Add Memory Bandwidth Monitoring(MBM) for VMs. Two types of monitoring
are supported: total and local memory bandwidth monitoring. To use it,
CMT should be enabled in hypervisor.
Signed-off-by: Chao Peng
---
docs/man/xl.pod.1 |9 +
tools/libxc/include/xenctrl.h |2 +
tools
This is the tools side wrapper for XEN_SYSCTL_PSR_CMT_get_l3_event_mask
of XEN_SYSCTL_psr_cmt_op.
Signed-off-by: Chao Peng
---
tools/libxc/include/xenctrl.h |1 +
tools/libxc/xc_psr.c | 17 +
tools/libxl/libxl.h |1 +
tools/libxl/libxl_psr.c |
Make some internal routines common so that total/local memory bandwidth
monitoring in the next patch can make use of them.
Signed-off-by: Chao Peng
---
tools/libxl/libxl_psr.c | 44 -
tools/libxl/xl_cmdimpl.c | 54 +++--
L3 event mask indicates the event types supported in host, including
cache occupancy event as well as local/total memory bandwidth events
for Memory Bandwidth Monitoring(MBM). Expose it so all these events
can be monitored in user space.
Signed-off-by: Chao Peng
Reviewed-by: Andrew Cooper
Acked-
- space: remove space after '(' or before ')' in 'if' condition;
- indention: align function definition/call arguments;
Signed-off-by: Chao Peng
Acked-by: Wei Liu
---
tools/libxc/include/xenctrl.h |8
tools/libxc/xc_psr.c | 10 +-
tools/libxl/libxl.h |
Changes from v2:
* Remove the usage of "static" to cache data in xc;
NOTE: Other places that already existed before are not touched due to
the needs for API change. Will fix in separate patch if desirable.
* Coding style;
Changes from v1:
* Move event type check from xc to xl;
* Add retr
From: "Palik, Imre"
In Dom0's the use of the TSC clocksource (whenever it is stable enough to
be used) instead of the Xen clocksource should not cause any issues, as
Dom0 VMs never live-migrated. The TSC clocksource is somewhat more
efficient than the Xen paravirtualised clocksource, thus it sho
flight 33382 libvirt real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/33382/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-armhf-libvirt 5 libvirt-build fail REGR. vs. 32648
build-i386-libvirt
On Mon, Jan 12, Ian Campbell wrote:
> @devs -- we obviously need to do something about this (too late for 4.5,
> but for 4.6 + backport). Perhaps there is some alternative systemd
> construction which disassociates the actual path from the abstract
> service "xenstored dir mounted"?
I dont think
There was no v2 (v2 was not created properly).
Yes, 1/4 was the cover letter. And 4/4 was not correct.
Thank you for applying the patches.
On Mon, Jan 12, 2015 at 11:29 PM, Ian Campbell wrote:
> On Tue, 2014-10-21 at 18:04 +0100, George Dunlap wrote:
>
> Just getting back to these after the free
>>> On 12.01.15 at 18:36, wrote:
> On 01/12/2015 02:00 AM, Jan Beulich wrote:
> On 10.01.15 at 00:04, wrote:
>>> On 01/09/2015 02:41 PM, Andrew Cooper wrote:
Having some non-OS part of the guest swap the EPT tables and
accidentally turn a DMA buffer read-only is not going to end wel
On Tue, 13 Jan 2015, Sander Eikelenboom wrote:
>
> Monday, January 12, 2015, 4:01:00 PM, you wrote:
>
> > On 12/01/15 13:39, Jiang Liu wrote:
> >> Commit b81975eade8c ("x86, irq: Clean up irqdomain transition code")
> >> breaks xen IRQ allocation because xen_smp_prepare_cpus() doesn't invoke
> >
flight 33377 linux-linus real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/33377/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-i386-qemuu-rhel6hvm-intel 5 xen-boot fail REGR. vs. 32879
test-amd64-i386-xl-cr
flight 33379 xen-4.3-testing real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/33379/
Failures :-/ but no regressions.
