> confirm this made a huge effect on a Xeon E3-1220, while it
(obviously) made no difference on a Ryzen 7 2700.
Bhyve is missing support for AMD AVIC, which is interrupt h/w assist
similar to Intel vAPIC. AVIC I think is available only in Ryzen gen1. It is
on my TODO list.
-Anish
On Wed, Ma
:
$kldload cpuctl
$ cpucontrol -m 0xc0010114 /dev/cpuctlX <- Read VM_CR MSR
MSR 0xc0010114: 0x 0x0008
X is any CPU core number.
On my AMD/Ryzen system, bit4 is clear and I can load bhyve.
-Anish
On Sun, Apr 26, 2020 at 12:07 PM Anish wrote:
>
> Hi Christian,
>
>
MSR 0xc0010114: 0x 0x0008
X is any CPU core number.
On my AMD/Ryzen system, bit4 is clear and I can load bhyve. I have attached
a screenshot from AMD SDM Vol2, which you may find useful.
[image: Screen Shot 2020-04-26 at 12.01.14 PM.png]
-Anish
On Sun, Apr 26, 2020 at 6:27 AM Christian Jean
:
$kldload cpuctl
$ cpucontrol -m 0xc0010114 /dev/cpuctlX <- Read VM_CR MSR
MSR 0xc0010114: 0x 0x0008
X is any CPU core number.
On my AMD/Ryzen system, bit4 is clear and I can load bhyve.
-Anish
On Sun, Apr 26, 2020 at 6:27 AM Christian Jeannot wrote:
> I have checked MSI s
>If SVM is actually disabled, I have no idea what impact this
change will have, so beware.
As per the AMD document, it will cause GP fault if it has been implemented
as per specification. If we are lucky and it works, that will be an
interesting finding.
-Anish
On Sun, Apr 26, 2020 at 10:53
these
intercepts always enabled even when migration is not started?
-Anish
___
freebsd-virtualization@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-virtualization
To unsubscribe, send any mail to
"freebsd-virtualization-uns
erning, as per AMD docs, bit6 is for illegal vector which
indicate CPU is getting invalid interrupts.
Can you share hw.vmm sysctl output?
-Anish
On Thu, Feb 28, 2019 at 9:59 AM Kevin Day wrote:
>
>
> > On Feb 27, 2019, at 11:34 PM, Rodney W. Grimes <
> freebsd-...@pdx.r
dule.
bhyvectl provide stats related to processor vmx/svm from vmm.ko and is the
first thing you want to run for performance regression. It will be nice to
include it as part of bhyve perf tool/dashboard that you are intended to
build.
-Anish
On Mon, Aug 27, 2018 at 8:20 AM Rodney W. G
ETTING, VM_ENTRY_CTLS_ZERO_SETTING,
_ctls);
-Anish
On Wed, Feb 21, 2018 at 11:08 AM, Fabian Freyer <
fabian.fre...@physik.tu-berlin.de> wrote:
> Hi Peter,
>
> thanks for your response!
>
> On 21 Feb 2018, at 17:59, Peter Grehan wrote:
> > >exit_reason 33
> >
Thanks and regards,
Anish
On Thu, Nov 16, 2017 at 12:51 PM, Andriy Gapon <a...@freebsd.org> wrote:
> On 14/11/2017 06:22, Anish wrote:
> > [root@ryzen /home/anish/FreeBSD/head]# vmstat -ia |grep ivh
> > irq256: ivhd0:fault0 0
> > irq25
Hi Andriy,
Newer AMD IOMMU has EFR capability which is missing here, something like:
ivhd0: PCI cap 0x190b640f@0x40 feature:19<IOTLB,EFR>
Is this an old chipset? Also ivhd has fault interrupt enabled which is very
helpful in debugging:
[root@ryzen /home/anish/FreeBSD/head]# vmstat -ia |gr
Hi Andriy,
>What's suspicious is that there are no interrupts either for the
ppt device on the host or the ahci device in the guest.
I think most likely AMD IOMMU/amdvi is not enabled, by default it is
disabled and can be confirmed by sysctl.
[root@ryzen /usr/home/anish]# sysctl hw.vmm.am
ost:~ # bhyvectl --get-stats --vm= --cpu= | grep
external
vm exits due to external interrupt 27273
root@svmhost:~ #
Regards,
Anish
On Sun, Jun 11, 2017 at 2:51 AM, Harry Schmalzbauer <free...@omnilan.de>
wrote:
> Bezüglich Harry Schmalzbauer's Nachricht vom 09.06.2017 10:2
ing for these performance benchmarking?
Can you run a simple experiment, assign pptdev interrupts to core that's
not running guest/vcpu? This will reduce #VMEXIT on vcpu which we know is
expensive.
Regards,
Anish
On Wed, Jun 7, 2017 at 11:01 AM, Harry Schmalzbauer <free...@omnilan.de&
anish added inline comments.
INLINE COMMENTS
> ept.c:58-59
> +#define INVVPID_REQUIRED_TYPES_MASK \
> + ((1UL << (INVVPID_TYPE_SINGLE_CONTEXT + 40)) | \
> + (1UL << (INVVPID_TYPE_ALL_CONTEXTS + 40)))
> +#define INVVPID_R
I just tested bvmdebug with FreeBSD HEAD and it worked fine.
