From pgf at laptop.org Sat May 2 15:04:23 2009 From: pgf at laptop.org (pgf at laptop.org) Date: Sat, 02 May 2009 11:04:23 -0400 Subject: package review needed for olpc-kbdshim Message-ID: <28058.1241276663@foxharp.boston.ma.us> i'd like to get olpc-kbdshim upstreamed so that more people can try it, etc. https://bugzilla.redhat.com/show_bug.cgi?id=495902 if anyone can do me the honor of taking a look, and gently telling me all that i've done wrong, or not at all, i'd appreciate it. (i'll be submitting olpc-powerd for review as well, but since i've surely made the same mistakes in both, and olpc-kbdshim is independent of olpc-powerd, i'm doing it first...) thanks! paul =--------------------- paul fox, pgf at laptop.org From pgf at laptop.org Sat May 2 16:51:26 2009 From: pgf at laptop.org (pgf at laptop.org) Date: Sat, 02 May 2009 12:51:26 -0400 Subject: package review needed for olpc-kbdshim In-Reply-To: <28058.1241276663@foxharp.boston.ma.us> References: <28058.1241276663@foxharp.boston.ma.us> Message-ID: <2108.1241283086@foxharp.boston.ma.us> i wrote: > i'd like to get olpc-kbdshim upstreamed so that more people can try it, etc. someone has gently pointed out that i'm actually trying to get it "downstreamed". duh. :-) paul > > https://bugzilla.redhat.com/show_bug.cgi?id=495902 > > if anyone can do me the honor of taking a look, and gently > telling me all that i've done wrong, or not at all, i'd > appreciate it. > > (i'll be submitting olpc-powerd for review as well, but since > i've surely made the same mistakes in both, and olpc-kbdshim is > independent of olpc-powerd, i'm doing it first...) > > thanks! > > paul > =--------------------- > paul fox, pgf at laptop.org =--------------------- paul fox, pgf at laptop.org From mikus at bga.com Fri May 8 13:57:39 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Fri, 08 May 2009 09:57:39 -0400 Subject: fedora-xo 20090507 does not boot on XO-1 Message-ID: <4A043A53.40907@bga.com> Using 'copy-nand', flashed ~cjb/fedora-xo 20090507.img to nand. Booted. At the point where the login screen normally shows up, the (blank) screen seemed to "pulse" repetitively (possibility: something invisible was started, crashed, started, crashed, etc). When "pulsing" eventually stopped, the screen was blank (black) and the XO-1 was completely unresponsive - only thing that would work was the button to power off. {Last "boot-up" line that showed was "starting sshd: [OK]" } mikus p.s. When I tried to boot with an USB hub (with devices) plugged in, "boot-up" would usually (but not always) stall earlier, after the line "creating devices". From mikus at bga.com Fri May 8 14:52:07 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Fri, 08 May 2009 10:52:07 -0400 Subject: fedora-xo 20090507 does not boot on XO-1 Message-ID: <4A044717.4000802@bga.com> I get the same result (XO-1 goes completely unresponsive at about the point where it should show the logon screen; preceded by "pulsing" of a blank screen) when I boot from an USB stick created by 'livecd-iso-to-disk' from ~cjb/rawhide-xo 20090507.iso [However (when booting from the USB stick), instead of the "non-pretty" boot-up messages scrolling by, I am entertained by an image of what looks like a penis inside a hand-waving hot dog bun.] mikus From sebastian at when.com Fri May 8 18:25:46 2009 From: sebastian at when.com (Sebastian Dziallas) Date: Fri, 08 May 2009 20:25:46 +0200 Subject: fedora-xo 20090507 does not boot on XO-1 In-Reply-To: <4A044717.4000802@bga.com> References: <4A044717.4000802@bga.com> Message-ID: <4A04792A.8000707@when.com> Mikus Grinbergs wrote: > I get the same result (XO-1 goes completely unresponsive at about the > point where it should show the logon screen; preceded by "pulsing" of a > blank screen) when I boot from an USB stick created by > 'livecd-iso-to-disk' from ~cjb/rawhide-xo 20090507.iso > > [However (when booting from the USB stick), instead of the "non-pretty" > boot-up messages scrolling by, I am entertained by an image of what > looks like a penis inside a hand-waving hot dog bun.] > > mikus Heh. Well, this is because of plymouth and the generic-logos package. We're currently using generic-logos here and those contain some kind of unfavorable boot screen. We ran into the same issue with Sugar on a Stick and rebuilt that package. Anyway, it's at least good to hear that you're getting an image instead of the text messages now. --Sebastian From pbrobinson at gmail.com Sat May 9 11:05:11 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sat, 9 May 2009 12:05:11 +0100 Subject: fedora-xo 20090507 does not boot on XO-1 In-Reply-To: <4A04792A.8000707@when.com> References: <4A044717.4000802@bga.com> <4A04792A.8000707@when.com> Message-ID: <5256d0b0905090405x1a449aaeoe5f76cdea1d05dcb@mail.gmail.com> >> I get the same result (XO-1 goes completely unresponsive at about the >> point where it should show the logon screen; preceded by "pulsing" of a >> blank screen) when I boot from an USB stick created by >> 'livecd-iso-to-disk' from ~cjb/rawhide-xo 20090507.iso >> >> [However (when booting from the USB stick), instead of the "non-pretty" >> boot-up messages scrolling by, I am entertained by an image of what >> looks like a penis inside a hand-waving hot dog bun.] >> >> mikus > > Heh. Well, this is because of plymouth and the generic-logos package. We're > currently using generic-logos here and those contain some kind of > unfavorable boot screen. We ran into the same issue with Sugar on a Stick > and rebuilt that package. Anyway, it's at least good to hear that you're > getting an image instead of the text messages now. I've been working with the plymouth guys to get the equivalent of the olpc boot anim into plymouth. Not quite there yet, but hopefully before too long we'll have something native we can use in Fedora. Peter From mikus at bga.com Sat May 9 13:35:46 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Sat, 09 May 2009 09:35:46 -0400 Subject: fedora-xo 20090507 does not boot on XO-1 Message-ID: <4A0586B2.1090705@bga.com> > I've been working with the plymouth guys to get the equivalent of the > olpc boot anim into plymouth. Not quite there yet, but hopefully > before too long we'll have something native we can use in Fedora. See 'Subject:' this version does not currently boot at all for me. If the problem was caused by instabilities in Rawhide itself - o.k., I would expect someone soon to be looking to fix it. But if not being able to boot 20090507 were to be associated with the XO environment, have we gotten away from the 'Subject:', and instead into the arrangement of the deck chairs on the Titanic? mikus From pbrobinson at gmail.com Sat May 9 15:10:19 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sat, 9 May 2009 16:10:19 +0100 Subject: fedora-xo 20090507 does not boot on XO-1 In-Reply-To: <4A0586B2.1090705@bga.com> References: <4A0586B2.1090705@bga.com> Message-ID: <5256d0b0905090810h773f4e34p87f0b93e18665cd@mail.gmail.com> >> I've been working with the plymouth guys to get the equivalent of the >> olpc boot anim into plymouth. Not quite there yet, but hopefully >> before too long we'll have something native we can use in Fedora. > > See 'Subject:' ?this version does not currently boot at all for me. > > > If the problem was caused by instabilities in Rawhide itself - o.k., I would > expect someone soon to be looking to fix it. > > But if not being able to boot 20090507 were to be associated with the XO > environment, have we gotten away from the 'Subject:', and instead into the > arrangement of the deck chairs on the Titanic? I had a problem with a kernel oops on boot the other day. Not sure what day it was as I was using my own build. I logged a ticket with the kernel guys and it looks to be fixed since then (at least it boots now for me). Chris can you possibly gen a new snapshot to see if this fixes the issue? Peter From pgf at laptop.org Sat May 9 21:27:33 2009 From: pgf at laptop.org (pgf at laptop.org) Date: Sat, 09 May 2009 17:27:33 -0400 Subject: bootanim on fedora (was Re: ... does not boot...) In-Reply-To: <5256d0b0905090810h773f4e34p87f0b93e18665cd@mail.gmail.com> References: <4A0586B2.1090705@bga.com> <5256d0b0905090810h773f4e34p87f0b93e18665cd@mail.gmail.com> Message-ID: <11109.1241904453@foxharp.boston.ma.us> peter wrote: > >> I've been working with the plymouth guys to get the equivalent of the > >> olpc boot anim into plymouth. Not quite there yet, but hopefully > >> before too long we'll have something native we can use in Fedora. this is great. which version of bootanim are they starting with? (cscott's act-gui version, or bobby's bootanim version?) paul =--------------------- paul fox, pgf at laptop.org From mikus at bga.com Sun May 10 19:31:25 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Sun, 10 May 2009 15:31:25 -0400 Subject: fedora-xo 20090510 does not start X server Message-ID: <4A072B8D.1020407@bga.com> Installed ~cjb/rawhide-xo 20090510.img with 'copy-nand' onto an XO-1 system. Booted. Boot process went normally up to the point where X would be used to draw the background and the "user logon" screen. Instead, the contents of the text console were repeatedly shown, blanked, shown, blanked, etc. - as the system repeatedly tried to start X, but failed. Eventually, the boot process stalled at that point. By switching to another text console, I was able to log in and examine the contents of memory. The only thing of interest was many many copies (to all intents and purposed identical) of Xorg._.log in /var/log . Copied the most recent of those, and put it into https://bugzilla.redhat.com/show_bug.cgi?id=500086 , which I created for that purpose. mikus From pbrobinson at gmail.com Sun May 10 20:51:40 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sun, 10 May 2009 21:51:40 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <4A072B8D.1020407@bga.com> References: <4A072B8D.1020407@bga.com> Message-ID: <5256d0b0905101351q6a7ce7detb18d0a7a5db34bdd@mail.gmail.com> Thanks for reporting that Mikus, I'm seeing exactly the same problem on my XO. I've added some notes to the bug as well as adding as a F11 blocker. Peter On Sun, May 10, 2009 at 8:31 PM, Mikus Grinbergs wrote: > Installed ~cjb/rawhide-xo 20090510.img with 'copy-nand' onto an XO-1 system. > ?Booted. ?Boot process went normally up to the point where X would be used > to draw the background and the "user logon" screen. > > Instead, the contents of the text console were repeatedly shown, blanked, > shown, blanked, etc. - as the system repeatedly tried to start X, but > failed. ?Eventually, the boot process stalled at that point. > > By switching to another text console, I was able to log in and examine the > contents of memory. ?The only thing of interest was many many copies (to all > intents and purposed identical) of Xorg._.log in /var/log . ?Copied the most > recent of those, and put it into > https://bugzilla.redhat.com/show_bug.cgi?id=500086 , which I created for > that purpose. > > mikus > > > _______________________________________________ > Fedora-olpc-list mailing list > Fedora-olpc-list at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-olpc-list > From cjb at laptop.org Mon May 11 00:34:37 2009 From: cjb at laptop.org (Chris Ball) Date: Sun, 10 May 2009 20:34:37 -0400 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <4A072B8D.1020407@bga.com> (Mikus Grinbergs's message of "Sun, 10 May 2009 15:31:25 -0400") References: <4A072B8D.1020407@bga.com> Message-ID: Hi Mikus/Peter, > Installed ~cjb/rawhide-xo 20090510.img with 'copy-nand' onto an > XO-1 system. Booted. Boot process went normally up to the point > where X would be used to draw the background and the "user logon" > screen. I'm curious about how you managed to get that far -- on an XO here, the 20090510.img prints "mount: unknown filesystem type 'jffs2'" and doesn't progress further. Does it not do the same for you? Thanks, - Chris. -- Chris Ball From cjb at laptop.org Mon May 11 01:13:48 2009 From: cjb at laptop.org (Chris Ball) Date: Sun, 10 May 2009 21:13:48 -0400 Subject: fedora-xo 20090510 does not start X server In-Reply-To: (Chris Ball's message of "Sun, 10 May 2009 20:34:37 -0400") References: <4A072B8D.1020407@bga.com> Message-ID: Hi, > I'm curious about how you managed to get that far -- on an XO > here, the 20090510.img prints "mount: unknown filesystem type > 'jffs2'" and doesn't progress further. Does it not do the same > for you? Oh, I see, it's some kind of race; a subsequent boot worked okay. I think there are several bugs here: * We get "mount: unknown filesystem type 'jffs2'" sporadically when booting, even though jffs2.ko is obviously in the initrd (since other boot attempts succeed). I'm not sure which BZ component to file initrd bugs against. * The already filed X bug. (BZ #500086) * When boot does progress, haldaemon sometimes fails to start, with a kernel protection fault in dmesg. (Filed against kernel as BZ #500096.) * Even when haldaemon does start, X doesn't seem to find any pointer devices. Hard to find out more about this while X is segfaulting; I guess it's possible that the segfault is related.. - Chris. -- Chris Ball From mikus at bga.com Mon May 11 03:56:14 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Sun, 10 May 2009 23:56:14 -0400 Subject: fedora-xo 20090510 does not start X server Message-ID: <4A07A1DE.2010006@bga.com> > * Even when haldaemon does start, X doesn't seem to find any pointer > devices. Hard to find out more about this while X is segfaulting; > I guess it's possible that the segfault is related.. Compared /etc/X11/xorg.conf in rawhide-xo 20090426 (X ok on boot) with that same file in rawhide-xo 20090510 (X fails on boot). The contents of that file are identical in the two builds. Also compared /var/log/Xorg._.log in rawhide-xo 20090510 (for which I wrote BZ 500086) against the equivalent file in rawhide-xo 20090426 (on which X worked) -- BOTH log files had the line: "(II) Cannot locate a core pointer device." Yet in 20090426, for both Sugar and Gnome, the XO-1 touchpad (and also an external USB mouse) work fine -- making me think that this informational line in the xorg.log is not significant. mikus From pbrobinson at gmail.com Mon May 11 08:15:37 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Mon, 11 May 2009 09:15:37 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: References: <4A072B8D.1020407@bga.com> Message-ID: <5256d0b0905110115hc1623ecv28c6b00ce4b4ec2e@mail.gmail.com> Hi, > ? > I'm curious about how you managed to get that far -- on an XO > ? > here, the 20090510.img prints "mount: unknown filesystem type > ? > 'jffs2'" and