[00:06] RAOF: what should i be looking from in xinput --test-xi2 [00:06] Whether you get multitouch events. I'm not *entirely* sure what they look like, though :) [00:07] My magic touchpad is in my missing luggage. [00:07] when its busted, i get 8 valuators going off while 2 finger scrolling, when its not i get 1 [00:09] well 8 flags [00:10] it will be hard as hell to capture spurious 4 finger taps while 2 finger scrolling in xinput --test-xi2, it happens like once every 5-10 minutes [00:10] and thats so much info printed in the meantime [00:17] I'd just dump the logs into a pastebin and throw it at whot :) [00:19] ok next time i want to inflict that crap upon myself (tomorrow morning) i'll do it, right now no mt gestures work cus i rmmod/modprobed bcm5974 so 2 finger scroll is usable :) [00:19] :) [00:20] * Sarvatt cant wait to get a limited piece of crap elantech touchpad so these things go away when ux31 comes with 1080p screens with ivybridge :) [06:37] bryceh: hey, looks like totals-precise-workqueue.svg is having some issues? === chrisccoulson_ is now known as chrisccoulson === yofel_ is now known as yofel [14:09] tjaalton, yep [14:14] bryceh: was known already? :) [14:15] yeah, hardware failure [14:16] ah :) [14:18] I moved all the scripts over to new hw. everything works except the graphs for some reason [14:20] ok [14:21] no rush, I like to think everything is smooth and working [14:21] (and not look at the bugmail folders..) [14:21] :) [14:22] heh [14:22] yeah, probably something simple; looks like the data is all generating properly, it's just gnuplot being cranky [14:25] tjaalton, I have things set up that whenever a script fails, it emails me. So I have an INBOX full of errors this morning. Hard to miss :-) [14:26] bryceh: heh :) [14:26] semi-fixed [14:35] looks good, or is the timescale a bit limited? [14:39] missing historical data; the scripts are not parsing the older json files. I think the json parser on the new machine is stricter on syntax [14:39] stand by [14:39] yeah that's why totals.svg looks the way it does [14:39] (need to figure out a good way to delete one comma from 408 files) [14:40] sed?-) [15:33] hah, launchpad is reporting quantal is now the current development codename [15:33] and is not listing precise as a supported release [15:34] oh good, that was fast [15:35] right, so with that, back online: http://www.bryceharrington.org/Arsenal/ubuntu-x-swat/Reports/totals-precise-workqueue.svg [15:36] totals.svg looks funny :) [15:36] oneiric is also not listed as supported. odd [15:37] tjaalton, hmm yeah what's up with totals.svg... [15:48] ok, here the problem is too *much* historical data [15:48] bryceh: not using all the data? looks like it leaves gaps [15:48] ah [15:49] maybe filter it so that the timescale is still the same but "resolution" is smaller [15:52] nah, just have to be more selective in the data files to use [15:52] tjaalton, ok that one should be fixed. going to get breakfast and will poke around for other breakages after that. [15:54] tjaalton, btw I also fixed it so workqueue graphs continue to get generated and updated on past releases, although a few look wrong [15:54] bryceh: thanks :) [15:54] e.g. so we can keep using the precise workqueue graph for tracking LTS work [15:54] yeah that's good [16:56] tjaalton, http://humber/Arsenal/ubuntu-x-swat/Reports/package_status.html is the new versions-current; decided package_status was a clearer name. [16:58] that script had also been on the failed system, but was more severely broken since it requires a chroot, so had to build a new one on the new system yesterday. [16:58] tjaalton, I updated it for quantal while I was at it, but looks like there's no data yet since dev hasn't opened [16:58] http://humber? [16:59] ok http://bryceharrington.org/Arsenal/ubuntu-x-swat/Reports/package_status.html [16:59] :) [16:59] (for those not on my home network) [16:59] hmm. no mesa there? [17:55] bryceh: ah, noted [17:55] jcristau: seems to be there now [18:59] why does wine want to deinstall debhelper et al [19:00] likely not wine but one of it's deps [19:00] maybe it pulls in some i386 that uninstall 64-bits versions? [19:00] could be === JanC_ is now known as JanC [20:24] cnd, I was just doing some more debugging with the bcm5974. I'm getting the weird behavior, but the number of touches x-x-i-s sees is correct. [20:28] cnd, whether or not I get left- or right-click behavior definitely seems to correspond to where my finger is on the touchpad, but the num_touches still always shows the correct value [20:33] cnd, I just added this information to the bug as well === RAOF_ is now known as RAOF