Older blog entries for rodrigo (starting at number 57)

D-Bus optimizations

In the last month and a half, I have been working, as part of my work at Collabora, on optimizing D-Bus, which even though is a great piece of software, has some performance problems that affect its further adoption (specially on embedded devices).

Fortunately, we didn’t have to start from scratch, since this has been an ongoing project at Collabora, where previous research and upstream discussions had been taking place.

Based on this great work (by Alban Créquy and Ian Molton, BTW), we started our work, looking first at the possible solutions for the biggest problems (context switches, as all traffic in the bus goes through the D-Bus daemon, as well as multiple copies of messages in their trip from one peer, via the kernel, then to the daemon, to end up in the peer the message is targeted to), which were:

  • AF_DBUS work from Alban/Ian: while it improved the performance of the bus by a big margin, the solution wasn’t very well accepted in the upstream kernel mailing list, as it involved having lots of D-Bus-specific code in the kernel (all the routing).
  • Shared memory: this has no proof-of-concept code to look at, but was a (maybe) good idea, as it would mean peers in the bus would use shared memory segments to send messages to each other. But this would mean mostly a rewrite of most of the current D-Bus code, so maybe an option for the future, but not for the short term.
  • Using some sort of multicast IPC that would allow peers in the bus to send messages to each other without having all messages go through the daemon, which, as found out by several performance tests, is the biggest bottleneck in current D-Bus performance. We had a look at different options, one of them being AF_NETCAST, which mostly provides all that is needed, although it has some limitations, the biggest one being that it drops packets when the receiver queue is full, which is not an option for the D-Bus case.
    UDP/IP multicast has been mentioned also in some of the discussions, but this seems to be too much overhead for the D-Bus use, as we would have to use eth0 or similar, as multicast on loopback device doesn’t exist (hence no D-Bus in computers without a network card). Also, losing packets is another caveat of this solution, as well as message order guarantee.

So, the solution we have come up with is to implement multicast on UNIX sockets, and make it support what we need for it in D-Bus, and, of course, make use of that in the D-Bus implementation itself. So, here’s what we have right now (please note that this is still a work in progress):

The way this works is better seen on a diagram, so here it is. First, how the current D-Bus architecture works:

and how this would be changed:

That is, when a peer wants to join a bus, it would connect to the daemon (exactly as it does today), authenticate, and, once the daemon knows the peer is authenticated, it would join the accept‘ed socket to the multicast group (this is important, as we don’t want to have peers join by themselves the multicast group, so it’s the daemon’s job to do that). Once the peer has joined the multicast group, it would use socket filters to determine what traffic it wants to receive, so that it only gets, from the kernel, the messages it really is interested in. The daemon would do the same, just setting its filters so that it only gets traffic to the bus itself (org.freedesktop.DBus well-known name).

In this multicast solution, we might have to prevent unauthorized eavesdropping, even though peers need to authenticate through the daemon to join the multicast group. For this, we have been thinking about using Linux Security Modules. It is still not 100% clear how this would be done, so more information on this soon.

The above-mentioned branches work right now, but as I said before, they are still a work in progress, so they still need several things before we can call this work finalized. For now, we have succeeded in making the daemon not get any traffic at all apart from what it really needs to get, so a big win there already as we are avoiding the expensive context switches, but the socket filters still need a lot of work, apart from other minor and not so minor things.

Right now, we are in the process of getting the kernel part accepted, which is in progress, and to finish the D-Bus branch to be in an upstreamable form. Apart from that, we will provide patches for all the D-Bus bindings we know about (GLib, QtDBus, python, etc).

Comments/suggestions/ideas welcome.

Syndicated 2012-02-27 12:02:42 from Rodrigo Moya

New beginning

I guess it is time to announce that since yesterday I am working at Collabora, a UK-based company very well known for its work in several free software projects, like Telepathy, Farstream, GStreamer and others.

