[08:22] <infinity> If someone's around to review, the above glibc is needed on arm64.  mwhudson may have tested one patch and then sent me another. :/
[08:22] <infinity> (This one was definitely tested this time... *sigh*)
[09:19] <knome> infinity, in the Release Notes, under "New Features in 14.04", does the subheading "Ubuntu" mean "Ubuntu Desktop" or "Common" (which i'd imagine what "Updated Packages" are)?
[09:39] <cjwatson> infinity: done
[09:52] <knome> infinity, the more i look at the release notes page, and the various ways how teams (want to?) do their release notes, the more i think we should just keep the flavors out of the generic release notes and make sure the generic page either (1) stays clean enough to be sane to link directly (2) has comments that allow "including" parts of the pages to the flavor-specific ones
[09:59] <Laney> cjwatson: did you want to unblock it too?
[10:01] <cjwatson> oh duh, yeah, done
[10:02]  * cjwatson uploads the source change necessary for ghc/ppc64el to unstable
[10:02] <cjwatson> last bootstrap stages should have finished by the time I can sync that
[11:15] <knome> what's the general thought about fixing a low, but annoying bug along with a high/critical bug during the final freeze?
[11:15] <knome> (in the same package)
[11:19] <cjwatson> I'd probably be OK with it as long as the diff was clear and we could make a good estimate from it that the risk is low
[11:19] <cjwatson> (but away for a while now)
[11:19] <knome> ok, cheers
[11:33] <zequence> We might need to make a change in our seeds, and update our meta source (bluetooth stuff). Also, it would be great if we could get this uploaded Bug 1291675
[11:33] <ubot2> Launchpad bug 1291675 in lmms (Ubuntu) "[FFe] LMMS 1.0.0" [Wishlist,Triaged] https://launchpad.net/bugs/1291675
[11:33] <zequence> (we == Ubuntu Studio)
[11:34] <knome> zequence, you should be in touch with cyphermox about the bluetooth stuff, i think he could take care of that...
[11:34] <knome> since it is, ultimately, a side-effect of what we're doing
[11:34] <zequence> knome: Yep. I'll talk with him later. A bit in a hurry atm :)
[11:34] <knome> yep
[11:35] <zequence> anyways, we'll most probably need to rebuild our ISO
[11:35] <knome> ^ re "bluetooth stuff", we're planning to (re)demote gnome-bluetooth in network-manager-gnome to "suggests" to avoid installing unity-control-center + ibus
[13:28] <xnox> infinity: what's up?
[13:31] <tumbleweed> I'm completely out of touch these days, but ginggs asked me to look at bug 1289463. it's a tiny transition, so I'm tempted to approve it. any other opinions? (we often release with a broken cuda stack, and then fix it with SRUs, so screwing up here wouldn't be the end of the world)
[13:31] <ubot2> Launchpad bug 1289463 in starpu-contrib (Ubuntu) "[FFe] Merge nvidia-cuda-toolkit 5.5 from Debian, transition to libcuda 5.5" [Undecided,Confirmed] https://launchpad.net/bugs/1289463
[15:02] <ScottK> tumbleweed: I'd go for it.
[15:03] <maclin_> slangasek, there is a critical problem about ubuntu kylin software center: Bug #1306871. I have added first-run update funciton. Can we update this version to solve this problem?
[15:03] <ubot2> Launchpad bug 1306871 in Ubuntu Kylin Software Center "software center only shows few apps before apt update(软件中心只显示少部分软件，更新源后正常)" [High,Fix committed] https://launchpad.net/bugs/1306871
[15:04] <tumbleweed> ScottK: thanks
[15:37] <bluesabre> release team, is it possible to get this bugfix release in before final release?
[15:37] <bluesabre> https://bugs.launchpad.net/ubuntu/+source/abiword/+bug/1306952
[15:37] <ubot2> Launchpad bug 1306952 in abiword (Ubuntu) "Abiword bug fix update for trusty" [Undecided,Confirmed]
[15:37] <bluesabre> I understand that knome discussed this with cjwatson yesterday
[16:05] <knome> well not the specific issue but generally pulling a few low-priority, low regression potential fixes in with the high/critical fix
[16:05] <knome> and this would be the bunch ^
[17:03] <slangasek> maclin: hi, looking at the bug now. do you need sponsorship+review, or just review?
[17:04] <slangasek> maclin: well, I'm not able to load the bug page for some reason.  Do you have a code branch you can point me at for the fix?