doesn't progress further. ?Does it not do the same > ? > for you? > > Oh, I see, it's some kind of race; a subsequent boot worked okay. > > I think there are several bugs here: > > * We get "mount: unknown filesystem type 'jffs2'" sporadically when > ?booting, even though jffs2.ko is obviously in the initrd (since > ?other boot attempts succeed). ?I'm not sure which BZ component to > ?file initrd bugs against. I would probably file one against either the kernel or mkinitrd. The kernel guys are generally pretty quick to respond and reallocate it as appropriate. The other thing to do is put the bug as a blocker to "F11Blocker" as all of these are regressions from the last week or so and IMO are definitely a blocker. Peter From cjb at laptop.org Mon May 11 16:01:38 2009 From: cjb at laptop.org (Chris Ball) Date: Mon, 11 May 2009 12:01:38 -0400 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <5256d0b0905110115hc1623ecv28c6b00ce4b4ec2e@mail.gmail.com> (Peter Robinson's message of "Mon, 11 May 2009 09:15:37 +0100") References: <4A072B8D.1020407@bga.com> <5256d0b0905110115hc1623ecv28c6b00ce4b4ec2e@mail.gmail.com> Message-ID: Hi, > I would probably file one against either the kernel or > mkinitrd. Filed as https://bugzilla.redhat.com/show_bug.cgi?id=500196 , thanks. - Chris. -- Chris Ball From pbrobinson at gmail.com Mon May 11 16:18:31 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Mon, 11 May 2009 17:18:31 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: References: <4A072B8D.1020407@bga.com> <5256d0b0905110115hc1623ecv28c6b00ce4b4ec2e@mail.gmail.com> Message-ID: <5256d0b0905110918s295f03ffu6f3d0f476fce21d9@mail.gmail.com> > Hi, > > ? > I would probably file one against either the kernel or > ? > mkinitrd. > > Filed as https://bugzilla.redhat.com/show_bug.cgi?id=500196 , thanks. Chris, out of interest how is the NAND attached wrt to bus etc to the rest of the system. There's been some issues recently (mostly with usb I think) with some of the fast boot features that have come from the Intel Moblin patchset. I'm not sure whether this could be coming into play or whether I have got completely the wrong end of the stick. This link is quite a good article on LWN with further details. http://lwn.net/Articles/330378/ Cheers, Peter From cjb at laptop.org Mon May 11 16:36:03 2009 From: cjb at laptop.org (Chris Ball) Date: Mon, 11 May 2009 12:36:03 -0400 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <5256d0b0905110918s295f03ffu6f3d0f476fce21d9@mail.gmail.com> (Peter Robinson's message of "Mon, 11 May 2009 17:18:31 +0100") References: <4A072B8D.1020407@bga.com> <5256d0b0905110115hc1623ecv28c6b00ce4b4ec2e@mail.gmail.com> <5256d0b0905110918s295f03ffu6f3d0f476fce21d9@mail.gmail.com> Message-ID: Hi, > Chris, out of interest how is the NAND attached wrt to bus etc to > the rest of the system. It speaks to our custom CaFE controller, via the "cafe_nand" kernel module. > There's been some issues recently (mostly with usb I think) with > some of the fast boot features that have come from the Intel > Moblin patchset. I'm not sure whether this could be coming into > play or whether I have got completely the wrong end of the > stick. The part that has me most confused is that the specific error message we get from mount should only ever appear if jffs2 is not in /proc/filesystems, which should only ever happen if the kernel module wasn't loaded. The kernel module obviously is loaded sometimes, since boot works sometimes, but somehow not on the failing boots. I wonder what's responsible for loading it; whether it's the initrd scripts or mount itself that calls modprobe jffs2. > http://lwn.net/Articles/330378/ Thanks for the link. We should try passing rootdelay= in olpc.fth, even though the problem doesn't seem too similar. - Chris. -- Chris Ball From pbrobinson at gmail.com Mon May 11 17:00:48 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Mon, 11 May 2009 18:00:48 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: References: <4A072B8D.1020407@bga.com> <5256d0b0905110115hc1623ecv28c6b00ce4b4ec2e@mail.gmail.com> <5256d0b0905110918s295f03ffu6f3d0f476fce21d9@mail.gmail.com> Message-ID: <5256d0b0905111000g31e784d4te71fd315567ec5f6@mail.gmail.com> > ? > There's been some issues recently (mostly with usb I think) with > ? > some of the fast boot features that have come from the Intel > ? > Moblin patchset. I'm not sure whether this could be coming into > ? > play or whether I have got completely the wrong end of the > ? > stick. > > The part that has me most confused is that the specific error > message we get from mount should only ever appear if jffs2 is not in > /proc/filesystems, which should only ever happen if the kernel module > wasn't loaded. ?The kernel module obviously is loaded sometimes, since > boot works sometimes, but somehow not on the failing boots. ?I wonder > what's responsible for loading it; whether it's the initrd scripts or > mount itself that calls modprobe jffs2. If you install the nash package (if its not there anyway) there's a lsinitrd command that gives you the contents of the initrd package including the initrd script that's run. Maybe there's been a change in the order of the script or something. Peter From pbrobinson at gmail.com Mon May 11 18:54:09 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Mon, 11 May 2009 19:54:09 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <4A080E60.6090302@bga.com> References: <4A080E60.6090302@bga.com> Message-ID: <5256d0b0905111154t5209a26g966ff58363eeade@mail.gmail.com> Hi Mikus, >> I would probably file one against either the kernel or mkinitrd. The >> kernel guys are generally pretty quick to respond and reallocate it as >> appropriate. The other thing to do is put the bug as a blocker to >> "F11Blocker" as all of these are regressions from the last week or so >> and IMO are definitely a blocker. > > The "unknown filesystem type" error is definitely not a regression from the > last week or so. ?I first told someone about seeing this in mid-April, > meaning I would have been running fedora-xo 20090416 (or even earlier) when > I grumbled. Can you try adding a "rootdelay=300" to the kernel command line in the olpc.fth and see if that helps the issues your seeing when you have usb devices attached? Peter From mikus at bga.com Mon May 11 21:05:54 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Mon, 11 May 2009 17:05:54 -0400 Subject: fedora-xo 20090510 does not start X server Message-ID: <4A089332.3020501@bga.com> > Can you try adding a "rootdelay=300" to the kernel command line in the > olpc.fth and see if that helps the issues your seeing when you have > usb devices attached? Added that to three different XO-1 machines: * rawhide-xo 20090416 (USB): This system was giving me lots of trouble this morning: first with stalls during boot-up at the "unknown filesystem" message, and later in the morning with stalls during boot-up after showing the "Loading ramdisk" message. Just now did multiple boot-ups after adding "rootdelay=300". When I had taken the system down with 'Restart', boot-up came up into Sugar the way it was supposed to. But when I then took the system down with 'Shutdown', the system would always stall during boot-up after showing the "Loading ramdisk" message. Had to completely re-flash that system (I now installed rawhide-xo 20090426, since that was the most recent rawhide-xo that worked). On the first couple of boot-up attempts, the new build stalled at the "unknown filesystem" message. By repeatedly trying, I got past that point and added "rootdelay=300". After that, it did not stall with the "unknown filesystem" message. This system still all too often gets the boot-up stall after showing the "Loading ramdisk" message. Luckily, pulling the USB devices (and trying and trying) eventually allowed me to boot-up. Since it takes me more than an hour of my time to re-flash (and re-customize) the system once I can't make headway against these boot-up stalls, please don't ask me again to try to create such error situations. * rawhide-xo 20090426 (nothing in USB sockets): This system has not recently experienced any stalls during boot-up. But this morning boot-up on this system had trouble starting Sugar (but not Gnome). Added "rootdelay=300" and now several boot-ups went correctly into Sugar. [This experience shows the __difficulty__ of looking for patterns of behavior -- I have no idea whether Sugar now being able to start easily was happenstance, or was caused by something having changed within this system.] * rawhide-xo 20090510 (USB): The last time this system showed me the "unknown filesystem" message was once - right after installing 20090510. Since then, boot-up has not stalled for me on this system, neither before nor after adding "rootdelay=300". But for me 20090510 does not start the X server, whether or not "rootdelay=300" has been added. Bottom line: Seems to help; does not hurt; needs more experience. mikus From mikus at bga.com Mon May 11 23:01:07 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Mon, 11 May 2009 19:01:07 -0400 Subject: fedora-xo 20090511 does not start X server Message-ID: <4A08AE33.7070709@bga.com> Using 'copy-nand', flashed ~cjb/rawhide-xo 20090511.img to nand. Booted. At the point where the X server would normally start up, the text console repetitively went blank, "flashed" text, went blank, "flashed" text, etc. -- eventually the "flashing" stopped (text showing), and text console 1 just sat there. Went to text console 2 and logged in. Last "boot-up" line that showed on the text console 1 was: Starting NetworkManager daemon: [ OK ] For me, rawhide-xo 20090511 behaves just the same as 20090510 - including similar X errors being logged to /var/log. mikus From cjb at laptop.org Tue May 12 00:08:06 2009 From: cjb at laptop.org (Chris Ball) Date: Mon, 11 May 2009 20:08:06 -0400 Subject: fedora-xo 20090511 does not start X server In-Reply-To: <4A08AE33.7070709@bga.com> (Mikus Grinbergs's message of "Mon, 11 May 2009 19:01:07 -0400") References: <4A08AE33.7070709@bga.com> Message-ID: Hi Mikus, > Using 'copy-nand', flashed ~cjb/rawhide-xo 20090511.img to nand. > Booted. At the point where the X server would normally start up, > the text console repetitively went blank, "flashed" text, went > blank, "flashed" text, etc. -- eventually the "flashing" stopped > (text showing), and text console 1 just sat there. Went to text > console 2 and logged in. Since you seem to be able to beat the jffs2 race better than the rest of us, it would be extremely helpful if you (or someone else) could work out which X server RPM introduced the X crash. The technique for doing so is: * go to http://koji.fedoraproject.org/koji/packageinfo?packageID=63 * choose a build somewhere between 0415 and 0508 * click on it * install the xorg-x11-server-Xorg.i586 download link, e.g. rpm --force -Uvh http://kojipkgs.fedoraproject.org/packages/xorg-x11-server/1.6.1/6.fc11/i586/xorg-x11-server-Xorg-1.6.1-6.fc11.i586.rpm http://kojipkgs.fedoraproject.org/packages/xorg-x11-server/1.6.1/6.fc11/i586/xorg-x11-server-common-1.6.1-6.fc11.i586.rpm * try starting X, perhaps by switching between "init 3" and "init 5". * if the earliest build tried works, attempt to bisect your way to the one that broke everything, and reply here with that RPM's version. Thanks, - Chris. -- Chris Ball From cjb at laptop.org Tue May 12 00:59:42 2009 From: cjb at laptop.org (Chris Ball) Date: Mon, 11 May 2009 20:59:42 -0400 Subject: fedora-xo 20090511 does not start X server In-Reply-To: (Chris Ball's message of "Mon, 11 May 2009 20:08:06 -0400") References: <4A08AE33.7070709@bga.com> Message-ID: Hi, > Since you seem to be able to beat the jffs2 race better than the > rest of us, it would be extremely helpful if you (or someone > else) could work out which X server RPM introduced the X crash. Sometimes it takes more time to explain something than to do it yourself. :) xorg-x11-server-1.6.1-7.fc11 works, xorg-x11-server-1.6.1-8.fc11 introduced the crash. I'll update the bug. - Chris. -- Chris Ball From pbrobinson at gmail.com Tue May 12 06:08:05 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Tue, 12 May 2009 07:08:05 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <4A089332.3020501@bga.com> References: <4A089332.3020501@bga.com> Message-ID: <5256d0b0905112308y65b1f752v9a129bed3cefbbc0@mail.gmail.com> Hi Mikus, >> Can you try adding a "rootdelay=300" to the kernel command line in the >> olpc.fth and see if that helps the issues your seeing when you have >> usb devices attached? > > Added that to three different XO-1 machines: > > > ?* ?rawhide-xo 20090416 (USB): ?This system was giving me lots of trouble > this morning: ?first with stalls during boot-up at the "unknown filesystem" > message, and later in the morning with stalls during boot-up after showing > the "Loading ramdisk" message. ?Just now did multiple boot-ups after adding > "rootdelay=300". ?When I had taken the system down with 'Restart', boot-up > came up into Sugar the way it was supposed to. > > But when I then took the system down with 'Shutdown', the system would > always stall during boot-up after showing the "Loading ramdisk" message. > ?Had to completely re-flash that system (I now installed rawhide-xo > 20090426, since that was the most recent rawhide-xo that worked). ?On the > first couple of boot-up attempts, the new build stalled at the "unknown > filesystem" message. ?By repeatedly trying, I got past that point and added > "rootdelay=300". ?After that, it did not stall with the "unknown filesystem" > message. > > This system still all too often gets the boot-up stall after showing the > "Loading ramdisk" message. ?Luckily, pulling the USB devices (and trying and > trying) eventually allowed me to boot-up. ?Since it takes me more than an > hour of my time to re-flash (and re-customize) the system once I can't make > headway against these boot-up stalls, please don't ask me again to try to > create such error situations. > > > ?* ?rawhide-xo 20090426 (nothing in USB sockets): ?This system has not > recently experienced any stalls during boot-up. ?But this morning boot-up on > this system had trouble starting Sugar (but not Gnome). ?Added > "rootdelay=300" and now several boot-ups went correctly into Sugar. ?[This > experience shows the __difficulty__ of looking for patterns of behavior -- I > have no idea whether Sugar now being able to start easily was happenstance, > or was caused by something having changed within this system.] > > > ?