I need to catch up on some older stuff, so here is a longer list of recent updates: libarchive to 3.0.2, xz to 5.0.3, mfi(4) and mfiutil(8) (LSI MegaRAID driver) updated, ATI SB7x0/SB8x0/SB9x0 AHCI devices (on motherboards I assume) updated, and the PHY ID for the Atheros F1 added.  Thanks to everyone who did the work!  I bet I missed something.

That’s Managed System Interrupts, for when your hardware is passing a lot of data and generating a lot of corresponding hardware interrupts.  MSI is what deals with all that traffic.  High-bandwidth (10G) network cards, for instance.  Anyway, Sepherosa Ziehau’s made more commits than what I’m linking to here, for support with various devices.

There’s many other MSIs out there, oddly enough.

Sepherosa Ziehau has added updated the ‘ecc’ device, for Intel E3-1200 series systems.  What’s it do?  It will report on memory errors, and potentially fix them.

You should have ECC memory in your server already.  If not, you oughta.

Update: as Sascha Wildner pointed out, ecc(4) already existed, but didn’t support Intel controllers.  Also, the Xeon X3400 series is supported now too.

I think a lot of people don’t realize rcrun exists, and run files in /etc/rc.d directly.  If you’re one of those people, read the rcrun(8) manpage.  Then, notice that Sascha Wildner has extended rcrun to work with the etc/rc.d in /usr/local and /usr/pkg too.

Matthew Dillon has written a contiguous memory mapper, which is designed to fix problems with video cards and USB drives that need a big chunk of memory to keep.  This can affect booting or later on, when disconnecting/reconnecting a USB drive.  If this still doesn’t fix the problem for you, try adjusting the sysctl ‘vm.dma_reserved’ to something bigger, like 64M.  It defaults to 16M.

(Normal mailarchive isn’t updating because of an ongoing upgrade to crater.dragonflybsd.org – sorry!)

John Marino added tuning support within GCC 4.4 for the Geode CPU.   Waaaay back when, these were x86 -compatible Cyrix chips.  Nowadays I think they are most common in single-board computers.

I did not realize this, but MMC/SD cards are not supported in the default DragonFly kernel.  Or at least, they weren’t until now.  (also committed to 2.12)

Update: PCI-based MMC/SD readers, specifically.  USB ones were already recognized as umass devices.

It looks like Sepherosa Ziehau is working on hardware support being split up per-CPU, judging by this commit – one of many, recently.

Sascha Wildner has updated ndis(4), the wrapper that makes Windows network drivers usable on DragonFly, with an extensive description of what’s changed.

Sascha Wildner has added safe(4), which will help if you have a SafeNet chip on your crypto accelerator card.  Untested, so you know what to do if you have this hardware.

If DragonFly/x86_64 fails to install on your system, but DragonFly/i386 works, try again.  Sepherosa Ziehau has a fix for the keyboard controller that may make x86_64 systems boot DragonFly when previously they did not.

Tim Bisson has posted a new batch of patches putting TRIM support into DragonFly.  He has a graph in there too!

It sounds like I’m about to mention something pirate-themed, doesn’t it?  Brendan Kosowski needs the rum(4) driver, for (I think) Ralink RT2501USB and RT2601USB wireless.  He’s willing to offer a bounty of $100 to anyone who can get it working before the next DragonFly release.  Work on it if you can port, or add money if you can use it.

Sepherosa Ziehau continues his relentless network feature improvement/porting: this time, adding the ability for DragonFly to recognize more varieties of Broadcom hardware.

Sepherosa Ziehau has been making a lot more changes to the msk(4) driver for Marvell Ethernet chipsets.  I link to this commit adding support for Yukon Supreme cards, but there’s a great deal of work from him, recently added.

Sascha Wildner has committed version 3.981 of the mfi(4) driver, for a variety of LSI MegaRAID SAS 92XX devices.  Read the commit message for details on the model numbers.

Sepherosa Ziehau has added support for a wider range of Marvell network interfaces; specifically the chips on board, not just card models.  If you’ve got the right chips but they aren’t working for you, you know what to do.

Thanks to Michael Neumann, there’s more supported Broadcom network card chipsets.  There’s some wierdness in setup, though, so look at his commit message.

Tim Bisson has another status report on supporting TRIM in DragonFly.   It supports UFS and Hammer slices, and trimming swap too.  I’m not sure what else could be done; that sounds pretty complete to me…  In any case, if you have a SSD, his code is available to try right now.

If you happen to use a LG P-500 smartphone to get online via USB, as ‘Romick’ does, he’s got a patch that makes that device work under DragonFly.  (Sorry, the original users@ email seems to have gone missing.)

Francois Tigeot has repeated his benchmarking, this time changing out the CPU instead of the operating system.  There’s still more graphs, yay!

Sepherosa Ziehau has a firmware update for bce(4) (Broadcom NetXtreme II) cards.  He’s been doing a lot more incidental network hardware updates I haven’t linked; thanks, Sephe!

Sepherosa Ziehau has been committing a bunch of changes for em/emx(4) and bce(4).  You may have hardware that has suddenly become supported, for instance. Also, credit is due to David Christensen and Broadcom for sending hardware to test out.

Do you have a Intel PRO/Wireless 2200BG, 2225BG, or 2915ABG wireless card?  The driver is iwi(4).  It requires a kernel module and some downloadable firmware, which makes it slightly more troublesome to set up.   Luckily, ‘ferz’ has written up just how to get it working.

Do you have a DragonFly workstation?  That you play audio on?  Do you have headphones hooked up?  Is it using Intel High Definition Audio?  (snd_hda)  Does connecting the headphones disable the system speaker?

You can probably guess exactly what I’m trying to troubleshoot given the above questions.

