Introduction to the new Haiku scheduler, and other tidbits

Hi, folks!

For those who don’t know me (or my GSoC assignment) already, I’m the one assigned to ticket #1069, namely:

Create an O(1) thread scheduler with CPU affinity and soft real-time support which targets desktop responsiveness.

I’d like to dedicate my next few blog entries to introducing myself and discussing how I got here, why I wanted to tackle this specific task, what background I have regarding the subject of thread scheduling, how I failed miserably to realise that my first attempt at designing an algorithm suitable for Haiku’s needs had fundamental flaws, how far I am at my second attempt, and the obligatory comparison to Ingo Molnár’s Completely Fair Scheduler that has been making the news in the Linux world.

Current ICMP implementation in HAIKU kernel

Hi all,

According to current Haiku source ICMP implementation is only a … framework. Only ICMP echo request messages are processed.

There are essentially two RFCs that should be considered when implementing ICMP (for IPv4). RFC792 - ‘INTERNET CONTROL MESSAGE PROTCOL’ and RFC1122 - ‘Requirements for Internet Hosts – Communication Layers’.

According to RFC1122, paragraph 3.2.2. there are two types of ICMP messages a host must process:

1. ICMP Error Messages

• Destination Unreachable
• Redirect
• Source Quench
• Time Exceeded
• Parameter Problem

2. ICMP query messages:

• Echo
• Information
• Timestamp
• Address Mask

First of all, I plan to implement a generic icmp_send_data() that should be used whenever sending ICMP messages. Then implement processing of received ICMP packets accordingly.

UHCI isochronous support half done

Actually is more than half. This quick post is just to inform you that I wrote the part that schedule an isochronous request in the UHCI driver. I’ve already sent the patch to Michael for his review. The only part that is missing is the code that remove the request once it has been processed or canceled, which is not as trivial as I thought.

The Package format

Personal rant: my university examination session draws near and with it all credit tests as well. I’m doing my best in time management not to put any of my current tasks and projects into starvation, but exactly as Ryan wrote to me - it’s not easy.

Going back to more Haiku-specific topics, last week I was mostly analyzing the .pkg format Be Inc. left us behind. After some tests, most crucial parts of it are clear to me now. I must say most of their design solutions for this particular chosen concept had logical and technical grounds. To tell the truth, when I started hacking on this format I thought to myself: “Good god, who did this to you?”. I was wrong. Well… it’s a very interesting one nevertheless. I’ll try to throw a little light on the internals of this system. Keep in mind that all this information comes from Ryan’s and my personal investigations – in this stage of development, it might not be as accurate as everyone would like it to be.

Haiku Admin Meeting Summary - June 4, 2007

Haiku Admin Meeting 2007-06-04:

• There was some discussion about the status of several of the GSoC students/mentors now that GSoC has officially begun (May 28th).
• A couple of minor notes about some website changes were mentioned.

Haiku Admin Meeting Summary - May 28, 2007

Haiku Admin Meeting 2007-05-28:

• Discussion about new "Admin organizer" internal-use website for managing tasks, admin votes, etc.
• Some discussion took place about possibly organizing a physical "admin gathering" somewhere.
• Michael Lotz has volunteered, and is currently a primary candidate for "Project management" type tasks.
• Mention of the need to assign/complete the Haiku, Inc. 2006 financial publication.
• Discussed Haiku Compatible Badges:
It appears many are currently in favor of Stephan's designs. Mention of possible "Haiku powered" or "Haiku inside" on a badge. Should open-vote be used similar to the icon sets? Further discussion about decision will be moved to Admin mailing list. Currently collecting all the "entries" from the General Mailing List.
• Some WalterCon scheduling options discussed:
Discussed what types of content should be provided for attendees. Reminder of survey results. (Ed note: Survey should now be closed as of June 1, 2007) Note that most attendees could fall into "User" and "Application Developer" categories. Possibly an audio driver API presentation?
• Discussion about Haiku Compatibility List:
Should it remain as an external site - or try to move it to the official site? If hosted internally, should it still be primarily a community-driven effort? This would probably ensure it remains up-to-date. Karl should be contacted to discuss options.
• Discussed possibility of opening a "bounties" section on the official site:
Bounties should be reasonable. Could both Haiku-sponsored bounties and community-initiated bounties be supported? If yes, they should be clearly separated to avoid confusion about sponsorship.

Coding Style

As many of you know, I’ve started working even before the SoC started officially. I’ve already sent two patches to both my mentor (Oliver R. Dorantes) and Michael Lotz for review. One of them has already been commited by mmu_man (thanks). The second one is under review. With this latest one, the usb stack manager should be complete, as the QueueIsochronous method has been implemented, along with the CalculateBandwidth. My next move is to implement the UHCI isochronous method. Once I’ve done that, testing can be made. As for now, there seem to be a lack of drivers with which I can test the code. Oliver has offered himself to write some simple bluetooth driver just to test the code. Isochronous UHCI Tester are obviously welcome.

Today is May 28^_^

Hi all,

My name is JiSheng Zhang. I come from China. I am one of the Haiku’s GSOC2007 students. My work is implementing a FireWire stack with support for mass storage and DV cams. I plan to port the FreeBSD FireWire stack. I will do my development under debian.

After google anouncement on April 11, I spent time in building Haiku, running on the real PC and reading the source code of haiku’s PCI and SCSI sub system and other code about driver writing.

Haiku Admin Meeting Summary - May 21, 2007

Haiku Admin Meeting 2007-05-21:

• Admins discussed OpenBFS Team broadening - it has currently been renamed to File System Team and should probably include all file system-related aspects of Haiku.
• There was some further discussion about other OSS projects and their similar struggles with "Openness" - Mozilla's recent scrutiny was cited as an example.
• There was a request for any final comments on the list of non-dev/non-admin tasks posted on the Admin mailing list.
• It was discussed that the Admin mailing list needs to be relocated - and preferably self-hosted on the Haiku-os.org server.
• It was mentioned that once the current T-shirt supply is exhausted, they will be looking for a new store/shop to create the next round of T-shirts (and possibly other merchandise). The hunt is on for high-quality shops cheaper than Cafepress - with the company that Mozilla currently uses as a possible prospect.
• The Admins discussed adopting a new method for managing meeting agendas in the future.

Report on the API Documentation Team Meeting of May 19th 2007

This is a report of the meeting of the API documentation team that was held on May 18th 2007 at 19:00 GMT on freenode.org in the #haiku-doc channel. The full IRC log is hosted by John Drinkwater.

Here are some of the highlights for those that don't want to read the complete document. In this meeting people introduced themselves to the group. There was discussion on the working process, and it was decided to create a three phase working process and to start on the support kit. Communications will be done on the mailing list and on a wiki page. There were decisions on authors versus proofreaders and some small amendments to the existing API documentation guidelines were decided on.