* ?rawhide-xo 20090510 (USB): ?The last time this system showed me the > "unknown filesystem" message was once - right after installing 20090510. > ?Since then, boot-up has not stalled for me on this system, neither before > nor after adding "rootdelay=300". ?But for me 20090510 does not start the X > server, whether or not "rootdelay=300" has been added. > > > Bottom line: ?Seems to help; ?does not hurt; ?needs more experience. Thanks for your testing. I wouldn't have expected the rootdelay to have any effect one systems that get past the kernel and into init. The cases I expect it to possibly have a positive effect are the cases where it can't find the jffs2 FS or the ramdisk issues. I understand the time it takes to reflash a device and appreciate the time you've taken. If its easy to change the rootdelay you might like to increase it to 500 for the XOs that you see the issues with the USB devices. What this does is waits a period of time before the kernel tries to mouth the rootfs to ensure that everything there that's needed to mount it. It might turn out to be completely a non issue. Peter From pbrobinson at gmail.com Tue May 12 08:18:54 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Tue, 12 May 2009 09:18:54 +0100 Subject: fedora-xo 20090511 does not start X server In-Reply-To: References: <4A08AE33.7070709@bga.com> Message-ID: <5256d0b0905120118u1f7ebb16t8ee6fb3a36315d10@mail.gmail.com> On Tue, May 12, 2009 at 1:59 AM, Chris Ball wrote: > Hi, > > ? > Since you seem to be able to beat the jffs2 race better than the > ? > rest of us, it would be extremely helpful if you (or someone > ? > else) could work out which X server RPM introduced the X crash. > > Sometimes it takes more time to explain something than to do it > yourself. ?:) > > xorg-x11-server-1.6.1-7.fc11 works, xorg-x11-server-1.6.1-8.fc11 > introduced the crash. ?I'll update the bug. Might be worthwhile installing the associated debuginfo package as it could provide a better traceback for the X guys as well. Peter From pbrobinson at gmail.com Tue May 12 10:36:10 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Tue, 12 May 2009 11:36:10 +0100 Subject: csound on Fedora 11/rawhide Message-ID: <5256d0b0905120336q231dea3bn48313c3ae0765926@mail.gmail.com> Hello All, I've just updated csound in F12/devel to csoound 5.10.1 and would like to get some testing (don't worry about the F12 designator its the same as F11) to see if everything works as expected for the OLPC stuff so when I push it into F11 and ultimately we can obsolete olpcsound. You can find a build to play with here http://koji.fedoraproject.org/koji/taskinfo?taskID=1349952 Peter From pbrobinson at gmail.com Tue May 12 12:17:16 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Tue, 12 May 2009 13:17:16 +0100 Subject: csound on Fedora 11/rawhide In-Reply-To: References: <5256d0b0905120336q231dea3bn48313c3ae0765926@mail.gmail.com> Message-ID: <5256d0b0905120517p29898ae6g5010807aaac2bbf0@mail.gmail.com> If we can have some testing on that build to confirm I've got the update vaguely correct would be great before I do builds for F-9 and F-10. Alternatively if that won't happen I can possibly do some scratch builds for F-10/F-9. Peter On Tue, May 12, 2009 at 1:01 PM, wrote: > Peter > thanks a lot for this. Is it possible for you > to build the package for FC9 & FC10 targets. This way > I could try getting more csound users to test it. > Regards > Victor > > ----- Original Message ----- > From: Peter Robinson > Date: Tuesday, May 12, 2009 11:38 am > Subject: csound on Fedora 11/rawhide > To: Devel's in the Details , fedora-olpc-list > > >> Hello All, >> >> I've just updated csound in F12/devel to csoound 5.10.1 and >> would like >> to get some testing (don't worry about the F12 designator its >> the same >> as F11) to see if everything works as expected for the OLPC >> stuff so >> when I push it into F11 and ultimately we can obsolete olpcsound. >> >> You can find a build to play with here >> http://koji.fedoraproject.org/koji/taskinfo?taskID=1349952 >> >> Peter >> _______________________________________________ >> Devel mailing list >> Devel at lists.laptop.org >> http://lists.laptop.org/listinfo/devel > > Dr Victor Lazzarini, Senior Lecturer, Dept. of Music,National University of > Ireland, Maynooth > > From cjb at laptop.org Tue May 12 15:56:17 2009 From: cjb at laptop.org (Chris Ball) Date: Tue, 12 May 2009 11:56:17 -0400 Subject: fedora-xo 20090511 does not start X server In-Reply-To: (Chris Ball's message of "Mon, 11 May 2009 20:59:42 -0400") References: <4A08AE33.7070709@bga.com> Message-ID: Hi, > xorg-x11-server-1.6.1-7.fc11 works, xorg-x11-server-1.6.1-8.fc11 > introduced the crash. I'll update the bug. This is fixed now, but won't be in a build until tomorrow. For the impatient, run: rpm -Uvh http://kojipkgs.fedoraproject.org/packages/xorg-x11-drv-geode/2.11.2/1.fc11/i586/xorg-x11-drv-geode-2.11.2-1.fc11.i586.rpm X works fine now. We just need to work on what's going on with this mounting jffs2 bug. - Chris. -- Chris Ball From pbrobinson at gmail.com Tue May 12 16:05:34 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Tue, 12 May 2009 17:05:34 +0100 Subject: fedora-xo 20090511 does not start X server In-Reply-To: References: <4A08AE33.7070709@bga.com> Message-ID: <5256d0b0905120905p74470240mf84863dd065da517@mail.gmail.com> Hi, > ? > xorg-x11-server-1.6.1-7.fc11 works, xorg-x11-server-1.6.1-8.fc11 > ? > introduced the crash. ?I'll update the bug. > > This is fixed now, but won't be in a build until tomorrow. ?For the > impatient, run: > > rpm -Uvh http://kojipkgs.fedoraproject.org/packages/xorg-x11-drv-geode/2.11.2/1.fc11/i586/xorg-x11-drv-geode-2.11.2-1.fc11.i586.rpm Great news. BTW to fix your commit rights to the package you can go to https://admin.fedoraproject.org/pkgdb/packages/name/xorg-x11-drv-geode and request to get added to the package. The maintainer can then auth your access. Cheers, Peter From mikus at bga.com Tue May 12 16:58:26 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Tue, 12 May 2009 12:58:26 -0400 Subject: csound on Fedora 11/rawhide Message-ID: <4A09AAB2.7070500@bga.com> > I've just updated csound in F12/devel to csoound 5.10.1 and would like > to get some testing (don't worry about the F12 designator its the same > as F11) to see if everything works as expected for the OLPC stuff so > when I push it into F11 and ultimately we can obsolete olpcsound. plus > I understand that the sound on rawhide won't work. I was more wanting > confirmation that things like TamTam (or whatever the Activities are > that use csound) work and don't crash (well, work as well as can be > expected with no sound..... so don't crash etc). ok. It's Activities you are asking about. [Activities are at the bottom of my priority list.] Since the latest rawhide-xo on which Sugar runs for me is 20090426, that's where I tested: Speak runs with both old 'olpcsound' and with new 'csound'. Am attaching (edited) log outputs from both cases. [Also ran on both: Distance; Words.] -------- Some Activities which make use of sound did *not* run (launch) with 'olpcsound'. I then removed olpcsound and installed csound (plus fltk). Now the four TamTams, and Bounce, launched successfully. By the way, it appeared to be a two-step process for me to stop an Activity using 'csound'. First I clicked on the "stop sign" in the activity's header. After waiting a bit, I intervened and clicked on 'Stop' in the drop-down palette of the activity's icon in Frame. My opinion: csound works on rawhide-xo. Of course, rawhide-xo sound itself does not yet work. mikus -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Speak.logo URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Speak.logn URL: From mikus at bga.com Tue May 12 17:05:42 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Tue, 12 May 2009 13:05:42 -0400 Subject: fedora-xo 20090510 does not start X server Message-ID: <4A09AC66.6000001@bga.com> > Can you try adding a "rootdelay=300" to the kernel command line in the > olpc.fth and see if that helps the issues your seeing when you have > usb devices attached? The machine which had for several weeks not given me any "unknown filesystem" errors on boot-up, now gave me a bunch -- despite having "rootdelay=300" (and later, '500') in olpc.fth. I just kept repeating the power-on sequence, until the boot-up finally worked. mikus From mikus at bga.com Tue May 12 19:12:39 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Tue, 12 May 2009 15:12:39 -0400 Subject: fedora-xo 20090512 does not boot on XO-1 Message-ID: <4A09CA27.4000303@bga.com> Using 'copy-nand', flashed ~cjb/fedora-xo 20090512.img to nand. Booted. A number of boot-time messages scrolled by; then the the XO-1 became completely unresponsive (i.e., the boot process stalled) - only thing that would work was the button to power off. Last "boot-up" line that was output was "creating devices" mikus From pbrobinson at gmail.com Tue May 12 21:29:14 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Tue, 12 May 2009 22:29:14 +0100 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <4A09AC66.6000001@bga.com> References: <4A09AC66.6000001@bga.com> Message-ID: <5256d0b0905121429s29fc4e42gf2f90448b72627a0@mail.gmail.com> >> Can you try adding a "rootdelay=300" to the kernel command line in the >> olpc.fth and see if that helps the issues your seeing when you have >> usb devices attached? > > The machine which had for several weeks not given me any "unknown > filesystem" errors on boot-up, now gave me a bunch -- despite having > "rootdelay=300" (and later, '500') in olpc.fth. ?I just kept repeating the > power-on sequence, until the boot-up finally worked. I don't think the rootdelay is the issue then. It would fix it or not. Back to the drawing board for this. Peter From cjb at laptop.org Tue May 12 21:55:27 2009 From: cjb at laptop.org (Chris Ball) Date: Tue, 12 May 2009 17:55:27 -0400 Subject: fedora-xo 20090510 does not start X server In-Reply-To: <5256d0b0905121429s29fc4e42gf2f90448b72627a0@mail.gmail.com> (Peter Robinson's message of "Tue, 12 May 2009 22:29:14 +0100") References: <4A09AC66.6000001@bga.com> <5256d0b0905121429s29fc4e42gf2f90448b72627a0@mail.gmail.com> Message-ID: Hi, > I don't think the rootdelay is the issue then. It would fix it or > not. Back to the drawing board for this. The module-load-and-mount both happen as part of the mount operation in the kernel, and rootdelay just delays when that operation occurs. The problem seems to be a race between those two things, even though that shouldn't be possible. - Chris. -- Chris Ball From mikus at bga.com Tue May 12 22:12:05 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Tue, 12 May 2009 18:12:05 -0400 Subject: fedora-xo 20090511 does not start X server Message-ID: <4A09F435.6070808@bga.com> > rpm -Uvh http://kojipkgs.fedoraproject.org/packages/xorg-x11-drv-geode/2.11.2/1 > .fc11/i586/xorg-x11-drv-geode-2.11.2-1.fc11.i586.rpm With rawhide-xo 20090512, I just kept on repeating the power-on sequence over and over. On about the 11th try (I kid you not), the boot process got past the 'creating devices' point and let me log in at a text console. Applied the geode .rpm. After another eight or so repeats of the power-on sequence (I expect the "power" button will soon become worn out), the updated system booted -- and I had X (and thereby Sugar and Gnome) in 20090512 (from nand) on the XO-1. Chris, thank you mikus From mikus at bga.com Wed May 13 01:49:24 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Tue, 12 May 2009 21:49:24 -0400 Subject: yum has problems with proxy Message-ID: <4A0A2724.3010107@bga.com> On my XO, I am a heavy user of yum. The communications setup at my home goes through a proxy to reach the internet. I have empirically seen that when told to go from the XO to the net through a proxy, 'opera' handles "https:" URLs correctly, whereas 'yum' does not. Package 'fedora-release-10.92' (included in rawhide-xo) set up one yum control file ('fedora-rawhide.repo') with "https:" URLs. I had to go into that file and manually change those URLs to "http:" in order for my XO to be able to use yum from home. Now the upcoming package 'fedora-release-11' sets up __all__ the yum control files with "https:" URLs -- I now had to modify four .repo files to be able to use yum in f11 on the XO when I'm at home. mikus From cjb at laptop.org Wed May 13 02:56:44 2009 From: cjb at laptop.org (Chris Ball) Date: Tue, 12 May 2009 22:56:44 -0400 Subject: yum has problems with proxy In-Reply-To: <4A0A2724.3010107@bga.com> (Mikus Grinbergs's message of "Tue, 12 May 2009 21:49:24 -0400") References: <4A0A2724.3010107@bga.com> Message-ID: Hi, > On my XO, I am a heavy user of yum. The communications setup at > my home goes through a proxy to reach the internet. I have > empirically seen that when told to go from the XO to the net > through a proxy, opera' handles "https:" URLs correctly, whereas > 'yum' does not. Have you tried:? # export https_proxy=http://yourproxy:port/ # yum ... - Chris. -- Chris Ball From NoiseEHC at freemail.hu Wed May 13 05:47:23 2009 From: NoiseEHC at freemail.hu (NoiseEHC) Date: Wed, 13 May 2009 07:47:23 +0200 Subject: fedora-xo 20090510 does not start X server In-Reply-To: References: <4A09AC66.6000001@bga.com> <5256d0b0905121429s29fc4e42gf2f90448b72627a0@mail.gmail.com> Message-ID: <4A0A5EEB.5080207@freemail.hu> It can be stupid but I had a boot problem last year and it turned out to be a bad block in the flash. Can it be that one block of the flash can be read unreliably? Chris Ball wrote: > Hi, > > > I don't think the rootdelay is the issue then. It would fix it or > > not. Back to the drawing board for this. > > The module-load-and-mount both happen as part of the mount operation > in the kernel, and rootdelay just delays when that operation occurs. > The problem seems to be a race between those two things, even though > that shouldn't be possible. > > - Chris. > From bochecha at fedoraproject.org Wed May 13 09:07:21 2009 From: bochecha at fedoraproject.org (Mathieu Bridon (bochecha)) Date: Wed, 13 May 2009 11:07:21 +0200 Subject: yum has problems with proxy In-Reply-To: References: <4A0A2724.3010107@bga.com> Message-ID: <2d319b780905130207g7fa66b39p5b1256fd14c8e573@mail.gmail.com> Hi, > ? > On my XO, I am a heavy user of yum. ?The communications setup at > ? > my home goes through a proxy to reach the internet. ?I have > ? > empirically seen that when told to go from the XO to the net > ? > through a proxy, opera' handles "https:" URLs correctly, whereas > ? > 'yum' does not. > > Have you tried:? > > # export https_proxy=http://yourproxy:port/ > # yum ... Last time I checked (was Fedora 10), yum didn't use the env var for