Tests which are failing intermittently (not blocking):
test-amd64-amd64-xl-sedf-pin 15 guest-localmigrate/x10 fail pass in 9
test-amd64-amd64-xl-qemuu-win7-
On 01/12/2015 07:36 PM, Roger Pau Monné wrote:
> El 12/01/15 a les 8.09, Bob Liu ha escrit:
>>
>> On 01/09/2015 11:51 PM, Roger Pau Monné wrote:
>>> El 06/01/15 a les 14.19, Bob Liu ha escrit:
When there is no enough free grants, gnttab_alloc_grant_references()
will fail and block reques
On 01/12/2015 07:36 PM, Roger Pau Monné wrote:
> El 12/01/15 a les 8.09, Bob Liu ha escrit:
>>
>> On 01/09/2015 11:51 PM, Roger Pau Monné wrote:
>>> El 06/01/15 a les 14.19, Bob Liu ha escrit:
When there is no enough free grants, gnttab_alloc_grant_references()
will fail and block reques
On 13/01/15 08:14, Imre Palik wrote:
> From: "Palik, Imre"
>
> In Dom0's the use of the TSC clocksource (whenever it is stable enough to
> be used) instead of the Xen clocksource should not cause any issues, as
> Dom0 VMs never live-migrated. The TSC clocksource is somewhat more
> efficient than
>>> On 13.01.15 at 09:02, wrote:
> L3 event mask indicates the event types supported in host, including
> cache occupancy event as well as local/total memory bandwidth events
> for Memory Bandwidth Monitoring(MBM). Expose it so all these events
> can be monitored in user space.
>
> Signed-off-by:
On 13/01/15 02:23, Jiang Liu wrote:
> On 2015/1/12 23:01, David Vrabel wrote:
>> On 12/01/15 13:39, Jiang Liu wrote:
>>> Commit b81975eade8c ("x86, irq: Clean up irqdomain transition code")
>>> breaks xen IRQ allocation because xen_smp_prepare_cpus() doesn't invoke
>>> setup_IO_APIC(), so no irqdom
On Tue, 2015-01-13 at 13:52 +0800, Wen Congyang wrote:
> On 12/01/2014 10:21 PM, Euan Harris wrote:
> > Tools debug builds are built with optimization level -O1, inherited from
> > the CFLAGS definition in StdGNU.mk. Optimizations confuse the debugger,
> > and the comment justifying -O1 in StdGNU
>>> On 12.01.15 at 17:45, wrote:
> There is race when we clear the STATE_SCHED in the softirq
> - which allows the 'raise_softirq_for' to progress and
> schedule another dpci. During that time the other CPU could
> receive an interrupt and calls 'raise_softirq_for' and put
> the dpci on its per-cp
On Mon, Jan 12, 2015 at 7:31 PM, Ed White wrote:
> On 01/12/2015 10:00 AM, Ian Jackson wrote:
>> Ed White writes ("Re: [PATCH 00/11] Alternate p2m: support multiple copies
>> of host p2m"):
>>> The hypercalls are all there. My testing is all done in a Windows
>>> domU with the tests running insid
On Tue, 2015-01-13 at 09:45 +0100, Olaf Hering wrote:
> On Mon, Jan 12, Ian Campbell wrote:
>
> > @devs -- we obviously need to do something about this (too late for 4.5,
> > but for 4.6 + backport). Perhaps there is some alternative systemd
> > construction which disassociates the actual path fro
>>> On 12.01.15 at 18:35, wrote:
>> --- a/xen/drivers/passthrough/io.c
>> +++ b/xen/drivers/passthrough/io.c
>> @@ -804,7 +804,17 @@ static void dpci_softirq(void)
>> d = pirq_dpci->dom;
>> smp_mb(); /* 'd' MUST be saved before we set/clear the bits. */
>> if ( test_and_
>>> On 13.01.15 at 02:59, wrote:
> Is there any support from Xen for supporting MTRRs or any similar caching
> mechanisms such as PAT? How does Xen handle I/O ranges because different
> memory regions may have different caching requirements and any domU can be
> allocated I/O regions? Is the contr
On 01/13/2015 06:11 PM, Ian Campbell wrote:
> On Tue, 2015-01-13 at 13:52 +0800, Wen Congyang wrote:
>> On 12/01/2014 10:21 PM, Euan Harris wrote:
>>> Tools debug builds are built with optimization level -O1, inherited from
>>> the CFLAGS definition in StdGNU.mk. Optimizations confuse the debugge
On Mon, 2015-01-12 at 19:59 -0600, ayush ruia wrote:
> Hi all,
> Hoping that you could answer a few of my queries regarding cache
> properties for Xen.
