On Wed, Sep 17, 2008 at 03:11:06PM +0300, Sami Liedes wrote: > On Wed, Sep 17, 2008 at 03:34:00AM +0200, Arno Wagner wrote: > > I have about 45% CPU on a very low end AMD Sempron(tm) > > Processor LE-1250 for 30MB/s. You probably should have > > bought faster cores instead of more. > > > > Incidentially, I believe your request has about zero prospect > > of being sucessful. It is a lot of effort for basically > > very few people having any gain. I would not do it unless fully > > paid, but thet _wpuld_ be expensive. > > Hmm, ok, thanks for your point of view. Didn't realize it's that much > work, although I do know that threads do have a very bad habit of > complicating everything (after all I'm a developer myself, albeit with > only rather limited experience of kernel hacking (and none for the > last six years)). And I hadn't thought of the solution of raid on top > of dm-crypt before :) > > So perhaps it's just best to forget it. Or alternatively to start a > project on micropledge and see how many want to pay (no, I'm not > overly optimistic, but you never know) and if there will be any > takers. From what you said, I gather the developer would need to have > quite an amount of familiarity with dm-crypt to have any real chance > of success? I do realize that would limit the number of interested > developers quite dramatically. Actually I expect, this might need a major filesystem layer rewrite, Maybe even a major kernel rewrite. > Still, I'd like one more opinion from you: If you think it shouldn't > be done from a dm-crypt developer point of view because it makes the Noooo, I am not a dm-crypt developer ;-) The most I have contribuited so far is a tony bit of informal security review quite some time ago. > code a mess for little gain, I'll give a lot of weight to that opinion > and just forget the idea, because I realize you are the dm-crypt > expert here and I genuinely believe you know better what's good for > dm-crypt (and since from a more pragmatic POV I guess that would > reduce the chance of patches being merged). The raid-on-crypt idea > probably solves most of the problem anyway, I really need to try that. Can't believe I did not think of that. With a bit of luck, you will effectively get one CPU per disk. > Thanks for this discussion, and I'm sorry for already having taken > more of your time than I probably should have if this has close to > zero chances of success :-) Hey, no problem. Arno -- Arno Wagner, Dr. sc. techn., Dipl. Inform., CISSP -- Email: arno@xxxxxxxxxxx GnuPG: ID: 1E25338F FP: 0C30 5782 9D93 F785 E79C 0296 797F 6B50 1E25 338F ---- Cuddly UI's are the manifestation of wishful thinking. -- Dylan Evans If it's in the news, don't worry about it. The very definition of "news" is "something that hardly ever happens." -- Bruce Schneier --------------------------------------------------------------------- dm-crypt mailing list - http://www.saout.de/misc/dm-crypt/ To unsubscribe, e-mail: dm-crypt-unsubscribe@xxxxxxxx For additional commands, e-mail: dm-crypt-help@xxxxxxxx
