If you are using an Intel 10G Ethernet card with a 82598GB chipset, you’re using ixgbe(4). You may want to set the net.inet.tcp.sosend_agglim sysctl to a value over 12 in certain circumstances, as described by Francois Tigeot.
These are small, but they make life easier: Hammer now has a scoreboard file, for viewing of mirror-streams running in the background. There’s also a ssh-remote directive, so you can use ssh without enabling an interactive shell, and a HAMMER_RSH environment variable so different remote shells can be used. These are all for Hammer 1.
If you ever wanted to read an extensive discussion about the scheduler, today’s your day. Mihai Carabas, who posted the details of a long discussion he had with Matthew Dillon about how the scheduler works. You may recall Mihai’s name from the very successful GSoC scheduler project that recently finished.
(look, a link to the new Mailman archive!)
If you’re on any of the dragonflybsd.org mailing lists, I’m converting them over from bestserv to Mailman. I’ve done bugs@, commits@, hammer@, and test@ so far, and I’ll move the old archives over to the same format as soon as I find an actual mbox file with the old messages in it. The remaining lists should be tomorrow.
(If you got a note tonight from a list you were sure you were unsubscribed from, that was my fault; sorry! I didn’t understand the format of the bestserv user lists.)
DragonFly user varialus has created a page on the DragonFly website (it’s a wiki, after all) with all the notes taken from trying installation, etc. There’s far more notes than I expected there, so it’s worth a read.
Much of this new document has been around in other forms for a while, but now, there’s a brief guide on porting drivers to DragonFly in the source tree.
I’ve uploaded DragonFly 3.0.3 disk images, both ISO and IMG. They should start appearing on a mirror site near you in the next 24 hours. This took a while after the tagging, I know, but I wanted to make sure every one of them booted. I didn’t on a previous release, and regretted it.
Francois Tigeot benchmarked several different operating systems using Postgres 9.2b3, including DragonFly, and published the results. I have a local copy of the PDF since the attachment didn’t really survive the archiving. Follow the thread for discussion. The Linux results look abnormally high, so it is possible that something different is happening on that platform…
Do you happen to have the saved messages for one or more @dragonflybsd.org mailing lists sitting around? Hopefully in mbox format? I’m working on getting Mailman installed to replace bestserv, and being able to bring in the old messages would be nice.
I’m working on building new images, but: DragonFly 3.0.3 has been tagged. If you’re running 3.0, you can update and get some of the recent bug fixes.
I noticed that this recent commit from Sepherosa Ziehau is a bug fix for jme(4). The commit thanks a JMicron employee for help. It’s always appreciated when a vendor is helpful to an open-source project for hardware support. It’s also something you should consider the next time you are shopping for computer parts.
John Marino has been on a tear fixing pkgsrc packages, and he posted a list of what he considers the most necessary packages to get working on DragonFly. Several people have already stepped up and fixed them if you follow the thread. If one of these packages is something you use, it’s worth looking at.
John Marino is working on updating tcl in pkgsrc. It’s apparently quite messy to update, which may be why it has sat out of date for some time. Never one to rest, he’s also been making FUSE filesystems work on DragonFly. (Here’s a FUSE explanation, if you need it.)
Also this. Someday I’m going to write a “games on DragonFly” feature, or series.
If you wanted to try Hammer 2 and you have several DragonFly-current systems around (virtual or not), Matthew Dillon has the instructions. Keep in mind that this is not something ready for use; it can’t actually free up space, for instance. It’s neat that you can have multiple systems passing data back and forth already, though!
John Marino finished another bulk build of pkgsrc, and reports a 96.4% package success rate, using DragonFly and pkgsrc-current. We’re just a week or so from the next quarterly pkgsrc freeze, come to think of it…
3.2 is the next major release of DragonFly, which will be relatively soon by the every-6-months release schedule. John Marino’s put together another catch-all bug report for that release.
I recently completed a bulk build of pkgsrc-2012Q2 on 64-bit DragonFly, though I still haven’t had a successfuly 32-bit build. However, John Marino has a report of how many packages are working on DragonFly in pkgsrc-current. (Answer: more than 95%)
Matthew Dillon recent posted a status report for Hammer 2. Of interest is the spanning tree protocol being built to handle messages between Hammer volumes. As he says in the message:
For example, we want to be able to have millions of diskless or cache-only clients be able to connect into a cluster and have it actually work…
(No, it doesn’t do this, yet.)
Hammer 2 (or is it HAMMER2?) is nowhere near ready to test. But! For laughs, I think it could be set up just so you can watch the messages go back and forth. Someone want to set up a few DragonFly-current VMs and try?
Pierre Abbat is curious about using Hammer on an SSD. The discussion that came from that has some useful points, including notes that a straightforward SSD as disk works for most anything with Hammer other than very intensive database use, due to the history retention. If space is an issue, swapcache on the SSD and attaching a normal HDD is a fine alternative. A SSD with Hammer can leave some features off, though I’d argue that dedup is totally worth is. Also, SSD speed is directly correlated with size.
