Category Archives: Open Source

Open Source, Free Software, and similar

C Code Commandments

I’m an old school C programmers guy and I stay true to some of the older and commonly used rules present in many open source and similar projects. Since I sometimes rant about this to people, I thought I’d amuse my surrounding by stating them here for public use/ridicule. Of course heavily inspired by the great and superior The Ten Commandments for C Programmers. My commandments are not necessarily in any prio order.

Thy Code Shall Be Narrow

Only in very rare situations should code be allowed to be wider than 80 columns. I want my two or three windows next to each other horizontally and still see the code fine. Not to mention the occasional loading up in an editor in a 80 columns terminal and that is should be possibly print nicely (for reviews etc). Wide code is also harder to read I think, quite similarly to how very wide texts in web pages etc aren’t kind to your eyes either.

Thou Shall Not Use Long Symbol Names

To be able to keep the code easily readable by human eyes so that you quickly get an overview and understand things, you simply need to keep the function and variable names fairly short. Not to mention that the code gets harder to keep within 80 columns if you use ridiculously long names.

Comments Shall Be Plenty

Yes, this is something we know everyone says and few live up to. In statistical analyzes of my own C code I usually reach around 25-27% comments and I’m usually happy with that amount. Comments should explain what is otherwise not obvious in the code.

No Hiding What’s Really Happening

I’m not a fan of overloaded operators or snazzy macros that do fancy stuff without it being noticeable in the code. It should be clear when reading the code what it does. That’s also one of the reasons you don’t catch me doing a lot of C++ work…

Thou Shalt Hunt Down and Kill Compiler Warnings

Compiler warnings may be significant and in some cases they are not. Either way, it is our duty to silence them at all times. Firstly because it is often simpler to fix the code to not warn than to figure out if the warning is indeed right or not, but perhaps primarily because it makes it harder to see new warnings appearing if the old ones have been left there.

Write Portable Code Unless Forced by Evil

You may first believe that your code will live on forever on this single platform with this single compiler, but soon and very soon you will learn otherwise. Then you will cheer this rule as it makes you consider unaligned memory accesses, assuming byte-order of binary data or the size of your ‘long’ variable type.

Repeat Not, Use Functions

I see a lot of “copy and paste” programming in my daily life and I’ve learned that sooner or later such practices lead to sorrow. If you paste the same code on multiple places it not only makes it repetitive and boring to update it when an API or something changes, more seriously it increases the risk that you address bugs only on one out of many places or that the fix differ etc. It also makes the code larger and thus harder to follow and understand.

Thou Shalt Not Typedef Away Pointers

A really nasty habit to be seen in some source codes is when people use typedefs to define their own types that is simply a pointer to something. Like with ‘typedef struct whatever * whatever_t’. While I’m in general against excessive typedefing, I’m fine with them in many cases but not when used to hide pointers to look like “ordinary” types. It makes code harder to follow.

Defines, no fixed numbers

Code that relies on zero and non-zero can get away without this, but as soon as you start relying on more numbers in the code you must start using #defines or possibly enums to make them appear with names in the code. Using names is more clever than hardcoded numbers since you can avoid having to explain the number in a comment, and of course it’ll be easier to change the actual number in the code at a later point without it being a painful search-and-replace operation.

libssh2 1.1

I’m happy to announce that we now have a version 1.1 of libssh2 released! Noticeable changes this time include:

  • Downloads using SCP or SFTP are now significantly faster
  • Added a Libtool -export-symbols-regex flag to reduce the number of exported symbols in shared libraries.
  • Added a bunch of new man pages and renamed some of the previous ones
  • Enhanced download performance
  • Made libssh2_scp_recv() and libssh2_scp_send() deal with spaces in filenames
  • Fixed the bad randomness and off-by-one in libssh2_channel_x11_req_ex()
  • Added libssh2_version()
  • Fixed libssh2_channel_direct_tcpip_ex() to not fail when called a second time
  • Fixed libssh2_channel_write_ex problems in blocking situations
  • ‘make check’ runs fine on cygwin
  • Added libssh2_channel_receive_window_adjust2() and deprecated  libssh2_channel_receive_window_adjust()
  • better socket error handling internally on win32
  • libssh2 now always set the socket non-blocking internally and deals with the interface as blocking or non-blocking set by libssh2_session_set_blocking.

