SM X11DPG-QT - not enough MMIO resources for SR-IOV

CyklonDX · May 13, 2026

Thus I was trying to get sr-iov working again on b60, on trusty x11 mobo.

Only to be met with disappointment

May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: PF: VF1..VF2 provisioned with 2090397696 (1.95 GiB) GGTT
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: PF: VF1..VF2 provisioned with 12754878464 (11.9 GiB) LMEM
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: PF: VF1..VF2 provisioned with 32639 GuC context IDs
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: PF: VF1..VF2 provisioned with 128 GuC doorbell IDs
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: PF: VF1..VF2 provisioned with 32639 GuC context IDs
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: PF: VF1..VF2 provisioned with 128 GuC doorbell IDs
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: not enough MMIO resources for SR-IOV
May 13 03:52:33 KVMHost kernel: xe 0000:b1:00.0: [drm] PF: Failed to enable 2 VFs (-ENOMEM)

After some research

lspci -v -s b0:01.0
b0:01.0 PCI bridge: Intel Corporation Device e2f0 (prog-if 00 [Normal decode])
Subsystem: Intel Corporation Device 0000
Flags: bus master, fast devsel, latency 0, NUMA node 1, IOMMU group 18
Bus: primary=b0, secondary=b1, subordinate=b1, sec-latency=0
I/O behind bridge: [disabled] [16-bit]
Memory behind bridge: ee400000-ee5fffff [size=2M] [32-bit]
Prefetchable memory behind bridge: 5b000000000-5b800ffffff [size=32784M] [32-bit]
Capabilities: [40] Express Downstream Port (Slot-), IntMsgNum 0
Capabilities: [a0] Power Management version 3
Capabilities: [94] Subsystem: Intel Corporation Device 0000
Capabilities: [100] Advanced Error Reporting
Capabilities: [220] Access Control Services
Kernel driver in use: pcieport

I came to think that bridge is constrained to [32-bit] instead of [64-bit] addressing rules. This hard-caps the available space at 32 GiB; but B60 wants to assign some more... baseline 32 GiB window (16000000000-167ffffffff) for its primary physical VRAM pool, plus an additional 384 GiB window (3ffa00000000-3fffffffffff) to handle core architecture execution logic.

May 13 03:50:58 KVMHost kernel: pci 0000:af:00.0: PCI bridge to [bus b0-b2]
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: [8086:e211] type 00 class 0x030000 PCIe Endpoint
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: BAR 0 [mem 0x5be17000000-0x5be17ffffff 64bit pref]
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: BAR 2 [mem 0x5be00000000-0x5be0fffffff 64bit pref]
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: ROM [mem 0xee400000-0xee5fffff pref]
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: PME# supported from D0 D3hot
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: VF BAR 0 [mem 0x5be10000000-0x5be10ffffff 64bit pref]
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: VF BAR 0 [mem 0x5be10000000-0x5be16ffffff 64bit pref]: contains BAR 0 for 7 VFs
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: VF BAR 2 [mem 0x5b000000000-0x5b1ffffffff 64bit pref]
May 13 03:50:58 KVMHost kernel: pci 0000:b1:00.0: VF BAR 2 [mem 0x5b000000000-0x5bdffffffff 64bit pref]: contains BAR 2 for 7 VFs
May 13 03:50:58 KVMHost kernel: pci 0000:b0:01.0: PCI bridge to [bus b1]
May 13 03:50:58 KVMHost kernel: pci 0000:b2:00.0: [8086:e2f7] type 00 class 0x040300 PCIe Endpoint
May 13 03:50:58 KVMHost kernel: pci 0000:b2:00.0: BAR 0 [mem 0xee600000-0xee603fff 64bit]
May 13 03:50:58 KVMHost kernel: pci 0000:b2:00.0: PME# supported from D3hot D3cold
May 13 03:50:58 KVMHost kernel: pci 0000:b0:02.0: PCI bridge to [bus b2]
May 13 03:50:58 KVMHost kernel: pci_bus 0000:ae: on NUMA node 1
I tried quite few things (in bios, and in grub) going from 56T mmio to 1-2-4T etc, above 4G etc ... but i didn't get anywhere.

(from syslog as its starting - in ref to pcie port / bridge, and the card itself.)

