=== ben_r_ is now known as ben_r
=== ben_r_ is now known as ben_r
=== manjo` is now known as manjo
=== setuid is now known as setuid_
=== dannf` is now known as dannf
[22:13] <dmj_s76> apw: kamal: Still need to test a potential fix, but there's a quite annoying bug affecting installs on Nvidia systems with 10 Series (Pascal) GPUs.  When using Nouveau with the 4.8 kernel, the cursor is only drawn stuck in the top left corner.
[22:14] <dmj_s76> The actual mouse pointer can be moved and clicked anywhere (though you can't see where it really is)
[22:15] <dmj_s76> This makes it essentially impossible for people to use their mouse during install.
[22:15] <bjf> dmj_s76, after install is it still broken?
[22:16] <dmj_s76> I've talked with the Nouveau bugs and this is fixed in 4.10 apparently, and they pointed to a couple commits that implement the needed functionality.
[22:17] <bjf> dmj_s76, is there an LP bug about it?
[22:18] <dmj_s76> bjf: It's a problem any time you're running nouveau.  It's fixed once you install the proprietary driver.
[22:18] <dmj_s76> bjf: I did find an existing bug report: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-nouveau/+bug/1646574
[22:18] <ubot5> Ubuntu bug 1646574 in xserver-xorg-video-nouveau (Ubuntu) "Mouse cursor invisible or does not move" [Undecided,Confirmed]
[22:22] <bjf> dmj_s76, ok, i think that makes it a kernel bug and not xserver
[22:24] <bjf> jsalisbury, LP: #1646574 can i trouble you to whip up a test kernel with the 3 commits in comment #23?
[22:24] <ubot5> Launchpad bug 1646574 in linux (Ubuntu) "Mouse cursor invisible or does not move" [Undecided,Confirmed] https://launchpad.net/bugs/1646574
[22:25] <bjf> dmj_s76, if you would kindly test that kernel when it's added to the bug that would be fantastic. we can get the fix into the next SRU cycle for the kernel
[22:26] <dmj_s76> bjf: hopefully we can fix this for the 16.04.2 iso.  I'm able to test the fix on a dozen nvidia machines.
[22:26] <dmj_s76> bjf: Great!
[22:26] <bjf> dmj_s76, nope, missed the window for that i'm afraid
[22:26] <dmj_s76> hrm
[22:27] <bjf> yeah, that sucks
[22:28] <dmj_s76> bjf: There isn't even an iso with the kernel.
[22:28] <dmj_s76> Nobody's been able to actually test how the iso works with 4.8 and the new graphics stack.
[22:29] <bjf> dmj_s76, the kernel that will be used for the .2 release is the kernel that is currently in -updates
[22:29] <bjf> infinity, ^ is true isn't it?
[22:30] <jsalisbury> bjf, no problem
[22:30] <dmj_s76> bjf: Yes, but there have been issues in the past that were only caught by testing the iso.  14.04.5 wouldn't boot with NVIDIA a week before release.
[22:31] <bjf> dmj_s76, i'm just telling you what i know right now
[22:31] <dmj_s76> bjf: Fair enough, not trying to be argumentative
[22:32] <bjf> dmj_s76, if we are using the kernel that is in -proposed for the .2 release there is a chance we could get this fixed and into the .2 release
[22:32] <bjf> dmj_s76, but i need infinity to tell me what he's using to build the .2 isos with
[22:34] <bjf> dmj_s76, didn't mean to imply you were arguing. i understand your concern
[22:35] <dmj_s76> bjf: Okay...yeah, there is a 4.8 kernel in -updates (which we've been testing with for our production image testing) but there hasn't been a daily iso yet with 4.8, which means there very well could be blocking issues.
[22:37] <bjf> dmj_s76, we don't have any know regressions right now that would cause a respin of the 4.8 kernel but if this fixes your issue i'd probably respin to pick it up and have it in -proposed
[22:40] <dmj_s76> This is a regression that does affect the yakkety iso (*way* too late...there was a lot going on at that time) but the 16.04 and 16.04.1 isos have worked well.  We're quite sure this will change when 4.8 lands.
[22:40] <dmj_s76> bjf: Thanks, hopefully this works like a charm.
[22:40] <bjf> yeah
[23:04] <infinity> bjf: The final ISOs will ship with whatever kernel you and Andy tell is sane. :P
[23:04] <bjf> infinity, hmm... ok
[23:04] <infinity> dmj_s76: If this affected yakkety, why sit on it for 3 months before pointing out its criticality again? :/
[23:04] <bjf> will discuss that with apw
[23:05] <infinity> bjf: Andy's been twiddling linux-hwe for other reasons, so squeezing in a few other commits IF TESTED would be entirely fine.
[23:05] <bjf> infinity, ack
[23:06] <dmj_s76> infinity: We did drop the ball a little...didn't notice in time for yakkety and I got quite busy with other fires.
[23:06] <dmj_s76> infinity: Thanks, we'll try to be a little more on the ball next time.