-Anish
*./vmrun.sh -g 6466 vm1*
**
*root@bsdguest:~ # uname -a*
*FreeBSD bsdguest 12.0-CURRENT FreeBSD 12.0-CURRENT #1 r314641M: Sat Apr 15
19:53:23 PDT 2017 root@svmhost:/usr/obj/root/head/sys/BHYVE amd64*
*root@bsdguest
are to be assigned to guest at virtual PCI
slot 10, assignment can be:
4/0/1 -> 10/0
4/0/2 -> 10/1
Regards,
Anish
On Sun, Apr 2, 2017 at 3:09 AM, Daniel Tihanyi <tetra...@fastmail.fm> wrote:
> Hello,
>
> I have a FreeBSD host running VMs on bhyve. I already have a few PCI-X
>
Hi Shunsuke Mie
Awesome! Feel free to also ping me for any question you have.
-Anish
On 3/28/17 8:27 AM, Shunsuke Mie wrote:
Hello bhyve hackers.
My name is Shunsuke Mie.
I go to graduate school in this April. I'm interesting in kernel, virtual
machine, hardware design and open some of code
Please share dmesg output. I have not run into this, will give you some patch
to try.
-Anish
Sent from my iPhone
> On Oct 24, 2016, at 9:00 PM, Aryeh Friedman <aryeh.fried...@gmail.com> wrote:
>
> After I did a clean install of 11 on a machine, HP Pavilion p7-1234, that
>
>ppt0@pci0:0:20:0: class=0x0c0330 card=0x21f317aa chip=0x1e318086
Passthrough stub driver, part of vmm, controls the USB controller.
Can you share output of /usr/sbin/devinfo from FreeBSD host, highlighting
the usb mass/flash device in tree?
Regards,
Anish
On Mon, Dec 21, 2015 at 7:44
>Is there any way to "change" it when the guest is already running?
You can specify at boot time, for example in vmrun.sh
-s $(($nextslot+1)):0,virtio-blk,${isofile}
\
'isofile' is specified using '-I' argument
$./vmrun.sh -I vm1
Once system is booted, you
Sorry, you need to have NRIP support in AMD processor for bhyve. I thought
of supporting it but all new processors do have NRIP so got pushed out. If
someone wants to do it, will be happy to help.
Regards,
Anish
On Wed, Sep 30, 2015 at 10:58 AM, C. L. Martinez <carlopm...@gmail.com>
-X configuration: 2/0/0
and quits.
ppt0@pci0:2:0:0:class=0x02 card=0x06471025 chip=0x16b514e4
rev=0x10 hdr=0x00
I will try to reproduce the problem if I get hold of this device.
-Anish
On Nov 29, 2014, at 6:44 AM, Nikos Vassiliadis nv...@gmx.com wrote:
On 11/29/14 06:02
Can you try cleaning old VM before restarting VM with same name?
bhyveload …testvm
bhyve……….testvm
bhyvectl —destroy —vm=testvm
bhyveload …testvm
bhyve……….testvm
-Anish
On Sep 26, 2014, at 10:36 AM, Denis Menshikov da...@inbox.ru wrote:
Hello,
for greater clarity, will present at once
that are in bhyve_svm and
then apply the patch. bhyve HEAD exposed vlapic related interfaces along
with some other changes, this patch will enable vlapic interfaces for SVM.
Thanks and regards,
Anish
On Tue, May 20, 2014 at 2:03 PM, Willem Jan Withagen w...@digiware.nlwrote:
On 15-5-2014 17:56, Anish wrote:
Hi
lockup - CPU#0 stuck for 23s! [swapper/0:1]
[ 30.596366] Modules linked in:
Additionally, It produces a lot of MSR requests:
Yes, on AMD Linux is touching more MSRs( AMD specific -address 0xC00)
compared to Intel.
Thanks and regards,
Anish
On Fri, May 16, 2014 at 2:17 PM, Nils Beyer n
and regards,
Anish
On Thu, May 15, 2014 at 2:10 AM, Andriy Gapon a...@freebsd.org wrote:
It seems that the bhyve_svm branch is a bit behind the latest interface
changes
in head. Is anyone working on sync-ing up the branch with the head?
Some examples:
- change of init method in vmm_ops
- addition
Phenom box.
bhyveload…..vm1
bhyve…..vm1
…
bhyvectl --destroy --vm=vm1
bhyveload …vm1
bhyve…..vm1
….
Thanks for your help in testing bhyve_svm and bhyve.
-Anish
On Feb 17, 2014, at 8:27 PM, Aryeh Friedman aryeh.fried...@gmail.com wrote:
Sorry for taking so long (forgot to say orginally
Is it just de vmm driver, or is there a lot more?
Yes, it only enable vmm AMD support, nothing else.
-Anish
On Thu, Oct 10, 2013 at 12:09 PM, Willem Jan Withagen w...@digiware.nlwrote:
Op 10 okt. 2013 om 19:00 heeft Peter Grehan gre...@freebsd.org het
volgende geschreven:
Hi,
I'm
(Anish - is nRIP required ?)
Yes, nRIP is required and it will be hard to find AMD processor with SVM
support that doesn't have it. We used Phenom II for bhyve port on AMD which
is 2+ years old and has RVI and nRIP support. I use
http://instlatx64.atw.hu/ to check SVM CPUID capability and looks
30 matches
Mail list logo