Haven’t had much time really to transition (and relax) from Canonical to Collabora, apart from last week, which I spent skiing, but hey, new year, new life, as we say in Spain, so the sooner you start with your new life, the better.

Syndicated 2012-01-03 15:17:10 from Rodrigo Moya

Leaving Canonical

Today marks the beginning of my last week at Canonical, where I’ve been working for the last 2.5 years. Because of the conflicts between the direction the company is driving to and my personal interests (GNOME), I have decided it is time for me to move on.

Since I am a positive person, I would just remember the good things of these 2.5 years, which have been, mainly, the nice people I’ve been working with, with a special mention to the Ubuntu Desktop team, composed of very great people. Also, some good projects I’ve worked on, like the Ubuntu One music store or the work at the Desktop team.

I can’t say yet publically where I’ll be working next, but I’ll continue being around GNOME.

Syndicated 2011-12-19 13:28:35 from Rodrigo Moya

Fix PDFs hack

I recently bought a new ebook reader (Wolder miBuk ALFA 7.0 Color) because my previous one was very bad at reading comics. It looked really great in the shop, but as soon as I copied my entired e-book collection to a memory card and inserted it on the reader, I found its 1st problem: it doesn’t have the option to display the books collection by file name, but it gets the PDF metadata and uses that. So, since lots of my books didn’t have correct metadata, it was very hard to find books in the library view.

But thanks to the help of Carlos García Campos (famous Evince/poppler hacker), I cooked up a patch for Poppler to add API to be able to set the metadata, and, right after that, wrote a very simple GTK program to allow me to “fix” my ebook collection.

The Poppler patch is still not ready to be pushed upstream (my fault, lack of time in the last couple of weeks, but will fix it soon), but posting this now just in case it is useful for someone.

Syndicated 2011-05-06 12:18:08 from Rodrigo Moya

Unofficial GNOME3 on Ubuntu PPA

A friend of mine was having problems with the GNOME3 packages in Ubuntu, and after some questioning, he told me he was using a PPA from this Launchpad team:

https://launchpad.net/ubuntugnome

The GNOME3 PPA for that team seems to be just a copy of the official GNOME3 PPA, but just in case, this is a public announcement to let people know that they shouldn’t use that PPA (unless they really want to, of course), but use the official one instead, which is at:

https://launchpad.net/~gnome3-team/+archive/gnome3

That is, the official team is the gnome3-team, so please make sure to check your sources.list if you really want to use the official one.

Syndicated 2011-04-11 11:56:34 from Rodrigo Moya

Internet hoaxes

As the number of my computer-illiterate friends that get an email address grows and grows, the number of mails containing hoaxes that I receive from them increases every day (things like “please forward this mail or the child would die”, “this music group helps financing a terrorist group”, “Mars will be as big in the sky as the moon”, etc). So, yesterday I got one about a restaurant charging 250€ instead of 2.50€ for the recipe of some cookies, giving the name of a real restaurant in Spain. Yolanda did a quick search for that restaurant and found a forum where people were complaining about that, and where one (clever) person pointed everyone to a page explaining the same hoax (word by word) for some restaurant in the US.

So yeah, a typical Internet hoax, you would say, but if I’m blogging about it is because I wondered yesterday what the purpose of these hoaxes is. Is it really just making fun of people? sociological studies? or using this for a revenge against a restaurant/shop/etc? There are clear cases, where you are asked to keep all people in the CC when answering, which seem, to me, related to getting email addresses for spammers, but all these hoaxes where people are just asked to forward the mail to their friends, what’s the purpose of them?

Please ask quick, as I couldn’t sleep last night because of this existential doubt :-D Another thing for further study would be how is it that so many people believe those hoaxes, but I’ll leave that for another time…

Syndicated 2011-03-24 11:01:28 from Rodrigo Moya

GNOME3 on Ubuntu

I already blogged about this some time ago but since some people keep asking, I’d thought about giving it more publicity.