Code:

May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Attempting to resize bar from 256MiB -> 32768MiB
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 0 [mem 0x5be17000000-0x5be17ffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 2 [mem 0x5be00000000-0x5be0fffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 0 [mem 0x5be10000000-0x5be16ffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 2 [mem 0x5b000000000-0x5bdffffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: pcieport 0000:b0:01.0: bridge window [mem 0x5b000000000-0x5be17ffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem 0x5b000000000-0x5be17ffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: bridge window [mem 0x5b000000000-0x5be187fffff 64bit pref]: was not released (still contains assigned
resources)
May 13 03:50:58 KVMHost kernel: pcieport 0000:b0:01.0: bridge window [mem size 0x801000000 64bit pref] to [bus b1] add_size e07000000 add_align 800000000
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem size 0x801000000 64bit pref] to [bus b0-b2] add_size e07000000 add_align 800000000
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem size 0x1608000000 64bit pref]: can't assign; no space
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem size 0x1608000000 64bit pref]: failed to assign
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem 0x5b000000000-0x5b800ffffff 64bit pref]: assigned
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem 0x5b000000000-0x5b800ffffff 64bit pref]: failed to expand by 0xe07000000
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem 0x5b000000000-0x5b800ffffff 64bit pref]: failed to add optional e07000000
May 13 03:50:58 KVMHost kernel: pcieport 0000:b0:01.0: bridge window [mem 0x5b000000000-0x5b800ffffff 64bit pref]: assigned
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 2 [mem 0x5b000000000-0x5b7ffffffff 64bit pref]: assigned
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 2 [mem size 0xe00000000 64bit pref]: can't assign; no space
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 2 [mem size 0xe00000000 64bit pref]: failed to assign
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 0 [mem 0x5b800000000-0x5b800ffffff 64bit pref]: assigned
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 0 [mem size 0x07000000 64bit pref]: can't assign; no space
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 0 [mem size 0x07000000 64bit pref]: failed to assign
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 2 [mem 0x5b000000000-0x5b7ffffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 0 [mem 0x5b800000000-0x5b800ffffff 64bit pref]: releasing
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 2 [mem 0x5b000000000-0x5b7ffffffff 64bit pref]: assigned
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: BAR 0 [mem 0x5b800000000-0x5b800ffffff 64bit pref]: assigned
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 2 [mem size 0xe00000000 64bit pref]: can't assign; no space
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 2 [mem size 0xe00000000 64bit pref]: failed to assign
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 0 [mem size 0x07000000 64bit pref]: can't assign; no space
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: VF BAR 0 [mem size 0x07000000 64bit pref]: failed to assign
May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: PCI bridge to [bus af-b2]
May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: bridge window [mem 0xee400000-0xee6fffff]
May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: bridge window [mem 0x5b000000000-0x5be187fffff 64bit pref]
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: PCI bridge to [bus b0-b2]
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem 0xee400000-0xee6fffff]
May 13 03:50:58 KVMHost kernel: pcieport 0000:af:00.0: bridge window [mem 0x5b000000000-0x5b800ffffff 64bit pref]
May 13 03:50:58 KVMHost kernel: pcieport 0000:b0:01.0: PCI bridge to [bus b1]
May 13 03:50:58 KVMHost kernel: pcieport 0000:b0:01.0: bridge window [mem 0xee400000-0xee5fffff]
May 13 03:50:58 KVMHost kernel: pcieport 0000:b0:01.0: bridge window [mem 0x5b000000000-0x5b800ffffff 64bit pref]
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] BAR2 resized to 32768M
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Running in SR-IOV PF mode
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Found battlemage (device ID e211) discrete display version 14.01 stepping B0
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] VISIBLE VRAM: 0x000005b000000000, 0x0000000800000000
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] VRAM[0]: Actual physical size 0x0000000600000000, usable size exclude stolen 0x00000005fa000000, CPU
accessible size 0x00000005fa000000
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] VRAM[0]: DPA range: [0x0000000000000000-600000000], io range: [0x000005b000000000-5b5fa000000]
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] VRAM[0]: Actual physical size 0x0000000600000000, usable size exclude stolen 0x00000005fa000000, CPU
accessible size 0x00000005fa000000
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] VRAM[0]: DPA range: [0x0000000000000000-600000000], io range: [0x000005b000000000-5b5fa000000]
May 13 03:50:58 KVMHost kernel: intel_rapl_common: Found RAPL domain package
May 13 03:50:58 KVMHost kernel: intel_rapl_common: Found RAPL domain dram
May 13 03:50:58 KVMHost kernel: intel_rapl_common: package-0:dram:long_term locked by BIOS
May 13 03:50:58 KVMHost kernel: intel_rapl_common: Found RAPL domain package
May 13 03:50:58 KVMHost kernel: intel_rapl_common: Found RAPL domain dram
May 13 03:50:58 KVMHost kernel: intel_rapl_common: package-1:dram:long_term locked by BIOS
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=none
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: Using GuC firmware from xe/bmg_guc_70.bin version 70.65.0
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Finished loading DMC firmware i915/bmg_dmc.bin (v2.6)

