Re: idea for running out of RAM
On Mon, 29 Sep 2008, Hal Murray wrote: One would adjust the shading in each activities portion of the circle. White for not used and black for used. (Red for over allocation?) The other would be to use the applications chunk of the circle as a pie shaped bar graph. The black section would show what % of the allocated memory was currently in use. Maybe a ring with shared memory protruding inside and allocated but not-paged-in memory protruding outside? -- Alex ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Stability and Memory Pressure in 8.2
On Tue, Sep 30, 2008 at 4:13 AM, S Page [EMAIL PROTECTED] wrote: On September 8th, Michael Stone wrote: Kim, Greg, and I have concluded that the instability we experience under memory-pressure in 8.2-759 and similar is the single hard issue that we wish to _attempt_ to address before releasing 8.2 on current timeframes. How did it go? I was going through my journal in 8.2-763. Browse and Paint open, accidentally started Read, suddenly the cursor stopped moving and XO completely unresponsive. I assume it's memory, but we never learned how to tell. Over two minutes later the first page of the PDF appeared and *then* immediately Sugar restarted. Just one datapoint. Thanks, Read has serious memory problems because renders whole pages into memory, regardless of what is the viewed area. Any chance the first pages of the PDF you opened contained big images? Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Signed candidate-765 and gg-765-2 builds available for testing.
So you are going to put those on a USB flash drive and run sudo olpc-update -usb from build 711 (release 8.1.2)? Yes! - the update process complete successfully. - at reboot the firmware got updated - at the start of the sugar session been asked to check for update activities (I had to say no because I'm not connected to network) - at first glance the system seems ok (username color scheme preserved) - my locale (it) get preserved - I open control panel (the widget layout is weird ): - Information of my XO widget says: Build 765 Sugar 8.2.1 Firmware Q2E18 - Browse starts - Journal get preserved more on it later... ._) please let me know if you need some specific test ! ciao e grazie Carlo Falciola - Messaggio originale - Da: S Page [EMAIL PROTECTED] A: Carlo Falciola [EMAIL PROTECTED] Inviato: Martedì 30 settembre 2008, 11:28:15 Oggetto: Re: Signed candidate-765 and gg-765-2 builds available for testing. Carlo Falciola wrote: I just need to re-download and I'll let you know asap (did you already update the link in the wiki-home box, did you?) Yes, I updated the wiki. sudo olpc-update candidate-765 from 8.2-763 worked fine for me. I'm getting os765.usb .toc from http://download.laptop.org/xo-1/os/candidate/765/jffs2/ . So you are going to put those on a USB flash drive and run sudo olpc-update -usb from build 711 (release 8.1.2)? Please let me know if you have any suggestions or problems with the steps in http://wiki.laptop.org/go/Olpc-update Thanks, -- =S Page (just someone trying to improve the wiki) Scopri il blog di Yahoo! Mail: Trucchi, novità e la tua opinione. http://www.ymailblogit.com/blog ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: New release8.2 build 764
C. Scott Ananian [EMAIL PROTECTED] writes: Do you mean the activities listed here: http://wiki.laptop.org/go/Activities/G1G1/8.2 The /8.2 version of the page are the 'stable' versions. When we get into 9.1 development, the [[Activities/G1G1]] page will contain the latest 9.1 or development versions, while [[Activities/G1G1/8.2]] will only be updated monthly or so, with those activity versions which are expected to work well on 8.2. At the moment, they should be identical. All right, thanks. I was in the process of updating my scripts to make the USB-ready zip files contain the G1G1 activities. To including activities in the stable and joyride zips, I respectively use these two pages: stable: http://mock.laptop.org/repos/local.update1/XOS/index.html joyride: http://xs-dev.laptop.org/~cscott/repos/joyride But it appears that the Help activity (for example) is not listed in these pages. What URL should I use? Thanks again, -- Bastien ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: New release8.2 build 764
On Tue, Sep 30, 2008 at 6:57 AM, Bastien [EMAIL PROTECTED] wrote: C. Scott Ananian [EMAIL PROTECTED] writes: Do you mean the activities listed here: http://wiki.laptop.org/go/Activities/G1G1/8.2 The /8.2 version of the page are the 'stable' versions. When we get into 9.1 development, the [[Activities/G1G1]] page will contain the latest 9.1 or development versions, while [[Activities/G1G1/8.2]] will only be updated monthly or so, with those activity versions which are expected to work well on 8.2. At the moment, they should be identical. All right, thanks. I was in the process of updating my scripts to make the USB-ready zip files contain the G1G1 activities. To including activities in the stable and joyride zips, I respectively use these two pages: stable: http://mock.laptop.org/repos/local.update1/XOS/index.html joyride: http://xs-dev.laptop.org/~cscott/repos/joyride But it appears that the Help activity (for example) is not listed in these pages. What URL should I use? You should use the URL listed on the [[Activities/G1G1]] page. The data is machine-readable; there is a parser at: http://dev.laptop.org/git?p=users/cscott/sugar-update-control;a=blob;f=bitfrost/update/microformat.py;h=4104bfdae43532125fbd1a94faf4ce489a6359ea;hb=HEAD and the format is described at http://wiki.laptop.org/go/Activity_microformat --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [sugar] Signed candidate-765 and gg-765-2 builds available for testing.