proxies. Instead, you should modify the /etc/yum.conf file. Add the following : proxy=http://your_proxy/ proxy_username=userxxx proxy_password=passxxx ---------- Mathieu Bridon (bochecha) From mikus at bga.com Wed May 13 18:24:01 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Wed, 13 May 2009 14:24:01 -0400 Subject: yum has problems with proxy Message-ID: <4A0B1041.1070604@bga.com> Thanks, folks, for your suggestions. > modify the /etc/yum.conf file. Add the following : > proxy=3Dhttp://your_proxy/ Did not get that working on my XO. [The error message changed.] Without modifying /etc/yum.conf, the error message I get is: | Could not get metalink https://mirrors.fedoraproject.org/metalink?repo=rawhide&arch=i386 error was | [Errno 4] IOError: # export https_proxy=3Dhttp://yourproxy:port/ Interestingly enough, this is currently working on my 20090512 XO. The trick seems to lie in re-orienting the application (yum) to use 'http' where the application control file has specified 'https'. Thanks, mikus From mikus at bga.com Thu May 14 19:09:45 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Thu, 14 May 2009 15:09:45 -0400 Subject: fedora-xo 20090514 has great difficulty booting on XO-1 Message-ID: <4A0C6C79.80701@bga.com> Using 'copy-nand', flashed ~cjb/fedora-xo 20090514.img to nand. Booted. A number of boot-time messages scrolled by; then the the XO-1 became completely unresponsive (i.e., the boot process stalled) - only thing that would work was the button to power off. Last "boot-up" line that was output was "creating devices" Had to repeat the power-on sequence five times before the boot-up process happened to get past the "creating devices" point. mikus From cjb at laptop.org Thu May 14 20:19:35 2009 From: cjb at laptop.org (Chris Ball) Date: Thu, 14 May 2009 16:19:35 -0400 Subject: fedora-xo 20090514 has great difficulty booting on XO-1 In-Reply-To: <4A0C6C79.80701@bga.com> (Mikus Grinbergs's message of "Thu, 14 May 2009 15:09:45 -0400") References: <4A0C6C79.80701@bga.com> Message-ID: Hi, > Last "boot-up" line that was output was "creating devices" This is the same as the "unknown filesystem type" error (which we still don't know how to fix), but with more verbose output. - Chris. -- Chris Ball From cjb at laptop.org Fri May 15 20:23:23 2009 From: cjb at laptop.org (Chris Ball) Date: Fri, 15 May 2009 16:23:23 -0400 Subject: OLPC's XO-1.5 software plan. Message-ID: We have some good news: OLPC has decided to base its software release for the new XO-1.5 laptop on Fedora 11. Unlike previous releases, we plan to use a full Fedora desktop build, booting into Sugar but giving users the option to switch into a standard GNOME install instead. (This will mostly be useful for older kids in high school.) I'm particularly happy about this plan because it will allow us to catch up with the awesome work present in the Sugar community's most recent release, Sugar 0.84, as well as merging the latest Fedora work and including GNOME into the mix for the first time. The new machines will have 1GB of RAM and 4GB of flash, so we have enough room for both environments at once. We think we'll need to use our own kernel and initrd, but the other base packages we expect to need are present in Fedora already, including Sugar; in fact, we already have an F11+Sugar+GNOME build for the XO-1 using pure Fedora packages. That build will get better as a result of this work (although OLPC's focus will be on getting the XO-1.5 running) and it will form the basis for the XO-1.5 build. If you're interested in contributing, we'd certainly love your help, and you can find us on the fedora-olpc mailing list?, and freenode IRC's #fedora-olpc channel. Our existing F11 build images for the XO-1 are here?, and we'll soon begin publishing images for the XO-1.5 too. XO-1.5 beta machines will start to be manufactured over the next few months, and will be available to contributors as part of our Contributors Program? once the hardware's up and running. Finally, thanks are due to the volunteer Fedora packagers and testers who helped us get to the point of being able to commit to Fedora 11 for this new build, in particular: Fabian Affolter, Kushal Das, Greg DeKoenigsberg, Martin Dengler, Scott Douglass, Sebastian Dziallas, Mikus Grinbergs, Bryan Kearney, Gary C. Martin, Steven M. Parrish, and Peter Robinson. Thanks! - Chris, for the OLPC techteam. ?: http://www.redhat.com/mailman/listinfo/fedora-olpc-list ?: http://dev.laptop.org/~cjb/rawhide-xo/ ?: http://wiki.laptop.org/go/Contributors_program -- Chris Ball From pbrobinson at gmail.com Sat May 16 08:35:16 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sat, 16 May 2009 09:35:16 +0100 Subject: OLPC's XO-1.5 software plan. In-Reply-To: References: Message-ID: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> Hi Chris, This is great news! A couple of queries about this. What are the plans with getting existing kernel patches upstream? What is the status of rainbow within this? Of the difference between olpc-4 and mainline packages, patches to various packages for rainbow support are now by far the biggest part of the remaining difference. Also for the initrd, is dracut of use in that regard? [1][2] Cheers, Peter [1] https://fedoraproject.org/wiki/Features/Dracut [2] http://lwn.net/Articles/317793/ On Fri, May 15, 2009 at 9:23 PM, Chris Ball wrote: > We have some good news: ?OLPC has decided to base its software release > for the new XO-1.5 laptop on Fedora 11. ?Unlike previous releases, we > plan to use a full Fedora desktop build, booting into Sugar but giving > users the option to switch into a standard GNOME install instead. > (This will mostly be useful for older kids in high school.) > > I'm particularly happy about this plan because it will allow us to > catch up with the awesome work present in the Sugar community's most > recent release, Sugar 0.84, as well as merging the latest Fedora work > and including GNOME into the mix for the first time. ?The new machines > will have 1GB of RAM and 4GB of flash, so we have enough room for both > environments at once. > > We think we'll need to use our own kernel and initrd, but the other > base packages we expect to need are present in Fedora already, > including Sugar; in fact, we already have an F11+Sugar+GNOME build > for the XO-1 using pure Fedora packages. ?That build will get better > as a result of this work (although OLPC's focus will be on getting > the XO-1.5 running) and it will form the basis for the XO-1.5 build. > > If you're interested in contributing, we'd certainly love your help, > and you can find us on the fedora-olpc mailing list?, and freenode > IRC's #fedora-olpc channel. ?Our existing F11 build images for the > XO-1 are here?, and we'll soon begin publishing images for the XO-1.5 > too. ?XO-1.5 beta machines will start to be manufactured over the next > few months, and will be available to contributors as part of our > Contributors Program? once the hardware's up and running. > > Finally, thanks are due to the volunteer Fedora packagers and testers > who helped us get to the point of being able to commit to Fedora 11 > for this new build, in particular: Fabian Affolter, Kushal Das, Greg > DeKoenigsberg, Martin Dengler, Scott Douglass, Sebastian Dziallas, > Mikus Grinbergs, Bryan Kearney, Gary C. Martin, Steven M. Parrish, > and Peter Robinson. ?Thanks! > > - Chris, for the OLPC techteam. > > ?: ?http://www.redhat.com/mailman/listinfo/fedora-olpc-list > ?: ?http://dev.laptop.org/~cjb/rawhide-xo/ > ?: ?http://wiki.laptop.org/go/Contributors_program > > -- > Chris Ball ? > > _______________________________________________ > Fedora-olpc-list mailing list > Fedora-olpc-list at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-olpc-list > From pbrobinson at gmail.com Sat May 16 10:26:13 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sat, 16 May 2009 11:26:13 +0100 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> Message-ID: <5256d0b0905160326i555bf9b5q7553aa9e1b3f4e13@mail.gmail.com> Oh, and the other query I forgot to add was about moving to 802.11s for the mesh networking? Peter On Sat, May 16, 2009 at 9:35 AM, Peter Robinson wrote: > Hi Chris, > > This is great news! > > A couple of queries about this. > > What are the plans with getting existing kernel patches upstream? > > What is the status of rainbow within this? Of the difference between > olpc-4 and mainline packages, patches to various packages for rainbow > support are now by far the biggest part of the remaining difference. > > Also for the initrd, is dracut of use in that regard? [1][2] > > Cheers, > Peter > > [1] https://fedoraproject.org/wiki/Features/Dracut > [2] http://lwn.net/Articles/317793/ > > On Fri, May 15, 2009 at 9:23 PM, Chris Ball wrote: >> We have some good news: ?OLPC has decided to base its software release >> for the new XO-1.5 laptop on Fedora 11. ?Unlike previous releases, we >> plan to use a full Fedora desktop build, booting into Sugar but giving >> users the option to switch into a standard GNOME install instead. >> (This will mostly be useful for older kids in high school.) >> >> I'm particularly happy about this plan because it will allow us to >> catch up with the awesome work present in the Sugar community's most >> recent release, Sugar 0.84, as well as merging the latest Fedora work >> and including GNOME into the mix for the first time. ?The new machines >> will have 1GB of RAM and 4GB of flash, so we have enough room for both >> environments at once. >> >> We think we'll need to use our own kernel and initrd, but the other >> base packages we expect to need are present in Fedora already, >> including Sugar; in fact, we already have an F11+Sugar+GNOME build >> for the XO-1 using pure Fedora packages. ?That build will get better >> as a result of this work (although OLPC's focus will be on getting >> the XO-1.5 running) and it will form the basis for the XO-1.5 build. >> >> If you're interested in contributing, we'd certainly love your help, >> and you can find us on the fedora-olpc mailing list?, and freenode >> IRC's #fedora-olpc channel. ?Our existing F11 build images for the >> XO-1 are here?, and we'll soon begin publishing images for the XO-1.5 >> too. ?XO-1.5 beta machines will start to be manufactured over the next >> few months, and will be available to contributors as part of our >> Contributors Program? once the hardware's up and running. >> >> Finally, thanks are due to the volunteer Fedora packagers and testers >> who helped us get to the point of being able to commit to Fedora 11 >> for this new build, in particular: Fabian Affolter, Kushal Das, Greg >> DeKoenigsberg, Martin Dengler, Scott Douglass, Sebastian Dziallas, >> Mikus Grinbergs, Bryan Kearney, Gary C. Martin, Steven M. Parrish, >> and Peter Robinson. ?Thanks! >> >> - Chris, for the OLPC techteam. >> >> ?: ?http://www.redhat.com/mailman/listinfo/fedora-olpc-list >> ?: ?http://dev.laptop.org/~cjb/rawhide-xo/ >> ?: ?http://wiki.laptop.org/go/Contributors_program >> >> -- >> Chris Ball ? >> >> _______________________________________________ >> Fedora-olpc-list mailing list >> Fedora-olpc-list at redhat.com >> https://www.redhat.com/mailman/listinfo/fedora-olpc-list >> > From dsd at laptop.org Sat May 16 14:35:08 2009 From: dsd at laptop.org (Daniel Drake) Date: Sat, 16 May 2009 10:35:08 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> Message-ID: <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> To the best of my knowledge: 2009/5/16 Peter Robinson : > What are the plans with getting existing kernel patches upstream? Deepak is working on it, and me and Chris agree that getting *all* XO-1.5 kernel stuff upstream early on should be a sensible and achievable goal. But time will tell, and it seems certain that we'll be using a forked kernel tree during development. > What is the status of rainbow within this? Of the difference between > olpc-4 and mainline packages, patches to various packages for rainbow > support are now by far the biggest part of the remaining difference. We'll start without it, since thats how Fedora 11 will ship. It could possibly be added later, based on time and other things... but if we have to fork a lot of packages, that would be a significant vote against it. > Also for the initrd, is dracut of use in that regard? [1][2] We discussed this already, and agreed in principle that it would be a good target for upstreaming OLPC's antitheft stuff... but Chris was loosely under the belief that dracut isn't being developed further. Is that correct? > Oh, and the other query I forgot to add was about moving to 802.11s for the mesh networking? I haven't heard any plans along these lines. Daniel From bobbypowers at gmail.com Sat May 16 15:56:52 2009 From: bobbypowers at gmail.com (Bobby Powers) Date: Sat, 16 May 2009 11:56:52 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> Message-ID: <23e2e54b0905160856n36fa2069xd65933c838fba61e@mail.gmail.com> On Sat, May 16, 2009 at 10:35 AM, Daniel Drake wrote: > To the best of my knowledge: > > 2009/5/16 Peter Robinson : >> What are the plans with getting existing kernel patches upstream? > > Deepak is working on it, and me and Chris agree that getting *all* > XO-1.5 kernel stuff upstream early on should be a sensible and > achievable goal. But time will tell, and it seems certain that we'll > be using a forked kernel tree during development. > >> What is the status of rainbow within this? Of the difference between >> olpc-4 and mainline packages, patches to various packages for rainbow >> support are now by far the biggest part of the remaining difference. > > We'll start without it, since thats how Fedora 11 will ship. It could > possibly be added later, based on time and other things... but if we > have to fork a lot of packages, that would be a significant vote > against it. > >> Also for the initrd, is dracut of use in that regard? [1][2] > > We discussed this already, and agreed in principle that it would be a > good target for upstreaming OLPC's antitheft stuff... but Chris was > loosely under the belief that dracut isn't being developed further. Is > that correct? I found git://dracut.git.sourceforge.net/gitroot/dracut from: http://fedoraproject.org/wiki/Features/Dracut While development seemed to have stalled for a while, there's been more than a dozen commits in the past week. bp >> Oh, and the other query I forgot to add was about moving to 802.11s for the mesh networking? > > I haven't heard any plans along these lines. > > Daniel > > _______________________________________________ > Fedora-olpc-list mailing list > Fedora-olpc-list