So, in case you don’t know, next Ubuntu version won’t ship GNOME 3, but we have been working in the last few months on providing GNOME 3 packages for anyone interested in running GNOME 3 on Ubuntu. The packages are in the GNOME 3 PPA, and although it still doesn’t include everything GNOME 3ish, it includes the stuff that has changed the most, like the new control center, gnome-shell and other core desktop things and some applications. Thanks to Allan Day, here are some instructions on how to use a PPA.

It still misses lots of apps and some core desktop things, like gnome-session, but should be ready for daily usage (using it myself on my systems).

You can report any problem you find on the PPA via the GNOME 3 team mailing list or directly to me, as you like.

Syndicated 2011-03-03 13:16:26 from Rodrigo Moya

“GNOME 3″ on Ubuntu

With the great work from Robert Ancell and Sebastien Bacher, who worked on packaging the new GLib/GTK3 stack, and with the recent packaging of a few GNOME 3 applications (eog, Nautilus, the new control center, …), you can start testing what will be GNOME 3 on Ubuntu (Natty) by using this PPA.

Please note that this is a work very much in progress, which means that, apart from the usual problems of running unstable software, it’s got the unstability of new packages added, so please USE WITH CARE. I would suggest to use a virtual machine for testing this, but please test it and report any problems you might find. It seems to be running ok for me (on a virtual machine), but please don’t risk your every day desktop :-D

Syndicated 2010-11-12 11:15:45 from Rodrigo Moya

Ubuntu One SyncDaemon API

Ubuntu One’s SyncDaemon (the process that takes care of synchronizing your files between the cloud and the desktop) has a DBus interface for applications to control and get notifications from it about what it is doing. This interface was being used in the Nautilus plugin and other desktop tools contained in ubuntuone-client itself. Even though powerful and straightforward, we haven’t seen many applications using it, since usually 3rd party applications don’t want to be dealing with the low level DBus API themselves.

But this is now history, since I introduce you to libsyncdaemon, a GObject wrapper on top of the DBus API which makes it very easy to communicate with SyncDaemon, as well as improving a lot the performance compared to accessing the DBus API directly (thanks to its use of caches, to avoid DBus calls repeating when no needed) and containing high level functions that would allow applications to, for instance, publish a file on Ubuntu One with just one call:

SyncdaemonDaemon *daemon = syncdaemon_daemon_new ();
syncdaemon_daemon_publish_file (daemon, "/home/user/myphoto.jpg");

But before you run to another place after seeing this C code, note that it includes bindings for many languages (Python, JavaScript, etc) for free, thanks to gobject-introspection. So, the same code in Python, for instance, would be:

daemon = Syncdaemon.Daemon ()
daemon.publish_file ("/home/user/myphoto.jpg")

So now, the next step is to start adding support for sharing/publishing files in Ubuntu One to many desktop applications, like, for instance (ideas stolen from Stuart Langridge):

  • Publish screenshots directly from gnome-screenshot tool
  • Sharing photos from f-spot/shotwell/etc
  • etc…

This, along with the already existing desktop APIs (desktopcouch, couchdb-glib, libubuntuone), makes integration of 3rd party applications into Ubuntu One a very easy thing.

Syndicated 2010-07-06 12:04:33 from Rodrigo Moya

Facebook account disabled

Last night, I was writing a message to a friend’s wall on Facebook, about telescopes, and when submitting the message, I got a message saying that the content of my message had been reported to be abusive by some Facebook users ?? After that, I couldn’t log in back again, it just said my account was disabled.

So, if you rely on Facebook for contacting me, please don’t, send me mail, which is still enabled, and even allows all sorts of abusive content (you can write me about telescopes if you want, no ultra-paranoid filters there :-D ).

I already wrote to the Facebook staff about the issue, but I’m not sure what they will decide.

Syndicated 2010-03-26 12:52:42 from Rodrigo Moya

48 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!