Important: has anyone successfully upgraded/installed to the signed candidate-765? 1. I retried `sudo olpc-update candidate-765` from 8.2-763 on a secure (developer.sig moved away) XO. This time it worked fine. 2. I followed http://wiki.laptop.org/go/Clean-install_procedure to go back to build 650 (for that Christmas 2007 ship.1 feel :-) ) on a secure XO. su -l olpc-update candidate-765 succeeded, after errors in irsync_pristine and irsync dirty but apparently no out-of-memory problems. This wasn't a perfect test because I didn't revert firmware as well, I was on latest q2e18 firmware throughout. After reboot Software update's Checking for updates hung in Loading groups... , just as Walter Bender reported to testing list. I think this is bug 8681, fixed in build 766. shell.log had two WARNING root: Activity directory lacks a MANIFEST file. then exception from actinfo.py, line 138, in get_activity_group_urls with open( USER_GROUPS_FILE, 'w') that No such file or directory: '/home/olpc/Activities/.groups' 3. Carlo Falciola updated from 711 (8.1.2) to candidate-765 using olpc-update -usb, see separate e-mail. I changed the wiki Sunday night and I think all the links and banners mentioning candidate-765 work. Nobody replied to my question about the gg-765-2 image below. Michael Stone wrote: I have also published gg-765-2, a signed G1G1 candidate composite image, created by Scott. gg-765-2 is similar to what we hope to put into manufacturing next week. ... http://download.laptop.org/xo-1/custom/g1g1/gg-765-2/ (G1G1 composite) Where should this be mentioned? Should someone doing http://wiki.laptop.org/go/Clean-install_procedure copy this image? -- =S Page user:skierpage ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [sugar] Signed candidate-765 and gg-765-2 builds available for testing.
On Tue, Sep 30, 2008 at 8:23 AM, S Page [EMAIL PROTECTED] wrote: After reboot Software update's Checking for updates hung in Loading groups... , just as Walter Bender reported to testing list. I think this is bug 8681, fixed in build 766. shell.log had two WARNING root: Activity directory lacks a MANIFEST file. then exception from actinfo.py, line 138, in get_activity_group_urls with open( USER_GROUPS_FILE, 'w') that No such file or directory: '/home/olpc/Activities/.groups' Yes, this is fixed in 766 (although testing confirmation welcome). I have also published gg-765-2, a signed G1G1 candidate composite image, created by Scott. gg-765-2 is similar to what we hope to put into manufacturing next week. ... http://download.laptop.org/xo-1/custom/g1g1/gg-765-2/ (G1G1 composite) Where should this be mentioned? Should someone doing http://wiki.laptop.org/go/Clean-install_procedure copy this image? We have not yet released 765. In fact, we just made 766. So let's not jump the gun. We'll update all the pages when we actually release 766. (Basically, we'll meet on Friday, and if we haven't found any blocking bugs in 766 by that time, we'll declare that to be 8.2 and partying will ensure. If we find severe bugs, we'll fix them, make a new build, and reset the testing clock.) --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: devkeys, prettyboot, and G1G1
On 29 Sep 2008, at 06:45, John Gilmore wrote: Requesting dev keys should not be difficult! How can we fix that problem? We could consider shipping the next G1G1 batch with developer keys already included (disable-security). The only reason any G1G1 user would prefer a lockdown laptop is because it won't do pretty boot if it's jailbroken. That's a bug (http://dev.laptop.org/ticket/7896), and there's a kludge circumvention for it (insert dcon-freeze as 2nd line of /boot/ olpc.fth). I wonder if it's worth putting something like this into the OS build, bypassed by a game key or something. This would give 99% of the effect of pretty-boot, without the lockdown, for the upcoming G1G1. Thanks John. Wow, it is pretty, that's the first time I've ever seen the boot animation :-) I've had a few folks take sharp intakes of breath at boot before, and had to explain that they were looking at a development build XO – not what kids would usually see – though screeds of boot text seems to scare adults more than kids. Think I'll leave this on for a bit :-) --Gary ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
olpc-update -v 8.2-766 gets fakeroot error
Attempting to run olpc-update -v 8.2-766 from root gets : @ERROR: Unexpected server greeting: fakeroot error, while creating message channels: Invaalid argument rsync error: error starting client-server protocol (code 5) at main.c (1383) [receiver=2.6.9] ... This is from build 656 using olpc-update version 2.7 running from root. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
Hi Tomeu, I don't think we need to question Latu to test again. Its more a matter of explaining to end users what is the best way to get a good experience with Flash. Can you layout the steps people need to take to try out the new Flash? If you can post that (preferably in Spanish) to a wiki somewhere we can ask the Sur list to try it out. We may need an explanation of how to try it with 8.2 release candidate, 708 (Peru image) and 656 (Uruguay image) if the instructions are different for each. Thanks, Greg S Tomeu Vizoso wrote: 2008/9/16 Greg Smith [EMAIL PROTECTED]: Hi All, We need to characterize the performance and support of Flash in release 8.2. Technical people in Uruguay did a tests a while ago on the Flash games at this web site: http://www.minijuegos.com/ They used the Gnash shipped with 656 (0.8.1-1) and a Flash plugin (9.0.124-0). They tested Castle Wars: http://www.minijuegos.com/juegos/jugar.php?id=5552 Freekick Fusion http://www.minijuegos.com/juegos/html/index.php?id=4570 Hulk Central Smahdown http://www.minijuegos.com/juegos/jugar.php?id=6769 I believe those were chosen as examples of Flash capabilities. They weren't necessarily things the ministry of education wanted to use in class :-) In all three Gnash failed completely and Flash ran slowly. If anyone has time to try those again with the 8.2 candidate build I would be very interested in the results. Any other comments, URLs or examples about our Flash support are appreciated. Just tried the Flash 10 RC and it's working much better. Should we ask Uruguay to repeat their tests with that version? Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