>
> Is there any support from Xen for supporting MTRRs or any similar
> caching mechanisms such as PAT? How does Xen handle I/O ranges because
> d
On Tue, Jan 13, 2015 at 10:00:58AM +, Jan Beulich wrote:
> >>> On 13.01.15 at 09:02, wrote:
> > L3 event mask indicates the event types supported in host, including
> > cache occupancy event as well as local/total memory bandwidth events
> > for Memory Bandwidth Monitoring(MBM). Expose it so a
Acquiring a page for the source or destination of a grant copy is an
expensive operation. A common use case is for two adjacent grant copy
ops to operate on either the same source or the same destination page.
Instead of always acquiring and releasing destination and source pages
for each operati
On Mon, 2014-12-15 at 00:59 +0100, Samuel Thibault wrote:
> Martin Lucina, le Thu 11 Dec 2014 13:10:00 +0100, a écrit :
> > I've done this in the updated patch, take a look and let me know what you
> > think.
>
> I like that :)
>
> > From 3880f59159bf0a05b47b6e723091b1e7e4fb6814 Mon Sep 17 00:00:
> From: George Dunlap
> Sent: Monday, January 12, 2015 10:20 PM
>
> On Mon, Jan 12, 2015 at 12:28 PM, Tian, Kevin wrote:
> >> From: George Dunlap
> >> Sent: Monday, January 12, 2015 8:14 PM
> >>
> >> On Mon, Jan 12, 2015 at 11:22 AM, Tian, Kevin
> wrote:
> >> >> From: Jan Beulich [mailto:jbeul..
Ian Campbell, le Tue 13 Jan 2015 11:01:06 +, a écrit :
> Applied, thanks.
BTW, it is perhaps worth backporting to previous versions of Xen.
Samuel
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
On 01/13/2015 06:11 PM, Ian Campbell wrote:
> On Tue, 2015-01-13 at 13:52 +0800, Wen Congyang wrote:
>> On 12/01/2014 10:21 PM, Euan Harris wrote:
>>> Tools debug builds are built with optimization level -O1, inherited from
>>> the CFLAGS definition in StdGNU.mk. Optimizations confuse the debugge
This was disabled by default in Xen 4.4. Since xend has now been removed from
the tree I don't believe anything is using it. Note this is blktap1
only, blktap2 is untouched.
I've split this into two patches, the first disconnects all the blktap1
bits from the build, the second removes the actual t
Ed White writes ("Re: [PATCH 00/11] Alternate p2m: support multiple copies of
host p2m"):
> On 01/12/2015 10:00 AM, Ian Jackson wrote:
> > To support this code in-tree, I think we will need Open Source code
> > for exercising it, surely ?
>
> I'm hoping that, as Andrew says, there will be people
This was disabled by default in Xen 4.4. Since xend has now been
removed from the tree I don't believe anything is using it.
We need to pass an explicit CONFIG_BLKTAP1=n to qemu-xen-traditional
otherwise it defaults to y and doesn't build.
This patch does all the ground work, the tools/blktap dir
Now that it is unhooked we can just remove it.