at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-olpc-list > From pbrobinson at gmail.com Sat May 16 18:09:52 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sat, 16 May 2009 19:09:52 +0100 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> Message-ID: <5256d0b0905161109h176b2379u3d95e508d1845d37@mail.gmail.com> >> Also for the initrd, is dracut of use in that regard? [1][2] > > We discussed this already, and agreed in principle that it would be a > good target for upstreaming OLPC's antitheft stuff... but Chris was > loosely under the belief that dracut isn't being developed further. Is > that correct? Incorrect :) It will be used for the initrd, or is at least proposed at this point, Fedora 12. See the fedoraproject page I linked above. As a result will no doubt see massive development by the RedHat Fedora devs over the coming months as a result, so might be the perfect timing from the OLPC perspective as I know its had some testing on F11 and as some have plans for it to be the kernel initrd across all distros it could be a worthwhile dev target as it would be upstream moving forward and hence be less maintenance going forward. Peter From pbrobinson at gmail.com Sat May 16 18:16:31 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Sat, 16 May 2009 19:16:31 +0100 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> Message-ID: <5256d0b0905161116r449a592at8b7799ca0fb1a3f1@mail.gmail.com> >> Oh, and the other query I forgot to add was about moving to 802.11s for the mesh networking? > > I haven't heard any plans along these lines. See http://www.open80211s.org/ for more details. I believe its now upstream and supports more chipsets than just the libertas and support is now in the mac80211 in mainline kernel along with a number of drivers. Peter From dsd at laptop.org Sat May 16 19:47:18 2009 From: dsd at laptop.org (Daniel Drake) Date: Sat, 16 May 2009 15:47:18 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <5256d0b0905161116r449a592at8b7799ca0fb1a3f1@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> <5256d0b0905161116r449a592at8b7799ca0fb1a3f1@mail.gmail.com> Message-ID: <818423da0905161247s7973d1e6na3a13876190caf87@mail.gmail.com> 2009/5/16 Peter Robinson : > See http://www.open80211s.org/ for more details. I believe its now > upstream and supports more chipsets than just the libertas and support > is now in the mac80211 in mainline kernel along with a number of > drivers. The XO wireless hardware is somewhat unusual and the driver doesn't use those kernel implementations (because the equivalent happens in the firmware/hardware). So any changes in this area would need firmware changes if not more.. the functionality/operation is not influenced by developments in that layer of the kernel. Daniel From bobbypowers at gmail.com Sat May 16 20:38:51 2009 From: bobbypowers at gmail.com (Bobby Powers) Date: Sat, 16 May 2009 16:38:51 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <818423da0905161247s7973d1e6na3a13876190caf87@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> <5256d0b0905161116r449a592at8b7799ca0fb1a3f1@mail.gmail.com> <818423da0905161247s7973d1e6na3a13876190caf87@mail.gmail.com> Message-ID: <23e2e54b0905161338q405fad31h54513ba8f2b287bb@mail.gmail.com> On Sat, May 16, 2009 at 3:47 PM, Daniel Drake wrote: > 2009/5/16 Peter Robinson : >> See http://www.open80211s.org/ for more details. I believe its now >> upstream and supports more chipsets than just the libertas and support >> is now in the mac80211 in mainline kernel along with a number of >> drivers. > > The XO wireless hardware is somewhat unusual and the driver doesn't > use those kernel implementations (because the equivalent happens in > the firmware/hardware). So any changes in this area would need > firmware changes if not more.. the functionality/operation is not > influenced by developments in that layer of the kernel. The original announcement [1] mentioned that, at least initially, the new wireless wouldn't support 802.11s, but that OLPC was 'exploring options'. Would mesh be something that could be added in the driver, absent it being added to the firmware by marvell or a contractor? I don't know how any kind of generic 802.11s could interact with the rest of the libertas wireless stack. bobby 1 - http://lists.laptop.org/pipermail/devel/2009-April/024127.html > Daniel > > _______________________________________________ > Fedora-olpc-list mailing list > Fedora-olpc-list at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-olpc-list > From dsd at laptop.org Sat May 16 20:42:00 2009 From: dsd at laptop.org (Daniel Drake) Date: Sat, 16 May 2009 16:42:00 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <23e2e54b0905161338q405fad31h54513ba8f2b287bb@mail.gmail.com> References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> <5256d0b0905161116r449a592at8b7799ca0fb1a3f1@mail.gmail.com> <818423da0905161247s7973d1e6na3a13876190caf87@mail.gmail.com> <23e2e54b0905161338q405fad31h54513ba8f2b287bb@mail.gmail.com> Message-ID: <818423da0905161342n3b9f00ffi8334f30294f863be@mail.gmail.com> 2009/5/16 Bobby Powers : > The original announcement [1] mentioned that, at least initially, the > new wireless wouldn't support 802.11s, but that OLPC was 'exploring > options'. I think these are 2 different questions. Peter asked: Can the new system move away from nearly-802.11s-mesh to real-802.11s-mesh? Whereas the announcement says that mesh (in any form) won't work on early production models but will be implemented later. Daniel From cjb at laptop.org Sat May 16 21:50:05 2009 From: cjb at laptop.org (Chris Ball) Date: Sat, 16 May 2009 17:50:05 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <818423da0905161247s7973d1e6na3a13876190caf87@mail.gmail.com> (Daniel Drake's message of "Sat, 16 May 2009 15:47:18 -0400") References: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> <818423da0905160735j6072b8ebuceccc50f7351e3db@mail.gmail.com> <5256d0b0905161116r449a592at8b7799ca0fb1a3f1@mail.gmail.com> <818423da0905161247s7973d1e6na3a13876190caf87@mail.gmail.com> Message-ID: Hi, > The XO wireless hardware is somewhat unusual and the driver > doesn't use those kernel implementations (because the equivalent > happens in the firmware/hardware). So any changes in this area > would need firmware changes if not more.. the functionality / > operation is not influenced by developments in that layer of the > kernel. Actually, that's not quite true -- my understanding is that the libertas_tf thin firmware *does* provide a mac80211 driver. Won't work with the main CPU off, though, which I think was your main point. http://linuxwireless.org/en/users/Drivers/libertastf - Chris. -- Chris Ball From michael at laptop.org Sat May 16 21:59:11 2009 From: michael at laptop.org (Michael Stone) Date: Sat, 16 May 2009 17:59:11 -0400 Subject: OLPC's XO-1.5 software plan. In-Reply-To: <5256d0b0905160135m1be16b0emdc104240b4f921af@mail.gmail.com> Message-ID: <20090516215911.GA9268@heat> > What is the status of rainbow within this? Do you mean "rainbow", "Sugar activity isolation", or something broader still? If you actually mean "rainbow", then you could you please to review http://wiki.laptop.org/go/Rainbow and ask a more specific question, so that I can answer it there? (I believe that I've updated it several times since we last spoke...) If, on the other hand, you're really asking about "Sugar activity isolation", or, more generally, "Sugar security", then you're in a bit of a bind... because, as far as I can tell, there is no consensus within the Sugar community about what security model defines the minimum standard of security that Sugar itself must provide by default in all settings in which it is deployed. (I, understandably, advocate for a far-reaching minimum standard.) Evidence: * "activity isolation", as implemented in sugar-0.82.* in terms or rainbow, was removed from sugar-0.84 without, so far as I can tell, much mention * no public replies: http://lists.sugarlabs.org/archive/sugar-devel/2009-May/014273.html http://lists.sugarlabs.org/archive/sugar-devel/2009-April/013732.html http://lists.sugarlabs.org/archive/sugar-devel/2009-March/012387.html * lots of replies, no action in the subsequent three months in response to my documentation improvements and further queries listed above: http://lists.sugarlabs.org/archive/sugar-devel/2009-February/012187.html Finally, if you mean "something broader" like, say, "what security features do OLPC's customers require (sic. desire) from their distro?", then could you please elaborate? Regards, Michael From martin.langhoff at gmail.com Mon May 18 12:48:46 2009 From: martin.langhoff at gmail.com (Martin Langhoff) Date: Mon, 18 May 2009 14:48:46 +0200 Subject: The XO-1.5 software plan. In-Reply-To: References: Message-ID: <46a038f90905180548k5c982f2ag208b720b1574ae73@mail.gmail.com> On Fri, May 15, 2009 at 10:17 PM, Chris Ball wrote: > We think we'll need to use our own kernel and initrd, but the other > base packages we expect to need are present in Fedora already, One area we'll also need help with is the "under a tree" networking scenario. If you've used an XO, you know what it works like: by default the OS automatically forms an ad-hoc network between the machines present using wifi but not relying on an AP. People refer to this as 'mesh' colloquially but it doesn't actually require 802.11s (as long as all the XOs are nearby). In theory at least. In practice, the ad-hoc network facility is tied to our use of a patched NM and our 'msh0' devices. The current plans don't include using 802.11s, and there are hopes to ship a more vanilla NM. This means that the 'under a tree' scenario needs help in NM integration and a bit of elbow grease. Ad-hoc networks can work pretty well for small numbers of nodes -- I suspect that that Fedora users (specially laptop users) would benefit from an easy way to run an ad-hoc network amongst machines, without the need of a 'hostap'-able driver. Cerebro has interesting code in this area -- a more ambitious goal would be to integrate it into our stack, as it can mimic some of the 802.11s mesh behaviour. But even without magic routing and path discovery, small ad hoc networks can and do work. cheers, m -- martin.langhoff at gmail.com martin at laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff From pbrobinson at gmail.com Mon May 18 14:05:45 2009 From: pbrobinson at gmail.com (Peter Robinson) Date: Mon, 18 May 2009 15:05:45 +0100 Subject: The XO-1.5 software plan. In-Reply-To: <46a038f90905180548k5c982f2ag208b720b1574ae73@mail.gmail.com> References: <46a038f90905180548k5c982f2ag208b720b1574ae73@mail.gmail.com> Message-ID: <5256d0b0905180705g1ff997caoecc3e23b9ce4566b@mail.gmail.com> >> We think we'll need to use our own kernel and initrd, but the other >> base packages we expect to need are present in Fedora already, > > One area we'll also need help with is the "under a tree" networking scenario. > > If you've used an XO, you know what it works like: by default the OS > automatically forms an ad-hoc network between the machines present > using wifi but not relying on an AP. People refer to this as 'mesh' > colloquially but it doesn't actually require 802.11s (as long as all > the XOs are nearby). > > In theory at least. In practice, the ad-hoc network facility is tied > to our use of a patched NM and our 'msh0' devices. > > The current plans don't include using 802.11s, and there are hopes to > ship a more vanilla NM. This means that the 'under a tree' scenario > needs help in NM integration and a bit of elbow grease. > > Ad-hoc networks can work pretty well for small numbers of nodes -- I > suspect that that Fedora users (specially laptop users) would benefit > from an easy way to run an ad-hoc network amongst machines, without > the need of a 'hostap'-able driver. NetworkManager 0.7 supports Ad-hoc wifi networks quite well. I've used it on a number of occasions to share my 3G dongle between a number of users. In that regard I suspect all that's needed is to be able to have some form of gui for it. > Cerebro has interesting code in this area -- a more ambitious goal > would be to integrate it into our stack, as it can mimic some of the > 802.11s mesh behaviour. But even without magic routing and path > discovery, small ad hoc networks can and do work. Do you have a link for Cerebro? Peter From martin.langhoff at gmail.com Tue May 19 13:08:25 2009 From: martin.langhoff at gmail.com (Martin Langhoff) Date: Tue, 19 May 2009 15:08:25 +0200 Subject: Help with testing a patch for xulrunner... Message-ID: <46a038f90905190608v12f331aao6e87e2c716e6f330@mail.gmail.com> Hi list, while working on a webbased project (as part of the OLPC School Server), we're hitting on a performance bug / oddity with Browse.xo (which is the sugarised face of xulrunner on the XO) on the 8.2 release. Rob O'Callahan proposes we test a patch (attached) to see if things get better. Can anyone help with this? The xulrunner we ship on 8.2 has some OLPC-specific patches, and I'd hope to add this patch to the pile. I couldn't find the spec or srpm, but here's the Koji task. Presumably it's retrievable from there? http://koji.fedoraproject.org/koji/taskinfo?taskID=783777 cheers, martin ---------- Forwarded message ---------- From: Robert O'Callahan Date: Fri, May 15, 2009 at 1:07 PM Subject: Re: Browse.xo performance & resolution - Hulahop 200dpi vs Browse 134dpi To: Mihai Sucan Cc: Martin Langhoff On Fri, May 15, 2009 at 10:54 PM, Mihai Sucan wrote: > > Le Fri, 15 May 2009 13:22:15 +0300, Robert O'Callahan a ?crit: > >> Another thing we could do is implement what I suggested here: >> https://bugzilla.mozilla.org/show_bug.cgi?id=394103#c53 >> That would let you set the dpi to the correct value (so physical units, >> including fonts in pt, display correctly) but force images and canvases (and >> other CSS px units) to not be scaled. They'd be too small, but fast :-). > > This is what would be best. OK, I'll do that. > Another idea is to have some CSS property to disable scaling entirely for the canvas elements I want. I don't really want to do that, because I don't want Web developers to write applications that depend on zoom or dpi. If you're willing to assume you're 200dpi, you can just write ... suboptimal though. I don't know if you can apply a patch to Gecko, but if you can, the attached patch would let us render the in a local surface instead of using X. That should provide a speedup in your benchmarks, where X isn't giving much hw acceleration and you're doing a large amount of drawing operations and hardly any displaying to the screen. I'd be interested to hear what