Hi, have added it to my queue, but I'm pretty busy with the DS and will be traveling tomorrow, so if someone could take this task instead of me, I will be very grateful. It's basically following adobe's instructions from http://labs.adobe.com/downloads/flashplayer10.html . Thanks, Tomeu On Tue, Sep 30, 2008 at 6:55 PM, Greg Smith [EMAIL PROTECTED] wrote: Hi Tomeu, I don't think we need to question Latu to test again. Its more a matter of explaining to end users what is the best way to get a good experience with Flash. Can you layout the steps people need to take to try out the new Flash? If you can post that (preferably in Spanish) to a wiki somewhere we can ask the Sur list to try it out. We may need an explanation of how to try it with 8.2 release candidate, 708 (Peru image) and 656 (Uruguay image) if the instructions are different for each. Thanks, Greg S Tomeu Vizoso wrote: 2008/9/16 Greg Smith [EMAIL PROTECTED]: Hi All, We need to characterize the performance and support of Flash in release 8.2. Technical people in Uruguay did a tests a while ago on the Flash games at this web site: http://www.minijuegos.com/ They used the Gnash shipped with 656 (0.8.1-1) and a Flash plugin (9.0.124-0). They tested Castle Wars: http://www.minijuegos.com/juegos/jugar.php?id=5552 Freekick Fusion http://www.minijuegos.com/juegos/html/index.php?id=4570 Hulk Central Smahdown http://www.minijuegos.com/juegos/jugar.php?id=6769 I believe those were chosen as examples of Flash capabilities. They weren't necessarily things the ministry of education wanted to use in class :-) In all three Gnash failed completely and Flash ran slowly. If anyone has time to try those again with the 8.2 candidate build I would be very interested in the results. Any other comments, URLs or examples about our Flash support are appreciated. Just tried the Flash 10 RC and it's working much better. Should we ask Uruguay to repeat their tests with that version? Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
I just checked Flash10 with 8.2-765. The results are the same as with earlier versions of Flash: Did this really have to be sent to olpc-sur without a cursory check first ? Castle Wars is doesn't function correctly, and is unplayable. The video in Hulk Central Smashdown shows only two frames, and the game is unplayable due to slowness in rendering. Bummer, wad On Sep 30, 2008, at 1:05 PM, Tomeu Vizoso wrote: Hi, have added it to my queue, but I'm pretty busy with the DS and will be traveling tomorrow, so if someone could take this task instead of me, I will be very grateful. It's basically following adobe's instructions from http://labs.adobe.com/downloads/flashplayer10.html . Thanks, Tomeu On Tue, Sep 30, 2008 at 6:55 PM, Greg Smith [EMAIL PROTECTED] wrote: Hi Tomeu, I don't think we need to question Latu to test again. Its more a matter of explaining to end users what is the best way to get a good experience with Flash. Can you layout the steps people need to take to try out the new Flash? If you can post that (preferably in Spanish) to a wiki somewhere we can ask the Sur list to try it out. We may need an explanation of how to try it with 8.2 release candidate, 708 (Peru image) and 656 (Uruguay image) if the instructions are different for each. Thanks, Greg S Tomeu Vizoso wrote: 2008/9/16 Greg Smith [EMAIL PROTECTED]: Hi All, We need to characterize the performance and support of Flash in release 8.2. Technical people in Uruguay did a tests a while ago on the Flash games at this web site: http://www.minijuegos.com/ They used the Gnash shipped with 656 (0.8.1-1) and a Flash plugin (9.0.124-0). They tested Castle Wars: http://www.minijuegos.com/juegos/jugar.php?id=5552 Freekick Fusion http://www.minijuegos.com/juegos/html/index.php? id=4570 Hulk Central Smahdown http://www.minijuegos.com/juegos/jugar.php? id=6769 I believe those were chosen as examples of Flash capabilities. They weren't necessarily things the ministry of education wanted to use in class :-) In all three Gnash failed completely and Flash ran slowly. If anyone has time to try those again with the 8.2 candidate build I would be very interested in the results. Any other comments, URLs or examples about our Flash support are appreciated. Just tried the Flash 10 RC and it's working much better. Should we ask Uruguay to repeat their tests with that version? Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Filesystem path ordering overrated.
[I found this email in my 'saved drafts' folder; I think I've mentioned my unordered tags experiments before, but I figured I'd send this off anyway.] Discussing next-gen journal designs, I often bring up 'ordered tags' as a necessary way to make the journal cooperate better with filesystems. The response usually is that additional context is sufficient to disambiguate tag sets, you don't actually need ordering. That is, it's okay if a/b is indistinguishable from b/a -- in practice one will really be c/a/b and the other will be b/a/d or whatever, and you can use the extra tag 'c' or 'd' to disambiguate. I ran a script over the complete contents of my laptop's filesystem, and I was rather surprised that the responders were correct: I have few files on my filesystem which would be indistiguishable with completely unordered paths. More details at http://wiki.laptop.org/go/Experiments_with_unordered_paths --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
On Tue, Sep 30, 2008 at 7:37 PM, John Watlington [EMAIL PROTECTED] wrote: I just checked Flash10 with 8.2-765. The results are the same as with earlier versions of Flash: The flash games in the ceibal tests are still unplayable, but their performance is greatly improved, just go back to the Flash 9 plugin and you will see it. Youtube videos become viewable with Flash 10 although it doesn't play totally smoothly. In my opinion, Flash 10 improves significantly the flash experience on the XO and it would be worth studying if it should be recommended instead of Flash 9. Did this really have to be sent to olpc-sur without a cursory check first ? As I said above, I actually did compared both versions. Castle Wars is doesn't function correctly, and is unplayable. The video in Hulk Central Smashdown shows only two frames, and the game is unplayable due to slowness in rendering. I don't think we should expect games like Hulk Central Smashdown to become playable at any time soon. Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: olpc-update -v 8.2-766 gets fakeroot error
