All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Andryuk <jason.andryuk@amd.com>
To: Mykyta Poturai <Mykyta_Poturai@epam.com>,
	"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>
Cc: "Roger Pau Monné" <roger.pau@citrix.com>,
	"Stewart Hildebrand" <stewart.hildebrand@amd.com>,
	"Andrew Cooper" <andrew.cooper3@citrix.com>,
	"Anthony PERARD" <anthony.perard@vates.tech>,
	"Michal Orzel" <michal.orzel@amd.com>,
	"Jan Beulich" <jbeulich@suse.com>,
	"Julien Grall" <julien@xen.org>,
	"Stefano Stabellini" <sstabellini@kernel.org>,
	"Daniel P. Smith" <dpsmith@apertussolutions.com>
Subject: Re: [PATCH v3 0/7] Implement SR-IOV support for PVH
Date: Fri, 25 Sep 2026 12:55:54 -0400	[thread overview]
Message-ID: <dbd7fdab-3833-4b65-b8b6-d92b5beb9e17@amd.com> (raw)
In-Reply-To: <cover.1775742115.git.mykyta_poturai@epam.com>

On 2026-04-09 10:01, Mykyta Poturai wrote:
> This series enables support for PCI SR-IOV capability for PVH domains.
> It allows Dom0 to enable and use SR-IOV virtual functions and for these
> functions to be passed to guests.
> 
> To achieve this, add handlers for SRIOV_CONTROL registers, simplified handlers
> for VFs headers. Xen relies on dom0 to enable SR-IOV and call
> PHYSDEVOP_pci_device_* to inform about addition/removal of VFs.
> 
> Core functionality is based on previous work[1].
> 
> Tested on R-Car Spider board with Samsung NVMe SSD Controller 980 and Intel
> X550T ethernet card.

Hi Mykyta,

I'm testing these patches from the patchew branch 
https://gitlab.com/xen-project/patchew/xen/-/tree/patchew/cover.1775742115.git.mykyta._5Fpoturai@epam.com?ref_type=tags
with an Intel 82576 card which uses the igb driver.  I think your X550T 
testing uses the ixgbe drvier.

SR-IOV capabilities are exposed, and I can create VFs (PF c4:00.0, VF 
c4:10.0).  However, the VF network interfaces don't work.  They can't 
get an IP address.  The PF reports "Detected Tx Unit Hang" as seen in 
the log snippet below.  I trimmed out some "Spoof event" and "invalid 
MAC filter messages" from Xen.  I also included some baremetal logs 
which don't have "Spoof event" messages.  On baremetal, the interface 
gets an IP and is usable.

This is an x86 PVH dom0, but you tested with ARM?  Do you know if there 
are any other dependencies for this patch series?  I'll keep looking at 
it, but I figured I'd ask first.

Thanks,
Jason

PVH dom0:
12:20:34: igb 0000:c4:00.0: 1 VFs allocated
12:20:34: pci 0000:c4:10.0: [8086:10ca] type 00 class 0x020000 PCIe Endpoint
12:20:34: igbvf 0000:c4:10.0: enabling device (0000 -> 0002)
12:20:34: igbvf 0000:c4:10.0: MAC address not assigned by administrator.
12:20:34: igbvf 0000:c4:10.0: Assigning random MAC address.
12:20:34: igbvf 0000:c4:10.0: Intel(R) 82576 Virtual Function
12:20:34: igbvf 0000:c4:10.0: Address: a6:26:55:1d:6b:92
12:20:34: igbvf 0000:c4:10.0 enp196s0f0v0: renamed from eth0
12:20:34: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:20:34: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:20:34: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:20:37: igb 0000:c4:00.0 enp196s0f0: igb: enp196s0f0 NIC Link is Up 
1000 Mbps Full Duplex, Flow Control: RX/TX
12:20:39: igbvf 0000:c4:10.0: Link is Up 1000 Mbps Full Duplex
12:20:39: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00

12:22:17: igb 0000:c4:00.0: Spoof event(s) detected on VF 0
12:22:51: igb 0000:c4:00.0 enp196s0f0: NETDEV WATCHDOG: CPU: 1: transmit 
queue 0 timed out 5004 ms
12:22:51: igb 0000:c4:00.0: Detected Tx Unit Hang
             Tx Queue             <0>
             TDH                  <78>
             TDT                  <83>
             next_to_use          <83>
             next_to_clean        <78>
           buffer_info[next_to_clean]
             time_stamp           <ffffa303>
             next_to_watch        <ffff88812e03c790>
             jiffies              <ffffbca5>
             desc.status          <160200>
12:22:51: igb 0000:c4:00.0 enp196s0f0: Reset adapter
12:22:53: igbvf 0000:c4:10.0: Link is Down