effect it has. However, in real-world usage I imagine the canvas is drawn to the screen far more often compared to the number of drawing operations, so this patch might not make sense for actual users. Rob -- "He was pierced for our transgressions, he was crushed for our iniquities; the punishment that brought us peace was upon him, and by his wounds we are healed. We all, like sheep, have gone astray, each of us has turned to his own way; and the LORD has laid on him the iniquity of us all." [Isaiah 53:5-6] -- martin.langhoff at gmail.com martin at laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff -------------- next part -------------- A non-text attachment was scrubbed... Name: patch Type: application/text Size: 1004 bytes Desc: not available URL: From mikus at bga.com Tue May 19 22:04:53 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Tue, 19 May 2009 18:04:53 -0400 Subject: rawhide-xo 20090519 has great difficulty booting on XO-1 Message-ID: <4A132D05.3060606@bga.com> Using 'copy-nand', flashed ~cjb/fedora-xo 20090519.img to nand. Booted. A number of boot-time messages scrolled by; then the the XO-1 became completely unresponsive (i.e., the boot process stalled) - only thing that would work was the button to power off. Last two "boot-up" lines that were output (before the stall) were: "starting udevd" "creating devices" Had to repeat the power-on sequence many many times before the boot-up process happened to get past the "creating devices" point. mikus p.s. During the boot-up process, the boot-time messages scroll by. This scrolling gets 'interrupted' by a from-top-down "erase" of the lines on the screen (as though a squeegee had been used to wipe out what had been written to the screen). This "screen cleaning" action usually occurs twice before the rawhide boot-up process gets to the "creating devices" point (where all too often the process stalls). However (and I can *not* swear to how consistent this is), in those cases where the boot-up process got past that stall point (and the system came up as it should), I've noticed that the "screen cleaning" action (i.e., the squeegeeing) occurred *three* times (not two times) before the boot-up got to that "creating devices" point. From cjb at laptop.org Mon May 25 13:43:55 2009 From: cjb at laptop.org (Chris Ball) Date: Mon, 25 May 2009 09:43:55 -0400 Subject: Daily builds are no longer livecd images Message-ID: Hi, Starting with today's 20090525 build, the images are no longer livecd images, so changes will persist across reboots. http://dev.laptop.org/~cjb/rawhide-xo/20090525/ (The sporadic "creating devices" mount hang will still be present for jffs2; there's no need to report that again.) Thanks to Sebastian for the script to make installed images! - Chris. -- Chris Ball From rms at 1407.org Mon May 25 13:54:19 2009 From: rms at 1407.org (Rui Miguel Silva Seabra) Date: Mon, 25 May 2009 14:54:19 +0100 Subject: Daily builds are no longer livecd images In-Reply-To: References: Message-ID: <20090525135419.GA14860@roque.1407.org> On Mon, May 25, 2009 at 09:43:55AM -0400, Chris Ball wrote: > Hi, > > Starting with today's 20090525 build, the images are no longer livecd > images, so changes will persist across reboots. > > http://dev.laptop.org/~cjb/rawhide-xo/20090525/ > > (The sporadic "creating devices" mount hang will still be present for > jffs2; there's no need to report that again.) > > Thanks to Sebastian for the script to make installed images! Bless you, guys! :) Rui From mikus at bga.com Mon May 25 19:36:25 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Mon, 25 May 2009 15:36:25 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A1AF339.1030803@bga.com> > Starting with today's 20090525 build, the images are no longer livecd > images, so changes will persist across reboots. > http://dev.laptop.org/~cjb/rawhide-xo/20090525/ > > Thanks to Sebastian for the script to make installed images! Chris, I'm not sure what this post is telling me. I visually compared the 20090525 subdirectory with the 20050519 subdirectory - they both contain the same number of files (of comparable size). And when I used 'copy-nand' to install the 20090525.img file, my XO-1's nand contents did not seem organized any different from when I installed 20090519.img. What does "no longer livecd images" mean ? [Having now tried 20090525, I see that directory '/home/liveuser' has now reverted to '/home/user'. Oh well - I'll just change all my customization scripts back to updating '/home/user'.] Thanks, mikus p.s. On my Ubuntu 9.04 system, I'm unable to run the 'livecd-iso-to-xo' script (mounting the squashfs fails). Could you perhaps ask Sebastian to get together with the Soas2 folks - they currently release only a .iso file, which I could only install to an USB stick (where changes didn't persist). It would be useful to me to be able to install Soas2 builds onto XO-1 nand, and be able to reboot and have changes persist from the previous time. From bochecha at fedoraproject.org Mon May 25 20:41:43 2009 From: bochecha at fedoraproject.org (Mathieu Bridon (bochecha)) Date: Mon, 25 May 2009 22:41:43 +0200 Subject: Daily builds are no longer livecd images In-Reply-To: <4A1AF339.1030803@bga.com> References: <4A1AF339.1030803@bga.com> Message-ID: <2d319b780905251341n4fee34ey5314f245e50087ba@mail.gmail.com> > What does "no longer livecd images" mean ? > > [Having now tried 20090525, I see that directory '/home/liveuser' has now > reverted to '/home/user'. ?Oh well - I'll just change all my customization > scripts back to updating '/home/user'.] It means that the default user is not liveuser any more :P No, it should mean that this is only a ? normal ? iso, not an iso you could burn on a CD and use as a live CD (if the XO had a CD drive) or burn on a USB and use as a live USB. Kind of the same difference as between the Fedora liveCD and the install DVD isos. Of course, if you install it, the result should be the same. ---------- Mathieu Bridon (bochecha) From mikus at bga.com Mon May 25 19:57:00 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Mon, 25 May 2009 15:57:00 -0400 Subject: rawhide-xo 20090525 problems Message-ID: <4A1AF80C.80503@bga.com> Using 'copy-nand', flashed ~cjb/rawhide-xo 20090525.img to nand. Have just started to customize it (have to change all references to /home/liveuser back to /home/olpc), but in the meantime have already encountered two problems that 20090519 did not have: - In Gnome (and on the logon screen), buttons that have had the focus (e.g., by being clicked on) are given a black background. [There is already a FC11 bug for this.] (Changing the focus appears to give back the appearance the button ought to have.) - In 'Terminal' (in Sugar), a click in the scrollbar gets "multiplied" -- a single click (for "scroll page up") results in the display scrolling *many* pages. [This is an old OLPC problem, which however is for me now showing up in rawhide-xo for the first time.] mikus From cjb at laptop.org Tue May 26 00:13:27 2009 From: cjb at laptop.org (Chris Ball) Date: Mon, 25 May 2009 20:13:27 -0400 Subject: Daily builds are no longer livecd images In-Reply-To: <4A1AF339.1030803@bga.com> (Mikus Grinbergs's message of "Mon, 25 May 2009 15:36:25 -0400") References: <4A1AF339.1030803@bga.com> Message-ID: Hi, > What does "no longer livecd images" mean ? On previous builds, changes to the root filesystem disappeared at reboot because the filesystem was refreshed from the same image every boot. This no longer happens, so if you (for example): # touch /foo # touch /etc/bar # reboot The files will still be there after reboot. -- Chris Ball From sebastian at when.com Tue May 26 04:56:06 2009 From: sebastian at when.com (Sebastian Dziallas) Date: Tue, 26 May 2009 06:56:06 +0200 Subject: Daily builds are no longer livecd images In-Reply-To: <4A1AF339.1030803@bga.com> References: <4A1AF339.1030803@bga.com> Message-ID: <4A1B7666.6090506@when.com> Mikus Grinbergs wrote: >> Starting with today's 20090525 build, the images are no longer livecd >> images, so changes will persist across reboots. >> http://dev.laptop.org/~cjb/rawhide-xo/20090525/ >> >> Thanks to Sebastian for the script to make installed images! > > Chris, I'm not sure what this post is telling me. > > I visually compared the 20090525 subdirectory with the 20050519 > subdirectory - they both contain the same number of files (of comparable > size). And when I used 'copy-nand' to install the 20090525.img file, my > XO-1's nand contents did not seem organized any different from when I > installed 20090519.img. > > What does "no longer livecd images" mean ? Hi, I'm jumping in here, as there's one other I would like to highlight: Before, *every* time you booted the XO with Rawhide-XO, there was a number of scripts initiated, for example to create the liveuser or to check for changed hardware. This is of course nonsense, as the XO's hardware won't change. And this means that it should be enough to run a slimmed down version of these scripts *once*. That's what we're now doing - we've kicked the livecd scripts. Obviously, those files aren't really that big, but it's still a big technical change for us. > [Having now tried 20090525, I see that directory '/home/liveuser' has > now reverted to '/home/user'. Oh well - I'll just change all my > customization scripts back to updating '/home/user'.] > > Thanks, mikus > > > p.s. On my Ubuntu 9.04 system, I'm unable to run the 'livecd-iso-to-xo' > script (mounting the squashfs fails). That's maybe some squashfs version mismatch? I'm not sure... > Could you perhaps ask Sebastian to get together with the Soas2 folks - > they currently release only a .iso file, which I could only install to > an USB stick (where changes didn't persist). It would be useful to me to > be able to install Soas2 builds onto XO-1 nand, and be able to reboot > and have changes persist from the previous time. Heh. Actually, I'm doing the SoaS builds. And yup, we release an .iso file, allowing the user to put it wherever he wants to. You can also run livecd-iso-to-xo on the .iso image and copy-nand the resulting .img file then to XO. But note that SoaS is not especially prepared for the XO. It contains more drivers than needed, as well as the livecd parts. Martin Dengler has done great work to improve its experience on the XO and those patches have been ported to fedora-xo, though! --Sebastian From sebastian at when.com Tue May 26 12:16:27 2009 From: sebastian at when.com (Sebastian Dziallas) Date: Tue, 26 May 2009 14:16:27 +0200 Subject: [SoaS] Important Schedule Changes - Please Read! Message-ID: <4A1BDD9B.4010106@when.com> Hi everybody, please read this carefully, as it concerns major schedule changes regarding our upcoming Sugar on a Stick release in the end of June. The original plan was have a Release Candidate based on F11 Final and Sugar 0.84 at that time and a Final Version later in Q3. After serious consideration, it looks way more sensible to do the following: => Omit the RC release and replace it with our Final Release! This means that Sugar on a Stick is going be released in June, on 2009-06-24. Now, why? Well, there were quite some reasons: * Fedora 11 will be released on June 2 and Sugar 0.84 has already had it's release some time ago. By moving our final release later into the year, we'd be either forced to use some outdated or unstable components, as the next major Sugar version will be 0.86, which is targeted for Fedora 12. We're preventing this by having our release now just a month after F11's. * It helps us a lot to get feedback from students over the summer break, so that we're increasing the likelihood of gaining more users. * Sugar on a Stick is rather stable right now - I'll outline this in a separate e-mail! The updated roadmap is located here: http://wiki.sugarlabs.org/go/Sugar_on_a_Stick/Roadmap Package Maintainers will receive reminders to update their RPM packages soonish! Again, please make sure to follow the deadlines. The last date for changes is 2009-06-10. Afterwards, dev team's approval is required. Please contact me with any concerns you may have - also off-list, if needed! Best Regards, --Sebastian Dziallas From gary at garycmartin.com Wed May 27 17:42:59 2009 From: gary at garycmartin.com (Gary C Martin) Date: Wed, 27 May 2009 18:42:59 +0100 Subject: Daily builds are no longer livecd images In-Reply-To: References: Message-ID: <79158625-4269-4594-80B0-972DB5B3B99B@garycmartin.com> Hi Chris, On 25 May 2009, at 14:43, Chris Ball wrote: > Starting with today's 20090525 build, the images are no longer livecd > images, so changes will persist across reboots. > > http://dev.laptop.org/~cjb/rawhide-xo/20090525/ Just tried testing 20090525 on an XO-1 as a clean install (OF copy- nand of .img with .crc) but couldn't get the XO to boot afterwards. Tried multiple reboots but it always stopped pretty soon in the console at "creating devices". On closer inspection, a handful of lines up, is the error "mount: unknown filesystem type 'jffs2' which seems pretty fatal ;-) Regards, --Gary > (The sporadic "creating devices" mount hang will still be present for > jffs2; there's no need to report that again.) > > Thanks to Sebastian for the script to make installed images! > > - Chris. > -- > Chris Ball > > _______________________________________________ > Fedora-olpc-list mailing list > Fedora-olpc-list at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-olpc-list From kevin at sonney.com Wed May 27 18:41:21 2009 From: kevin at sonney.com (Kevin Sonney) Date: Wed, 27 May 2009 14:41:21 -0400 Subject: Daily builds are no longer livecd images In-Reply-To: <79158625-4269-4594-80B0-972DB5B3B99B@garycmartin.com> References: <79158625-4269-4594-80B0-972DB5B3B99B@garycmartin.com> Message-ID: On Wed, May 27, 2009 at 1:42 PM, Gary C Martin wrote: > Just tried testing 20090525 on an XO-1 as a clean install (OF copy-nand of > .img with .crc) but couldn't get the XO to boot afterwards. Tried multiple > reboots but it always stopped pretty soon in the console at "creating > devices". On closer inspection, a handful of lines up, is the error "mount: > unknown filesystem type 'jffs2' which seems pretty fatal ;-) I had it boot from the NAND exactly once, and have been seeing the same thing on all subsequent boots. However, before that, it was looking pretty good. What I'm seeing on screen after USB initializes : udev: starting version 141 input: OLPC HGPK ALPS HGPK as /devices/platform/i8042/serio1/input/input2 JFFS2 version 2.2 (NAND) (SUMMARY) 2001-2006 Red Hat, Inc. sdhci: Secure Digital Host Controller Interface driver sdhci: Copyright(c) Pierre