This happened a while ago, and I’m just catching up to it: the virtio_scsi(4) driver has been added to DragonFly; ported from FreeBSD and worked on by a number of people. ‘man virtio‘ if you want background.
If you are running DragonFly on a Ryzen CPU, this commit will fix (work around?) a hardware bug. I have not looked at how other operating systems may be addressing it, but it may be interesting to contrast.
Remember the firmware in userland idea I linked to before? Here’s the written plan/proposal.
If you’ve got a Skylake CPU, setting P state won’t save you as much energy as powerd(8)‘s -c option, according to Sepherosa Ziehau.
Bryan C. Everly eventually figured out how to configure his ThinkPad x230 so that the TrackPoint worked in xorg, and he wrote it down.
Do you have an isp(4) device? That would be a Qlogic SCSI/Fibre adapter. If you do, firmware handling has changed internally, thanks to Jan Sucan. I think configuration is unchanged, however.
There’s an update for the i915 (Intel) driver that is mostly of interest to you if you have a more recent processor. It looks to be mostly bugfixes.
If you’re looking for hardware RAID on DragonFly, here’s some brief notes on what should work. Areca and LSI are the hardware names to watch for. The person who asked the original question pointed to earlier benchmarks, which I may have linked before.
Thanks to Imre Vadasz, your Haswell, Atom SoCs, Skylake, and Kaby Lake CPUs will now recognize the ig4 device on the associated motherboard. I think it does something with ACPI? I have always been hazy on smbus device functions.
Radeon hardware support in DragonFly has been moved up to match what’s in the Linux 4.7.10 kernel. If you have a R9 290 GPU, there’s some tweaks you may need.
(n.b. may be unnecessary now from later commits; I don’t have the hardware to check.)
For some reason I am completely unfamiliar with this standard, but UHS-1 support for Secure Digital cards has been ported to DragonFly by Imre Vadasz, for a limited range of models. UHS stands for “Ultra High Speed”, so perhaps it’s clear what that standard does for you.
If your DragonFly system’s Intel network device doesn’t seem to pick up on DHCP, try turning on polling. This may already be a nonissue, but it doesn’t hurt to mention it.
Update: fixed.
Imre Vadasz added support for ADMA2 transfers in DragonFly. It doesn’t lead to a huge performance boost – yet. It can be turned on and off, but requires Intel chipsets.
Mobile Skylake CPUs appear to have issues with power management and direct video rendering. There’s potential fixes on the horizon, but until then, be aware if you have that specific hardware and software mix.
I noted commits about this before, but here’s the instructions: how to use DragonFly on a Macbook Pro with dual GPUs.
There’s an update for Radeon DRM that matches it up (mostly) to what’s in Linux kernel 4.7.10. If it gives you problems, there’s some workarounds. Remember, this is in DragonFly-current, so anyone running 4.8 is unaffected.
I am late in mentioning this, because it was added just before the DragonFly 4.8 branch: there’s a new ‘efisetup(8)‘ script added to DragonFly. Use to to perform a complete a UEFI-bootable installation to a given disk.
Now that we’re past the DragonFly 4.8 release, Francois Tigeot has added an update to the i915 driver, bringing it to match what’s in the Linux 4.7.10 kernel. He also committed Peeter Must’s port of the vga_switcheroo module.
Matthew Dillon picked up a number of different NVMe SSD drives, and tested them. He wrote up the entire test, but the immediate summary is: buy Samsung.