[17:04] <maclin> slangasek, thanks, I need both of that:)
[17:06] <maclin> slangasek, the code branch is: https://code.launchpad.net/~ubuntukylin-members/ubuntu-kylin-software-center/trunk
[17:06] <xnox> slangasek: should plymouth in shutdown mode, be started with "--attach-to-session" like we do on boot, and thus redirect all console output? (possibly further improving message less shut-down, no?)
[17:07] <xnox> slangasek: maclin: https://code.launchpad.net/~xnox/ubuntu-kylin-software-center/invalid-syntax/+merge/215325 hasn't been merged. So if we want another upload, i'd like that to be included as well.
[17:07] <xnox> slangasek: i can review trunk and make another release.
[17:07] <slangasek> xnox: that'd be great, thanks
[17:07] <slangasek> xnox: --attach-to-session> I confess that the usage of this on startup but not shutdown predates me; not sure if that was a latent bug or if the semantics of the option changed since it was first deployed
[17:09] <xnox> slangasek: i've used attach-to-session, and removed all lightdm "clear > /dev/tty7" and I still have messages-less shutdown. I'll open a bug about upgrading to that configuration and test it more. I guess it's for u-cycle, with possibly sru's.
[17:10] <slangasek> xnox: are we currently starting plymouthd on shutdown before lightdm is killed, and have we determined that this is safe?  (The last time I tried this I got bad things on shutdown)
[17:10] <slangasek> xnox: if not, then there's definitely a window between lightdm stop and plymouthd start where messages *may* be left on console
[17:11] <maclin> xnox,  i will check the merge
[17:11] <xnox> slangasek: we explicetely enforce that lightdm stops, before plymouthd for shutdown is execed. I've added wait-for-state stopped in plymouth-shutdown.conf pre-start.
[17:12] <slangasek> xnox: in this case, I don't think you should drop the 'clear' from the lightdm job
[17:12] <xnox> slangasek: right, i see. Was the problem with exec plymouthd, or executing "show-splash" whilst lightdm was still using it?
[17:12] <slangasek> xnox: since the 'clear' is there to deal with any kernel messages that may have been printed underneath lightdm while running
[17:13] <slangasek> xnox: I don't remember, sorry
[17:13] <slangasek> I think I was seeing problems with just the exec plymouthd while lightdm was running, no show-splash
[17:13] <xnox> slangasek: re:lightdm kernel messages -> right makes sense.
[17:13] <xnox> ok.
[17:14] <infinity> zul: Your swift upload drops a patch as "no longer needed", but I see no indication of a fix in the new upstream.
[17:14] <slangasek> xnox: can confirm here that the serial console output here from plymouth-upstart-bridge is much prettier now, thanks
[17:14] <infinity> zul: Also,
[17:14] <infinity> +Breaks: swift-object (<< 1.12.0-0ubuntu2), swift-object (<< 1.13.1~rc1-0ubuntu1~)
[17:15] <infinity> +Replaces: swift-object (<< 1.12.0-0ubuntu2), swift-object (<< 1.13.1~rc1-0ubuntu1~)
[17:15] <xnox> infinity: what were your questions about upstart patch? i've tested it locally from test-build before uploading and I saw no ill effects. are there bugs that got uncovered somewhere?
[17:15] <infinity> zul: That makes no sense.  "foo (<< 1), foo (<< 2)" just means "foo (<< 2)"
[17:15] <xnox> slangasek: excellent! =)
[17:16] <infinity> xnox: Hrm?  No, I haven't noticed any ill effects, I was just curious how carefully you reviewed it, since it was something I just banged out in a few minutes due to frustration, I didn't test or anything. :P
[17:17]  * slangasek looks at https://bugs.launchpad.net/ubuntu/+source/cryptsetup/+bug/1305961/comments/4
[17:17] <ubot2> Launchpad bug 1153661 in cryptsetup (Ubuntu) "duplicate for #1305961 The disk drive for /dev/mapper/cryptswap1 is not ready yet or not present" [Medium,Triaged]
[17:17] <slangasek> seriously, how do you think that's "confirming" a bug?
[17:19]  * slangasek finds himself losing time studying the Chinese characters in the description of that uksc bug, and decides he should probably find something more productive to do with his time this morning
[17:25] <infinity> slangasek: You could come do my laundry.  I wouldn't say no.
[17:26] <slangasek> infinity: oh, given those options I think I'll stay here and be amused that "software" has a literal translation
[17:26] <xnox> slangasek: i need help with efibootmgr vs UEFI spec =) where do ACPI(a0341d0,0)PCI(4,0)03171000010000000000000000000000HD(1,800,f3800,9265fded-5be9-4265-a2c7-d5ebeba33954)File(\EFI\ubuntu\shimx64.efi) values come from / what is their definition?
