Intel Xeon D-1500 Series Discussion

Paul Braren

New Member
Nov 10, 2012
22
6
3
Wethersfield, CT
TinkerTry.com
There's a new bios for the supermicro X10SDV series that is supposed to make M.2 SSDs bootable (specifically samsung SM951, and likely XP941). They just emailed me about it months after my initial email about there being a bug with it. I guess that's better than never emailing me.
FYI, I installed the 5028D-TN4T/X10SDV-TLN4F BIOS 1.0b, then installed Windows 10 to my M.2 SM951 with BIOS in UEFI mode. It went fine, bootable.
Curious if you have more info the bug. Thanks!
 

paulsperch

New Member
May 10, 2011
7
1
3
I am in the market for a new board and have settled on the Supermicro X10SDV-4C-TLN2F. Since stock levels have been extremely low for it (almost non-existent at decent prices) I decided to ask Supermicro about it. The tech support rep I spoke with asked the team responsible for that board and they stated that there are not any known issues with the it however they currently have a new revision in beta testing. Just thought I would pass along what I learned (or at least what I was told). Since my urgency for a new board is lower now than when I started looking I might wait another 2-3 months to see if any new boards or revisions are released.
 
  • Like
Reactions: AmshTemp

Paul Braren

New Member
Nov 10, 2012
22
6
3
Wethersfield, CT
TinkerTry.com
FYI, I was asked "SM951 is not 1 type.
is yours MZVPV (NVMe) or MZHPV (AHCI)?"

Excellent question, and a very good point. I'll be testing 1.0b UEFI boot from an Intel 750 Series next, which is NVMe.

I should have mentioned that my SM951 is MZHPV128HDGM (pictured here), yep, the AHCI model, not the (elusive) NVMe model.

In the BIOS, it doesn't show under the 6 SATA devices, nor would I expect it to. You can truly have a 7th (M.2) and 8th (PCIe NVMe) drive with this system, which I've tested.

Hopefully someday fairly soon the Samsung 950 PRO (M.2 NVMe) will be available!
 
  • Like
Reactions: AmshTemp

AmshTemp

New Member
May 28, 2015
18
0
1
42
Thanks Paul.
Indeed, I removed the question because I saw somewhere BIOS 1.0b has fixed the issue.
Not sure where though.


Also a side question: x557 is useless in ESXi, but this motherboard has two i350 ports.
So are these two i350 ports dependent on each other? In other words, can you passthrough ONLY one of the two i350 ports available?

Eagerly waiting for your NVMe tests.
 

Davewolfs

Active Member
Aug 6, 2015
337
31
28
I was told that 1.0b has micro code changes to address a C3/C6 sleep issue which can cause CPU to crash. My microcode revision did not change when I applied the update.
 

Dom

New Member
Oct 22, 2015
5
0
1
45
Just spoke to Asrock in California. No info on the Asrock D1540D4U-2T8R, expected release info will be mid November.

I'm getting quoted $1200+ for the SM "equivalent" X10SDV-8C-TLN4F, which is just gouging so I'll refuse to justify it by not buying one. (In Canada)

It's nice to see Raid, SPF/RJ 10G, muticore low voltage solutions All-In-One.. About time we can have our cake and eat it to.

Unfortunately I have $16K in budget for the remainder of the year but can't spend a dime until these new boards come out. However the EPC612D4U-2T8R is in stock @ 650 ish, add another 400 for CPU, a bit of easier pill to swallow vs. SM.
 
Last edited:

RchGrav

Member
Aug 21, 2015
44
28
18
50
Just spoke to Asrock in California. No info on the Asrock D1540D4U-2T8R, expected release info will be mid November.

I'm getting quoted $1200+ for the SM "equivalent" X10SDV-8C-TLN4F, which is just gouging so I'll refuse to justify it by not buying one.

It's nice to see Raid, SPF/RJ 10G, muticore low voltage solutions All-In-One.. About time we can have our cake and eat it to.

