<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div>Hmm, my intention is to replicate the physical Intel Core i7-4770 which has 4 cores with 8 threads in a single socket. I just double checked if I use <span style="background-color: rgba(255, 255, 255, 0);">cpus=8,cores=4,threads=2,sockets=1 my </span>Windows 10 detects 8 processors in the Device Manager, 8 virtual processors in Task Manager, and CPU-Z also detects 4 cores with 8 threads just like I want to.</div><div id="AppleMailSignature"><br></div><div id="AppleMailSignature">If I change the config to <span style="background-color: rgba(255, 255, 255, 0);">cpus=8,cores=4,threads=1,sockets=2, Device Manager detects the same, CPU-Z detects 2 socket of 4 single threaded cores, but Task Manager is smart it knows that it only runs in 1 socket with 8 virtual processors.</span></div><div id="AppleMailSignature"><br>Best regards,<div>Okky Hendriansyah</div></div><div><br>On Oct 1, 2015, at 12:05, Blank Field <<a href="mailto:ihatethisfield@gmail.com">ihatethisfield@gmail.com</a>> wrote:<br><br></div><blockquote type="cite"><div><p dir="ltr">Just a side-note:<br>
AFAIR -smp option ignores topology specs(sockets-cores-threads) if abregular value is specified.<br>
And doing cores=8 threads=2 should give you a 16 (logical) core vcpu. I'm pretty sure that's not what you want to do. Consult qemu-system-x86_64 help to make sure.</p>
<div class="gmail_quote">On Oct 1, 2015 3:26 AM, "Okky Hendriansyah" <<a href="mailto:okky@nostratech.com">okky@nostratech.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div style="font-family:Helvetica,Arial;font-size:13px;color:rgba(0,0,0,1.0);margin:0px;line-height:auto">Hi Alex,</div><div style="font-family:Helvetica,Arial;font-size:13px;color:rgba(0,0,0,1.0);margin:0px;line-height:auto"><br></div><div style="font-family:Helvetica,Arial;font-size:13px;color:rgba(0,0,0,1.0);margin:0px;line-height:auto">I have experienced my setup is also not working as it used to after upgrading from 4.1.6 to 4.2.1. After reading on the list mentioned I googled around and found [1].</div><div style="font-family:Helvetica,Arial;font-size:13px;color:rgba(0,0,0,1.0);margin:0px;line-height:auto"><br></div><div style="color:rgb(0,0,0);margin:0px"><span style="font-family:Helvetica,Arial;font-size:13px">Changing my processor and memory settings from </span><font face="Courier">-smp cpus=8,cores=4,threads=2,sockets=1 -m size=16G</font> to <font face="Courier">-smp cpus=4,cores=4,threads=1,sockets=1 -m size=8G</font> resulting the VM successfully boots like normal. But changing to other values did not always succeeded. I did not now the formula so I kind of brute forcing my way. I think I managed to boot the VM also with <span style="font-family:Courier">-smp cpus=8,cores=4,threads=2,sockets=1 -m size=4G</span><font face="Helvetica">. I forgot the details.</font></div><div style="color:rgb(0,0,0);margin:0px"><br></div><div style="color:rgb(0,0,0);margin:0px">Currently I’m rolling back to 4.1.6, but if there’s something that I could provide to help finding out the issue, I can upgrade it and test it again. Please let me know.</div><div style="font-family:Helvetica,Arial;font-size:13px;color:rgba(0,0,0,1.0);margin:0px;line-height:auto"><br></div><div style="font-family:Helvetica,Arial;font-size:13px;color:rgba(0,0,0,1.0);margin:0px;line-height:auto">[1] <a href="https://github.com/tianocore/edk2/issues/21" target="_blank">https://github.com/tianocore/edk2/issues/21</a></div> <br> <div><font face="helvetica, arial">-- </font><br><span style="color:rgb(2,1,3)"><font face="Calibri"><b>Okky Hendriansyah</b></font></span></div><p>On September 28, 2015 at 23:53:08, <a href="mailto:globalgorrilla@fastmail.fm" target="_blank">globalgorrilla@fastmail.fm</a> (<a href="mailto:globalgorrilla@fastmail.fm" target="_blank">globalgorrilla@fastmail.fm</a>) wrote:</p> <blockquote type="cite"><span><div><div></div><div>Alex,<br><br>I reported this on 08/18. It's been echoed several times on this list <br>since.