| CIA-34 | ubiquity: cjwatson * r4294 ubiquity/ (6 files in 3 dirs): Remove lpia architecture support. | 00:11 |
|---|---|---|
| persia | Thanks! Apologies for missing that. | 00:16 |
| cjwatson | that's ok | 00:26 |
| CIA-34 | usb-creator: superm1 * r323 usb-creator/ (3 files in 3 dirs): | 02:00 |
| CIA-34 | usb-creator: Grow support for installing GRUB to USB sticks if it's detected in the image | 02:00 |
| CIA-34 | usb-creator: rather than isolinux. | 02:00 |
| CIA-34 | ubiquity: superm1 * r4293 ubiquity/ (debian/changelog src/panel/panel.c): | 06:40 |
| CIA-34 | ubiquity: Set the panel indicators to show up on the right to match the rest | 06:40 |
| CIA-34 | ubiquity: of the desktop (LP: #632592) | 06:40 |
| CIA-34 | ubiquity: superm1 * r4295 ubiquity/ (debian/changelog src/panel/panel.c): | 06:42 |
| CIA-34 | ubiquity: Set the panel indicators to show up on the right to match the rest | 06:42 |
| CIA-34 | ubiquity: of the desktop (LP: #632592) | 06:42 |
| superm1 | that's weird it didn't cia-submit the first time | 06:43 |
| NCommander | cjwatson: just a friendly reminder ping, new patches went to d-boot. Since all d-boot subscirbers just got ~2000 pieces of mail from BTS, wanted to make sure it didn't get lost :-) | 09:23 |
| CIA-34 | ubiquity: cjwatson * r4296 ubiquity/ (4 files in 2 dirs): | 09:50 |
| CIA-34 | ubiquity: Handle grub-efi when installing on amd64/efi or i386/efi | 09:50 |
| CIA-34 | ubiquity: subarchitectures (LP: #632642). | 09:50 |
| CIA-34 | grub-installer: cjwatson * r867 ubuntu/ (debian/changelog grub-installer): | 10:30 |
| CIA-34 | grub-installer: Purge grub-efi* when $grub_package is grub or grub-pc, and purge grub, | 10:30 |
| CIA-34 | grub-installer: grub-legacy, and grub-pc when $grub_package is grub-efi (LP: #632642). | 10:30 |
| CIA-34 | grub-installer: cjwatson * r868 ubuntu/debian/changelog: releasing version 1.55ubuntu3 | 10:31 |
| cjwatson | NCommander: you have mail - I know that my Mail-Followup-To probably says otherwise, but if you could CC me I'll see it more quickly | 10:55 |
| CIA-34 | ubiquity: evand * r4297 trunk/ (4 files in 4 dirs): (log message trimmed) | 10:55 |
| CIA-34 | ubiquity: * Update the KDE partitioning UI to reflect changes to ubi-partman and | 10:55 |
| CIA-34 | ubiquity: partman-auto. | 10:55 |
| CIA-34 | ubiquity: - Use the already existent ubiquity variants of the d-i "Guided -" strings | 10:55 |
| CIA-34 | ubiquity: (LP: #628864). | 10:55 |
| NCommander | cjwatson: gah. whoops :-). | 10:58 |
| cjwatson | you weren't to know about my odd mail arrangements | 10:58 |
| NCommander | cjwatson: probably just as odd as mine. Thanks for the review. I tried using a weak symbol, but it doesn't like LD's versioning voodoo | 10:59 |
| NCommander | cjwatson: thanks for not biting off my head with the review for some of the newbie mistakes with d-i hacking :-) | 11:00 |
| cjwatson | np. library versioning can be less than obvious anyway | 11:03 |
| NCommander | cjwatson: no kidding :-/ | 11:08 |
| === rgreening_ is now known as ghost | ||
| === ghost is now known as rgreening | ||
| ev | I'd like to commit the following: <http://paste.ubuntu.com/490342/>. Are there any objections? I realize we're well past feature freeze, but I think this is immensely important to have, given how much of a better experience you get in the installer with an Internet connection. Equally, aside from the few lines of code to construct the widget, it's all very well tested, having been ripped out of gnome-panel. | 14:04 |
| cjwatson | so this basically lets us run network-manager-applet? | 14:10 |
| ev | exactly | 14:10 |
| ogra | add it !!! add it !!! | 14:11 |
| ev | heh | 14:11 |
| cjwatson | wfm | 14:14 |
| ev | awesome | 14:37 |
