The Media Industry is once again geared to screw consumers over. They are pushing hardware/software makers to include Digital Rights Management (DRM) protection on DVD Players, future versions of Microsoft Operating Systems, CPUs, Motherboards and even Graphic Cards. DRM protection will restrict our use of media/content to what the Media Industry calls “Fair Use”.
Example 1: I can’t record a TV broadcast in HDTV and play it on a different computer. This is not piracy. I have multiple computers networked at home. I store all my content in the server and want to play it on any of my other computers.
Reading an article on Linux Gazette and reporting his own experience Jan Engelhardt asked on the lkml why reiserfs has the longest mount time among the filesystems that were tested. Jeffrey Mahoney explained:
ReiserFS caches bitmaps on mount and
for large file systems it can take a quite a long time. I've heard
reports of up to 15 minutes on multi-TB file systems
But there are patches that solve this issue.
O/S : Linux
Distro : kubuntu
Program : VMware
Title : Installing VMware on Dapper (Kernel 2.6.15)
Just got done taking care of a bit of an albatross. Notably, getting the "production" version of Oracle (10.2g) to install on RHAS 4 (update 2) for PPC. Anyway, here are the snags I hit and how to fix them.
This is about attempt # 69 to put up another blog entry. I keep starting them and never really finishing. So in my favorite, unordered list format, are quick updates:
I run a small linux server at home. Its a box I put together from scratch using better quality components hoping that will last a few years running 24x7. It's been running almost two years without a glitch. The main problem has been that, as my needs change, I need to add new things. My biggest need for the past nine months has been disk space. All I'll say is that I became a "torrentaddict" and the 144Gig of my two WD74 raptors disintegrated in my first few weeks.
Working on dynticks for i386 reminds me how every project in the kernel is 1000 times more intricate than it first appears. Put together your code and it looks fine but then you have to rewrite it twenty times or so for things you didn't anticipate. Then you chip away at the unexpected bugs one by one like trying to carve an ice cube out of an iceberg with a toothpick.
what I understood by learning about Linux VM.
What is LRU?
We have a cache which can accomodate limited number of members at a time. So our job is to admit members in the cache so as to increase the cache hit ratio. LRU is one of the effective caching policies. Let's consider the scenario of page level caching. We have a cache which is already full and a request comes for a page that is not present in the cache. Now to admit this new requested page, we have to throw away one of the cached guys. The candidate should be selected which has the least possibility of access request among all of them. LRU(Least Recently Used) suggests us to throw away the page whose last access was least recent. This strategy proves effective in most of the cases(obviously there are cases in which this algo totally fails.... can u tell me such cases?)
Started having kernel panics resulting from libata-core.c for some reason. It started when I upgraded to 2.6.13-gentoo-r3 (I think). Backed off to vanilla 2.6.12 and it seemed to be fine, but no fbsplash. Haven't tried backing off to earlier Gentoo kernels yet. Errors are totally intermittent. The system is rock-solid stable when running. Read some on Gentoo forums this may be caused by the Seagate SATA drive, but no idea if this is rumor, conjecture or what.
Just upgraded to dovecot 1.0alpha3 on the mail machine (which is an OpenBSD box). It seems weird that the author considers it *the* version to use, yet won't remove the alpha tag. You'd think it'd be a beta. Anyway, had a few growing pains in the config file (the new one doesn't care for some of the old config options) so I had to take the example config for the new dovecot and bring it up to speed. The biggest thing I found once it was up and running was with mailbox locking....I had to add the group that owned /var/mail to the mail_extra_groups line.
I'm installing Ubuntu on the Powerbook to see if it solves my sleep woes. I've narrowed down what's up (I think). The system will sleep and wake up once, and it will go to sleep a second time, but it will never wake up. The Flashing LED stops, but the screen never turns on, hard disk doesn't start running (that I can tell), etc. Ubuntu looks like it's got a decent amount of active development.
I'm trying to find out some decent algorithm to detect memory leaks in the code written for kernel space (never exits).
There are so many tools to detect memory leaks but none promise to detect leaks from code wriiten in kernel space i.e. code which never exists.
1. Probably the problem can only be solved by static analysis of the code but how can we simulate external dependencies (I/O or mem volatility)..?
Wouv .. this is my first blog!
Been reading away for days now .. the feeling of internet after six months without it :)
Brushing off my knowledge of linux at the moment
Reading Linux 2.4 internals right now and re-reading Linux Device Drivers
while trying to get up to speed with the latest kernel development.
was that 50 words ?
Now that I've done this....twice, I figure I might as well put this info on my blog so I (or maybe someone) else can retrieve it in case of problems in the future.
After screwing around with various OS's, I decided to settle on Yellowdog 4 for my PowerBook. Unfortunately, Yellowdog isn't what it used to be. YDL 3 installed very well and had few problems. YDL 4, on the other hand, does have a few problems when installed on this machine.