Signed-off-by: Ian Campbell
---
.gitignore |5 -
.hgignore |5 -
tools/blktap/Makefile | 13 -
tools/blktap/README | 122 --
tools/blktap/dri
On 09/01/15 21:26, Ed White wrote:
> Add the basic data structures needed to support alternate p2m's and
> the functions to initialise them and tear them down.
>
> Although Intel hardware can handle 512 EPTP's per hardware thread
> concurrently, only 10 per domain are supported in this patch for
>
Jan Beulich writes ("Re: [Xen-devel] [PATCH 00/11] Alternate p2m: support
multiple copies of host p2m"):
> On 12.01.15 at 18:36, wrote:
> > Should some version of this patch series be accepted, my hope is that
> > someone who does understand xsm policies would put the appropriate checks
> > in pl
On Tue, 2015-01-13 at 19:17 +0800, Wen Congyang wrote:
> On 01/13/2015 06:11 PM, Ian Campbell wrote:
> > On Tue, 2015-01-13 at 13:52 +0800, Wen Congyang wrote:
> >> On 12/01/2014 10:21 PM, Euan Harris wrote:
> >>> Tools debug builds are built with optimization level -O1, inherited from
> >>> the CF
On Tue, 2015-01-13 at 11:16 +, Ian Campbell wrote:
Ugh, this really is v2, despite the subject (because of trimming patch
#2 I had to do the format-patch and send-mail separately, which broke my
usual finger macros)
> This was disabled by default in Xen 4.4. Since xend has now been
> removed
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Linus,
Please git pull the following tag:
git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip.git
stable/for-linus-3.19-rc4-tag
xen: bug fixes for 3.19-rc4
- - Several critical linear p2m fixes that prevented some hosts from
booting.
Thanks.
On Mon, 2014-11-03 at 14:16 +, Julien Grall wrote:
> Hi Frediano,
>
> FYI, this is only force DOM0 to use the normal GICv2 drivers.
>
> Do you have any plan to support hi-silicon vGIC in Xen? This would allow
> guest running with more than 8 cores.
I'm against exposing any non-standardised v
On Mon, 2014-11-03 at 10:11 +, Frediano Ziglio wrote:
> The GIC in this platform is mainly compatible with the standard
> GICv2 beside:
> - ITARGET is extended to 16 bit to support 16 CPUs;
> - SGI mask is extended to support 16 CPUs;
> - maximum supported interrupt is 510.
I'm not super keen
>>> On 13.01.15 at 12:03, wrote:
> Then I hope you understand now about our discussion in libxl/xen/
> hvmloader, based on the fact that conflict may not be avoided.
> That's the major open in original discussion with Jan. I'd like to
> give an example of the flow here per Jan's suggestion, star
On Mon, 2014-11-03 at 10:11 +, Frediano Ziglio wrote:
> Add this new platform to Xen.
> This platform require specific code to initialize CPUs.
What is the "bootwrapper"? Are you running this on real silicon or on an
emulator? Can the platform be made to do PSCI instead?
> +np_fab = dt_fi
On 09/01/15 21:26, Ed White wrote:
> Implement and hook up the code to enable VMX support of VMFUNC and #VE.
>
> VMFUNC leaf 0 (EPTP switching) and #VE are emulated on hardware that
> doesn't support them.
>
> Signed-off-by: Ed White
> ---
> xen/arch/x86/hvm/vmx/vmx.c | 138
> +++
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Tuesday, January 13, 2015 7:56 PM
>
> >>> On 13.01.15 at 12:03, wrote:
> > Then I hope you understand now about our discussion in libxl/xen/
> > hvmloader, based on the fact that conflict may not be avoided.
> > That's the major open in origi
Introduce a arch-independent routine to generate one vmemrange per
vnode. Also introduce arch-dependent routines for different
architectures because part of the process is arch-specific -- ARM has
yet have NUMA support and E820 is x86 only.
