Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.debian.kernel > #88755 > unrolled thread

Bug#1110812: PCI passthrough device visible in guest, but not functional

Started byKarel Van Hecke <bugreport@karelvanhecke.com>
First post2025-08-11 10:10 +0200
Last post2025-08-11 13:10 +0200
Articles 3 — 3 participants

Back to article view | Back to linux.debian.kernel


Contents

  Bug#1110812: PCI passthrough device visible in guest, but not functional Karel Van Hecke <bugreport@karelvanhecke.com> - 2025-08-11 10:10 +0200
    Processed: Re: Bug#1110812: PCI passthrough device visible in  guest, but not functional "Debian Bug Tracking System" <owner@bugs.debian.org> - 2025-08-11 13:10 +0200
    Bug#1110812: PCI passthrough device visible in guest, but not functional Salvatore Bonaccorso <carnil@debian.org> - 2025-08-11 13:10 +0200

#88755 — Bug#1110812: PCI passthrough device visible in guest, but not functional

FromKarel Van Hecke <bugreport@karelvanhecke.com>
Date2025-08-11 10:10 +0200
SubjectBug#1110812: PCI passthrough device visible in guest, but not functional
Message-ID<LivLr-6HP9-3@gated-at.bofh.it>

[Multipart message — attachments visible in raw view] — view raw

Package: linux-image-amd64
Version: 6.12.38-1

Host system: Debian 13
Guest system: tested various operating systems, all with the same result.

The following device can be passed through to a guest, but isn't functional: 03:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS3008 PCI-Express Fusion-MPT SAS-3 (rev 02)
DeviceName: Integrated RAID
Subsystem: Dell HBA330 Mini
Flags: bus master, fast devsel, latency 0, IRQ 53, NUMA node 0, IOMMU group 28
I/O ports at 2000 [size=256]
Memory at 93700000 (64-bit, non-prefetchable) [size=64K]
Memory at 92600000 (64-bit, non-prefetchable) [size=1M]
Expansion ROM at <ignored> [disabled]
Capabilities: [50] Power Management version 3
Capabilities: [68] Express Endpoint, IntMsgNum 0
Capabilities: [a8] MSI: Enable- Count=1/1 Maskable+ 64bit+
Capabilities: [c0] MSI-X: Enable+ Count=96 Masked-
Capabilities: [100] Advanced Error Reporting
Capabilities: [1e0] Secondary PCI Express
Capabilities: [1c0] Power Budgeting <?>
Capabilities: [190] Dynamic Power Allocation <?>
Capabilities: [150] Single Root I/O Virtualization (SR-IOV)
Capabilities: [148] Alternative Routing-ID Interpretation (ARI)
Kernel driver in use: vfio-pci
Kernel modules: mpt3sas

None of the drives show up in the guest system. The device seems to hang on:
[ +0.006783] mpt3sas_cm0: sending port enable !!

Expected behavior:
All drives show up in the guest system and I should see the following output:
[ +0.006783] mpt3sas_cm0: sending port enable !!

[ +0.003790] mpt3sas_cm0: hba_port entry: 00000000b1fe5e82, port: 255 is added to hba_port list
[ +0.001898] mpt3sas_cm0: host_add: handle(0x0001), sas_addr(0x54cd98f081d36a00), phys(8)
[ +0.002040] mpt3sas_cm0: expander_add: handle(0x0009), parent(0x0001), sas_addr(0x500056b3dd56c6ff), phys(29)
[ +0.003458] mpt3sas_cm0: port enable: SUCCESS
[ +0.003190] mpt3sas_cm0: handle(0xa) sas_address(0x500056b3dd56c6c0) port_type(0x1)
[ +0.000211] mpt3sas_cm0: handle(0xb) sas_address(0x500056b3dd56c6c1) port_type(0x1)
[ +0.000212] mpt3sas_cm0: handle(0xc) sas_address(0x500056b3dd56c6c2) port_type(0x1)
[ +0.000168] mpt3sas_cm0: handle(0xd) sas_address(0x500056b3dd56c6c3) port_type(0x1)
[ +0.000149] mpt3sas_cm0: handle(0xe) sas_address(0x500056b3dd56c6c4) port_type(0x1)
[ +0.000229] mpt3sas_cm0: handle(0xf) sas_address(0x500056b3dd56c6c5) port_type(0x1)[ +0.000953] mpt3sas_cm0: handle(0x10) sas_address(0x500056b3dd56c6fd) port_type(0x1)

I haven't yet tried to reproduce the issue with another device.
The issue was never present on Debian 12. Using a Debian 12 kernel on the Debian 13 system also resolves the issue, which is what led me to think it's a Linux kernel issue.