Unfortunately I have $16K in budget for the remainder of the year but can't spend a dime until these new boards come out. However the EPC612D4U-2T8R is in stock @ 650 ish, add another 400 for CPU, a bit of easier pill to swallow vs. SM.
PM Sent, not sure if that will help you.
 

ggg

Member
Jul 2, 2015
35
1
8
42
Here's the pastebin with the output above (2.6 kernel, so not quite everything)...
Below is what I see about the the SR-IOV and ACS support for then passing through the VFs.

I'm not familiar with this chipset and how DMI is used in it. I do too want to know if I could use SR-IOV here.

I would like to see how IOMMU groups are created with this:

$ find /sys/kernel/iommu_groups/ -type l

Maybe you could post that too?

Also the version of ixgbe that is upstream in the kernel now (4.2+?) should support the X552/X557-AT.

Perhaps you might boot off a Ubuntu 15.10 or latest Arch USB disk and try this:

$ echo 7 > /sys/class/net/eth0/device/sriov_numvfs

Where 'eth0' is replaced by one of the I350 and X552/X557-AT adapters.

The network controllers (4) and (6) support SR-IOV. The root port (3) upstream of (4) supports ACS.

However the host bridge (1) upstream of everything and chipset root port (5) upstream of (6) don't report ACS capabilities.

The ACS support (or potential absence of actual or reported support for it) might not prevent the VFs from being created. It would prevent the created VFs from being passed through if the created VFs are not at least all in one IOMMU separate from the PF (i.e. devices 03:00.0 and 03:00.1).

(1) 00:00.0 Host bridge [0600]: Intel Corporation Xeon E7 v4/Xeon E5 v4/Xeon E3 v4/Xeon D DMI2 [8086:6f00] (rev 02)
...
(2) \-[0000:00]-+-00.0 Intel Corporation Xeon E7 v4/Xeon E5 v4/Xeon E3 v4/Xeon D DMI2
...
(3,4) +-02.2-[03]--+-00.0 Intel Corporation Ethernet Connection X552/X557-AT 10GBASE-T
...
(5,6) +-1c.0-[05]--+-00.0 Intel Corporation I350 Gigabit Network Connection

(1) 00:00.0 Host bridge [0600]: Intel Corporation Xeon E7 v4/Xeon E5 v4/Xeon E3 v4/Xeon D DMI2 [8086:6f00] (rev 02)
[No ACS Capabilities Reported]
...
(2) 00:01.0 PCI bridge [0604]: Intel Corporation Xeon E7 v4/Xeon E5 v4/Xeon E3 v4/Xeon D PCI Express Root Port 1 [8086:6f02] (rev 02) (prog-if 00 [Normal decode])
...
Capabilities: [110 v1] Access Control Services
ACSCap: SrcValid+ TransBlk+ ReqRedir+ CmpltRedir+ UpstreamFwd+ EgressCtrl- DirectTrans-
ACSCtl: SrcValid+ TransBlk- ReqRedir+ CmpltRedir+ UpstreamFwd+ EgressCtrl- DirectTrans-
...
(3) 00:02.2 PCI bridge [0604]: Intel Corporation Xeon E7 v4/Xeon E5 v4/Xeon E3 v4/Xeon D PCI Express Root Port 2 [8086:6f06] (rev 02) (prog-if 00 [Normal decode])
...
Capabilities: [110 v1] Access Control Services
ACSCap: SrcValid+ TransBlk+ ReqRedir+ CmpltRedir+ UpstreamFwd+ EgressCtrl- DirectTrans-
ACSCtl: SrcValid+ TransBlk- ReqRedir+ CmpltRedir+ UpstreamFwd+ EgressCtrl- DirectTrans-
...
(4) 03:00.0 Ethernet controller [0200]: Intel Corporation Ethernet Connection X552/X557-AT 10GBASE-T [8086:15ad]
...
Capabilities: [160 v1] Single Root I/O Virtualization (SR-IOV)
IOVCap: Migration-, Interrupt Message Number: 000
IOVCtl: Enable- Migration- Interrupt- MSE- ARIHierarchy+
IOVSta: Migration-
Initial VFs: 64, Total VFs: 64, Number of VFs: 0, Function Dependency Link: 00
VF offset: 128, stride: 2, Device ID: 15a8
Supported Page Size: 00000553, System Page Size: 00000001
Region 0: Memory at 0000000000000000 (64-bit, non-prefetchable)
Region 3: Memory at 0000000000000000 (64-bit, non-prefetchable)
VF Migration: offset: 00000000, BIR: 0
...
(5) 00:1c.0 PCI bridge [0604]: Intel Corporation 8 Series/C220 Series Chipset Family PCI Express Root Port #1 [8086:8c10] (rev d5) (prog-if 00 [Normal decode])
[No ACS Capabilities Reported]
(6) 05:00.0 Ethernet controller [0200]: Intel Corporation I350 Gigabit Network Connection [8086:1521] (rev 01)
...
Capabilities: [160 v1] Single Root I/O Virtualization (SR-IOV)
IOVCap: Migration-, Interrupt Message Number: 000
IOVCtl: Enable- Migration- Interrupt- MSE- ARIHierarchy-
IOVSta: Migration-
Initial VFs: 8, Total VFs: 8, Number o Capabilities: [110 v1] Access Control Services
ACSCap: SrcValid+ TransBlk+ ReqRedir+ CmpltRedir+ UpstreamFwd+ EgressCtrl- DirectTrans-
ACSCtl: SrcValid+ TransBlk- ReqRedir+ CmpltRedir+ UpstreamFwd+ EgressCtrl- DirectTrans-f VFs: 0, Function Dependency Link: 01
VF offset: 384, stride: 4, Device ID: 1520
Supported Page Size: 00000553, System Page Size: 00000001
Region 0: Memory at 0000000090040000 (64-bit, prefetchable)
Region 3: Memory at 0000000090060000 (64-bit, prefetchable)
VF Migration: offset: 00000000, BIR: 0
 

