Truly curious how they did this without you sending it in to them...I got in touch with Intel, they were able to confirm that it was genuine.
Truly curious how they did this without you sending it in to them...I got in touch with Intel, they were able to confirm that it was genuine.
i can tell you one i learned through pain staking trial and error, Dell FKHKC Intel Ethernet X550-T2 doesn't support it =(Is there a definitive list to which Dell card part numbers don't support SR-IOV?
And second question, what are the physical differences between T4 and T4V2 cards if I only have photo of the top side with no sticker indicating version present on that side?
I haven't looked into any of the more recent cards since I originally found this thread years ago. For the specifics, I would search for each of the brand's part numbers and SRV-IO and see what search results return. If there are any issues, someone would be discussing it online somewhere.@Samir do you happen to know?![]()
Says "Virtual Channel" but not "Single Root I/O Virtualization (SR-IOV)". So I am going to say "no". Card has the latest firmware from Dell from 2024. And all ports land in the same IOMMU group. Some Reddit posts claim there are i350 with SR-IOV disabled.I would like a card that can do both PCI pass-through and SR-IOV. Can the Broadcom card you mentioned do this?
# lshw -c network -businfo
Bus info Device Class Description
==========================================================
pci@0000:17:00.0 enp23s0f0 network NetXtreme BCM5719 Gigabit Ethernet PCIe
pci@0000:17:00.1 enp23s0f1 network NetXtreme BCM5719 Gigabit Ethernet PCIe
pci@0000:17:00.2 enp23s0f2 network NetXtreme BCM5719 Gigabit Ethernet PCIe
pci@0000:17:00.3 enp23s0f3 network NetXtreme BCM5719 Gigabit Ethernet PCIe
pci@0000:1a:00.0 enp26s0f0 network Ethernet Connection X722 for 1GbE
pci@0000:1a:00.1 enp26s0f1 network Ethernet Connection X722 for 1GbE
pci@0000:1a:00.2 enp26s0f2 network Ethernet Connection X722 for 1GbE
pci@0000:1a:00.3 enp26s0f3 network Ethernet Connection X722 for 1GbE
pci@0000:b3:00.0 enp179s0 network MT27500 Family [ConnectX-3]
usb@1:11 wlp0s20f0u11 network Wireless interface
pci@0000:b3:00.0 enp179s0d1 network MT27500 Family [ConnectX-3]
# lspci -vs 0000:17:00.0
17:00.0 Ethernet controller: Broadcom Inc. and subsidiaries NetXtreme BCM5719 Gigabit Ethernet PCIe (rev 01)
Subsystem: Broadcom Inc. and subsidiaries 4-port 1Gb Ethernet Adapter
Flags: bus master, fast devsel, latency 0, IRQ 55, NUMA node 0, IOMMU group 8
Memory at 27fff4b0000 (64-bit, prefetchable) [size=64K]
Memory at 27fff4a0000 (64-bit, prefetchable) [size=64K]
Memory at 27fff490000 (64-bit, prefetchable) [size=64K]
Expansion ROM at c5ec0000 [disabled] [size=256K]
Capabilities: [48] Power Management version 3
Capabilities: [50] Vital Product Data
Capabilities: [58] MSI: Enable- Count=1/8 Maskable- 64bit+
Capabilities: [a0] MSI-X: Enable+ Count=17 Masked-
Capabilities: [ac] Express Endpoint, IntMsgNum 0
Capabilities: [100] Advanced Error Reporting
Capabilities: [13c] Device Serial Number XX-XX-XX-XX-XX-XX-XX-XX
Capabilities: [150] Power Budgeting <?>
Capabilities: [160] Virtual Channel
Kernel driver in use: tg3
Kernel modules: tg3
My opinion is that the top one is definitly fake, an older one too when they didn't copy so well. The bottom two could be fake or real--the Dell and HP fakes are identical to the real ones so you'll only know when you use both in the same system and one is acting funny--that one will be the fake. 'Acting funny' was how the fakes were first spotted as fake Intels would have problems with iscsi, but that was years ago. Today's fakes are very, very good in terms of looking real. The fakers have seen this thread and changed their tactics.I have 3 Intel I350 cards...two 4 port and one 2 port card. Can someone please tell me if these are fake or authentic.
You might as well have had a flakey original Intel card, where somebody ripped off some components when uninstalling or handling the card and you didn't (and couldn't, without super close inspection) see that. The capacitors and other SMD parts, on the other side from where the chip is, are really tiny. Smash card on table too hard and now you are missing three, one gone completely and the other two cracked.Today's fakes are very, very good in terms of looking real
That can happen too, but this was from extensive reading on others' experiences and warranty attempts with Intel. I personally avoided the whole issue by getting some not commonly faked at the time HP cards that were the same as the Intels.You might as well have had a flakey original Intel card, where somebody ripped off some components when uninstalling or handling the card and you didn't (and couldn't, without super close inspection) see that. The capacitors and other SMD parts, on the other side from where the chip is, are really tiny. Smash card on table too hard and now you are missing three, one gone completely and the other two cracked.
I guess the rabbit hole here is, where did he get those cards. Or is there some way newer component on a card with a date code far far away into the future from say 2013. This would be a hint towards a clone. Also back in 2013 Dell and even Intel used to have something called quality control teams. When Dell disabled SR-IOV it might not even have been product differentiation, but unsolved issues with Intel's drivers, firmware, or bugs in the chip, or at least in an older stepping. Which Intel hid behind an NDA.
That could be it. For example the SR-IOV on the 'Devices settings' for each NIC will only appear and has the option to be enable or not when 'Integrated device' in Dell BIOS settings set to enabled it (SR-IOV). I did not test the card on other generic system but there's possibility. (I'll try and check it one day).bear in mind that its not necessarily all china in this, some manufacturers lock SR-IOV behind firmware which you can't change (or at least not that i've seen).
take my previous post as an example in this, that was a genuine Dell part but Dell decided to lock SR-IOV away.