Ossman NAND device: Manufacturer ID: 0xad, Chip ID: 0xdc (Hynix NAND 512MiB 3,3V 8-bit) 2 NAND chips detected cmdlinepart partition processing not available RedBoot partition parsing not available sdhci-pci 0000:00:0c.1: SDHCI controller found [11ab:4101] (rev 10) sdhci-pci 0000:00:0c.1 enabling device (0000 -> 0002) sdhci-pci 0000:00:0c.1: Invalid iomem size. You may experience problems. Registered led device mmc0:: mmc0: SDHCI COntroller on PCI [0000:00:0c.1] using DMA mount: unknown filesystem type 'jffs2' kernel commandline: root=mtd0 rootfstype=jffs2 console=tty0 console=ttyS0,115200 fbcon=font:SUN12x22 init=/sbin/init root=mtd0 rootflags= rootfstype=jffs2 root_ro=0 root_rw=0 starting udev creating devices And at this point, it hangs until I power it off. -- Kevin Sonney -- kevin at sonney.com "Be daring, be different, be impractical, be anything that will assert integrity of purpose and imaginative vision against the play-it-safers, the creatures of the commonplace, the slaves of the ordinary. " - Cecil Beaton I check email a couple times daily; to reach me sooner, click here: http://awayfind.com/ksonney From mikus at bga.com Wed May 27 19:29:14 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Wed, 27 May 2009 15:29:14 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A1D948A.2070109@bga.com> One of the posts to this topic said: > Before, *every* time you booted the XO with Rawhide-XO, there was a > number of scripts initiated, for example to create the liveuser or to > check for changed hardware. This is of course nonsense, as the XO's > hardware won't change. In my case, the XO hardware does change from boot to boot. I plug (or unplug) various "external" devices (hard disks or other storage devices, USB hubs, ethernet adapters, keyboards, etc. (even multiple SD cards)) as I need them. Also, sometimes the boot-up process stalls on me -- my experience has been that changing the external device configuration of the XO is sometimes enough to alter the "internal timings" of the boot-up process sufficiently to get past some instances of stalling. mikus From mikus at bga.com Wed May 27 20:22:13 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Wed, 27 May 2009 16:22:13 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A1DA0F5.5090007@bga.com> > > What does "no longer livecd images" mean ? > > On previous builds, changes to the root filesystem disappeared at > reboot because the filesystem was refreshed from the same image > every boot. This no longer happens, so if you (for example): > > # touch /foo > # touch /etc/bar > # reboot > > The files will still be there after reboot. I *still* do not understand why that should be so. The only machines available for me to experiment with are XO-1s. The above-described scenario does NOT work for me. -------- I did the zcat 20090525.bootable.gz > /dev/sdf1 (USB stick). That USB stick failed to boot on the XO. The messages from OFW were "Error: Unknown file system Can't open disk label package" Then I did the livecd-iso-to-disk.sh 20090525.iso > /dev/sdf1 (USB stick). That USB stick booted fine on the XO. Made some changes to the root filesystem, and rebooted. After the XO had completed rebooting (in the process, it reloaded its root filesystem from the USB stick), the changes I had previously made were NOT there. -------- I can understand how changes made to the "recording medium" will be there after a reboot. [For instance, if I boot from nand, then make changes to the filesystem in memory (and the system writes those changed memory pages back out to nand), then when I reboot (from nand) the changes will still be there.] But in the case of booting from an USB stick, I believe the system does *not* write the contents of changed memory pages back out to the USB stick. When I reboot from that USB stick (particularly if I've powered-down the system in the meantime), why would the changes still be there ?? Thanks, mikus From bochecha at fedoraproject.org Wed May 27 21:31:50 2009 From: bochecha at fedoraproject.org (Mathieu Bridon (bochecha)) Date: Wed, 27 May 2009 23:31:50 +0200 Subject: Daily builds are no longer livecd images In-Reply-To: <4A1DA0F5.5090007@bga.com> References: <4A1DA0F5.5090007@bga.com> Message-ID: <2d319b780905271431o1fc30b5bw2e27394f6d829c5@mail.gmail.com> > But in the case of booting from an USB stick, I believe the system does > *not* write the contents of changed memory pages back out to the USB stick. > ?When I reboot from that USB stick (particularly if I've powered-down the > system in the meantime), why would the changes still be there ?? Because USB sticks are not CDs. We use a persistent overlay on live USBs so that you can store your changes. Regards, ---------- Mathieu Bridon (bochecha) From mikus at bga.com Wed May 27 22:21:43 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Wed, 27 May 2009 18:21:43 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A1DBCF7.6080708@bga.com> >> But in the case of booting from an USB stick, I believe the system does >> *not* write the contents of changed memory pages back out to the USB stick. >> When I reboot from that USB stick (particularly if I've powered-down the >> system in the meantime), why would the changes still be there ?? > > Because USB sticks are not CDs. We use a persistent overlay on live > USBs so that you can store your changes. I am not seeing what you describe. [The only machines I have to experiment on are XO-1s.] Since I do not see the "changes still there" that you keep talking about, all I can figure out is that we are talking past each other. -------- If I look at the contents of an USB stick (df, ls -la) before using that USB stick to boot from; then boot the system and make changes; then shutdown -- when I again look at the contents of that USB stick, I do not notice those contents having been changed. If in fact the content of that USB stick has not changed, where does the "persistent overlay on live USBs" (that you describe) get kept ? Thanks, mikus p.s. I just did an experiment. I made a copy of the USB stick I created with 'livecd-iso-to-disk' from rawhide-xo 20090525.iso. I then booted an XO with the original USB stick, and made changes to / and to /etc (in memory) once Gnome was running. I then shut down the XO, and compared the original USB stick to the earlier copy. They were byte-for-byte identical. To me, that says that *nothing* got written to the original USB stick (used to boot from) when I made those changes to the booted system. From cjb at laptop.org Thu May 28 00:49:14 2009 From: cjb at laptop.org (Chris Ball) Date: Wed, 27 May 2009 20:49:14 -0400 Subject: Daily builds are no longer livecd images In-Reply-To: <4A1DA0F5.5090007@bga.com> (Mikus Grinbergs's message of "Wed, 27 May 2009 16:22:13 -0400") References: <4A1DA0F5.5090007@bga.com> Message-ID: Hi Mikus, > I did the zcat 20090525.bootable.gz > /dev/sdf1 (USB stick). > That USB stick failed to boot on the XO. The messages from OFW > were "Error: Unknown file system Can't open disk label package" I think that's because you should have used /dev/sdf, not /dev/sdf1. Using /dev/sdf1 fails to copy over the correct partition table to the start of the disk. > After the XO had completed rebooting (in the process, it reloaded > its root filesystem from the USB stick), the changes I had > previously made were NOT there. You're right -- I think the persistence only applies to NAND images generated with livecd-iso-to-xo at the moment. Does anyone have suggestions for how to go from a livecd ISO to an ext3 build that doesn't involve a squashfs/ext3 image, and just puts the files in place on the image directly? livecd-iso-to-xo already knows how to mount the ISO and then call mkfs.jffs2 on the root dir, so we need to work out how to do a similar thing but using mkfs.ext3: going from a root dir to a disk image file that contains a partition table/bootloader/directory tree in ext3. Any ideas? Thanks, - Chris. -- Chris Ball From jvonau at shaw.ca Thu May 28 01:43:07 2009 From: jvonau at shaw.ca (Jerry Vonau) Date: Wed, 27 May 2009 20:43:07 -0500 Subject: Daily builds are no longer livecd images In-Reply-To: References: <4A1DA0F5.5090007@bga.com> Message-ID: <1243474987.32131.26.camel@f9.vonau.ca> On Wed, 2009-05-27 at 20:49 -0400, Chris Ball wrote: > Hi Mikus, > > > I did the zcat 20090525.bootable.gz > /dev/sdf1 (USB stick). > > That USB stick failed to boot on the XO. The messages from OFW > > were "Error: Unknown file system Can't open disk label package" > > I think that's because you should have used /dev/sdf, not /dev/sdf1. > Using /dev/sdf1 fails to copy over the correct partition table to > the start of the disk. > > > After the XO had completed rebooting (in the process, it reloaded > > its root filesystem from the USB stick), the changes I had > > previously made were NOT there. > > You're right -- I think the persistence only applies to NAND images > generated with livecd-iso-to-xo at the moment. Does anyone have > suggestions for how to go from a livecd ISO to an ext3 build that > doesn't involve a squashfs/ext3 image, and just puts the files in > place on the image directly? > Isn't that what image-creator is used for? Martin is using that with a script to mount a yum repo to create XS-on-OX ext3 images. Have a look at: http://dev.laptop.org/git/projects/xs-livecd/tree/util/mkext3img > livecd-iso-to-xo already knows how to mount the ISO and then call > mkfs.jffs2 on the root dir, so we need to work out how to do a similar > thing but using mkfs.ext3: going from a root dir to a disk image file > that contains a partition table/bootloader/directory tree in ext3. > > Any ideas? Thanks, > Jerry From mikus at bga.com Thu May 28 02:41:12 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Wed, 27 May 2009 22:41:12 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A1DF9C8.8020300@bga.com> > > I did the zcat 20090525.bootable.gz > /dev/sdf1 (USB stick). > > That USB stick failed to boot on the XO. > > I think that's because you should have used /dev/sdf, not /dev/sdf1. > Using /dev/sdf1 fails to copy over the correct partition table to > the start of the disk. Thank you. That was the mistake I had made. Now, booting from USB stick created from zcat 20090525.bootable.gz consistently stalls on me subsequent to showing the message " mount: unknown filesystem type 'jffs2' " and the message " Bug in initramfs /init detected. Dropping to a shell. Good luck! ". [I repeated the power-on boot more than 10 times, and it stalled each time.] [In contrast, when the booting XO from nand (created by 'copy-nand' from 20090525.img) sequence stalls (typically subsequent to that " unknown filesytem type 'jffs2' " message having been shown), I often manage to get past that point by persistently re-trying the power-on boot sequence over and over - if I'm lucky, I eventually get no error message and no stall.] Thanks, mikus From cjb at laptop.org Thu May 28 05:28:31 2009 From: cjb at laptop.org (Chris Ball) Date: Thu, 28 May 2009 01:28:31 -0400 Subject: Daily builds are no longer livecd images In-Reply-To: <1243474987.32131.26.camel@f9.vonau.ca> (Jerry Vonau's message of "Wed, 27 May 2009 20:43:07 -0500") References: <4A1DA0F5.5090007@bga.com> <1243474987.32131.26.camel@f9.vonau.ca> Message-ID: Hi Jerry, > Isn't that what image-creator is used for? Martin is using that > with a script to mount a yum repo to create XS-on-OX ext3 > images. Have a look at: > http://dev.laptop.org/git/projects/xs-livecd/tree/util/mkext3img Thanks for the pointer. Seems like this might be only for install CDs rather than live CDs? I tried: # ./mkext3img 20090525.iso /home/cjb/git/fedora-xo/olpc-desktop.ks [pid 28624] stat64("/media/cdtmp.42DLUT/repodata/repomd.xml", 0xbfa36afc) = -1 ENOENT (No such file or directory) Unable to create image : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: cdrom. Please verify its path and try again - Chris. -- Chris Ball From mikus at bga.com Thu May 28 17:12:29 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Thu, 28 May 2009 13:12:29 -0400 Subject: rawhide-xo 20090528 Message-ID: <4A1EC5FD.9030000@bga.com> Using 'copy-nand', flashed ~cjb/rawhide-xo 20090528.img to XO nand. My experience with rawhide-xo 20090528 matches that with 20090525. One significant fix that first showed up in 20090525 is that the OLPC keyboard (e.g., the Frame key) is now being recognized. One problem that showed up in 20090525 is that Sugar does not pick up on the 'Desktop' font size specification set within Gnome. The result is that labels (e.g., in My Settings) are now really tiny. A bigger problem, which I first noticed with 20090525, is that Neighborhood View is empty (except for the 'me' icon in the middle). [In contrast, my rawhide-xo 20090519 system's Neighborhood View shows both existing wireless access points and existing wired XOs.] mikus From sayamindu at gmail.com Thu May 28 18:24:25 2009 From: sayamindu at gmail.com (Sayamindu Dasgupta) Date: Thu, 28 May 2009 23:54:25 +0530 Subject: rawhide-xo 20090528 In-Reply-To: <4A1EC5FD.9030000@bga.com> References: <4A1EC5FD.9030000@bga.com> Message-ID: On Thu, May 28, 2009 at 10:42 PM, Mikus Grinbergs wrote: > Using 'copy-nand', flashed ~cjb/rawhide-xo 20090528.img to XO nand. > > My experience with rawhide-xo 20090528 matches that with 20090525. > > One significant fix that first showed up in 20090525 is that the OLPC > keyboard (e.g., the Frame key) is now being recognized. > > One problem that showed up in 20090525 is that Sugar does not pick up on the > 'Desktop' font size specification set within Gnome. ?The result is that > labels (e.g., in My Settings) are now really tiny. > I noticed this as well - and apparently the screen dpi for the font-rendering mechanism is being set to 50 for some reasons. I tried a modified build, but it did not work. I looked at the source code of the part of GNOME which does this, and they have a sane-dpi-value enforcing mechanism, which may be kicking in. I'll try to go deeper into this. Thanks, Sayamindu -- Sayamindu Dasgupta [http://sayamindu.randomink.org/ramblings] From mikus at bga.com Fri May 29 04:36:25 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Fri, 29 May 2009 00:36:25 -0400 Subject: rawhide-xo 20090528 Message-ID: <4A1F6649.7040104@bga.com> > In Soas we are forcing a sane dpi for the XO by using a .Xresources > file. When I last looked into this, I think it pointed to the graphics > driver. Ran 'livecd-iso-to-disk' with Soas2-200905241902.iso to create a bootable USB stick. Then booted my XO from that stick. The labels in 'My Settings' using Soas2-20090524 are equally tiny as those same