Thanks in advance,

Karel Van Hecke

[toc] | [next] | [standalone]


#88763 — Processed: Re: Bug#1110812: PCI passthrough device visible in guest, but not functional

From"Debian Bug Tracking System" <owner@bugs.debian.org>
Date2025-08-11 13:10 +0200
SubjectProcessed: Re: Bug#1110812: PCI passthrough device visible in guest, but not functional
Message-ID<LiyzD-6JHD-7@gated-at.bofh.it>
In reply to#88755
Processing control commands:

> reassign -1 src:linux
Bug #1110812 [linux-image-amd64] PCI passthrough device visible in guest, but not functional
Bug reassigned from package 'linux-image-amd64' to 'src:linux'.
No longer marked as found in versions linux-signed-amd64/6.12.38+1.
Ignoring request to alter fixed versions of bug #1110812 to the same values previously set
> forcemerge 1109676 -1
Bug #1109676 [src:linux] linux-image-6.12.35+deb13-amd64: Regression in host kernel breaks VM VFIO graphics
Bug #1110722 [src:linux] linux: v6.12.35..v6.12.39 fail PCI (vfio) passthrough
Bug #1110812 [src:linux] PCI passthrough device visible in guest, but not functional
Set Bug forwarded-to-address to 'https://lore.kernel.org/linux-iommu/721D44AF820A4FEB+722679cb-2226-4287-8835-9251ad69a1ac@bbaa.fun/T'.
Marked as found in versions linux/6.12.35-1 and linux/6.12.38-1.
Added tag(s) fixed-upstream, pending, confirmed, and upstream.
Bug #1110722 [src:linux] linux: v6.12.35..v6.12.39 fail PCI (vfio) passthrough
Merged 1109676 1110722 1110812

-- 
1109676: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1109676
1110722: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1110722
1110812: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1110812
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems

[toc] | [prev] | [next] | [standalone]


#88764

FromSalvatore Bonaccorso <carnil@debian.org>
Date2025-08-11 13:10 +0200
Message-ID<LiyzD-6JHD-5@gated-at.bofh.it>
In reply to#88755
Control: reassign -1 src:linux
Control: forcemerge 1109676 -1

Hi,

On Mon, Aug 11, 2025 at 07:50:41AM +0000, Karel Van Hecke wrote:
> Package: linux-image-amd64
> Version: 6.12.38-1
> 
> Host system: Debian 13
> Guest system: tested various operating systems, all with the same result.
> 
> The following device can be passed through to a guest, but isn't functional: 03:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS3008 PCI-Express Fusion-MPT SAS-3 (rev 02)
> DeviceName: Integrated RAID
> Subsystem: Dell HBA330 Mini
> Flags: bus master, fast devsel, latency 0, IRQ 53, NUMA node 0, IOMMU group 28
> I/O ports at 2000 [size=256]
> Memory at 93700000 (64-bit, non-prefetchable) [size=64K]
> Memory at 92600000 (64-bit, non-prefetchable) [size=1M]
> Expansion ROM at <ignored> [disabled]
> Capabilities: [50] Power Management version 3
> Capabilities: [68] Express Endpoint, IntMsgNum 0
> Capabilities: [a8] MSI: Enable- Count=1/1 Maskable+ 64bit+
> Capabilities: [c0] MSI-X: Enable+ Count=96 Masked-
> Capabilities: [100] Advanced Error Reporting
> Capabilities: [1e0] Secondary PCI Express
> Capabilities: [1c0] Power Budgeting <?>
> Capabilities: [190] Dynamic Power Allocation <?>
> Capabilities: [150] Single Root I/O Virtualization (SR-IOV)
> Capabilities: [148] Alternative Routing-ID Interpretation (ARI)
> Kernel driver in use: vfio-pci
> Kernel modules: mpt3sas
> 
> None of the drives show up in the guest system. The device seems to hang on:
> [ +0.006783] mpt3sas_cm0: sending port enable !!
> 
> Expected behavior:
> All drives show up in the guest system and I should see the following output:
> [ +0.006783] mpt3sas_cm0: sending port enable !!
> 
> [ +0.003790] mpt3sas_cm0: hba_port entry: 00000000b1fe5e82, port: 255 is added to hba_port list
> [ +0.001898] mpt3sas_cm0: host_add: handle(0x0001), sas_addr(0x54cd98f081d36a00), phys(8)
> [ +0.002040] mpt3sas_cm0: expander_add: handle(0x0009), parent(0x0001), sas_addr(0x500056b3dd56c6ff), phys(29)
> [ +0.003458] mpt3sas_cm0: port enable: SUCCESS
> [ +0.003190] mpt3sas_cm0: handle(0xa) sas_address(0x500056b3dd56c6c0) port_type(0x1)
> [ +0.000211] mpt3sas_cm0: handle(0xb) sas_address(0x500056b3dd56c6c1) port_type(0x1)
> [ +0.000212] mpt3sas_cm0: handle(0xc) sas_address(0x500056b3dd56c6c2) port_type(0x1)
> [ +0.000168] mpt3sas_cm0: handle(0xd) sas_address(0x500056b3dd56c6c3) port_type(0x1)
> [ +0.000149] mpt3sas_cm0: handle(0xe) sas_address(0x500056b3dd56c6c4) port_type(0x1)
> [ +0.000229] mpt3sas_cm0: handle(0xf) sas_address(0x500056b3dd56c6c5) port_type(0x1)[ +0.000953] mpt3sas_cm0: handle(0x10) sas_address(0x500056b3dd56c6fd) port_type(0x1)
> 
> I haven't yet tried to reproduce the issue with another device.
> The issue was never present on Debian 12. Using a Debian 12 kernel on the Debian 13 system also resolves the issue, which is what led me to think it's a Linux kernel issue.

This sounds like #1109676 and #1110722, which will be resolved
shortly.

Regards,
Salvatore

[toc] | [prev] | [standalone]


Back to top | Article view | linux.debian.kernel


csiph-web