The library is rapidly maturing and is getting really usable. I’m happy to see that there’s a community slowly building up around this and I’m also grateful for my sponsor paying for parts of the fixes that contribute to make this release the best ever in libssh2’s history.

libssh2

libssh2.haxx.se

libssh2I’ve played around with a possible new design for the libssh2 web site and I’ve put it up on libssh2.haxx.se for everyone to play around with and comment (on the libssh2-devel list please).

The original and actual home page for the project is still over at www.libssh2.org but I’m not happy with that because of a few things:

The wiki duplicates info that we these days write in man pages and hardly anyone updates the wiki so it lags behind or just contains false or outdated info. I also think having the entire site a wiki a bit problematic for things like menus and generic site layout etc.

I want daily snapshots, web versions of the man pages and mailing list archives to be somewhat integrated in the site to be easier to find.

I already hosted libssh2-related stuff anyway so I’ve just packaged it in a slightly more friendly way. In my view.

What do you think?

HD is the thing

Thomson apparently brought the new mp3hd format for music the other day. “HD” is apparently the thing we need to have included when a new term is announced. Why does the world need another lossless music format?

It seems they’ve introduced a crafty dual-format thing where they stuff MPEG-4 SLS lossless encoded data in a new id3 XHD3 tag within the mp3 and then stuff a “regular” mp3 as the normal data in there. This way it is supposed to still work fine with existing and older mp3 players. Of course the total size of all id3v2 tags is limited to 256MB, which could be a limiting factor for it.

As usual, you can find a thriving discussion on this topic on hydrogenaudio.

Rockbox should of course be possible to at first use the mp3 parts and if this truly is an existing established lossless codec there’s a chance it might be able to play that part in the future.

Rockbox 3.2

The never-ending flow of creativity in the Rockbox project was today turned into a release that we label 3.2. The goodies this time include the things below. The three-months release cycle does prevent the list from growing terribly big…

  • Faster text/graphics rendering on colour targets and in the greyscale library, speeding up list scrolling noticeably on ipod Video.
  • PictureFlow supports all targets except Archos Player, and can function during playback on all non-Archos targets.
  • Add LCD sleep/wakeup for iPod Video (5G, 5.5G) which allows significant increase of battery runtime.
  • New game, Goban plugin.
  • Battery charging on Sansa e200v1/c200v1.
  • PictureFlow resizes cover art on load, and supports greyscale targets.
  • Preliminary support for Ipod accessories.

What didn’t get included:

  • The ‘natsort’ which sorts files with numbers as the number and not by ascii. This caused quite a lot of discussions and will be sorted out for 3.3
  • The Rockbox USB stack. It has been enabled in SVN build for several weeks already, but due to it causing some pretty drastic problems to some users we decided to play it safe and disable it in the release. We really hope it’ll be fine for 3.3.
  • Support for any new targets. The Gigabeat S, the Ondas and the AMS sansas aren’t terribly far away, but still not “there”.

A more detailed list can be found in the Release notes for 3.2.

Rockbox

Code re-use is fun

Back in 2003 I wrote up support for the HTTP NTLM authentication method for libcurl. Happy with my achievement, I later that year donated a GPL licensed version of my code to the Wget project (which also was my first contact with the signed paper stuff with the GNU/FSF to waive my copyright claims and instead hand them over). What was perhaps not so amusing with this code was when both curl and Wget 2005 were discovered to have the same security flaw due to my mistakes in this code shared by both projects!

Just recently, the neon project seems to be interested in taking on the version I adjusted somewhat for them, so possibly the third HTTP code is soon using this. Yeah I posted it on their mailing list back then so it has been sitting there in the archives maturing for some 6 years by now…

I also happened to fall over the SSH Tunnel Creator tool, which I’ve never used myself, that apparently snatched my neon donation (quite according to what the license allowed of course) and used it in their tool to do NTLM!

It’s actually not until recent years I discovered libntlm, and while I don’t know how good it was back in the days when I wrote my first NTLM stuff I generally think using existing libs is the better idea…

Rockbox in gsoc 2009

Gsoc 2009

For the third year in a row, the Rockbox project participates in Google’s Summer of Code and has just now officially been accepted as a mentor organization.