<br><br>You've said everything is working for you and you appear to have a very <br>similar setup to us others (passing through devices with VFIO to QEMU <br>and using OVMF).<br><br>Circumstantial evidence is that there is something in > 4.1 that (often) <br>breaks OVMF in QEMU.<br><br>How could we dig into this? Perhaps it's not related to vfio?<br><br>Regarding the MTRR patch, I had made the fix myself in the 4.2 RCs and I <br>believe the patch is merged in already to 4.2 +. I don't think that is <br>the culprit ...?<br><br>Thoughts?<br><br>On 20 Sep 2015, at 9:09, Kővágó Zoltán wrote:<br><br>> Hello,<br>><br>> I've been using vfio-pci to passthrough a GPU to virtual machine since <br>> some time now, and it worked great. But this weekend I've finally had <br>> enough time to update kernel, and things are completely broken with <br>> the new kernel...<br>><br>> I've been using the ACS override patch (and a quick-and-dirty fix for <br>> multiple GOPs, but created a proper-ish patch yesterday, see <br>> <a href="http://article.gmane.org/gmane.linux.kernel.efi/6332" target="_blank">http://article.gmane.org/gmane.linux.kernel.efi/6332</a> ), CSM disabled <br>> in UEFI and using OVMF virtual machines. The motherboard is an ASRock <br>> Z87M Extreme4, with two PCI video cards (an NVidia GT640 (the primary <br>> card, used for linux), for which I almost had to beg at Gigabyte <br>> support to send an UEFI compatible VBIOS, and a GTX980 (secondary <br>> card, to passthrough)). The integrated Intel GPU is diabled in UEFI <br>> settings. I'm not sure if it's supposed to work, but with 4.0.4 <br>> kernels it worked like a charm.<br>><br>> Now with 4.2.0, when I start qemu the monitor attached to the <br>> secondary card powers down, and then nothing happens, except qemu <br>> eating about 150% cpu. I've started bisecting the kernel, and found <br>> out that<br>><br>> d69afbc6b1b5d0579f13d1a6339d952c4f60a9f4 KVM: MMU: fix decoding cache <br>> type from MTRR<br>><br>> is the culprit. When mtrr is diabled, the old code returns 0xFF while <br>> the new returns MTRR_TYPE_UNCACHABLE. I have absolutely no idea what <br>> the hell is going on here, but changing that return statement back <br>> solves the problem, until<br>><br>> b18d5431acc7a2fd22767925f3a6f597aa4bd29e KVM: x86: fix <a href="http://CR0.CD" target="_blank">CR0.CD</a> <br>> virtualization<br>><br>> If I comment out the if kvm_read_cd0 part it will work.. until <br>> 4e241557fc1cb560bd9e77ca1b4a9352732a5427, which is a merge commit(!). <br>> I'm attaching a patch, it fixes the problem until <br>> f2ae45edbca7ba5324eef01719ede0151dc5cead for me. But as I said <br>> earlier I have no freakin' idea what's going on here.<br>><br>> I have recompiled OVMF from svn yesterday evening, and have a <br>> recent-ish qemu master (with some audio related patches). Tell me if <br>> you need any more information.<br>><br>> Thanks,<br>> Zoltan<br>><br>> [magic.patch]<br>> _______________________________________________<br>> vfio-users mailing list<br>> <a href="mailto:vfio-users@redhat.com" target="_blank">vfio-users@redhat.com</a><br>> <a href="https://www.redhat.com/mailman/listinfo/vfio-users" target="_blank">https://www.redhat.com/mailman/listinfo/vfio-users</a><br><br>_______________________________________________<br>vfio-users mailing list<br><a href="mailto:vfio-users@redhat.com" target="_blank">vfio-users@redhat.com</a><br><a href="https://www.redhat.com/mailman/listinfo/vfio-users" target="_blank">https://www.redhat.com/mailman/listinfo/vfio-users</a><br></div></div></span></blockquote></div><br>_______________________________________________<br>
vfio-users mailing list<br>
<a href="mailto:vfio-users@redhat.com">vfio-users@redhat.com</a><br>
<a href="https://www.redhat.com/mailman/listinfo/vfio-users" rel="noreferrer" target="_blank">https://www.redhat.com/mailman/listinfo/vfio-users</a><br>
<br></blockquote></div>
</div></blockquote></body></html>