On Tue, Sep 30, 2008 at 11:55 AM, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Attempting to run olpc-update -v 8.2-766 from root gets : @ERROR: Unexpected server greeting: fakeroot error, while creating message channels: Invaalid argument rsync error: error starting client-server protocol (code 5) at main.c (1383) [receiver=2.6.9] ... This is from build 656 using olpc-update version 2.7 running from root. That message just indicates high load on the updates server. It's on my to-do list to rewrite it to use rsync's --fake-super argument, which should alleviate these problems, but I haven't gotten a chance to do that yet. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
On Sep 30, 2008, at 1:53 PM, Tomeu Vizoso wrote: On Tue, Sep 30, 2008 at 7:37 PM, John Watlington [EMAIL PROTECTED] wrote: The video in Hulk Central Smashdown shows only two frames, and the game is unplayable due to slowness in rendering. I don't think we should expect games like Hulk Central Smashdown to become playable at any time soon. Personally, I agree, but LATU disagrees. Their report indicates that this video plays on Intel Classmates (but only under Fedora, not Windows!) and there shouldn't be such a gulf in performance. Flash is detecting and using hardware assist on the Classmate, but isn't happy with what it finds on the XO. wad ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
Maybe it is worth testing flash on a classmate running Sugar to determine if the problem lies with Sugar or the XO? Anyone have a Classmate out there? -walter On Tue, Sep 30, 2008 at 2:13 PM, John Watlington [EMAIL PROTECTED] wrote: On Sep 30, 2008, at 1:53 PM, Tomeu Vizoso wrote: On Tue, Sep 30, 2008 at 7:37 PM, John Watlington [EMAIL PROTECTED] wrote: The video in Hulk Central Smashdown shows only two frames, and the game is unplayable due to slowness in rendering. I don't think we should expect games like Hulk Central Smashdown to become playable at any time soon. Personally, I agree, but LATU disagrees. Their report indicates that this video plays on Intel Classmates (but only under Fedora, not Windows!) and there shouldn't be such a gulf in performance. Flash is detecting and using hardware assist on the Classmate, but isn't happy with what it finds on the XO. wad ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel -- Walter Bender Sugar Labs http://www.sugarlabs.org ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
Am 30.09.2008 um 11:13 schrieb John Watlington: On Sep 30, 2008, at 1:53 PM, Tomeu Vizoso wrote: On Tue, Sep 30, 2008 at 7:37 PM, John Watlington [EMAIL PROTECTED] wrote: The video in Hulk Central Smashdown shows only two frames, and the game is unplayable due to slowness in rendering. I don't think we should expect games like Hulk Central Smashdown to become playable at any time soon. Personally, I agree, but LATU disagrees. Their report indicates that this video plays on Intel Classmates (but only under Fedora, not Windows!) and there shouldn't be such a gulf in performance. Flash is detecting and using hardware assist on the Classmate, but isn't happy with what it finds on the XO. The Classmate has a significantly faster CPU and a lot less pixels than the XO. Which makes it more power-hungry and less crisp, but better for games. Still, the XO performance might be tuned - did someone get a couple of dev machines to the Adobe Linux hackers? - Bert - ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash tests
On Tue, Sep 30, 2008 at 8:18 PM, Walter Bender [EMAIL PROTECTED] wrote: Maybe it is worth testing flash on a classmate running Sugar to determine if the problem lies with Sugar or the XO? Anyone have a Classmate out there? I don't think a link to Sugar can be made. Also, LATU tested with Debian on XO in order to get one more reference point. My biggest suspects are: - Flash being tuned for several CPUs but not for the Geode, - slow graphics performance of X in the OLPC builds. Regards, Tomeu On Tue, Sep 30, 2008 at 2:13 PM, John Watlington [EMAIL PROTECTED] wrote: On Sep 30, 2008, at 1:53 PM, Tomeu Vizoso wrote: On Tue, Sep 30, 2008 at 7:37 PM, John Watlington [EMAIL PROTECTED] wrote: The video in Hulk Central Smashdown shows only two frames, and the game is unplayable due to slowness in rendering. I don't think we should expect games like Hulk Central Smashdown to become playable at any time soon. Personally, I agree, but LATU disagrees. Their report indicates that this video plays on Intel Classmates (but only under Fedora, not Windows!) and there shouldn't be such a gulf in performance. Flash is detecting and using hardware assist on the Classmate, but isn't happy with what it finds on the XO. wad ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel -- Walter Bender Sugar Labs http://www.sugarlabs.org ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [sugar] rendering test
Michel Dänzer wrote: On Sun, 2008-09-28 at 18:46 +0200, Bernie Innocenti wrote: Tomeu Vizoso wrote: On Sun, Sep 28, 2008 at 12:46 PM, Riccardo Lucchese [EMAIL PROTECTED] wrote: On Sun, 2008-09-28 at 12:43 +0200, Riccardo Lucchese wrote: * build 703, xorg driver = amd, redraws = 200 - pixbuf: 98.63s 96.96s 96.58s 97.14s 99.21s * build 703, xorg driver = fbdev, redraws = 200 - pixbuf: 55.81s 55.40s 55.22s 55.50s 55.63s * build 2489, xorg driver = amd, redraws = 200 - pixbuf: 84.21s 84.81s 81.94s 81.79s 85.29s * build 2489, xorg driver = fbdev, redraws = 200 - pixbuf: 62.83s 62.81s 62.81s 62.66s 63.14s - joyride regressed sensibly at rendering with cairo since 703 - rendering pixbufs is extremely slow on the xo - server side surfaces are awesome ;) and btw why is fbdev faster than the geode driver at rendering pixbufs ? My performance tests with X 1.3 and 1.4 had shown that turning on EXA makes many operations slower. It's hard to tell why, but it might have to do with loosing XShmPut() (MIT shared memory), EXA does support XShmPutImage(), just not SHM pixmaps. I was remembering the code. As a result of ee7c684f21d, the PutImage hook in ShmFuncs is no longer being used. Shall I commit a cleanup? Also note that the fbdev driver by default uses a shadow framebuffer in system RAM and only updates the visible screen contents at regular intervals. It might be fairer to compare with Option ShadowFB off, at least assuming the amd driver provides other desirable features the fbdev driver can't provide. Riccardo, could you try that? -- \___/ Bernie Innocenti - http://www.codewiz.org/ _| X | Sugar Labs Team - http://www.sugarlabs.org/ \|_O_| It's an education project, not a laptop project! ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