[17:27] <infinity> slangasek: Like "squishy goods"?
[17:27] <xnox> slangasek: e.g. efibootmgr is failing to generate "HD(1,800,f3800,9265fded-5be9-4265-a2c7-d5ebeba33954)" from my device, instead it generates e.g. HD(1,0,0,1b960301), which tianocore is not happy about.
[17:27] <xnox> (and does not boot)
[17:27] <slangasek> infinity: "soft goods", but yes
[17:28] <slangasek> xnox: ah, I don't know the answer to this
[17:28] <slangasek> xnox: is this nvme again?
[17:28] <xnox> slangasek: yes =)
[17:29] <slangasek> xnox: hmm, I thought you got that fixed already by virtuous upgrades
[17:30] <xnox> slangasek: oh, tianocore now knows about nvme, and our kernel/ubiquity/grub-installer do, the last portion to get it reboot out of the box is to get efibootmgr to generate the right entry. At the moment, i need to go into tianocore boot manager and add the right boot entry to boot it.
[17:30] <xnox> i guess instead of reading the UEFI spec i can check tianocore sources to see how it decides what "HD(1,800,f3800,9265fded-5be9-4265-a2c7-d5ebeba33954)" means.
[17:31] <slangasek> xnox: ok, so all I can suggest is to look in the efibootmgr source; I suspect it's getting these values from EFI itself, so it might be a remaining edk2 bug
[17:31] <xnox> slangasek: oh really, interesting.
[17:31] <slangasek> (I'm assuming you mean edk2 when you say tianocore, rather than one on a physical system?)
[17:31] <xnox> slangasek: correct, i mean edk2.
[17:31] <slangasek> xnox: well, the one bit looks an awful lot like a UUID, which I'm sure efibootmgr isn't generating on its own
[17:31] <maclin> slangasek: xnox:  I have merged https://code.launchpad.net/~xnox/ubuntu-kylin-software-center/invalid-syntax/+merge/215325, the new code is on https://code.launchpad.net/~ubuntukylin-members/ubuntu-kylin-software-center/trunk
[17:32] <xnox> slangasek: i see, i thought it would be GPT UUIDs or some-such but those didn't match.
[17:33] <xnox> slangasek: also can i disable 3second grub-efi timeout on boot? it's annoying.
[17:34] <slangasek> xnox: ah, so I see (comparing with my own entries here) that the first field in the parens is the partition number; not sure about the others
[17:34] <slangasek> xnox: you can't disable it /in the archive/, it's the only way to make the boot menu accessible in UEFI
[17:34] <slangasek> you can override it locally if you like
[17:35] <xnox> i've set timeout and hidden timouts to zero, but that's not it.
[17:35] <xnox> what's the "feature" name?
[17:35] <slangasek> not sure
[17:36] <slangasek> that's a cjwatson question :)
[17:37] <slangasek> (for my part, I think it was a mistake to upgrade to a version of grub that natively supports a rich scripting language, and then continue to generate the scripts from another language)
[17:39] <xnox> slangasek: found definition of the headers. So it looks like NVMe support in edk2 is bogus. Second field is "Type an integer between 0-255 or else the keyword MBR or GPT" of which "800" is not it =)
[17:39] <cjwatson> I don't know what timeout this is
[17:39] <slangasek> xnox: and the error is on the edk2 side, not the efibootmgr side?
[17:39] <cjwatson> (But I only just woke up from a nap, bit woozy)
[17:39] <slangasek> cjwatson: the EFI-can't-detect-shift-modifier timeout
[17:40] <slangasek> I seem to have that disabled here, but buggered if I know which of the settings in my /etc/default/grub actually did it
[17:40] <xnox> cjwatson: with GRUB_HIDDEN_TIMEOUT_QUIET=false and GRUB_HIDDEN_TIMEOUT=0 and GRUB_TIMEOUT=0 i get a 3second count-down on boot, when fastpath boot is enabled on my laptop.
[17:41] <slangasek> xnox: and with GRUB_HIDDEN_TIMEOUT_QUIET=true?
[17:41] <xnox> slangasek: 3s delay without countdown printed on screen.
[17:41] <slangasek> really?  I don't see any 3s delays in the source, only a 5s delay
[17:42] <cjwatson> OK, I really don't recall, sorry, would have to look when I'm properly awake
[17:42] <xnox> slangasek: i've tested it by setting a very visual grub2 background image which stays for 3 seconds. E.g.: cp /usr/share/backgrounds/warty-final-ubuntu.png /boot/grub && update-grub && reboot
[17:42] <cjwatson> (Which isn't going to be today at this rate, urgh)
[17:42] <infinity> cjwatson: Waking up on Saturdays is overrated.