For those x86 guests who care about machine E820 map (i.e
On Fri, 2015-01-09 at 14:32 +, Wei Liu wrote:
> A pkg-config file for libxl. It also contains two variables
> (xenfirmwaredir and libexec_bin) so that tools that are very keen on
> knowing the locations of Xen binaries (say, libvirt) can use them to
> determine the location of the binaries.
>
This function gets the machine E820 map and sanitize it according to PV
guest configuration.
This will be used in later patch. No functional change introduced in
this patch.
Signed-off-by: Wei Liu
Cc: Ian Campbell
Cc: Ian Jackson
Cc: Dario Faggioli
Cc: Elena Ufimtseva
---
tools/libxl/libxl_
This function is used to check whether vNUMA configuration (be it
auto-generated or supplied by user) is valid.
The checks performed can be found in the comment of the function.
This vNUMA function (and future ones) is placed in a new file called
libxl_vnuma.c
Signed-off-by: Wei Liu
Cc: Ian Cam
Signed-off-by: Elena Ufimsteva
Signed-off-by: Wei Liu
Cc: Jan Beulich
---
Changes in v3:
1. Constify struct vnuma_info.
2. Don't print amount of ram of a vmemrange.
3. Process softirqs when dumping information.
4. Fix format string.
Changes in v2:
1. Use unsigned int for loop vars.
2. Use strlc
>From libxc's point of view, it only needs to know vnode to pnode mapping
and size of each vnode to allocate memory accordingly. Add these fields
to xc_dom structure.
The caller might not pass in vNUMA information. In that case, a dummy
layout is generated for the convenience of libxc's allocation
Transform the user supplied vNUMA configuration into libxl internal
representations, and finally libxc representations. Check validity of
the configuration along the line.
Signed-off-by: Wei Liu
Cc: Ian Campbell
Cc: Ian Jackson
Cc: Dario Faggioli
Cc: Elena Ufimtseva
---
Changes in v3:
1. Add
A vnode consists of one or more vmemranges (virtual memory range). One
example of multiple vmemranges is that there is a hole in one vnode.
Currently we haven't exported vmemrange interface to libxl user.
Vmemranges are generated during domain build, so we have relevant
structures in domain build
A domain can contain several virtual NUMA nodes, hence we introduce an
array in libxl_domain_build_info.
libxl_vnode_info contains the size of memory in that node, the distance
from that node to every nodes, the underlying pnode and a bitmap of
vcpus.
Signed-off-by: Wei Liu
Cc: Ian Campbell
Cc:
Make XENMEM_increase_reservation and XENMEM_populate_physmap
vNUMA-aware.
That is, if guest requests Xen to allocate memory for specific vnode,
Xen can translate vnode to pnode using vNUMA information of that guest.
XENMEMF_vnode is introduced for the guest to mark the node number is in
fact virt
Hi all
This is version 3 of this series rebased on top of staging. More commit
messages have been written and I hope this can help make things clearer.
This patch series implements virtual NUMA support for both PV and HVM guest.
That is, admin can configure via libxl what virtual NUMA topology th
On Tue, Jan 13, 2015 at 11:16:10AM +, Ian Campbell wrote:
> This was disabled by default in Xen 4.4. Since xend has now been
> removed from the tree I don't believe anything is using it.
>
> We need to pass an explicit CONFIG_BLKTAP1=n to qemu-xen-traditional
> otherwise it defaults to y and d
On Wed, 2015-01-07 at 10:41 +, David Vrabel wrote:
> On 07/01/15 09:10, Olaf Hering wrote:
> > On Mon, Jan 05, Vitaly Kuznetsov wrote:
> >
> >> Wei Liu writes:
> >>
> >>> Olaf mentioned his concern about handling ballooned pages in
> >>> <20141211153029.ga1...@aepfle.de>. Is that point moot n
On Tue, Jan 13, Ian Campbell wrote:
> On Fri, 2015-01-09 at 14:32 +, Wei Liu wrote:
> > A pkg-config file for libxl. It also contains two variables
> > (xenfirmwaredir and libexec_bin) so that tools that are very keen on
> > knowing the locations of Xen binaries (say, libvirt) can use them to
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> Signed-off-by: Vitaly Kuznetsov
> ---
> xen/common/shutdown.c | 7 +++
> xen/include/public/sched.h | 3 ++-
> 2 files changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/xen/common/shutdown.c b/xen/common/shutdown.c
> in
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> Use letter 't' to indicate a domain in such state.
