Recent content by ericloewe

The U.2 connector was probably cheaper and/or more available for that production run. I bet the PCIe pins are all not connected to anything.

Because why not? They're cheap and cheerful and easily handle 3 Gb/s and 6 Gb/s per their spec. 12 Gb/s would get dodgy, though. Of course, single lane only, but you would never connect both lanes to the same host port, so there's no point in having a dual-lane SAS cable. As you've discovered...

Also, running compression with 4k writes is utterly pointless as all blocks with ashift=12 (you did set ashift=12 when creating the pool, right?) will be at least 4k anyway. But really, reconsider whether 4k sync writes are really necessary or representative of your future workload.

The .12 firmware update, IIRC, fixed some edge cases with SATA devices that would fail pretty obviously with .10. Probably not a big deal if Dell's firmware works without apparent issue.

I'm not sure I've seen it done with the SAS3108, but the process has been documented for some SAS2 controllers. It's definitely not officially supported and Broadcom's tools will fight you every step of the way, but it's also true that the hardware is essentially compatible (which is also how...

Since it wasn't obvious, my post #13 above was sarcastic.

The PCIe configuration space on these things has the SR-IOV flag set to true, what more support do you want?

Ok, but then that needs all sorts of plumbing that doesn't exist and is unlikely to exist (see the relatively poor handling of SR-IOV NICs, and those have been around for literally a decade and a half): The driver needs to handle all the housekeeping involved in this The host needs an interface...

Ok. What happens when you add an expander? An expander is always going to be taking at least four ports. Is the card just going to hot plug a truckload of new PCIe devices? Does it remove three channels and shove every disk behind the expander into the first channel? The concept breaks down...

I have to steal this one.

I think there are just too many moving parts for this to really be viable outside of a niche use case. Sure, you could assign disks to VFs, but then what about hot swap? What if you add a whole expander full of disks? Do you rely on SES to maintain a physical slot mapping table that can then be...

Well, it's rude not to share. If there's anything substantial to it, I'm sure a bunch of people would be interested in knowing more, me included.

Who says that SR-IOV is in any way supported? What would it even mean for an HBA? Google shows me a lot of talk about SR-IOV with LSI HBAs, but very little in terms of anyone acquiring the mythical firmware that enables it, much less what the VFs would expose and how.

3. Enjoy worse performance than if you'd plugged them into a standard PCIe bus.

You don't, unless you're a masochist who loves lower performance at higher cost.