May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: ccs2 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT0: ccs3 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: Using GuC firmware from xe/bmg_guc_70.bin version 70.65.0
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: Using HuC firmware from xe/bmg_huc.bin version 8.2.10
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vcs1 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vcs3 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vcs4 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vcs5 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vcs6 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vcs7 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vecs2 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Tile0: GT1: vecs3 fused off
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Can't init xe mei late bind missing mei component
May 13 03:50:58 KVMHost kernel: [drm] Initialized xe 1.1.0 for 0000:b1:00.0 on minor 1
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Cannot find any crtc or sizes
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Using mailbox commands for power limits
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] PL2 is supported on channel 0
May 13 03:50:58 KVMHost kernel: xe 0000:b1:00.0: [drm] Cannot find any crtc or sizes

Any tips? my guess i have to mod my bios?

RolloZ170 · May 13, 2026

CyklonDX said:
above 4G etc ... but i didn't get anywhere.

at least above 4G decoding enabled is a must have.
enable ACS Control
you may have to reinstall OS after enabling.
Kernel opts: `intel_iommu=on iommu=pti pci=realloc,assign-busses`
you need resources for every single VF. you run out of memory.

edit:
the X550 nic may have old formware (32bit)

RolloZ170 · May 13, 2026

maybe this helps:

Resolution

The BIOS is not providing enough MMIO space for VFs. Contact your hardware vendor for firmware or BIOS update.

As a workaround solution, one can pass "pci=realloc" to kernel 2.6.32-228.el6 during booting.

Unable to enable SR-IOV and receiving the message "not enough MMIO resources for SR-IOV" in Red Hat Enterprise Linux - Red Hat Customer Portal

Found following error messages in /var/log/dmesg when loading the igb driver with max_vfs=4 kernel: Intel(R) Gigabit Ethernet Network Driver - version 2.1.0-k2 kernel: Copyright (c) 2007-2009 Intel Corporation. kernel: igb 0000:01:00.0: PCI INT B -> GSI 17 (level, low) -> IRQ 17 kernel: igb...

access.redhat.com

CyklonDX · May 13, 2026

RolloZ170 said:
at least above 4G decoding enabled is a must have.
enable ACS Control
you may have to reinstall OS after enabling.
Kernel opts: `intel_iommu=on iommu=pti pci=realloc,assign-busses`

done those, and more - no luck

RolloZ170 said:
you need resources for every single VF. you run out of memory.

got 768G of ram on this box, i get this error even when i try to create 1 vf.

RolloZ170 said:
the X550 nic may have old formware (32bit)

this is on intel arc pro b60 -?> would old fimware on intel nic do anything? I mean i can enable vf's just fine on my x550 (I don't use it though)

RolloZ170 said:
maybe this helps:

tried those, no bueno.
I also doubt sm would release a bios patch for me -> to increase pcie bridge bar size beyond 32bit address space *(but its odd, there were some people with desktop platforms that succeeded just fine.) (sm these last few years became bunch of a's...) -- running latest bios, bmc from 2025

i was thinking about moving it to different slot - maybe onto slot 2
https://www.supermicro.com/manuals/motherboard/C620/MNL-1998.pdf (page 12)
*It currently has 4080 there (intel b60 is installed in slot6)

RolloZ170 · May 13, 2026

CyklonDX said:
this is on intel arc pro b60 -?> would old fimware on intel nic do anything?

bricks the pci recoure management.
e.g. one bad pcie and all following devices fail.