| ev | sorry for the delay, was just in a meeting with Amanda and mpt on the text for the non-free software option, which mpt is kindly improving based on her and Andrew's input. | 14:38 |
| CIA-34 | ubiquity: evand * r4298 trunk/ (14 files in 2 dirs): Add notification area support in the panel. | 15:02 |
| CIA-34 | ubiquity: evand * r4299 trunk/src/panel/panel.c: Pack notification area on the correct side. | 15:11 |
| ScottK | Is SQUASHFS error: Unable to read data cache entry, a few repeats of similar lines and then Bus error (core dumped) in a live session concerning? Should I file a bug and if so on what? | 15:13 |
| ev | ScottK: virtual machine? CD? USB disk? | 15:14 |
| ScottK | ev: CD on real hardware | 15:15 |
| ScottK | Worked on a different machine. | 15:16 |
| ev | sounds like a bad drive or CD. I'd check the MD5SUM of the disk for a start. | 15:16 |
| ScottK | OK. | 15:16 |
| ScottK | This machine worke fine with the beta CD last week and the CD worked fine in another machine. I'll try it again. | 15:17 |
| ev | oh and there's also md5sum -c /cdrom/md5sum.txt | 15:19 |
| superm1 | cjwatson, re you patch to ubiquity for grub-efi, shouldn't you also have grub-efi-ia32 as a package that is kept for when this is added or used in 32 bit disks since that's what grub-efi depends on in 32 bit currently? | 15:20 |
| ev | if you want to verify that your burner wrote the correct data | 15:20 |
| ScottK | Just rebooted into a new live session and told it to check the CD. | 15:22 |
| ScottK | Not sure that's actually working though. | 15:22 |
| ScottK | Ah, there we are. | 15:22 |
| ScottK | Worked fine after reboot with the same CD, so I guess "stuff" happens. | 15:36 |
| cjwatson | superm1: not planning on supporting UEFI for 32-bit, so I don't care much :-) | 15:39 |
| cjwatson | superm1: significantly mitigates the risk of multi-catalog images, and AMD were very clear to us that UEFI was undefined on non-64-bit | 15:39 |
| CIA-34 | ubiquity: evand * r4300 trunk/debian/changelog: Add LP reference. | 16:24 |
| ev | cjwatson: should silo-installer be in universe? | 16:39 |
| cjwatson | yep | 16:39 |
| cjwatson | sparc no longer being supported and all | 16:39 |
| ev | so then should we nuke it from ubiquity? | 16:39 |
| cjwatson | yeah, that and ia64, imo | 16:39 |
| cjwatson | s/supported/built/ | 16:39 |
| ev | http://paste.ubuntu.com/490418/ - that look okay? | 16:51 |
| superm1 | cjwatson, hum, is that bootloader and kernel behavior are undefined, or you are saying 32 bit uefi executable behavior is undefined? eg would it be possible to run a 64 bit uefi executable for grub and boot a 32 bit OS? | 16:52 |
| cjwatson | ev: looks fine | 16:53 |
| ev | thanks | 16:53 |
| CIA-34 | ubiquity: evand * r4301 trunk/ (9 files in 4 dirs): Drop ia64 and sparc. | 16:54 |
| cjwatson | superm1: I forget the exact wording but AMD basically just told us that 32-bit UEFI was thoroughly deprecated and we'd be best advised to steer clear of it (in response to me asking how to tell in advance whether a system had 32-bit or 64-bit firmware, and how that interoperated with 32-bit or 64-bit kernels) | 16:55 |
| cjwatson | superm1: on my test system a 32-bit OS works OK though (with grub-efi-amd64) | 16:55 |
| cjwatson | I have no idea about definedness | 16:55 |
| cjwatson | but AMD's message was very much that this was a you-get-to-keep-both-pieces kind of thing, as I understood it | 16:56 |
| superm1 | cjwatson, okay i'll ping my internal contact about it to see if he understands a little bit better the expected results. would you mind allowing grub-efi-amd64 to resolve as a dependency for grub-efi then too in the event that it has a possibility of working? | 16:57 |
| cjwatson | um | 16:57 |
| superm1 | (on 32 bit that is) | 16:57 |