re: Flash Tests
Hi guys! I'm a volunteer for OLPC Philippines and I'm very new here. Our project, OLPC XO as a Flash gaming/educational gaming platform (http://wiki.laptop.org/go/Projects/Flash_Gamedev) got approved and we just received 3 XO units Monday last week for testing and development. Anyway: 1) I recommend Flash 10 RC over Flash 9 because: - Flash 10 final should be coming out in a couple of months or so. Adobe just announced the launch of the CS4 product line Tue. last week (including Flash CS4). I guess that means they'll be shipping within 2 months or so, so expect Flash player 10 to be finalized by then. - Flash 9 has no V4L2 hardware support. The XO comes with a V4L2 webcam. Flash apps that use the webcam show nothing in Flash 9 (cam LED will light up, but only a black box appears in Flash). In Flash 10, it shows red green static but the static reacts if you wave your hand in front of the camera. I reported the cam behavior to Adobe, and they're looking into it. (tested in update.1 711 - will test in 766 if cam behavior improves). Anyway, my point is, we should really make Flash 10 our baseline rather than Flash 9, because it'll be coming out of RC soon. Btw, youtube played in Flash 9.0.124.0, but maybe that was because I was using Opera 9.52 in 711... 2) Gnash behavior: testing the pre-bundled Gnash 8.3 in the Sept.19 Livecd build (ftp://rohrmoser-engineering.de/pub/XO-LiveCD/XO-LiveCD_080919.iso) emulated in Virtualbox showed a lot of sound glitches with Flash 7 8 apps, and no sound at all in Youtube. I'm downloading RC 766 now, and will install it in one of the XOs in a while and report back. Currently, I only have update.1 711 stable installed in the XO units. Regarding Flash performance: I noticed that Flash runs faster in Opera 9.52 (latest desktop linux version of Opera runs fine on the XO) than in the Browse activivity that comes with 711. I think this is primarily caused by the fact that browser contents in the Browse activity are scaled up, and thus, it's taxing Flash's engine more. In general, the bigger the dimensions of the Flash applet, the more CPU horsepower it eats. Thus, using the same applet, if you zoom in on a page with Flash in Opera, you'll see performance drop and if you zoom out a page and thus shrink the flash applet size, you'll see performance improve. Oh, I heard from Ben Schwartz that the Browse activity in 8.2.0 is based on Firefox 3. If so, that would be interesting because my general experience is that the Flash plugin performs better in Firefox 3 than in MS IE or Opera 9.x. Anyway, am installing 766 and will report back on Flash/Gnash test results. -Naz -- Carlos Nazareno http://www.object404.com interactive media specialist zen graffiti studios [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash Tests
On Tue, Sep 30, 2008 at 4:37 PM, Carlos Nazareno [EMAIL PROTECTED] wrote: Hi guys! I'm a volunteer for OLPC Philippines and I'm very new here. Our project, OLPC XO as a Flash gaming/educational gaming platform (http://wiki.laptop.org/go/Projects/Flash_Gamedev) got approved and we just received 3 XO units Monday last week for testing and development. Anyway: 1) I recommend Flash 10 RC over Flash 9 because: - Flash 10 final should be coming out in a couple of months or so. Adobe just announced the launch of the CS4 product line Tue. last week (including Flash CS4). I guess that means they'll be shipping within 2 months or so, so expect Flash player 10 to be finalized by then. - Flash 9 has no V4L2 hardware support. The XO comes with a V4L2 webcam. Flash apps that use the webcam show nothing in Flash 9 (cam LED will light up, but only a black box appears in Flash). In Flash 10, it shows red green static but the static reacts if you wave your hand in front of the camera. I reported the cam behavior to Adobe, and they're looking into it. (tested in update.1 711 - will test in 766 if cam behavior improves). Anyway, my point is, we should really make Flash 10 our baseline rather than Flash 9, because it'll be coming out of RC soon. Btw, youtube played in Flash 9.0.124.0, but maybe that was because I was using Opera 9.52 in 711... 2) Gnash behavior: testing the pre-bundled Gnash 8.3 in the Sept.19 Livecd build (ftp://rohrmoser-engineering.de/pub/XO-LiveCD/XO-LiveCD_080919.iso) emulated in Virtualbox showed a lot of sound glitches with Flash 7 8 apps, and no sound at all in Youtube. I'm downloading RC 766 now, and will install it in one of the XOs in a while and report back. Currently, I only have update.1 711 stable installed in the XO units. A month or two ago on joyride I had gnash playing youtube videos with perfect sound, but very slow (.5ish FPS) video. For sound (and maybe video) you probably have to install the ffmpeg-gstreamer, and gstreamer-plugins-bad (and maybe gstreamer-plugins-dirty), all from the Livna repositories. The problem is that we've reverted to using an older version of gstreamer to get correct record behavior, and the livna gstreamer-plugins want the newer gstreamer. bobby Regarding Flash performance: I noticed that Flash runs faster in Opera 9.52 (latest desktop linux version of Opera runs fine on the XO) than in the Browse activivity that comes with 711. I think this is primarily caused by the fact that browser contents in the Browse activity are scaled up, and thus, it's taxing Flash's engine more. In general, the bigger the dimensions of the Flash applet, the more CPU horsepower it eats. Thus, using the same applet, if you zoom in on a page with Flash in Opera, you'll see performance drop and if you zoom out a page and thus shrink the flash applet size, you'll see performance improve. Oh, I heard from Ben Schwartz that the Browse activity in 8.2.0 is based on Firefox 3. If so, that would be interesting because my general experience is that the Flash plugin performs better in Firefox 3 than in MS IE or Opera 9.x. Anyway, am installing 766 and will report back on Flash/Gnash test results. -Naz -- Carlos Nazareno http://www.object404.com interactive media specialist zen graffiti studios [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [sugar] rendering test