Here’s two items I meant to post and for some reason did not:

• Sven Gaerner posted a short description of how he migrated his DragonFly system from a hard disk to a SSD.  This may be useful for anyone considering a move.  Decent-sized SSDs are reaching low prices these days.
• Tim Bisson posted an update on his work on TRIM support for DragonFly.  The code is available now if you’re feeling lucky.

Sepherosa Ziehau’s made it possible for uniprocessor kernels to use the LAPIC and IOAPIC functions on x86_64, which means better timer support, less need to fiddle with configs, and more supported hardware.  A win all around!  Set hw.lapic_enable=”0″ if there’s trouble.  The same changes for i386 are on the way.

I haven’t covered recent disk encryption work evenly, here, so I’ll point at a recent discussion instead.  Alex Hornung mentioned a cryptsetup(8) man page that may help, as does any dm-crypt tutorial out there on the Internet.  (DragonFly has the same userland tools.)   The DragonFly installer will create encrypted disks at install time, too.

The I/O APIC is now always on unless you say otherwise.  This may not make a clear difference to you, but enabling that kernel option has always been a somewhat iffy thing; working for some configurations and not others.  Now, it’s one less thing to worry about.

I posted something about this before, but now it’s definite: bleeding-edge users of DragonFly can boot a multiprocessor kernel on a single-processor machine.

If you’ve ever wanted to really make sure of all the network interfaces supported on your DragonFly system, you can create an exhaustive (and exhausting) list.

Samuel Greear has a totally untested update to the NVIDIA video driver available.  It may not work, but it’s not like that’ll be any different than the current state of the driver.

Tim Bisson has inital TRIM support working for UFS.  His lengthy posting talks about how it’s done, and shows how much it speeds things up.  He’s looking for testers, so please try it if you have a SSD.  (The usual warnings apply about testing code that specifically deletes things.)

For those not familiar with TRIM in SSD context, here’s the least annoying page with an explanation that I could find in a few seconds of Googling.

Francois Tigeot did some testing of various hardware RAID adapters (Areca, LSI, 3ware, and Adaptec) in DragonFly, and reported thoroughly on each.  This may come as no surprise, but it sounds like Areca adapters are worth the money.

Update: There’s an updated mpt(4) driver, and the performance issues are fixed by enabling write caching.

To go along with the recently-added suggested hardware page on the DragonFly website, Francois Tigeot puts in a good word for SuperMicro boards and DragonFly, and links to some good hardware combinations.

If you have a USB printer, you may not have been able to print since the kqueue changes came in for… DragonFly 2.8?  Anyway, Matthew Dillon’s made some changes to ulpt(4) that means USB and kqueue play nicely together.

Hey, it rhymes!  Matthew Dillon’s added support for 4-port Gigabit Ethernet PCI-E cards from Intel.  I wish I had one.

Matthew Dillon made some changes to swapcache(8).  Swapcache is now able to cache a lot more data, and the result is that general disk performance for _all_ disks is accelerated by an included SSD using swapcache.  Performance previously restricted to all-SSD systems or serious RAID setups is now possible with much less investment.

In addition to that, the long-term wear on the SSD appears to be less of a problem than expected.

Chris Turner is looking to implement something similar to OpenBSD’s mount_vnd(8) operation, where virtual disks can be mounted at boot.  He talks about some of the work and ideas at length.  If you don’t feel like reading about it, you can instead mess with it; he has a tarball of the current state of his work linked in his message.

Joe Talbott has some changes for both Intel and non-Intel wifi NICs; please try out his branch and report the results.

Sascha Wildner’s removed the meteor(4) code because it apparently no longer builds, and it’s unlikely anyone uses an actual video board that requires this driver, at this point.  If you do, speak up.

Do you have a Western Digital model 1021 external disk drive?  Matthias Rampke does, and he found he had to make some USB quirk entries to get it to work reliably.

Tim Bisson posted new network tests contrasting the virtio driver against emulated re(4) in virtual environments.  Previously, the virtio driver performed worse, but a more developed test suite seems to deliver more positive results.

Sascha Wildner re-added burncd(8); it still works for some people.  As Matthew Dillon pointed out, cdrecord is probably the better long-term bet.

Tim Bisson put together support for the RealTek 8168E network card, under the re(4) driver.  It’s in DragonFly now.

Sascha Wildner has continued his driver-adding run, bringing in mps(4).  This supports various LSI Logic SAS controllers, taken from FreeBSD.  Support isn’t complete or tested, but it’s enough to start with.

Tim Bisson posted a note on the progress he and Pratyush have made on a virtio driver for DragonFly, ported from NetBSD.  This is for use in virtualized environments; his post links to graphs (yay!) that show the performance improvement over emulated IDE.  His note also links to the code and documentation.

Sepherosa Ziehau fixed a clock issue with the JMicron JMC250/JMC260 chipset, used with the jme(4) driver, and apparently JMicron helped out with hardware for testing this fix.  So, thanks, Sephe, and thanks, JMicron!  (buy their stuff)

Sascha Wildner has brought in the mfi(4) and mfiutil(8) drivers from FreeBSD, adding support for a number of different RAID controllers – including the Dell PERC 5 and PERC 6.

Sascha Wildner has updated the arcmsr(4) driver, for you Areca users out there.  I think Areca was one of the vendors kind enough to test DragonFly on their hardware directly, so please consider them next time you are in the market for a SATA RAID card.

Matthew Dillon’s made several changes to improve support for AMD SB850 chipsets (for AHCI) and also for 880/890 chipsets.  If you have one of these systems, it may be bootable/more reliable.  Don’t start messing with the hot-plug capability yet, though.

Chris Turner is working on ral(4) support, specifically the eee901′s 2860 network chip.