| cjwatson | I think I'd rather have a way to tell which firmware is installed | 16:57 |
| cjwatson | and have grub-installer install grub-efi-ia32 or grub-efi-amd64 as appropriate | 16:57 |
| cjwatson | unlike kernels, if you pick the wrong one it simply won't work | 16:58 |
| superm1 | Oh i see. | 16:58 |
| cjwatson | if you try media with grub-efi-ia32 on it, a machine with 64-bit UEFI won't even show it as bootable | 16:58 |
| superm1 | then is there actually hardware that does the 32 bit uefi? if AMD is saying that's the undefined bit and MS even doesn't offer a 32-bit UEFI variant | 17:01 |
| cjwatson | yes, there is, notably some Mac versions | 17:01 |
| cjwatson | I don't know about non-Mac hardware; it's quite possible there is but I get the impression it's becoming rarer | 17:02 |
| cjwatson | but I think on 32-bit UEFI it might well be better to just do legacy | 17:02 |
| cjwatson | it would be worth finding out if you guys have any plans to ship 32-bit UEFI (and if so WHY ...) | 17:02 |
| superm1 | well the problem is we're not moving to a 64 bit OS as shipping yet, so my hope was that we could drop 64 bit UEFI on top of 32-bit OS for now | 17:03 |
| cjwatson | it *should* work, I think, but I honestly don't know all the constraints | 17:03 |
| cjwatson | of course right now this test machine *only* boots 32-bit Linux on top of 64-bit UEFI :-/ | 17:04 |
| cjwatson | I'm certainly inclined not to bother shipping 32-bit UEFI on CDs, ever | 17:04 |
| superm1 | so then it sounds like it would make sense to make grub-efi depend on grub-efi-amd64 even on 32 bit, and only treat 32-bit uefi as a special case. cody and team will be producing images with correct language packs in the pool and all for us again that are 32-bit, so then he could just have 64-bit binaries for uefi on there | 17:07 |
| cjwatson | I think it would make more sense to stop using grub-efi, ultimately; I just didn't want to do that for maverick | 17:10 |
| cjwatson | I mean, it is labelled as transitional | 17:10 |
| superm1 | is it much more than a few lines change in grub-installer to handle that? | 17:11 |
| cjwatson | it seems to me that doing it properly involves, as I said, detecting which firmware is available | 17:12 |
| cjwatson | OEM could certainly change it trivially | 17:12 |
| superm1 | if there isn't an intention to include 32 bit uefi on cds ever, is it that important to detect which firmware was on there? grub-installer wouldn't be installing grub-efi-ia32 unless it was booted from a 32 bit uefi executable | 17:21 |
| cjwatson | it can't tell whether it was booted from a 32-bit or a 64-bit uefi executable | 17:24 |
| cjwatson | I think it can be detected from within /sys/firmware/efi/ somewhere anyway | 17:24 |
| cjwatson | I just haven't got round to figuring out the details yet | 17:24 |
| * cjwatson has a quick look | 17:26 | |
| cjwatson | grr, can't easily check, need to replace this test initrd with a netboot one so that I can get hold of efivars | 17:28 |
| cjwatson | wait, it's supposed to be built-in, so where's /sys/firmware/efi? | 17:28 |
| cjwatson | hmm. I think this has booted in legacy mode. | 17:29 |
| cjwatson | I'll have to look again later | 17:30 |
| CIA-34 | ubiquity: evand * r4302 trunk/ (bin/ubiquity-dm debian/changelog): Run nm-applet in the ubiquity GTK session. | 17:32 |
| ev | http://people.canonical.com/~evand/tmp/ubiquity-nm-applet.png | 17:33 |
| superm1 | ooh. i might have to try to fix the interaction with xfwm4, that's pretty nice to have now. | 17:48 |
| ogra | so sweet ! | 17:50 |
| ogra | i will sooo steal that panel for the arm netbook session in natty :) | 17:50 |
| ev | haha | 17:55 |
| ev | I should probably tidy it up in N then | 17:55 |
| ogra | make it its own package :) | 17:57 |
