fujitsu.com
Reviewed-by: Andrew Cooper andrew.coop...@citrix.com
CC: Ian Campbell
On Tue, 2015-05-12 at 17:24 +0800, Yang Hongyang wrote:
On 05/12/2015 04:19 PM, Ian Campbell wrote:
On Tue, 2015-05-12 at 15:18 +0800, Yang Hongyang wrote:
On 05/11/2015 07:53 PM, Ian Campbell wrote:
On Fri, 2015-05-08 at 17:33 +0800, Yang Hongyang wrote:
Define a user pointer that
Hi,
On 12/05/15 11:07, Ian Campbell wrote:
On Mon, 2015-05-11 at 18:09 +, osstest service user wrote:
flight 55255 linux-linus real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55255/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could
On Tue, 12 May 2015, Ian Campbell wrote:
On Tue, 2015-05-12 at 10:20 +0100, Stefano Stabellini wrote:
Signed-off-by: Stefano Stabellini stefano.stabell...@eu.citrix.com
---
Changes in v4:
- introduce enable_raisin in mfi-common: only build raisin when building
xen-unstable
-
Split read/handle qemu info. The receiving of qemu info
should be done while we receive the migration stream,
handle_qemu will be called when the stream complete.
Otherwise, it will break Remus because read_record()
won't read qemu info and stream_complete will be called
at failover.
flight 55291 seabios real [real]
http://logs.test-lab.xenproject.org/osstest/logs/55291/
Failures :-/ but no regressions.
Regressions which are regarded as allowable (not blocking):
test-amd64-i386-libvirt 11 guest-start fail REGR. vs. 52630
Signed-off-by: Stefano Stabellini stefano.stabell...@eu.citrix.com
---
Changes in v4:
- introduce enable_raisin in mfi-common: only build raisin when building
xen-unstable
- start off from the default raisin config, then append osstest config
options to it
- do not write variable to the raisin
On Tue, May 12, 2015 at 3:18 AM, 蒋雄伟(蒋冲) xiongwei.ji...@alibaba-inc.com wrote:
I try again with -e 0xaf000, and run xenalyze --summary. The attached is
the output.
Again, please don't top-post: configure your mailer to quote properly,
and reply in-line (as I am doing now).
So it looks like
1 - 100 of 219 matches
Mail list logo