labels using rawhide-xo 20090528. So is the default text in 'Terminal' (in Sugar). mikus From martin at martindengler.com Fri May 29 09:19:42 2009 From: martin at martindengler.com (Martin Dengler) Date: Fri, 29 May 2009 10:19:42 +0100 Subject: rawhide-xo 20090528 In-Reply-To: <4A1EC5FD.9030000@bga.com> References: <4A1EC5FD.9030000@bga.com> Message-ID: <20090529091942.GT31940@ops-13.xades.com> On Thu, May 28, 2009 at 01:12:29PM -0400, Mikus Grinbergs wrote: > One significant fix that first showed up in 20090525 is that the OLPC > keyboard (e.g., the Frame key) is now being recognized. Yes, that should've shown up. When you press the Fn key + End (right arrow) in Pippy/Terminal/etc., do you also get an equals sign character inserted in the text before gets activated? > mikus Martin -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mikus at bga.com Fri May 29 14:34:06 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Fri, 29 May 2009 10:34:06 -0400 Subject: rawhide-xo 20090528 Message-ID: <4A1FF25E.8020201@bga.com> > When you press the Fn key + End (right > arrow) in Pippy/Terminal/etc., do you also get an equals sign > character inserted in the text before gets activated? Just tried it with rawhide-xo 20090528.iso (installed on nand): Pressing the Fn key generates an '=' character at the text cursor. [I normally use an external USB keyboard with my XO, to avoid this unwanted Fn key behavior.] mikus From martin at martindengler.com Fri May 29 15:36:45 2009 From: martin at martindengler.com (Martin Dengler) Date: Fri, 29 May 2009 16:36:45 +0100 Subject: rawhide-xo 20090528 In-Reply-To: <4A1FF25E.8020201@bga.com> References: <4A1FF25E.8020201@bga.com> Message-ID: <20090529153645.GC31940@ops-13.xades.com> On Fri, May 29, 2009 at 10:34:06AM -0400, Mikus Grinbergs wrote: >> When you press the Fn key + End (right >> arrow) in Pippy/Terminal/etc., do you also get an equals sign >> character inserted in the text before gets activated? > > Just tried it with rawhide-xo 20090528.iso (installed on nand): > > Pressing the Fn key generates an '=' character at the text cursor. Thanks. It does for me too, and I'm sure the fix will be trivial once I/someone finds it. > [I normally use an external USB keyboard with my XO, to avoid this > unwanted Fn key behavior.] Are you saying it happened before the Frame key started working? > mikus Martin -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mikus at bga.com Fri May 29 18:11:25 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Fri, 29 May 2009 14:11:25 -0400 Subject: rawhide-xo 20090528 Message-ID: <4A20254D.30001@bga.com> >>> When you press the Fn key + End (right >>> arrow) in Pippy/Terminal/etc., do you also get an equals sign >>> character inserted in the text before gets activated? > > Are you saying it happened before the Frame key started working? Yes. If I remember correctly, this has happened with every rawhide-xo build ever released. [It also happens with Soas2.] mikus From mikus at bga.com Fri May 29 19:56:33 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Fri, 29 May 2009 15:56:33 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A203DF1.8040503@bga.com> > When you use livecd-iso-to-disk to burn a liveUSB, you get a read-only > system on the USB stick. > > However, if you use the "--overlay-size-mb " to the > livecd-iso-to-disk script, it will create a unionfs persistent storage > of Mega Bytes _on the same USB stick_. Thank you. I added the "--overlay-size-mb " to the script invocation, so it now is: 'livecd-iso-to-disk.sh --format --overlay-size-mb 600 --xo --xo-no-home /media/disk/20090519.iso /dev/sdf1'. That caused additional file '/LiveOS/overlay-LIVE-AA00-0383' to be allocated on the USB stick when its content was being built. [But note that when I run 'livecd-iso-to-disk' on my Ubuntu 9.04 (Jaunty) system (with latest kernel released by Ubuntu), I'm still getting the error: "running kernel cannot mount the squashfs from the ISO file to extract it. The compressed squashfs will be copied to the USB stick."] When I booted the XO using that USB stick; made system changes; shutdown; powered off; rebooted with the USB stick -- the changes I had made had "evaporated". ---- What can I do to make my changes "persistent" when booting an XO from an USB stick ? [It is not realistic for me to acquire a Fedora desktop system to run 'livecd-iso-to-disk'.] Thanks, mikus p.s. On the USB stick, in the file '/boot/olpc.fth' (generated by 'livecd-iso-to-disk'), the line that sets up the boot parameters specifies: 'reset_overlay'. What does 'reset_overlay' as a parameter do ? Should it be there ? From martin at martindengler.com Sat May 30 02:05:23 2009 From: martin at martindengler.com (Martin Dengler) Date: Sat, 30 May 2009 03:05:23 +0100 Subject: rawhide-xo 20090528 In-Reply-To: <4A20254D.30001@bga.com> References: <4A20254D.30001@bga.com> Message-ID: <20090530020523.GD31940@ops-13.xades.com> On Fri, May 29, 2009 at 02:11:25PM -0400, Mikus Grinbergs wrote: >>>> When you press the Fn key + End (right >>>> arrow) in Pippy/Terminal/etc., do you also get an equals sign >>>> character inserted in the text before gets activated? >> >> Are you saying it happened before the Frame key started working? > > Yes. If I remember correctly, this has happened with every rawhide-xo > build ever released. [It also happens with Soas2.] Thanks - good to know we didn't break it with the Frame-key-fix. Now we'll need to fix it... > mikus Martin -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From jvonau at shaw.ca Sat May 30 14:45:02 2009 From: jvonau at shaw.ca (Jerry Vonau) Date: Sat, 30 May 2009 09:45:02 -0500 Subject: Daily builds are no longer livecd images In-Reply-To: <4A203DF1.8040503@bga.com> References: <4A203DF1.8040503@bga.com> Message-ID: <1243694702.32131.31.camel@f9.vonau.ca> On Fri, 2009-05-29 at 15:56 -0400, Mikus Grinbergs wrote: > > When you use livecd-iso-to-disk to burn a liveUSB, you get a read-only > > system on the USB stick. > > > > However, if you use the "--overlay-size-mb " to the > > livecd-iso-to-disk script, it will create a unionfs persistent storage > > of Mega Bytes _on the same USB stick_. > > Thank you. > > I added the "--overlay-size-mb " to the script invocation, so > it now is: 'livecd-iso-to-disk.sh --format --overlay-size-mb 600 > --xo --xo-no-home /media/disk/20090519.iso /dev/sdf1'. That caused > additional file '/LiveOS/overlay-LIVE-AA00-0383' to be allocated on > the USB stick when its content was being built. > > [But note that when I run 'livecd-iso-to-disk' on my Ubuntu 9.04 > (Jaunty) system (with latest kernel released by Ubuntu), I'm still > getting the error: "running kernel cannot mount the squashfs from > the ISO file to extract it. The compressed squashfs will be copied > to the USB stick."] > > When I booted the XO using that USB stick; made system changes; > shutdown; powered off; rebooted with the USB stick -- the changes > I had made had "evaporated". > > ---- > > What can I do to make my changes "persistent" when booting an XO > from an USB stick ? [It is not realistic for me to acquire a Fedora > desktop system to run 'livecd-iso-to-disk'.] > > Thanks, mikus > > > > p.s. On the USB stick, in the file '/boot/olpc.fth' (generated > by 'livecd-iso-to-disk'), the line that sets up the boot > parameters specifies: 'reset_overlay'. > > What does 'reset_overlay' as a parameter do ? Should it be > there ? That would re-create the "/LiveOS/overlay-LIVE-AA00-0383" file, thus wiping out the changes made prior to the reboot. I'd remove that from the olpc.fth file. Jerry From sebastian at when.com Sat May 30 17:28:36 2009 From: sebastian at when.com (Sebastian Dziallas) Date: Sat, 30 May 2009 19:28:36 +0200 Subject: Please update your Sugar packages for SoaS! Message-ID: <4A216CC4.6000300@when.com> Hi all, as we're quite quickly approaching the next release of SoaS, which is scheduled for June 24, it would be great if you could update the Sugar packages you maintain in Fedora to their latest versions. The whole roadmap is listed on our wiki, where you can also find the current version number of your package(s): http://wiki.sugarlabs.org/go/Sugar_on_a_Stick/Roadmap#Packaging In case you're unable to do the update, you can drop Steven (listed as CC) an e-mail, who has kindly agreed to help, if needed. Thanks a lot, --Sebastian From mikus at bga.com Sat May 30 19:49:01 2009 From: mikus at bga.com (Mikus Grinbergs) Date: Sat, 30 May 2009 15:49:01 -0400 Subject: Daily builds are no longer livecd images Message-ID: <4A218DAD.7040608@bga.com> >> On the USB stick, in the file '/boot/olpc.fth' (generated >> by 'livecd-iso-to-disk'), the line that sets up the boot >> parameters specifies: 'reset_overlay'. >> >> What does 'reset_overlay' as a parameter do ? Should it be >> there ? > > That would re-create the "/LiveOS/overlay-LIVE-AA00-0383" file, thus > wiping out the changes made prior to the reboot. I'd remove that from > the olpc.fth file. Thank you, thank you. Finally, I'm able to see that changes will "persist" after the reboot of an XO, which was booted off an USB stick. It was a long struggle to get this working - but with the help of the participants on the fedora-olpc-list it ended successfully. Thanks: Chris, Mathieu, Jerry, everyone. mikus From bochecha at fedoraproject.org Sat May 30 20:53:43 2009 From: bochecha at fedoraproject.org (Mathieu Bridon (bochecha)) Date: Sat, 30 May 2009 22:53:43 +0200 Subject: Daily builds are no longer livecd images In-Reply-To: <4A218DAD.7040608@bga.com> References: <4A218DAD.7040608@bga.com> Message-ID: <2d319b780905301353j6f1029cdtc99cde4e47b95f50@mail.gmail.com> >>> ?On the USB stick, in the file '/boot/olpc.fth' (generated >>> ?by 'livecd-iso-to-disk'), the line that sets up the boot >>> ?parameters specifies: 'reset_overlay'. >>> >>> ?What does 'reset_overlay' as a parameter do ? ?Should it be >>> ?there ? >> >> That would re-create the "/LiveOS/overlay-LIVE-AA00-0383" file, thus >> wiping out the changes made prior to the reboot. I'd remove that from >> the olpc.fth file. > > Thank you, ?thank you. > > Finally, I'm able to see that changes will "persist" after the reboot of an > XO, which was booted off an USB stick. I'm curious, does anyone know why exactly there is this parameter ? Don't we want changes to be persistent by default ? ---------- Mathieu Bridon (bochecha) From rms at 1407.org Sun May 31 15:02:03 2009 From: rms at 1407.org (Rui Miguel Silva Seabra) Date: Sun, 31 May 2009 16:02:03 +0100 Subject: Daily builds are no longer livecd images In-Reply-To: References: Message-ID: <20090531150203.GB4479@roque.1407.org> On Mon, May 25, 2009 at 09:43:55AM -0400, Chris Ball wrote: > http://dev.laptop.org/~cjb/rawhide-xo/20090525/ > > (The sporadic "creating devices" mount hang will still be present for > jffs2; there's no need to report that again.) This isn't sporadic for me, it's constant. 20090525 or 20090528 images both hang on "creating devices" every boot, for me. :( From cjb at laptop.org Sun May 31 15:35:29 2009 From: cjb at laptop.org (Chris Ball) Date: Sun, 31 May 2009 11:35:29 -0400 Subject: Daily builds are no longer livecd images In-Reply-To: <20090531150203.GB4479@roque.1407.org> (Rui Miguel Silva Seabra's message of "Sun, 31 May 2009 16:02:03 +0100") References: <20090531150203.GB4479@roque.1407.org> Message-ID: Hi, > This isn't sporadic for me, it's constant. 20090525 or 20090528 > images both hang on "creating devices" every boot, for me. It always works after about three tries here. If you get time, maybe you could unpack the initrd and see if you can make a modification that gets you to win the race every time? (Doing a "modprobe jffs2" much earlier than the mount should be one example of how to win it, but it would be better to work out a way to fix the real problem.) Thanks, - Chris. -- Chris Ball From rms at 1407.org Sun May 31 17:43:56 2009 From: rms at 1407.org (Rui Miguel Silva Seabra) Date: Sun, 31 May 2009 18:43:56 +0100 Subject: Daily builds are no longer livecd images In-Reply-To: References: <20090531150203.GB4479@roque.1407.org> Message-ID: <20090531174356.GC4479@roque.1407.org> On Sun, May 31, 2009 at 11:35:29AM -0400, Chris Ball wrote: > Hi, > > > This isn't sporadic for me, it's constant. 20090525 or 20090528 > > images both hang on "creating devices" every boot, for me. > > It always works after about three tries here. If you get time, maybe > you could unpack the initrd and see if you can make a modification > that gets you to win the race every time? (Doing a "modprobe jffs2" > much earlier than the mount should be one example of how to win it, > but it would be better to work out a way to fix the real problem.) Can I just mount the img and try to do that (what about the crc?), or should I generate an image from scratch? Rui From rms at 1407.org Sun May 31 18:58:00 2009 From: rms at 1407.org (Rui Miguel Silva Seabra) Date: Sun, 31 May 2009 19:58:00 +0100 Subject: Daily builds are no longer livecd images In-Reply-To: <20090531174356.GC4479@roque.1407.org> References: <20090531150203.GB4479@roque.1407.org> <20090531174356.GC4479@roque.1407.org> Message-ID: <20090531185759.GD4479@roque.1407.org> On Sun, May 31, 2009 at 06:43:56PM +0100, Rui Miguel Silva Seabra wrote: > On Sun, May 31, 2009 at 11:35:29AM -0400, Chris Ball wrote: > > Hi, > > > > > This isn't sporadic for me, it's constant. 20090525 or 20090528 > > > images both hang on "creating devices" every boot, for me. > > > > It always works after about three tries here. If you get time, maybe After another attempt it did boot. Then after a short time there was a system freeze and now I'm on the 11th unsuccessful boot :( Rui From cjb at laptop.org Sun May 31 23:47:48 2009 From: cjb at laptop.org (Chris Ball) Date: Sun, 31 May 2009 19:47:48 -0400 Subject: Daily builds are no longer livecd images In-Reply-To: <20090531174356.GC4479@roque.1407.org> (Rui Miguel Silva Seabra's message of "Sun, 31 May 2009 18:43:56 +0100") References: <20090531150203.GB4479@roque.1407.org> <20090531174356.GC4479@roque.1407.org> Message-ID: Hi, > Can I just mount the img and try to do that (what about the > crc?), or should I generate an image from scratch? Hm, it would be easier if you could get it to boot once -- then you could just unpack/repack the initrd while booted. If you have to make a new image, the following (from livecd-iso-to-xo) will get you a new CRC: mkfs.jffs2 -n -e128KiB -r $ROOT -o newjffs.img.tmp sumtool -n -p -e 128KiB -i newjffs.img.tmp -o newjffs.img crcimg newjffs.img (in Fedora, sumtool is in "mtd-utils" and crcimg is in "crcimg".) Thanks, - Chris. -- Chris Ball