State of hardware support in Linux
Submitted by srlinuxx on Sun, 11/30/2008 - 14:19.
I’m getting increasingly annoyed by the state of some aspects of hardware support in Linux.
My laptop (old Dell Latitude D610, Intel-based with ATI graphics) used to suspend/resume correctly with Linux 2.6.24 (ie, > 95% success rate). Later changes made the success rate drop significantly, and added a problem with kacpid taking 100% CPU because of an interrupt storm. And now, with 2.6.28-rc6, it’s completely broken. (partially documented in bug 11563, but I admit I gave up on this bug, because I’m going to change my laptop soon).
My desktop used to wake-on-lan correctly with 2.6.24 (but required some hacks, because it wouldn’t wake up if the NIC was DOWNed before shutdown), but changes in the r8169 driver broke it.
How many things are routinely broken during each kernel release cycle?
Two Linux technology specializations ? Bios based ? Novell ?
Two Linux technology not perfected are instant on and hibernation to save battery usage for notebooks.
Instant on has to use bios for splash screen on vga. Then click on icons to launch apps. The apps are in runtime codes previously installed from iso file in flash card. 3G modem on ARM cpu can also do instant on more easily.
Hibernation has to use the reusable file in bios to put cpu to sleep, then wake on peripheral data input. Application(cpu and display on LCD) switches on the sleep feature in bios even in smallest inactive time slices. The battery savings came from automatic power management codes in the bios which is sleep on 1 volt and wake to 5 volts for normal operation of the computer(note or netbooks). Battery life can be in a dashboard display, based on voltage of the battery.
Obviously, manufacturers of notebooks and netbooks know their bios, but open source community does not. Which is why manufacturers roll their own Linux operating system. Any distros such as Novell need to study bios codes to support HP and Dell on the above two Linux technology? If bios is not used for its reusable files, then compatibility can not be assured for each brand of netbooks.
We are stockholders of novl. Opinions here maybe biased.
Futility in preserving Linux integrity by free open source hacks
Moving from kernel 2.4 to 2.6 created 5 years of frustrations of bugs galore. At least in kernel 2.4.x, we have sweetspots at 1024x768 resolution in xvesa 791/792.
Kernel 2.6.x has not yet achieved its purpose of priority in distributed data synchronization. Unionfs and aufs are not used for database; rather people visioned them to be super synaptic on repositories of packages.
Anyway, Linux kernel versions are the source for dependencies. Packages must be fixed for missing bindings and blacklisting of hacked symlinks.
So, RPM has 2500 bugs with each kernel version change. Debian apt-get has 586 builds. Suse has 50,000 bugs. Many bugs are not fixable due to missing Linux legacy commands(reusable files).
For some of us, we are promoting stripped down distros to get rid of many apps not needed but maybe full of bugs. We also promote simplified hardware to eliminate driver bugs. But once stripped down, you run into missing reusable files; even more so, when you add packages.
But don't give up. We are now proposing stabilizing a kernel version for each new generation of netbooks, then add gui features on older version of apps to keep OS stable. This may do it fine as ASUS and Acer start to practice the restraint in open source coding(better no bugs than new kernels).