IRC log for #elinux on 20110516

01:50.47*** join/#elinux geep (~JoeBelow@Bas21-toronto12-1242559789.dsl.bell.ca)
03:58.09*** join/#elinux ashes (ashes@modemcable026.148-202-24.mc.videotron.ca)
05:14.46*** join/#elinux toi (~peter@vrt.nat.ibbt.be)
06:01.08*** join/#elinux vanner_ (~vanner_@nat2.rnd.stcnet.ru)
06:17.21*** join/#elinux lyakh (~lyakh@dslb-088-076-017-191.pools.arcor-ip.net)
06:30.57*** join/#elinux risca (~risca@130.236.250.26)
06:45.08*** join/#elinux risca (~risca@130.236.250.26)
07:03.25*** join/#elinux risca (~risca@130.236.133.218)
07:21.17*** join/#elinux toi (~peter@vrt.nat.ibbt.be)
07:26.44*** join/#elinux dijenerate (~dijenerat@204.212.240.167)
07:52.27*** join/#elinux speakman (~daniel@unaffiliated/speakman)
08:41.38*** join/#elinux dijenerate (~dijenerat@204.212.240.167)
11:43.52*** join/#elinux risca (~risca@h241n2-n-a31.ias.bredband.telia.com)
11:49.28*** join/#elinux cb22 (~federico@moinmoin/developer/federico)
11:50.19cb22Hi, I'm trying to compile an OE distro for
11:50.57cb22the Archos 101s - compilation and everything isn't an issue, but it appears as if NEON optimizations are causing every compiled binary to segfault when run on the device
11:51.24cb22My kernel has CONFIG_NEON enabled, and /proc/cpuinfo reports http://pastebin.com/WDWwFDLh
11:51.32cb22Anyone have any ideas?
11:51.58cb22(I'm basing my config off the BeagleBoard - they are very similar in terms of hardware specs)
11:56.42*** join/#elinux vanner_ (~vanner_@nat2.rnd.stcnet.ru)
12:25.21*** join/#elinux gustavoz (~gustavoz@host58.186-109-9.telecom.net.ar)
12:45.21wmatcb22: why do you conclude that NEON is causing the segfaults?
13:00.12cb22wmat, I'm actually compiling another toolchain now to test it (csl 2007q3) before I was using gcc 4.5. A binary compiled with -march=armv7-a -mtune=cortex-a8 -mfpu=neon -mfloat-abi=softfp while the same one compiled without those flags seem to be fine
13:00.45cb22However, I think the two might have been compiled with different compilers - hence compiling another one to make doubly sure first
13:01.10cb22I've also done a fair bit of searching, and seen that NEON issues can pop up quite often
13:02.20wmatyes, it can be tricky
13:42.55*** join/#elinux GPSFan (~kenm@64.92.145.112)
14:01.44*** join/#elinux dijenerate (~dijenerat@204.212.240.167)
14:06.39*** join/#elinux risca (~risca@h241n2-n-a31.ias.bredband.telia.com)
14:12.23cb22Hm - it would appears as if http://rcn-ee.net/deb/rootfs/natty/ubuntu-11.04-r0-minimal-armel.tar.xz yields the same results for me. I'd presume it works on the beagleboard... I'm using kernel 2.6.29
14:19.38*** join/#elinux tgall_foo (~tgall@206.9.88.154)
14:20.13*** join/#elinux tgall_foo (~tgall@linaro/tgall-foo)
14:22.11wmatcb22: you could ask in #beagle, as the guy who created that image usually is there
14:45.32*** join/#elinux roxfan (dunno@91.180.66.153)
15:06.54*** join/#elinux gustavoz (~gustavoz@host58.186-109-9.telecom.net.ar)
15:17.21*** join/#elinux vanner_ (~vanner_@nat2.rnd.stcnet.ru)
15:20.27*** join/#elinux risca (~risca@h241n2-n-a31.ias.bredband.telia.com)
15:47.59*** join/#elinux Russ (foobar@ip70-176-251-1.ph.ph.cox.net)
15:47.59*** mode/#elinux [+o Russ] by ChanServ
16:03.41*** join/#elinux DEADBEEF- (~DEADBEEF@65.49.73.214)
16:05.25cb22Looks like some success :) the 2011 CSL toolchain appears to work great!
16:05.46cb22At least, busybox didn't segfault, like it did before. Time to compile a few things more complicated
16:12.53*** join/#elinux toi (~peter@94-226-59-90.access.telenet.be)
16:46.48*** join/#elinux toi (~peter@94-226-59-90.access.telenet.be)
19:30.19*** join/#elinux krushia (~krushia@h69-131-92-179.cntcnh.dsl.dynamic.tds.net)
22:11.06*** join/#elinux roxfan (dunno@91.180.66.153)
23:14.58*** join/#elinux geep (~JoeBelow@Bas21-toronto12-1242559789.dsl.bell.ca)
23:54.14*** join/#elinux dijenerate (~dijenerat@204.212.240.159)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.