| ev | well yes, that too :) | 18:04 |
| hihihi100 | can u gimme a hand regarding the installation of alsa-driver-1.0.23? | 18:18 |
| cjwatson | this channel is about initial installation of the operating system; for help with individual packages, please use #ubuntu | 18:19 |
| ev | right, heading home. I have a patch to land for the prepare page text fixes, but they've uncovered a bug in wrap_fix (it doesn't like pango.SCALE_SMALL text) that I'm still trying to work through. | 18:20 |
| hihihi100 | can u at lest give me some tips regarding the slow speed of my 10.04? it needs a minute to fully load | 18:30 |
| hihihi100 | memory leak? | 18:30 |
| hihihi100 | it needs another full minute to shut down | 18:30 |
| cjwatson | hihihi100: I'm sorry, this isn't the right channel for that either | 18:36 |
| cjwatson | please ask in #ubuntu | 18:36 |
| cjwatson | we simply don't do things related to that | 18:37 |
| superm1 | ev i'm gonna get an upload together to get that new grub-installer fix in so we can see on tomorrow's disks if efi can get all the way through with it | 20:43 |
| CIA-34 | ubiquity: superm1 * r4303 ubiquity/ (d-i/manifest debian/changelog): | 20:44 |
| CIA-34 | ubiquity: Automatic update of included source packages: grub-installer | 20:44 |
| CIA-34 | ubiquity: 1.55ubuntu3. | 20:44 |
| CIA-34 | ubiquity: superm1 * r4303 ubiquity/ (d-i/manifest debian/changelog): | 20:44 |
| CIA-34 | ubiquity: Automatic update of included source packages: grub-installer | 20:44 |
| superm1 | cia seemed like it wasn't working^ | 20:44 |
| CIA-34 | ubiquity: superm1 * r4304 ubiquity/debian/changelog: releasing version 2.3.17 | 20:44 |
| ev | superm1: ah, yikes. Some of those things probably should've had a freeze exception. | 21:22 |
| ev | Not sure what we can do now though | 21:22 |
| ev | ugh, fail on my part for not remembering my own fixes | 22:03 |
| ev | any label with wrap set in ubiquity must also have yalign set to 0 in order for the wrap_fix to work | 22:03 |
| CIA-87 | ubiquity: evand * r4305 trunk/ (3 files in 2 dirs): | 22:07 |
| CIA-87 | ubiquity: Improve the text on the prepare page, based on input from Matthew | 22:07 |
| CIA-87 | ubiquity: Paul Thomas, Amanda Brock, and Andrew Sinclair. | 22:07 |
| * cjwatson goes grr at an LWN commenter saying that the new partitioner probably wasn't our work and should be credited to RH | 22:14 | |
| cjwatson | and manages to post a reasonably polite reply | 22:14 |
| cjwatson | (http://lwn.net/Articles/403837/) | 22:14 |
| ScottK | Reminds me I need to see about getting my subscription renewed. | 22:16 |
| ev | ugh | 22:21 |
| ev | oh my god, rounded corners, it must be Mac. | 22:22 |
| ev | seriously, no one said *anything* when we had the GtkVPane based rectangle layout back in the day | 22:23 |
| ev | *sigh* | 22:23 |
| ScottK | Would one of you mind giving me the "see this article for free link"? | 22:24 |
| ev | surely | 22:25 |
| ev | http://lwn.net/SubscriberLink/403837/0b3d34a37b65f118/ | 22:26 |
| ScottK | THanks | 22:26 |
| superm1 | ev, doh.... well looking through the ones that would most likely be in need of exception would have been the panel moving icons to the side and the nm-applet added to it. sorry i should have paid closer attention. | 22:44 |
| superm1 | if someone yells about anything i'll prepare diffs to revert those specific things until appropriate exceptions are added | 22:45 |
| ev | no worries, I'll just incorporate it in the freeze exception for the prepare page stuff | 22:45 |
| superm1 | i wasn't sure you were doing more fixes today for the prepare page stuff, and wanted to make sure the new grub stuff could make tomorrow's disks | 22:46 |
| ev | indeed, I understand completely | 22:47 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