>
> Signed-off-by: Vitaly Kuznetsov
> ---
> tools/libxl/libxl_types.idl | 1 +
> tools/libxl/xl_cmdimpl.c | 2 +-
> tools/python/xen/lowlevel/xl/xl.c | 1 +
> 3 files ch
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> Use letter 't' to indicate a domain in such state.
For rese(t) I suppose? How about "(S)oft reset" or "soft (R)eset"?
(depending on how closely you think this operation aligns as a special
case of reset).
_
This patch includes configuration options parser and documentation.
Please find the hunk to xl.cfg.pod.5 for more information.
Signed-off-by: Wei Liu
Cc: Ian Campbell
Cc: Ian Jackson
Cc: Dario Faggioli
Cc: Elena Ufimtseva
---
Changes in v2:
1. Make vnuma_vdistances mandatory.
2. Use nested l
From: Tiejun Chen
We need to act on some specific hypercall error numbers, so
require the hypervisor view on the errno.h value rather than
just the build environment's number. So here link this headfile
from xen.
Acked-by: Jan Beulich
Signed-off-by: Tiejun Chen
---
.gitignore
Transform user supplied vNUMA configuration into libxl internal
representations then libxc representations. Check validity along the
line.
Libxc has more involvement in building vmemranges in HVM case compared
to PV case. The building of vmemranges is placed after xc_hvm_build
returns, because it
Signed-off-by: Wei Liu
Cc: Jan Beulich
---
1. Coding style fix.
2. Fix an error code.
3. Use unsigned int for loop variable.
Changes in v2:
1. Adapt to new vNUMA retrieval routine.
2. Move SLIT very late in secondary table build.
---
tools/firmware/hvmloader/acpi/acpi2_0.h |8 +++
tools/
Signed-off-by: Wei Liu
Cc: Jan Beulich
---
Changes in v3:
1. Fix hard tabs.
2. Fix subject line.
3. Add extra hunk to handle -ENOBUFS (hence drop Reviewed-by:)
---
xen/common/compat/memory.c | 46
xen/include/xlat.lst |2 ++
2 files change
Signed-off-by: Wei Liu
Cc: Jan Beulich
---
Changes in v3:
1. Remove redundant variable.
2. Coding style fix.
3. Add assertion.
Changes in v2:
1. Remove explicit zero initializers.
2. Adapt to new vNUMA retrieval routine.
3. Move SRAT very late in secondary table build.
---
tools/firmware/hvmload
The algorithm is more or less the same as the one used for PV guest.
Libxc gets hold of the mapping of vnode to pnode and size of each vnode
then allocate memory accordingly.
And then the function returns low memory end, high memory end and mmio
start to caller. Libxl needs those values to constru
Hvmloader issues XENMEM_get_vnumainfo hypercall and stores the
information retrieved in scratch space for later use.
Signed-off-by: Wei Liu
Cc: Jan Beulich
---
Changes in v3:
1. Move init_vnuma_info before ACPI stuff.
2. Fix errno.h inclusion.
3. Remove upper limits and use loop.
---
tools/firm
This is done with three major changes:
1. Rework internal representation of setting.
2. Extend grammar of parser.
3. Introduce new APIs.
New APIs introduced:
1. xlu_cfg_value_type
2. xlu_cfg_value_get_string
3. xlu_cfg_value_get_list
4. xlu_cfg_get_listitem2
Previous APIs work as before.
