Bug 1654550

Summary: [svvp] [WS2019] Error occurred when running job "Verify IOMMU presence and configuration"
Product: Red Hat Enterprise Linux 8 Reporter: Yu Wang <wyu>
Component: qemu-kvmAssignee: Yvugenfi <yvugenfi>
Status: CLOSED CURRENTRELEASE QA Contact: Virtualization Bugs <virt-bugs>
Severity: medium Docs Contact:
Priority: medium    
Version: 8.0CC: ailan, blc, chayang, ddepaula, hhuang, jasowang, jinzhao, juzhang, knoel, lijin, mtessun, peterx, rbalakri, ribarry, virt-maint, vkuznets, wchadwic, wyu, ybendito, yvugenfi
Target Milestone: rc   
Target Release: 8.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
: 1664922 (view as bug list) Environment:
Last Closed: 2019-06-17 05:28:48 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1659244, 1664922    
Attachments:
Description Flags
iommu_log
none
hlk_logs
none
txt_log
none
html_logs
none
new_params
none
new_pass
none
iommu_amd_log none

Description Yu Wang 2018-11-29 04:36:13 UTC
Created attachment 1509733 [details]
iommu_log

Description of problem:

Version-Release number of selected component (if applicable):
kernel-4.18.0-40.el8.x86_64
qemu-kvm-3.0.0-2.module+el8+2208+e41b12e0.x86_64
seabios-bin-1.11.1-2.module+el8+2179+85112f94.noarch

How reproducible:
100%

Steps to Reproduce:
1. boot guest with/without iommu config

/usr/libexec/qemu-kvm \
-name SVVP_RHEL8_INTEL_SUT \
-M q35,kernel-irqchip=split \
-device intel-iommu,intremap=on,caching-mode=on,eim=on,device-iotlb=on \
-cpu Westmere,+kvm_pv_unhalt,hv_spinlocks=0x1fff,hv_relaxed,hv_vapic,hv_time \
-enable-kvm -nodefconfig -nodefaults \
-m 6G \
-smp 4 \
-uuid fdc472e5-06e8-4647-9283-569e657c41a1 \
-smbios type=1,manufacturer="Red Hat",product="RHEL OpenStack Platform Version 8 Fast Train & RHEV Version 8",version=8Server,serial="14F00000-BF99-11E1-BAC6-80 C84 BB17 B77",uuid=40366627-6ea0-455f-b773-b5edd5023c37 \
-nodefconfig \
-rtc base=localtime,clock=host,driftfix=slew \
-device pcie-root-port,bus=pcie.0,id=root1.0,slot=1 \
-object iothread,id=thread0 \
-drive file=160BLK201964FQ2,if=none,format=raw,cache=none,werror=stop,rerror=stop,id=drive-virtio-disk0,aio=native \
-device virtio-blk-pci,scsi=off,iothread=thread0,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1,bus=root1.0,bus=root1.0,iommu_platform=on,ats=on,disable-legacy=on,disable-modern=off \
-drive file=virtio-win-1.9.6-5.el8.iso,media=cdrom,id=cdrom,if=none -device ide-drive,drive=cdrom,bootindex=0 \
-device piix3-usb-uhci,id=usb \
-device usb-tablet,id=tablet0 \
-k en-us \
-monitor stdio \
-qmp tcp:0:4455,server,nowait \
-device usb-ehci,id=ehci0 \
-drive file=usb-disk-0.raw,if=none,id=drive-usb-2-0,media=disk,format=raw,cache=none,werror=stop,rerror=stop,aio=threads \
-device usb-storage,bus=ehci0.0,drive=drive-usb-2-0,id=usb-2-0,removable=on \
-vnc 0.0.0.0:0 \
-vga std \
-netdev tap,script=/etc/qemu-ifup1,id=hostnet1,vhost=on -device e1000,netdev=hostnet1,id=net1,mac=00:e2:52:20:16:d2 \



2. submit job "Verify IOMMU presence and configuration"
3.

Actual results:
Now we are running svvp for windows server 2019:
1 If I use "-device intel-iommu,intremap=on,caching-mode=on,eim=on,device-iotlb=on -device virtio-net-pci,netdev=hostnet2,...,iommu_platform=on,ats=on,disable-legacy=on,disable-modern=off \"
The HLK log shows"Segment 0 has no DMA remapping capable IOMMU units"

2 If not using iommu config above, the log shows "Could not find DMAR. Have you enabled IOMMU in the BIOS?" and "No IOMMU support found"



Expected results:
This job has manual errata, but now it support iommu now,so we hope it can pass without error

Additional info:
Detail log refer to attachment

Comment 2 Yu Wang 2018-12-12 06:49:21 UTC
We have errata(Manual Errata 7024) for job "Verify IOMMU presence and configuration"
when running ws2016, I'm not sure if it can be used for ws2019. If not, could you also
apply a manual errata for it?

HCK log please refer to the attachment

Comment 3 Yu Wang 2018-12-12 06:53:28 UTC
Created attachment 1513611 [details]
hlk_logs

Comment 8 lijin 2018-12-14 02:55:13 UTC
Here is msft document about this job:

This test does the following validation:

    Ensures that at least one IOMMU device is present on the system
    Performs basic validation of the IOMMU ACPI tables (DMAR/IVRS) looking for simple inconsistencies and/or divergence from the IOMMU specs
    Performs basic validation to ensure that there are no inconsistencies in how IOMMUs are reported in the MADT table vs. in the DMAR/IVRS tables

https://docs.microsoft.com/en-us/windows-hardware/test/hlk/testref/3b880cc8-516d-4ab4-9d7f-88b558bf4420

Comment 9 Yu Wang 2018-12-14 03:09:14 UTC
Created attachment 1514283 [details]
txt_log

Comment 10 Yu Wang 2018-12-14 03:23:03 UTC
Created attachment 1514288 [details]
html_logs

Comment 11 Yu Wang 2018-12-14 07:31:39 UTC
Created attachment 1514301 [details]
new_params

test with new params as peter told:
-device intel-iommu,intremap=on,x-aw-bits=48,caching-mode=off,eim=on \

The former subtest is passed. But it failed as
" IOMMU[0] - DWD (DMA write draining) not supported"
and "IOMMU[0] - DWD (DMA read draining) not supported"

Test log refer to attachment "new_params"

Thanks
Yu Wang

Comment 15 Yu Wang 2018-12-17 06:17:00 UTC
Created attachment 1514974 [details]
new_pass

Comment 17 Yu Wang 2018-12-19 03:17:38 UTC
Created attachment 1515460 [details]
iommu_amd_log

AMD iommu failed logs for ws2019(hlk1809)

Comment 20 Yvugenfi@redhat.com 2019-01-01 09:28:27 UTC
We can use manual errata to pass the test.

Below is the email from MS:

>>>
Hello Yan,
 
Please reference manual errata # 10745 in your submission readme file for the errors in "Verify IOMMU presence and configuration".
 
Thanks!
 
Best Regards,
Tianwei Yang

Comment 22 lijin 2019-01-10 02:15:42 UTC
Hi Amnon,

Could you ack this bug?

Thanks

Comment 39 Danilo de Paula 2019-01-21 18:24:51 UTC
Peter Xu Nacked his on patch, hence moving this back to assigned as there's no patch posted anymore.