CyklonDX said:
I also doubt sm would release a bios patch for me -> to increase pcie bridge bar size beyond 32bit address space

you are not alone with this, it happens on other platforms too.
32bit will be only 4GB.

RolloZ170 · May 13, 2026

CyklonDX said:
i was thinking about moving it to different slot - maybe onto slot 2
https://www.supermicro.com/manuals/motherboard/C620/MNL-1998.pdf (page 12)
*It currently has 4080 there (intel b60 is installed in slot6)

enabled ACS ?

CyklonDX · May 13, 2026

RolloZ170 said:
enabled ACS ?

yes, i had enabled all vtd features.

RolloZ170 said:
bricks the pci recoure management.
e.g. one bad pcie and all following devices fail.

i'll take a look, i did disable onboard x550, and i'm using x540-ta2 (in slot8 - if its blowing up on that cpu, it could pose issues - so moving b60 from slot 6 to slot 2 might do something)
firmware-version: 0x8000059e, 17.0.12

CyklonDX · May 13, 2026

but not sure if it would solve anything as memory behind bridge always is 32bit on all of those big x16 slots.

5d is where x540 sits on *and device itself barely reserves 4M for itself in memory
3a where 4080 sits at it takes some 256+32+16M *but its idling
ae is where b60 sits at, and it requests 32G but memory behind the bridge is like 3M only.

Code:

lspci -v -s b1:00.0
b1:00.0 VGA compatible controller: Intel Corporation Battlemage G21 [Arc Pro B60] (prog-if 00 [VGA controller])
        Subsystem: ASRock Incorporation Device 6023
        Flags: fast devsel, NUMA node 1, IOMMU group 20
        Memory at 5b800000000 (64-bit, prefetchable) [size=16M]
        Memory at 5b000000000 (64-bit, prefetchable) [size=32G]
        Expansion ROM at ee400000 [disabled] [size=2M]
        Capabilities: [40] Vendor Specific Information: Intel Capabilities v1
                CapA: Peg60Dis- Peg12Dis- Peg11Dis- Peg10Dis- PeLWUDis- DmiWidth=x4
                      EccDis- ForceEccEn- VTdDis- DmiG2Dis- PegG2Dis- DDRMaxSize=Unlimited
                      1NDis- CDDis- DDPCDis- X2APICEn- PDCDis- IGDis- CDID=0 CRID=0
                      DDROCCAP+ OCEn- DDRWrtVrefEn+ DDR3LEn+
                CapB: ImguDis- OCbySSKUCap- OCbySSKUEn- SMTCap- CacheSzCap 0x0
                      SoftBinCap- DDR3MaxFreqWithRef100=Disabled PegG3Dis-
                      PkgTyp- AddGfxEn- AddGfxCap- PegX16Dis- DmiG3Dis- GmmDis-
                      DDR3MaxFreq=2932MHz LPDDR3En-
        Capabilities: [70] Express Endpoint, IntMsgNum 0
        Capabilities: [ac] MSI: Enable- Count=1/1 Maskable+ 64bit+
        Capabilities: [d0] Power Management version 3
        Capabilities: [100] Alternative Routing-ID Interpretation (ARI)
        Capabilities: [110] Null
        Capabilities: [200] Address Translation Service (ATS)
        Capabilities: [420] Physical Resizable BAR
        Capabilities: [220] Virtual Resizable BAR
        Capabilities: [320] Single Root I/O Virtualization (SR-IOV)
        Capabilities: [400] Latency Tolerance Reporting
        Kernel modules: xe

here are top pcie slots on the box

Code:

root@KVMHost:~# lspci -v -s 5d:00.0
5d:00.0 PCI bridge: Intel Corporation Sky Lake-E PCI Express Root Port A (rev 04) (prog-if 00 [Normal decode])
        Subsystem: Super Micro Computer Inc Device 096d
        Flags: bus master, fast devsel, latency 0, IRQ 41, NUMA node 0, IOMMU group 0
        Bus: primary=5d, secondary=5e, subordinate=5e, sec-latency=0
        I/O behind bridge: [disabled] [16-bit]
        Memory behind bridge: c5a00000-c5efffff [size=5M] [32-bit]
        Prefetchable memory behind bridge: 4ffffa00000-4ffffefffff [size=5M] [32-bit]
        Capabilities: [40] Subsystem: Super Micro Computer Inc Device 096d
        Capabilities: [60] MSI: Enable+ Count=1/2 Maskable+ 64bit-
        Capabilities: [90] Express Root Port (Slot+), IntMsgNum 0
        Capabilities: [e0] Power Management version 3
        Capabilities: [100] Vendor Specific Information: ID=0002 Rev=0 Len=00c <?>
        Capabilities: [110] Access Control Services
        Capabilities: [148] Advanced Error Reporting
        Capabilities: [1d0] Vendor Specific Information: ID=0003 Rev=1 Len=00a <?>
        Capabilities: [250] Secondary PCI Express
        Capabilities: [280] Vendor Specific Information: ID=0005 Rev=3 Len=018 <?>
        Capabilities: [298] Vendor Specific Information: ID=0007 Rev=0 Len=024 <?>
        Capabilities: [300] Vendor Specific Information: ID=0008 Rev=0 Len=038 <?>
        Kernel driver in use: pcieport

root@KVMHost:~# lspci -v -s 3a:00.0
3a:00.0 PCI bridge: Intel Corporation Sky Lake-E PCI Express Root Port A (rev 04) (prog-if 00 [Normal decode])
        Subsystem: Super Micro Computer Inc Device 096d
        Flags: bus master, fast devsel, latency 0, IRQ 39, NUMA node 0, IOMMU group 3
        Bus: primary=3a, secondary=3b, subordinate=3b, sec-latency=0
        I/O behind bridge: 7000-7fff [size=4K] [16-bit]
        Memory behind bridge: b7000000-b80fffff [size=17M] [32-bit]
        Prefetchable memory behind bridge: 4bfe0000000-4bff1ffffff [size=288M] [32-bit]
        Capabilities: [40] Subsystem: Super Micro Computer Inc Device 096d
        Capabilities: [60] MSI: Enable+ Count=1/2 Maskable+ 64bit-
        Capabilities: [90] Express Root Port (Slot+), IntMsgNum 0
        Capabilities: [e0] Power Management version 3
        Capabilities: [100] Vendor Specific Information: ID=0002 Rev=0 Len=00c <?>
        Capabilities: [110] Access Control Services
        Capabilities: [148] Advanced Error Reporting
        Capabilities: [1d0] Vendor Specific Information: ID=0003 Rev=1 Len=00a <?>
        Capabilities: [250] Secondary PCI Express
        Capabilities: [280] Vendor Specific Information: ID=0005 Rev=3 Len=018 <?>
        Capabilities: [298] Vendor Specific Information: ID=0007 Rev=0 Len=024 <?>
        Capabilities: [300] Vendor Specific Information: ID=0008 Rev=0 Len=038 <?>
        Kernel driver in use: pcieport

root@KVMHost:~# lspci -v -s d7:00.0
d7:00.0 PCI bridge: Intel Corporation Sky Lake-E PCI Express Root Port A (rev 04) (prog-if 00 [Normal decode])
        Subsystem: Super Micro Computer Inc Device 096d
        Flags: bus master, fast devsel, latency 0, IRQ 53, NUMA node 1, IOMMU group 14
        Bus: primary=d7, secondary=d8, subordinate=d8, sec-latency=0
        I/O behind bridge: f000-ffff [size=4K] [16-bit]
        Memory behind bridge: fbd00000-fbefffff [size=2M] [32-bit]
        Prefetchable memory behind bridge: 5ffffe00000-5ffffffffff [size=2M] [32-bit]
        Capabilities: [40] Subsystem: Super Micro Computer Inc Device 096d
        Capabilities: [60] MSI: Enable+ Count=1/2 Maskable+ 64bit-
        Capabilities: [90] Express Root Port (Slot+), IntMsgNum 0
        Capabilities: [e0] Power Management version 3
        Capabilities: [100] Vendor Specific Information: ID=0002 Rev=0 Len=00c <?>
        Capabilities: [110] Access Control Services
        Capabilities: [148] Advanced Error Reporting
        Capabilities: [1d0] Vendor Specific Information: ID=0003 Rev=1 Len=00a <?>
        Capabilities: [250] Secondary PCI Express
        Capabilities: [280] Vendor Specific Information: ID=0005 Rev=3 Len=018 <?>
        Capabilities: [298] Vendor Specific Information: ID=0007 Rev=0 Len=024 <?>
        Capabilities: [300] Vendor Specific Information: ID=0008 Rev=0 Len=038 <?>
        Kernel driver in use: pcieport