Nested
Disallow memory relocation when vNUMA is enabled, because relocated
memory ends up off node. Further more, even if we dynamically expand
node coverage in hvmloader, low memory and high memory may reside
in different physical nodes, blindly relocating low memory to high
memory gives us a sub-optimal
On Thu, 2014-12-18 at 13:57 +, Jan Beulich wrote:
> >>> On 11.12.14 at 14:45, wrote:
> > --- a/xen/common/domctl.c
> > +++ b/xen/common/domctl.c
> > @@ -1177,6 +1177,39 @@ long do_domctl(XEN_GUEST_HANDLE_PARAM(xen_domctl_t)
> > u_domctl)
> > }
> > break;
> >
> > +case XEN_DOMC
On 2015/1/13 18:09, David Vrabel wrote:
> On 13/01/15 02:23, Jiang Liu wrote:
>> On 2015/1/12 23:01, David Vrabel wrote:
>>> On 12/01/15 13:39, Jiang Liu wrote:
Commit b81975eade8c ("x86, irq: Clean up irqdomain transition code")
breaks xen IRQ allocation because xen_smp_prepare_cpus() do
flight 33387 linux-next real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/33387/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-i386-xl3 host-install(3) broken REGR. vs. 7
test-amd64-amd64-xl-qe
On Tue, Jan 13, 2015 at 01:19:05PM +0100, Olaf Hering wrote:
> On Tue, Jan 13, Ian Campbell wrote:
>
> > On Fri, 2015-01-09 at 14:32 +, Wei Liu wrote:
> > > A pkg-config file for libxl. It also contains two variables
> > > (xenfirmwaredir and libexec_bin) so that tools that are very keen on
>
On Sat, 2014-11-29 at 14:33 -1000, Justin T. Weaver wrote:
> when deciding which run queue to assign each vcpu to.
>
> There are two main changes in functionality that this patch introduces.
>
> First, in function runq_tickle, it tries to find idle pcpus in other run
> queues that the vcpu would
2015-01-13 11:54 GMT+00:00 Ian Campbell :
> On Mon, 2014-11-03 at 10:11 +, Frediano Ziglio wrote:
>> The GIC in this platform is mainly compatible with the standard
>> GICv2 beside:
>> - ITARGET is extended to 16 bit to support 16 CPUs;
>> - SGI mask is extended to support 16 CPUs;
>> - maximum
On Tue, Jan 13, 2015 at 11:03 AM, Tian, Kevin wrote:
>> Right; so the "report" in this case is "report to the guest".
>>
>> As I said, I think that's confusing terminology; after all, we want to
>> report to the guest all holes that we make, and only the holes that we
>> make. The question isn't
On Tue, Jan 13, 2015 at 04:02:12PM +0800, Chao Peng wrote:
> Make some internal routines common so that total/local memory bandwidth
> monitoring in the next patch can make use of them.
>
> Signed-off-by: Chao Peng
Acked-by: Wei Liu
Could you please in your later patch submission include short
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> +gmfn = mfn_to_gmfn(d, mfn);
(I haven't thought about it super hard, but I'm taking it as given that
this approach to kexec is going to be needed for ARM too, since that
seems likely)
mfn_to_gmfn is going to be a bit pricey
On Tue, Jan 13, 2015 at 04:02:13PM +0800, Chao Peng wrote:
> Add Memory Bandwidth Monitoring(MBM) for VMs. Two types of monitoring
> are supported: total and local memory bandwidth monitoring. To use it,
> CMT should be enabled in hypervisor.
>
> Signed-off-by: Chao Peng
> ---
> docs/man/xl.pod.
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> New libxl__domain_soft_reset_destroy() is an internal-only
> version of libxl_domain_destroy() which follows the same domain
> destroy path with the only difference: xc_domain_destroy() is
> being avoided so the domain is not actually bei
Always fully coalesce guest Rx packets into the minimum number of ring
slots. Reducing the number of slots per packet has significant
performance benefits (e.g., 7.2 Gbit/s to 11 Gbit/s in an off-host
receive test).