Patrick

Administrator
Staff member
Dec 21, 2010
12,363
5,495
113
@ggg Just so you know, the official stance from Intel is that SR-IOV is not supported on the Xeon D NICs.

@Dom apologies I missed this on Thursday. I was actually in Taipei and met with the ASRock Rack folks. I did get to see a version of the board you mentioned and I urged them to get it here as fast as possible. I was told it would be coming soon. I also spoke with the US sales channel person there and stressed the need for availability. Look for more info tomorrow on the STH main site.
 

Patrick

Administrator
Staff member
Dec 21, 2010
12,363
5,495
113
Some News about Pentium/Xeon-D:
Intel Xeon D SoCs may have up to 16 cores and others upcoming models (cpu-world source)
Yes - we are still a bit away from seeing these in shipping product from what I am told.

The Xeon D line is taking over the low end no-integrated GPU role from the E3 for many applications.

Supply constraints are still present though.
 

gigatexal

I'm here to learn
Nov 25, 2012
2,870
577
113
Portland, Oregon
alexandarnarayan.com

Patrick

Administrator
Staff member
Dec 21, 2010
12,363
5,495
113
Thanks Patrick for the read. Is not having this a show stopper for you? It sounds like a rather important feature. Do you think future drivers will enable it or is Intel segmenting this feature out top recent cannibalizing its other products.
Not necessarily a show stopper but is why I am not overly keen on this as a heavy virtualization platform (e.g. ESXi.)

Also @Dom you may want to check this out: http://www.servethehome.com/asrock-rack-xeon-d-ultimate-motherboard-spotted/
 
  • Like
Reactions: KioskAdmin

whitey

Moderator
Jun 30, 2014
2,771
872
113
40
Damn, so ASRock is going to take my money for an SFP+ part instead of SM.. what a shame.
HELL YES they are, I am stoked a vendor is actually making what a LOT of us have been patiently waiting for...sure I would have liked it to be a SM part so maybe this ASRock Rack D1540D4U-2TO8R offering will put some pressure on them.
 
  • Like
Reactions: Kristian

abstractalgebra

Active Member
Dec 3, 2013
182
26
28
MA, USA