This is a great chance for every student or others who have been looking for a good chance to join in an open source project, and in particular if you’re interested in embedded systems developing for (relatively) low-resource devices.

The Rockbox project also has a very large developer base, a very large community and a healthy and friendly surrounding that is welcoming to newbies.

If you’re even just the slightest intrigued, please don’t hesitate to read further on the Rockbox gsoc 2009 page.

The two previous years we’ve had four students each year with 15-20 applications.

Let’s make this a great gsoc year!

libssh2 upped a notch

There have been some well-founded criticism against libssh2 for a long time for its bad transfer performance when doing SCP and SFTP based transfers. Tests have proved it to be significantly slower than the openssh based alternatives in comparisons done in similar conditions. We’re talking down to a tenth(!) of the speed for SFTP.

Luckily I have a unnamed (by agreement) sponsor who pays me for improving this.

Giving it some love

I basically started out reading the SFTP code and cleaned it up as I went over it, and I added some clarifying comments etc. I found some irregularities that I fixed. Soon I could spot an obvious performance boost, like perhaps 3-4 times the previous speed. But since SFTP was painfully slow originally, this was still very crappy compared to openssh.

I then switched over to plain SCP tests. SCP is basically just an “scp” command sent over SSH and then streaming the data over a plain SSH “channel”, while SFTP is a whole additional protocol layer on top. Thus SCP is more low-level, on the actual SSH level, and the foundation on which SFTP runs anyway so getting SCP faster was fundamental.

Make it speedier

My initial tests with libssh2 1.0 showed libssh2 to download data at roughly 25% of the speed of openssh when SCPing a 1GB file from an openssh server running on localhost. The openssh client shows roughly 40MB/sec on my test box.

Also, just checking my CPU load meter while doing the libssh2 transfers showed that it certainly wasn’t hitting the roof or anything. It was barely even noticeable! Of course something was really wrong but what was it?

SSH has a lower protocol layer that does the entire encryption thing, the transport layer, but on top of that is the “channel layer” that is packet based for sending data back and forth over the transport layer. This channel thing has a receive window concept, much like TCP itself has, which tells the remote side how much data it is allowed to send until it gets further notice.

libssh2 1.0 had a very conservative windowing logic. It started with a default window size of 64KB and it upped it at every read with the same amount that was read (which then could be 1K to 16KB something depending on the app).

My remake of this was to simplify the logic, read data from the network more evenly distributed over time, update the window size much less frequent and increase the window size by magnitudes! I found that when using a window size of 38MB (600 times the previous default size!!) things started flying.

Improved

With these modifications, libssh2 transfers SCP at close to 40MB/sec! SFTP is still left behind at a “mere” 14MB/sec on the same test setup but it has its own set of problems and solutions. Now this discussion on the libssh2 list is more about how to sensibly size the window to work the best way for different situations.

SFTP is a protocol that works more on file operations. The client sends OPEN, READ and CLOSE requests and the server replies with status and data. The READ request asks for N bytes starting at offset Z so a simple implementation like libssh2 asks for chunk after chunk in a serial manner, increasing the offset as it loops over the range. This causes a back-and-forth effect that certainly does not make optimized use of the network bandwidth.

SFTP ping pong

openssh has a nifty approach to enhance throughput for SFTP: it sends off and handles multiple outstanding READ requests in parallel so that it better can keep things busy (and the reverse when doing uploads). That concept is slightly harder to do with an API like the one libssh2 offers but it is of course still quite doable. I suspect that we might achieve results somewhat faster by simply use multiple connections as then we can remain using this simplistic approach but still use the full bandwidth. (Yes, I realize multiple connections may not be feasible for all applications.)

Previous tests we’ve done with SFTP uploads using multiple connections have proven libssh2 to be on par or even better than competitors on both Windows and Mac.

Please test

I’ll leave it like this for now. I’ll be very happy if people could test this version and report findings so that we make sure this is working and stable enough to release soonish. We’ll need to do something that offers window size controlling to apps, but we’ll discuss that further on the mailing list. Join in!

logo1-250