However, this does increase the number of grant ops per packet which
decreases pe
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> Add new xc_domain_soft_reset() function which performs so-called 'soft reset'
> for an HVM domain. It is being performed in the following way:
> - Save HVM context and all HVM params;
> - Devour original domain with XEN_DOMCTL_devour;
> -
2015-01-13 11:58 GMT+00:00 Ian Campbell :
> On Mon, 2014-11-03 at 10:11 +, Frediano Ziglio wrote:
>> Add this new platform to Xen.
>> This platform require specific code to initialize CPUs.
>
> What is the "bootwrapper"? Are you running this on real silicon or on an
> emulator? Can the platform
flight 33391 seabios real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/33391/
Failures :-/ but no regressions.
Tests which did not succeed, but are not blocking:
test-amd64-amd64-xl-pvh-intel 9 guest-start fail never pass
test-amd64-i386-libvirt 9 guest-start
On Mon, 2015-01-12 at 16:52 +, Ian Jackson wrote:
> Dario Faggioli writes ("[OSSTEST PATCH] make-flight: reorganize scheduling
> related test jobs"):
> > Scheduling related tests are ok to run on ARM, so do
> > not cut them off. They also do not depend on a
> > particular Dom0 architecture.
>
On Thu, 2014-12-11 at 14:45 +0100, Vitaly Kuznetsov wrote:
> Supported for HVM guests only.
Is it specifically PVHVM guests, or are unaware HVM guests also
supported? (I think the answer is that an unaware HVM guest has no way
to trigger a soft reset, so maybe it's moot...)
> diff --git a/tools/l
On Tue, 2015-01-13 at 12:56 +, Wei Liu wrote:
> On Tue, Jan 13, 2015 at 01:19:05PM +0100, Olaf Hering wrote:
> > On Tue, Jan 13, Ian Campbell wrote:
> >
> > > On Fri, 2015-01-09 at 14:32 +, Wei Liu wrote:
> > > > A pkg-config file for libxl. It also contains two variables
> > > > (xenfirmw
Hello all,
This is the third version of this patch series to add support for platform
device passthrough on ARM.
Compare to the previous version [1], the automatic mapping of MMIO/IRQ and
the generation of the device tree has been dropped.
Instead the user will have to:
- Map manually MMIO/I
Flask code already provides an helper to copy a string from guest. In a later
patch, the new DT hypercalls will need a similar function.
To avoid code duplication, copy the flask helper (flask_copying_string) to
common code:
- Rename into safe_copy_string_from_guest
- Add comment to explai
This new function will correctly initialize the IOMMU page table for the
current domain.
Also use it in iommu_assign_dt_device even though the current IOMMU
implementation on ARM shares P2M with the processor.
Signed-off-by: Julien Grall
Cc: Jan Beulich
---
Changes in v3:
- The ASS
Currently Xen only supports SPIs routing for guest, add a function
is_assignable_irq to check if we can assign a given IRQ to the guest.
Secondly, make sure the vIRQ is not the greater that the number of IRQs handle
to the vGIC and it's an SPIs.
Thirdly, when the IRQ is already assigned to the do
The structure pending_irq is initialized on the same way in 2 differents
place. Introduce vgic_init_pending_irq to avoid code duplication.
Also move the setting of the irq field in this function as we need to
initialize it once rather than every time an IRQ is injected to the guest.
Finally, use
The check to avoid mapping disabled device in DOM0 was added in the anticipation
of the device passthrough. But, a brand new property will be added later to mark
device which will passthrough. At the same time, remove the memory type
check because those nodes has been blacklisted.
Futhermore, some
Xen has to release IRQ routed to a domain in order to reuse later. Currently
only SPIs can be routed to the guest so we only need to browse SPIs for a
specific domain.
Futhermore, a guest can crash and let the IRQ in an incorrect state (i.e has
not being EOIed). Xen will have to reset the IRQ in o
1 - 100 of 266 matches
Mail list logo