On Tue, 2008-09-30 at 21:30 +0200, Bernie Innocenti wrote: Michel Dänzer wrote: On Sun, 2008-09-28 at 18:46 +0200, Bernie Innocenti wrote: Tomeu Vizoso wrote: On Sun, Sep 28, 2008 at 12:46 PM, Riccardo Lucchese [EMAIL PROTECTED] wrote: On Sun, 2008-09-28 at 12:43 +0200, Riccardo Lucchese wrote: * build 703, xorg driver = amd, redraws = 200 - pixbuf: 98.63s 96.96s 96.58s 97.14s 99.21s * build 703, xorg driver = fbdev, redraws = 200 - pixbuf: 55.81s 55.40s 55.22s 55.50s 55.63s * build 2489, xorg driver = amd, redraws = 200 - pixbuf: 84.21s 84.81s 81.94s 81.79s 85.29s * build 2489, xorg driver = fbdev, redraws = 200 - pixbuf: 62.83s 62.81s 62.81s 62.66s 63.14s - joyride regressed sensibly at rendering with cairo since 703 - rendering pixbufs is extremely slow on the xo - server side surfaces are awesome ;) and btw why is fbdev faster than the geode driver at rendering pixbufs ? My performance tests with X 1.3 and 1.4 had shown that turning on EXA makes many operations slower. It's hard to tell why, but it might have to do with loosing XShmPut() (MIT shared memory), EXA does support XShmPutImage(), just not SHM pixmaps. I was remembering the code. As a result of ee7c684f21d, the PutImage hook in ShmFuncs is no longer being used. Shall I commit a cleanup? Also note that the fbdev driver by default uses a shadow framebuffer in system RAM and only updates the visible screen contents at regular intervals. It might be fairer to compare with Option ShadowFB off, at least assuming the amd driver provides other desirable features the fbdev driver can't provide. Riccardo, could you try that? weird, testing with the ShadowFb option off slightly speeds up the test ;P avg time on 5 tries: ~57.5s (it was 62.83s) riccardo ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
New joyride build 2500
http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2500 Changes in build 2500 from build: 2499 Size delta: 0.00M -olpcrd 0.47-0 +olpcrd 0.48-0 -- This mail was automatically generated See http://dev.laptop.org/~rwh/announcer/joyride-pkgs.html for aggregate logs See http://dev.laptop.org/~rwh/announcer/joyride_vs_update1.html for a comparison ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash Tests
A month or two ago on joyride I had gnash playing youtube videos with perfect sound, but very slow (.5ish FPS) video. For sound (and maybe video) you probably have to install the ffmpeg-gstreamer, and gstreamer-plugins-bad (and maybe gstreamer-plugins-dirty), all from the Livna repositories. The problem is that we've reverted to using an older version of gstreamer to get correct record behavior, and the livna gstreamer-plugins want the newer gstreamer. Hey Bobby. How do I install these? I can't find docs at wiki.laptop.org or by googling on how to install these on the XO. I'm a fedora noob and googling, all I see are apt-getting for ubuntu. I just installed 766 on one of the XOs and I'd like to see how Gnash performs with sound. What's the easiest way to get sound working with Gnash on a just-clean-installed-766 + installed some activities environment? This page: http://wiki.laptop.org/go/Restricted_Formats doesn't help (sorry, I'm a linux newb, haven't had much experience with recompiling from sources. Have only mostly installed from apt-get and .debs in ubuntu and .rpms in fedora) Thanks! -Naz -- Carlos Nazareno http://www.object404.com interactive media specialist zen graffiti studios [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Greg Smith's Weekly Report
On Mon, Sep 29, 2008 at 5:41 AM, Greg Smith [EMAIL PROTECTED] wrote: Hi All, Here's my weekly report for week ending 9/26. *** Status against last week goals: 1 - Hound engineers to close all 8.2.0 blockers (see http://dev.laptop.org/report/28) and get a firm date for the Release Candidate build. Triage bugs and keep the release on schedule. GS - Done! In the end, no hounding was needed :-) Triaged bugs but fell a little behind the incoming rate this week. Also picked and tested a set of additional activities to add to the G1G1 default install. 2 - Clean up open bugs section of 8.2. release notes. Get release notes ready for final review. GS - Mostly done. Top section finished and ready for final review. Human readable explanation and categorization of key bugs started. Final edits and comments welcome: http://wiki.laptop.org/go/Release_notes/8.2.0 3 - Write 8.2 launch plan and technical overview presentation. GS - Not done. Wrote brief blurb on the release for re-use in promoting it. Still need to list communication vehicles, choose the right landing page (Release notes?) and refine the high level message for the release. First pass release message: 8.2 has major enhancements including: - A flexible Home view and Journal with several options for searching and organizing activities. - An enhanced Frame for accessing other XOs and peripherals and for switching between running activities. - A Graphical Control Panel for setting language, network, power and other defaults. - An automated Software Update tool which finds the latest version of activities and updates them over the Internet. - Capability to backup XOs to a school server and restore files to the Journal as needed. - A new manual shipped with the XO as an activity. - Many other bug fixes and enhancements. For more details, see the final draft of the Release Notes at: http://wiki.laptop.org/go/Release_notes/8.2.0 4 - Share and post 8.2.1 time frame and operating procedure (e.g. Trac queries). Start planning for Early Field Trial/Beta of 8.2.1. Keep pushing for an engineering leader/owner of 8.2.1. GS - Not done. Pushed for an engineering owner a little but not too hard until 8.2 is done. 5 - Review and finalize short 9.1 strategy description. Restructure requirements section to align with strategy. Fold in more deployment requirements. Keep pushing for an engineering leader/owner of 9.1. Stretch goal: prepare to write detailed requirements sections and start scrubbing bugs to create working Trac queries. GS - Mostly not done. Added a few more details, discussed strategy and country demands. Added some more detailed requirements for deployability at: http://wiki.laptop.org/go/9.1.0#Security.2C_Activation_and_Deployability 6 - Update deployments page. Update releases page and start using new semantic format (thanks to S Page for laying out the structure). Also update XS sections of releases page. GS - Not done but did get approval to repost statistics on XOs Delivered, shipped and ordered by country. See also this URL for new deployment and XO information: http://wiki.laptop.org/go/Learning_Workshops Some of this will be integrated in my next update of the Deployments page. ** Goals for next week in priority order: 1 - Get 8.2 to manufacturing. Catch up on bug triage. 