**** BEGIN LOGGING AT Thu Feb 11 02:59:58 2016
Feb 11 08:30:57 <mckoan> good morning
Feb 11 11:14:19 <rburton1> ndec: did you get piglit to work?
Feb 11 11:20:40 <ndec> rburton1: yes.
Feb 11 11:25:31 <ndec> rburton1: i will add piglit test in our CI loops. with freedreno we expect to have ~96% pass rate. this is the current status of the driver. we never ran piglit on OE, but that's what we get on debian/fedora.. 
Feb 11 11:26:15 <rburton1> is that excluding skips, i presume
Feb 11 11:26:24 <rburton1> in theory you'll get ~identical results right :)
Feb 11 11:26:39 <funman> koen: is variable expension possible from python functions? http://pastie.org/10717704
Feb 11 11:26:41 <ndec> yes, excluding skip.
Feb 11 11:27:04 <rburton1> funman: use d.getVar or d.expand
Feb 11 11:27:21 <ndec> rburton1: yes, it should be identical.. given the piglit version and the mesa version might not be completely aligned..
Feb 11 11:27:34 <funman> rburton1: rburton1 that's what I did for UBOOT_*IMG
Feb 11 11:28:00 <funman> but on another machine the variables are expanded correctly without that diff :/
Feb 11 11:28:21 <rburton1> ndec: i just posted an update of piglit so its a bit more recent now.  need to get python-six in oe-core to go any further
Feb 11 11:28:27 <rburton1> funman: yes, that was changed this week
Feb 11 11:28:42 <rburton1> if you track master, read the git log and mailing lists :)
Feb 11 11:28:42 <funman> ah ok, which change was it?
Feb 11 11:29:01 <ndec> rburton1: yes, i saw that. i will try it. not until next Tue, though.. 
Feb 11 11:29:19 <funman>     scripts/devtool: Add getVarFlag expand argument
Feb 11 11:29:40 <rburton1> "    bitbake: build/data: Don't expand python functions before execution [API change]"
Feb 11 11:29:50 <funman> was it on oe-core list?
Feb 11 11:29:54 <rburton1> bitbake
Feb 11 11:29:57 <rburton1> as its a bitbake change
Feb 11 11:30:27 <funman> I see
Feb 11 11:30:59 <rburton1> (cc'd oe-arch which is also worth reading if you're tracking master)
Feb 11 11:31:10 <funman> i couldnt use d.getVar("${S}/u-boot") though
Feb 11 11:31:27 <rburton> correct
Feb 11 11:31:30 <rburton> getVar takes a variable name
Feb 11 11:31:49 <rburton> so either getVar and then glue it with os.path.join, or use d.expand
Feb 11 11:32:10 <rburton> ie os.path.join(d.getVar("S", True), "u-boot")
Feb 11 11:32:18 <rburton> or d.expand("${S}/u-boot")
Feb 11 11:35:13 <funman> i had missed d.expand, looks perfect
Feb 11 17:11:21 <dv_> JaMa: about ne10: I get the textrel problem even if I explicitely add -fPIC
Feb 11 17:11:55 <dv_> I wonder if the assembly bits are responsible
Feb 11 17:14:32 <JaMa> in worse case you can INSANE_SKIP them, but I don't want more QA issues in build than we currently have
Feb 11 17:15:14 <dv_> understandable.
Feb 11 17:15:24 <dv_> btw, the textrel issue with ffmpeg is still unresolved?
Feb 11 17:16:11 <JaMa> afaik yes
Feb 11 19:24:13 <muep_> is it intended that instructions at http://www.openembedded.org/wiki/OE-Core_Standalone_Setup would most of the time also work for master?
Feb 11 19:25:26 <muep_> ah silly me, I guess I had bitbake out of date
**** ENDING LOGGING AT Fri Feb 12 02:59:58 2016