On Sun, 25 Sep 2005 14:29:47 +1000 David Seikel <[EMAIL PROTECTED]> babbled:
> On Sun, 25 Sep 2005 11:24:43 +0900 Carsten Haitzler (The Rasterman) > <[EMAIL PROTECTED]> wrote: > > > On Sat, 24 Sep 2005 03:10:12 +1000 David Seikel > > <[EMAIL PROTECTED]> babbled: > > > > > the background' tricks to X. At the moment, the bulk of getting > > > logged in graphically is spent in X before it starts the display > > > manager. > > > > actually - here's a hint. x actually loads EVERY module when it starts > > to find out whats there - that means every driver, every extension > > file etc. it may only end up using a few but it literally does a heap > > of disk work on them anyway (for interest - start x and put it under > > strace - log it and check the output and see all the disk io it does > > for modules etc.). > > This is why I am waiting for the Xorg modular tree to come out (soon, > soon, Real Soon Now they say). Once I have that I can start to classify > the modules into optional, needed, and needed later categories. It's a > lot more work doing that to the current monolithic X. Once classified, > I will then know what to start now, what to start in parallel after > starting entranced, what to leave out completely, and what to hack for > speed. > > > hmm - i still think you have some gross disk IO issues. i may have > > only a fraction - but e starts in like < 1 second for me (200 not > > 1600, but i dont get like a 20 / 8 (2.something) second extra lag. > > what fs are u using? ext2? ext3? xfs? reiser? what disk? whats its IO > > capacity? (hpardm -t /dev/device) ? > > ext3. As part of my "what can boot faster" work I did some timing > tests. Others where faster at writing, but booting is more read bound, > and ext3 won the boot race. Xfs was a close second, and I would > recommend it for multimedia work, since it gave more consistent times, > and it was designed for multimedia work. Reiser and JFS where both much > slower, but again there was not much difference between the two. I have > not tried Reiser 4 yet. I also only tried journalling file systems, so > ext2 was not part of the tests. > > cluster:/home/dvs1 # hdparm -t /dev/hda > > /dev/hda: > Timing buffered disk reads: 148 MB in 3.01 seconds = 49.19 MB/sec ok- thats fast enough. weird that it takes 20 seconds to load those files. weird. anyway - cache is now done. i have dne some levels of testing and it seems to work and update etc. feel free to try it out. it's in cvs now. > Are you reading in all of the eap at start up? For some strange reason no- only the header whihc will be a few hundred bytes at most, and then selected pieces of data from the file (it seeks to them and loads what it needS) in reality the data loaded is only like < 1kb of actual data, but with buffered file opes it may end up loading in a good 4 or 8k or more of the file - times each file and lots of seeks. > most of the e17genmenu generated eaps are around 92KB. On the other > hand, I put off a new HD purchase to buy my wireless modem, so space is > tight. Thousands of eaps may have got scattered all over the disk. E17 > startup with thousands of eaps may be seek bound on my computer. thats what i'm thinking. i have gobs of free space and i dont even notice it. :) my disk only pulls about 40mb/sec here on my desktop. > > btw - partt way through. my 200 eap's use about 6mb of disk. the cache > > file for that dir is a touch under 7kb. thats all e has to load on > > startup - so it should be massively faster for you in the future. i'm > > still workign on it and its not enabled in production code yet. :) > > Let me know when I can try it out. now - well wait 6-12hrs till it hits anon cvs :) > > > Oh well, back to struggling to get a cvs update of e17 out of > > > sourceforge. Man that can be painful some times. Yesterday I did a > > > cvs update out of sourceforge for ONE OF MY OWN PROJECTS, using my > > > account and everything. Took five hours for something that should > > > have taken minutes. Hmm, got e17, but can't get the misc module > > > yet. > > > > i havent seen anything that bad yet. are you pn dialup? :) > > Nope, I'm on something called iBurst, which is a new technology that may ooh! that one - yeah - the sydney wireless boardband - they bought up a bunch of 3g licenses unused for cheap and built a data network for internet access around it :) i know of it :) > be called something else if it exists at all in your country. It's > propietry wireless broadband that is more like mobile phones than > current WiFi. I don't have to worry about finding a hot spot, as it > works anywhere in the city. Australia is a small continent with most of > the population hugging the coast, and very large expanses of nothing in > the middle. Things like mobile phones and iBurst are not likely to work > 50 kms away from the coast. iBurst is new, so it doesn't yet have the > same coverage as mobiles, but they are working on it. On the plus side, > an iBurst tower has greater range than a GSM tower, so they will need > less of them. Getting decent telecommunications outside the main cities > is a big political bun fight at the moment. indeed :) > My Internet access is the equivilent of ADSL, but without the telephone that depends whihc adsl. here in japan standard adsl is about 10-50mbps. mine's 45mbps for exmaple and its only like $au30/month. but in .au - iburst is comaprable to current adsl offerings for sure :) > line, or any other wires. It even works while on the move, but since I > have a desktop, I would still need to plug into a power socket. Laptop > users can use it while driving on the freeways, and I could probably use > it on the City Cat ferries, catamerans that are part of the public > transport system in this city. If only they would let me use their > power. B-( > > At any rate, the really slow cvs was not due to limited bandwidth at my > end, as it spent most of that five hours just siting there doing > nothing. The e17 network monitor said "0B" for most of that time, > unless I was doing other net stuff. SourceForge sometimes has major > performance problems, and when they do they often give priority to > account holders over anonymous cvs, and they may even give priority to > paid subscribers, I wouldn't know, I'm not a paid subscriber. I don't > have cvs access to enlightenment from my SourceForge account, so when > updating e17 I get the anonymous priority. But I should get better for > my own projects, where I can get account priority. weird. totally weird. maybe the server your project is on was having progems. dev cvs server are split based on project name so different proejcts are on different cvs dev servers. > -- > Stuff I have no control over could be added after this line. > > Send instant messages to your online friends http://au.messenger.yahoo.com > > > > ------------------------------------------------------- > SF.Net email is sponsored by: > Tame your development challenges with Apache's Geronimo App Server. Download > it for free - -and be entered to win a 42" plasma tv or your very own > Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php > _______________________________________________ > enlightenment-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) [EMAIL PROTECTED] 裸好多 [EMAIL PROTECTED] Tokyo, Japan (東京 日本) ------------------------------------------------------- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42" plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