root@KVMHost:~# lspci -v -s ae:00.0
ae:00.0 PCI bridge: Intel Corporation Sky Lake-E PCI Express Root Port A (rev 04) (prog-if 00 [Normal decode])
        Subsystem: Super Micro Computer Inc Device 096d
        Flags: bus master, fast devsel, latency 0, IRQ 50, NUMA node 1, IOMMU group 16
        Bus: primary=ae, secondary=af, subordinate=b2, sec-latency=0
        I/O behind bridge: [disabled] [16-bit]
        Memory behind bridge: ee400000-ee6fffff [size=3M] [32-bit]
        Prefetchable memory behind bridge: 5b000000000-5be187fffff [size=57736M] [32-bit]
        Capabilities: [40] Subsystem: Super Micro Computer Inc Device 096d
        Capabilities: [60] MSI: Enable+ Count=1/2 Maskable+ 64bit-
        Capabilities: [90] Express Root Port (Slot+), IntMsgNum 0
        Capabilities: [e0] Power Management version 3
        Capabilities: [100] Vendor Specific Information: ID=0002 Rev=0 Len=00c <?>
        Capabilities: [110] Access Control Services
        Capabilities: [148] Advanced Error Reporting
        Capabilities: [1d0] Vendor Specific Information: ID=0003 Rev=1 Len=00a <?>
        Capabilities: [250] Secondary PCI Express
        Capabilities: [280] Vendor Specific Information: ID=0005 Rev=3 Len=018 <?>
        Capabilities: [298] Vendor Specific Information: ID=0007 Rev=0 Len=024 <?>
        Capabilities: [300] Vendor Specific Information: ID=0008 Rev=0 Len=038 <?>
        Kernel driver in use: pcieport

root@KVMHost:~#

submitted a ticket to sm, but again not sure what to expect.

RolloZ170 · May 13, 2026

CyklonDX said:
I came to think that bridge is constrained to [32-bit] instead of [64-bit] addressing rules. This hard-caps the available space at 32 GiB;

CyklonDX said:
but not sure if it would solve anything as memory behind bridge always is 32bit on all of those big x16 slots.

32bit is 4GB not 32GB

RolloZ170 · May 13, 2026

May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: PCI bridge to [bus af-b2]
May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: bridge window [mem 0xee400000-0xee6fffff]
May 13 03:50:58 KVMHost kernel: pcieport 0000:ae:00.0: bridge window [mem 0x5b000000000-0x5be187fffff 64bit pref]

Prefetchable memory behind bridge: 5b000000000-5b800ffffff [size=32784M] [32-bit]

0x5b000000000 is above 4G / 32bit address space
there may be a (fixed) space limit because of sorting.
so some kernel cmds could fix( "pci=realloc" )

but you should wait for SM answer. maybe they know the solution,
it can be a bug in latest BIOS also.

RolloZ170 · May 16, 2026

this what you search for ???

Increase MMIO space from 16-32GB to full usage of 512GB/39-bit range (Skylake/Kaby Lake/Coffee Lake)

Using UEFIPatch

Resizable BAR for (almost) any UEFI system. Contribute to xCuri0/ReBarUEFI development by creating an account on GitHub.

github.com

CyklonDX · May 16, 2026

yeah if sc support doesn't help - that would be my option.

Search

SM X11DPG-QT - not enough MMIO resources for SR-IOV

CyklonDX

Well-Known Member

RolloZ170

Well-Known Member

RolloZ170

Well-Known Member

Unable to enable SR-IOV and receiving the message "not enough MMIO resources for SR-IOV" in Red Hat Enterprise Linux - Red Hat Customer Portal

CyklonDX

Well-Known Member

RolloZ170

Well-Known Member

RolloZ170

Well-Known Member

CyklonDX

Well-Known Member

CyklonDX

Well-Known Member

RolloZ170

Well-Known Member

RolloZ170

Well-Known Member

RolloZ170

Well-Known Member

Using UEFIPatch

CyklonDX

Well-Known Member