Baremetal:
12:51:01: igb 0000:c4:00.0: 1 VFs allocated
12:51:02: pci 0000:c4:10.0: [8086:10ca] type 00 class 0x020000 PCIe Endpoint
12:51:02: pci 0000:c4:10.0: Adding to iommu group 35
12:51:02: igbvf 0000:c4:10.0: enabling device (0000 -> 0002)
12:51:02: igbvf 0000:c4:10.0: MAC address not assigned by administrator.
12:51:02: igbvf 0000:c4:10.0: Assigning random MAC address.
12:51:02: igbvf 0000:c4:10.0: Intel(R) 82576 Virtual Function
12:51:02: igbvf 0000:c4:10.0: Address: 5a:d0:88:3d:0d:44
12:51:02: igbvf 0000:c4:10.0 enp196s0f0v0: renamed from eth0
12:51:02: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:02: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:02: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0 enp196s0f0: igb: enp196s0f0 NIC Link is Up 
1000 Mbps Full Duplex, Flow Control: RX/TX
12:51:05: igbvf 0000:c4:10.0: Link is Up 1000 Mbps Full Duplex
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00
12:51:05: igb 0000:c4:00.0: VF 0 attempted to set invalid MAC filter 
00:00:00:00:00:00



      parent reply	other threads:[~2026-09-25 16:58 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-09 14:01 [PATCH v3 0/7] Implement SR-IOV support for PVH Mykyta Poturai
2026-04-09 14:01 ` [PATCH v3 2/7] vpci: rename and export vpci_guest_mem_bar_{read,write} Mykyta Poturai
2026-04-22 10:27   ` Roger Pau Monné
2026-04-09 14:01 ` [PATCH v3 1/7] vpci: rename and export vpci_modify_bars Mykyta Poturai
2026-04-09 14:01 ` [PATCH v3 3/7] vpci: Use pervcpu ranges for BAR mapping Mykyta Poturai
2026-04-22 11:00   ` Roger Pau Monné
2026-04-22 12:04     ` Jan Beulich
2026-04-22 14:20       ` Roger Pau Monné
2026-04-09 14:01 ` [PATCH v3 4/7] vpci: allow queueing of mapping operations Mykyta Poturai
2026-04-22 11:38   ` Roger Pau Monné
2026-07-09 15:10     ` Stewart Hildebrand
2026-04-09 14:01 ` [PATCH v3 5/7] vpci: add SR-IOV support for PVH Dom0 Mykyta Poturai
2026-04-09 15:27   ` Daniel P. Smith
2026-04-21 14:43   ` Jan Beulich
2026-04-23 10:12     ` Mykyta Poturai
2026-05-04  5:37       ` Jan Beulich
2026-05-06  9:39         ` Mykyta Poturai
2026-05-06 11:54           ` Jan Beulich
2026-05-07 20:40             ` Volodymyr Babchuk
2026-05-08  5:52               ` Jan Beulich
2026-05-11 14:10                 ` Volodymyr Babchuk
2026-05-12  6:20                   ` Jan Beulich
2026-05-12  7:32                     ` Mykyta Poturai
2026-05-12  8:58                       ` Roger Pau Monné
2026-05-12 10:28                         ` Volodymyr Babchuk
2026-05-12 13:22                           ` Roger Pau Monné
2026-05-12 10:44                         ` Jan Beulich
2026-05-12 11:11                           ` Roger Pau Monné
2026-05-08  5:52               ` Jan Beulich
2026-04-22 14:19   ` Roger Pau Monné
2026-04-28 20:05   ` Stewart Hildebrand
2026-04-09 14:01 ` [PATCH v3 6/7] vpci: add SR-IOV support for DomUs Mykyta Poturai
2026-04-21 14:55   ` Jan Beulich
2026-04-24  6:34     ` Mykyta Poturai
2026-04-09 14:01 ` [PATCH v3 7/7] docs: Update SR-IOV support status Mykyta Poturai
2026-04-21 14:56   ` Jan Beulich
2026-09-25 16:55 ` Jason Andryuk [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=dbd7fdab-3833-4b65-b8b6-d92b5beb9e17@amd.com \
    --to=jason.andryuk@amd.com \
    --cc=Mykyta_Poturai@epam.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=anthony.perard@vates.tech \
    --cc=dpsmith@apertussolutions.com \
    --cc=jbeulich@suse.com \
    --cc=julien@xen.org \
    --cc=michal.orzel@amd.com \
    --cc=roger.pau@citrix.com \
    --cc=sstabellini@kernel.org \
    --cc=stewart.hildebrand@amd.com \
    --cc=xen-devel@lists.xenproject.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.