[17:43] <xnox> cjwatson: no worries, you'll be able to experiment  / see it on my laptop next week =)
[17:43]  * knome pours cjwatson a big cup of coffee
[17:45] <infinity> cjwatson: Do we want a udeb britney block, as a subtle reminder that we might need to respin d-i (as we currently do for my glibc upload, for instance)
[17:45] <infinity> Well, we don't *need* to, but I really dislike d-i not matching the sources.
[17:45] <cjwatson> I guess it might be helpful
[17:45] <cjwatson> There's already code for it I think
[17:46] <infinity> Yeah, block-all-udeb
[17:46] <infinity> I think.
[17:47] <infinity> Oddly, not mentioned in the README, or I'm blind.
[17:48] <infinity> Okay, not mentioned in the source either, so I might be on something.
[17:49] <infinity> Meh, if I have to generate a "block-udeb" for every source that generates one, I think I just lost interest.
[17:53] <xnox> slangasek: i found NVMe in the UEFI 2.4A spec, which defines special device path nodes.
[18:08] <bluesabre> everybody appears when I walk away :)
[18:08] <bluesabre> cjwatson: we have a clean diff now, if you're interested
[18:08] <bluesabre> https://bugs.launchpad.net/ubuntu/+source/abiword/+bug/1306952
[18:08] <ubot2> Launchpad bug 1306952 in abiword (Ubuntu) "Abiword bug fix update for trusty" [Undecided,Confirmed]
[18:08] <bluesabre> (or awake enough) ;)
[19:01] <cjwatson> bluesabre: I was just answering briefly earlier, I'm not fit to be the reviewer right now (and wasn't volunteering earlier).  Hopefully somebody else can
[19:16] <zul> infinity:  crap can you reject swift please
[19:21] <ScottK> zul: Done.
[19:22] <knome> ScottK, any possibility you could look at the bug bluesabre mentioned?
[19:22] <ScottK> Don't have time to look at uploading stuff now, sorry.
[19:23] <knome> we'd be fine with a release team ACK that it's ok to land it
[20:12] <infinity> knome / bluesabre: Go ahead and upload.  We can review it in the queue if there looks to be anything amiss, but a quick once-over looks fine.
[20:13] <infinity> (Sorry, in a panic mode today, don't have the time for a solid review until I get my laundry done... Maybe I can look from the airport)
[20:15] <knome> infinity, cheers, now onto finding a sponsor :)
[20:18]  * ogra_ reboots after upgrade and hopes that the screensaver now stops keeping the focus in xchat instead of the password field when locking
[20:29] <apw> ogra_, we all hope for that ...
[20:29] <ogra_> heh, i'm not alone then :)
[20:33] <apw> ogra_, man no, though i know the signs now before hitting return
[20:34] <ogra_> change user works quite well as workaround
[20:34] <apw> in my case always type /b 1 first, then your password just ni case, and if there is no cursor, fiddle with the indicators to get it back
[20:34] <apw> i find opening and closing an indicator gets focus back
[20:34] <ogra_> ah
[20:34] <apw> but noticing is the key
[20:34] <ogra_> thats easier than change user
[20:34] <ogra_> yup
[20:35] <cjwatson> Right, that's ghc bootstrapped on ppc64el, so hopefully lots of rows will clear out of /ftbfs/
[20:36] <apw> cjwatson, you are a machine ...
[20:36] <cjwatson> Obsessive would be closer to the mark
[20:36] <cjwatson> Given it's a language I don't actually use :-)
[20:37] <apw> heh ... well i am impressed none the less :)
[20:39] <infinity> Insane is the word I'd use.
[20:39] <infinity> And have.
[21:57] <maclin> slangasek, xnor, could you take a time to review the ubuntu kylin software center for a new release: https://code.launchpad.net/~ubuntukylin-members/ubuntu-kylin-software-center/trunk?  I have merged https://code.launchpad.net/~xnox/ubuntu-kylin-software-center/invalid-syntax/+merge/215325.
[21:59] <maclin> sorry,  xnox,  i typed a wrong word^
[22:02] <maclin> This is mainly for the critical bug of cache update: #bug 1306871
[22:02] <ubot2> Launchpad bug 1306871 in Ubuntu Kylin Software Center "software center only shows few apps before apt update(软件中心只显示少部分软件，更新源后正常)" [Critical,Fix committed] https://launchpad.net/bugs/1306871