2 - Finish 8.2 Release notes and get final review of them from engineering. 3 - Finalize blurb used to promote the release and start sending it out to internal lists. Write more generally usable Release message. 4 - Update deployments page and releases page. 5 - Post more detailed 8.2.1 page and restructure 9.1 page. 6 - Once 8.2 ships, open a bottle of Champagne. One glass then start work on the next release :-) One glass is inefficient. You'll lose all the fizz if you don't finish the rest. Or, just share! :-) -- Sameer ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash Tests
Gnash is probably already installed by default. Type:rpm -qv gnash-plugin and see what it says. You could try: yum install gnash-plugin if it isn't already installed. wad On Sep 30, 2008, at 6:34 PM, Carlos Nazareno wrote: A month or two ago on joyride I had gnash playing youtube videos with perfect sound, but very slow (.5ish FPS) video. For sound (and maybe video) you probably have to install the ffmpeg-gstreamer, and gstreamer-plugins-bad (and maybe gstreamer-plugins-dirty), all from the Livna repositories. The problem is that we've reverted to using an older version of gstreamer to get correct record behavior, and the livna gstreamer-plugins want the newer gstreamer. Hey Bobby. How do I install these? I can't find docs at wiki.laptop.org or by googling on how to install these on the XO. I'm a fedora noob and googling, all I see are apt-getting for ubuntu. I just installed 766 on one of the XOs and I'd like to see how Gnash performs with sound. What's the easiest way to get sound working with Gnash on a just-clean-installed-766 + installed some activities environment? This page: http://wiki.laptop.org/go/Restricted_Formats doesn't help (sorry, I'm a linux newb, haven't had much experience with recompiling from sources. Have only mostly installed from apt-get and .debs in ubuntu and .rpms in fedora) Thanks! -Naz -- Carlos Nazareno http://www.object404.com interactive media specialist zen graffiti studios [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Flash Tests
On Wed, Oct 1, 2008 at 6:34 AM, Carlos Nazareno [EMAIL PROTECTED] wrote: Hey Bobby. How do I install these? I can't find docs at wiki.laptop.org or by googling on how to install these on the XO. I'm a fedora noob and googling, all I see are apt-getting for ubuntu. Hi Carlos, What you want to do is add the livna repositories: http://rpm.livna.org/rlowiki/ Jerome G. Website: http://www.gotangco.com Blog: http://engage.wordpress.com ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [Testing] Flash tests
On Wed, Oct 1, 2008 at 1:05 AM, Tomeu Vizoso [EMAIL PROTECTED] wrote: have added it to my queue, but I'm pretty busy with the DS and will be traveling tomorrow, so if someone could take this task instead of me, I will be very grateful. It's basically following adobe's instructions from http://labs.adobe.com/downloads/flashplayer10.html . Perhaps we can start adding these info to this project page: http://wiki.laptop.org/go/Projects/Flash_Gamedev Jerome G. Website: http://www.gotangco.com Blog: http://engage.wordpress.com ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [sugar] rendering test
On Sun, 2008-09-28 at 18:46 +0200, Bernie Innocenti wrote: Tomeu Vizoso wrote: On Sun, Sep 28, 2008 at 12:46 PM, Riccardo Lucchese [EMAIL PROTECTED] wrote: On Sun, 2008-09-28 at 12:43 +0200, Riccardo Lucchese wrote: * build 703, xorg driver = amd, redraws = 200 - pixbuf: 98.63s 96.96s 96.58s 97.14s 99.21s * build 703, xorg driver = fbdev, redraws = 200 - pixbuf: 55.81s 55.40s 55.22s 55.50s 55.63s * build 2489, xorg driver = amd, redraws = 200 - pixbuf: 84.21s 84.81s 81.94s 81.79s 85.29s * build 2489, xorg driver = fbdev, redraws = 200 - pixbuf: 62.83s 62.81s 62.81s 62.66s 63.14s - joyride regressed sensibly at rendering with cairo since 703 - rendering pixbufs is extremely slow on the xo - server side surfaces are awesome ;) and btw why is fbdev faster than the geode driver at rendering pixbufs ? My performance tests with X 1.3 and 1.4 had shown that turning on EXA makes many operations slower. It's hard to tell why, but it might have to do with loosing XShmPut() (MIT shared memory), EXA does support XShmPutImage(), just not SHM pixmaps. excessive migration of pixmaps to the framebuffer, and so on. Migration overhead is indeed often the cause of EXA performance issues. Also note that the fbdev driver by default uses a shadow framebuffer in system RAM and only updates the visible screen contents at regular intervals. It might be fairer to compare with Option ShadowFB off, at least assuming the amd driver provides other desirable features the fbdev driver can't provide. -- Earthling Michel Dänzer | http://tungstengraphics.com Libre software enthusiast | Debian, X and DRI developer ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: WPA-PEAP with MSCHAPv2
On Tue, Sep 30, 2008 at 01:01:11AM -0700, pato wrote: Pato, We are using a Cisco RADIUS server that implements WPA2 enterprise and WPA-PEAP with MSCHAPv2. The SSID is also hidden. Glad to hear from you -- PEAP/802.11i/802.1X and hidden SSIDs are not yet supported but will probably be supported for the next major release. If you're feeling impatient, then you can help get this feature working faster by helping to test Sugar as we port it to more recent versions of NetworkManager and as we implement support for negotiating network use authorization. (If you're a programmer, then you can also help with the implementation of these tasks.) Regards, Michael ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
why are removable storage devices just an adjunct ?
