Sepherosa Ziehau has added MSI-X support to igb(4), the Intel PRO/1000 gigabit network card. What does that mean? The commit message mentions a default transmit rate of 1.48Mpps small packets, which is good?
The usual weekly updates from Mihai Carabas, Vishesh Yadav, and Ivan Sichmann Freitas. Mihai has some interesting bugs found this past week by running his code on Matt Dillon’s 48-core system.
Francois Tigeot has added the Intel PRO/10GbE driver from FreeBSD, or ixgbe(4). A couple features are turned off, for now.
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.
This is a report for last week’s work, so this is week 6 we are in now, and the reports are week 5’s status. So:
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.)
Based on a suggestion from Venkatesh Srinivas, tmpfs now uses a red-black tree for directory lookups, and is also now faster. Credit goes to Johannes Hofmann for doing the testing.
Sascha Wildner has synced find(1) with what’s in FreeBSD, which means there’s a lot more options available – see the commit for details. Many of them are for GNU compatibility, and I’m sure I’ll forget them all. I seem to have issues remembering how to use find(1) successfully.
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.
Here’s your most recent weekly round of DragonFly/Google Summer of Code updates:
- Ivan Sichmann Freitas: 32 bit api status
- Vishesh Yadav: inotify and fs indexing service status
- Mihai Carabas: Add SMT/HT awareness to DragonFlyBSD scheduler
- Loganaden Velvindron: Privilege separation
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.
The i386-only doscmd(1) is gone from DragonFly. I don’t think I ever used it, as other emulators/systems are so prevalent and complete.
The cache coherency management code in DragonFly has been removed, but it’s coming back under HAMMER2, as part of how HAMMER2 maintains multiple master drives.
Venkatesh Srinivas, currently on his colossal bike ride, introduced a different way of creating a tmpfs. This was test code, and Johannes Hofmann benchmarked it (see same page). It’s interesting cause there are numbers, and nice to see one person jumping in to test someone else’s results/idea.
Week 2 Summer of Code status reports from Loganaden Velvindron, Mihai Carabas (plus followup), Vishesh Yadav, and Ivan Sichmann Freitas are available. Ivan Sichmann Freitas also has a RFC on changes to DragonFly’s 32-bit API.