Mihai Carabas has posted some more results from an 8-core system showing his efforts to make the scheduler multi-threading aware. The results are generally a 5% speed gain, which I think matches previous benchmarks on machines with less processors.
Sepherosa Ziehau’s added TSO support (that’s TCP Segmentation Offloading”, or “Large Segment Offload” going by Wikipedia) within IPv4 on DragonFly, pushing segmentation work from the CPU to the network card. There’s also some DragonFly-specific improvements.
There’s been a lot of commits from him lately focused around network card improvements; they haven’t been easily summarizable, but it’s worth watching if you are interested in high-bandwidth usage and the hardware to support it.
Not all flavors of Atom CPU support frequency scaling, as Sven Gaerner found out. This means more heat and more power usage. There’s further details scattered through the thread, but Sascha Wildner found what seems to be the definitive answer of which variants do and do not.
Pierre Abbat noticed that bc(1)‘s usage of
GNU readline something that wasn’t GNU readline made it harder to use; Sascha Wildner changed it to use libedit. Pierre’s other complaint, that BSD man page output stays on-screen when completed, is a positive feature. Linux systems that clear man page output enrage me, because I expect to be able to take advantage of my scroll buffer.
Juraj Sipos wrote me to describe MaheshaDragonFlyBSD, a live DragonFly image that has additional software preinstalled, and can easily be set to understand Sanskrit. It’s available in DVD and USB versions.
John Marino sent a nice email to users@ about the improvements in build success for pkgsrc since May – and I can’t find it in the mailarchive. I’ll paste a summary after the break.
At least for DragonFly, every current participant in Google Summer of Code passed the midterm evaluation. Yay!
I don’t, but I know there are people that do. That’s why I’m pointing out this discussion where it appears that TeXLive 2012 won’t support NetBSD, which may mean no DragonFly either. There’s the not-yet-packaged alternative kertex. TeXLive is in pkgsrc, so I don’t know if that means the package will be discontinued or just altered.
(Please correct me where I go wrong here; I’m not very familiar with this, but it sounds like a drastic enough change that it should be mentioned.)
Update: as several people pointed out, it’s just prebuilt binary versions that aren’t being provided upstream. The packages will all still be present in pkgsrc. So, no functional change for most everyone.
… because versions 3.0 and 3.3 will be leaving pkgsrc soon-ish. You’d probably want to update anyway, but this is just in case you haven’t been upgrading too vigorously.
John Marino has added a ‘gcc47′ compiler ccvar, so you can build world and kernel with it. ’It’ is actually gcc-aux, since it seems to work better than the basic (“vanilla”?) gcc47. You also get Ada support, though that wasn’t the driving reason to pick it. This is brand new so don’t try it unless you’re ready to discover issues.
Is there any other BSD able to use gcc 4.7 for world/kernel? Even 4.6? Most of the attention has been on clang.
If you want to put something towards DragonFly, and you don’t have time or hardware, cash is now an option. (It’s not tax-deductible.)
Nuno Antunes is still working on that netgraph upgrade. Among other changes, ng_tty has been added. What’s it do? Something with ppp, I think.
Someone trying DragonFly couldn’t get it to start, and appeared to have a confused disk. It looks like the system BIOS were at fault, and Matt Dillon has an explanation of this minefield. (Including some comments on 4k physical disk sectors.)
Attention students and mentors: the Summer of Code midterms open up on July 9th. This means students fill out an evaluation, and mentors also fill out an evaluation. Don’t forget, because completed evals from mentor and student both are necessary for a project to continue being funded.
More benchmarks, in this case a comparison of Debian GNU/kFreeBSD and DragonFly. I’m not even sure how to derive meaning from it.
If you have an Intel processor with multiple cores and hyperthreading support, you can compile a new kernel and try out Mihia Carabas’s GSoC work already; he’s created a test using the OpenSSL test case to time scheduling performance vs. number of threads.
Mihai Carabas posted some benchmarks for his work with the DragonFly default scheduler and hyperthreaded CPUs. The end result, for those who don’t like number analysis, is that CPU-dependent speeds are reliably constant because tasks are being evenly scheduled across available CPUs.
(Well, CPU threads, since this is hyperthreading, but you get the idea.)
I think it’s week four, at least.
Mihai Carabas, Vishesh Yadav, and Ivan Sichmann Freitas all have their weekly status reports up for Summer of Code. Unfortunately, Loganaden Velvindron received a great job offer out of the blue, so he no longer has time for Summer of Code. (He plans to continue involvement in DragonFly, however.)
Sascha Wildner has made it easier to use alternative syntax checking systems as a “lint” make target in DragonFly. His usage of coccinelle, as one of these alternatives, has already found many bugs – just today, for instance.
Is “alternative syntax checking systems” the right phrase for this? I don’t know. ”Correctness checker”? My phrases all sound like something you’d read on a government form.
Reading this HAMMER2 commit carefully shows some future plans: remote cluster control, and the ability to mount nonlocal HAMMER2 volumes. A reminder: those are future plans, not what you can do now.
It’s possible to accidentally truncate your password when using DES encryption and 0×80 in UTF-8 encoding. It’s fixed.