Applications which I intend to use in the near future I keep resident (Sugar Activities in /home/olpc/Activities, Linux applications on my permanent SD card). Those I access rarely I keep on a removable storage device. Just now was using Journal to access Activity bundles kept on a removable storage device. All I wanted to do was to run them once -- but Journal *installed* (in /home/olpc/Activities) each one that I clicked on. I had not expected that. The XO-1 does not have a lot of nand storage. What interests me is how best to off-load data *and programs* from nand. I had been told that it was possible to run Activities from a removable storage device -- but I now see that in the actual implementation it *still* requires nand to run an off-loaded Activity -- in other words, the removable storage device is just an adjunct, not a repository. There really ought to be a better way to deposit Activities which are not being accessed each week. Sooner rather than later, there simply will not be room in /home/olpc/Activities. mikus ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [Server-devel] tiny_mce Sugar theme
Hi Martin, Nothing else interesting I can think of, mostly working on cleaning up and organizing, and rebasing of off XS Moodle. And of course, making clean commits this time =). Thanks, Tarun Martin Langhoff wrote: On Mon, Sep 29, 2008 at 6:05 PM, Tarun Pondicherry [EMAIL PROTECTED] wrote: Most of it is self contained in lib/editor/tinymce/jscripts/tiny_mce/themes/sugar/. I suspected it'd be most :-) that's how it is usually. Those little changes... is specially good if you have them as a separate commit. Then you can say look at commit ABCD123, it has all you need, so then you can cherry-pick commits between branches. Anyway, this is great info, I'll have to come back to it later, but now that I have it, it'll be easier. Not as easy as applying a patch or a series of patches, but very good anyway. Thanks! Are there other interesting bits that are worth looking at in the Edublog branch? (Other than your oublog changes to add external blog support, I know about that :-) ) cheers, m ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
[Server-devel] What's cooking in the XS pot this week (2008-10--01)
Overall, XS 0.5 is looking shaping up nicely... and late. The F9 port took quite a bit more time to get finished off, perhaps because I tried too hard to get it work relatively well, and uncovered a whole lot of problems with it -- it was a good thing to do as we now have fixes for all of them (mostly thanks to an very responsive Fedora community), but it sure took time. Luckily, nobody is waiting for 0.5 -- not that we know of! do mention it if a delay of 0.5 impacts on your deployment -- so we'll take a bit more time to finish off the Moodle bits, SOTP and mapping out the ejabberd configuration. The idea is that XS-0.5 and later XS-0.6 should be the preferred pairing for XO-8.2, still uses the same interfaces as 0.4, but it does a whole lot more. Good nice features built on top of what we have will come with XS-0.6. This week and next I am working on Moodle from various a Moodle conferences -- my email traffic moves a bit to the moodle.org forums, unfortunately I can't CC those easily here. I might post a summary with links to interesting threads - but I'd recommend that people subscribe to 'Using Moodle' 'General Developer Forum', which is effectively 'moodle-dev' :-) Some notes from last week: - My week was focused on getting Moodle on the XS. Some progress around installation, configuration and themes, still a lot to go. (Was a short week anyway, I ran away for a long weekend sailing :-) ) - Douglas Bagnall spent quite a bit of the week trying to work out what resources ejabberd uses in various circumstances. Things did not always go his way, but he is getting there. At other times he fixed some remaining issues related to the leap from Fedoras 7 to 9, and in spare moments he worked on bug #8610 (Theme music for the XS) which might be opened as a competition once 8.2 is out of the way. On Friday he helped the Wellington testers update the XO pool to build 764, which had handily been announced just a few minutes before. cheers, martin -- [EMAIL PROTECTED] [EMAIL PROTECTED] -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
Re: [Server-devel] XS_0_4 install]
I can't remember precisely, but I thought I started ejabberd after running domain_config, but I could be mistaken. Tony, did u start ejabberd before or after running domain_config on ur latest install? On Wed, 2008-10-01 at 11:13 +1300, Martin Langhoff wrote: On Tue, Sep 30, 2008 at 5:18 PM, Bryan Berry [EMAIL PROTECTED] wrote: What's more concerning to me is that ejabberd has never once worked for me out-of-box. I have always had to reinstall it before getting it to work. Has anyone out there gotten ejabberd to work consistently on new XS installs w/ reinstalling it? That sounds like you might be trying to start it before running domain_config. It's a tiny change to the installation workflow outlined in the doco, but can trip people up. We could change the init script to check for that case and refuse to start. cheers, m -- Bryan W. Berry Technology Director OLE Nepal, http://www.olenepal.org ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
