[00:33:25] *** Styx has quit IRC
[00:48:14] *** magn3ts has joined #chromium-os
[01:16:56] *** sadrul has quit IRC
[01:20:37] *** sadrul has joined #chromium-os
[01:20:43] *** ChanServ sets mode: +v sadrul
[01:56:24] *** magn3ts has quit IRC
[01:57:27] *** lipsinV2 has joined #chromium-os
[02:25:42] *** lipsinV21 has joined #chromium-os
[02:25:42] *** lipsinV2 has quit IRC
[02:25:43] *** lipsinV21 has quit IRC
[02:25:43] *** lipsinV21 has joined #chromium-os
[04:11:30] *** yosafbridge has quit IRC
[04:11:41] *** yosafbridge has joined #chromium-os
[04:53:33] *** JasonO has quit IRC
[04:59:20] *** perezd has quit IRC
[05:06:11] *** Xavierdarkness_ has joined #chromium-os
[05:06:38] *** Xavierdarkness has quit IRC
[05:06:52] *** Xavierdarkness_ is now known as Xavierdarkness
[05:59:33] *** BladeFreak has quit IRC
[06:05:43] *** Justasic has joined #chromium-os
[06:11:28] *** Justasic has quit IRC
[06:11:29] *** Justasic has joined #chromium-os
[06:12:48] *** magn3ts has joined #chromium-os
[07:37:38] *** yusukes has joined #chromium-os
[07:41:39] *** Styx has joined #chromium-os
[08:00:17] *** vmil86 has joined #chromium-os
[08:05:47] *** phil2 has joined #chromium-os
[08:06:41] *** phil2 has quit IRC
[08:32:44] *** djkurtz has joined #chromium-os
[09:07:22] *** jujugre has joined #chromium-os
[09:19:32] *** cchristiansen has joined #chromium-os
[09:44:32] *** jochen__ has quit IRC
[09:44:37] *** jochen__ has joined #chromium-os
[09:44:37] *** ChanServ sets mode: +v jochen__
[10:12:14] *** magn3ts has quit IRC
[10:35:05] *** yusukes_ has joined #chromium-os
[11:22:29] *** patcito has quit IRC
[11:29:23] *** nick761_ has joined #chromium-os
[11:31:31] *** jochen___ has joined #chromium-os
[11:31:31] *** ChanServ sets mode: +v jochen___
[11:31:44] *** Sthebig_ has joined #chromium-os
[11:31:49] *** falken_ has joined #chromium-os
[11:34:50] *** lipsinV2 has joined #chromium-os
[11:36:11] *** jennb_ has joined #chromium-os
[11:36:12] *** ChanServ sets mode: +v jennb_
[11:36:25] *** jochen__ has quit IRC
[11:36:25] *** yosafbridge has quit IRC
[11:36:25] *** lipsinV21 has quit IRC
[11:36:25] *** jennb has quit IRC
[11:36:25] *** arun_ has quit IRC
[11:36:25] *** nick761 has quit IRC
[11:36:25] *** deshantm has quit IRC
[11:36:25] *** Sthebig has quit IRC
[11:36:25] *** falken has quit IRC
[11:36:25] *** eggie has quit IRC
[11:36:25] *** crosbot has quit IRC
[11:36:25] *** Hamlin has quit IRC
[11:36:26] *** Sthebig_ is now known as Sthebig
[11:36:29] *** jochen___ is now known as jochen__
[11:36:29] *** jennb_ is now known as jennb
[11:36:30] *** Hamlin_ has joined #chromium-os
[11:40:41] *** yosafbridge has joined #chromium-os
[11:40:44] *** lipsinV2 has quit IRC
[11:41:25] *** lipsinV2 has joined #chromium-os
[11:43:33] *** arun_ has joined #chromium-os
[11:44:22] *** eggie has joined #chromium-os
[11:44:27] *** deshantm has joined #chromium-os
[12:34:08] *** sadrul has quit IRC
[13:21:05] *** SignedAdam has joined #chromium-os
[13:27:49] <SignedAdam> hello could someone help me shrink H-STATE so i can make a other part for windows so i can deal boot chrome os and windows
[13:33:10] <SignedAdam> you guys never help me, you never talk about anything so why cant you all just give me somehelp, i know one of you must be jeal booting windows and chromium os
[13:50:03] <SignedAdam> awww, i resized it, in ubuntu
[13:50:22] <SignedAdam> right, and then fat32 the left over
[13:53:06] <SignedAdam> ummmmm,
[13:53:08] <SignedAdam> what the
[13:54:48] *** borios has quit IRC
[14:07:00] <SignedAdam> well i cant do it
[14:07:03] <SignedAdam> stupid thing
[14:10:51] *** SignedAdam has quit IRC
[14:26:47] *** petermayo has joined #chromium-os
[14:26:48] *** ChanServ sets mode: +v petermayo
[14:33:28] <petermayo> Good Morning.  Sheriff coming on duty .. was wondering if Hung-Te or anyone else had a status update (or document) for the current effort to greent he tree so we can open it?
[14:43:21] *** flackr has joined #chromium-os
[14:43:21] *** ChanServ sets mode: +v flackr
[15:16:43] *** uncola has joined #chromium-os
[15:16:58] *** uncola has left #chromium-os
[15:18:03] *** vmil86 has quit IRC
[15:27:51] *** phil2 has joined #chromium-os
[15:28:54] *** CrashTest_ has joined #chromium-os
[15:32:32] *** vmil86 has joined #chromium-os
[15:34:59] <seanpaul> Sooo.... looks like alsa is failing b/c the "panel" library is now missing
[15:35:07] <seanpaul> I have this library in my local chroot
[15:38:17] *** sadrul has joined #chromium-os
[15:38:18] *** ChanServ sets mode: +v sadrul
[15:38:47] *** rharrison_chrome has joined #chromium-os
[15:38:48] *** ChanServ sets mode: +v rharrison_chrome
[15:39:13] <petermayo> How do dependencies get installed in the chroot?
[15:39:36] <kliegs> petermayo: As part of make_chroot, setup_board or build_packages
[15:39:38] <seanpaul> petermayo: make_chroot emerges a bunch of host packages
[15:40:03] *** TW1920 has joined #chromium-os
[15:41:02] <petermayo> If you clobber yours and rebuild, does it come back?
[15:41:15] <kliegs> clobber what and rebuild how?
[15:41:29] <petermayo> make_chroot --delete
[15:41:41] <petermayo> make_chroot
[15:41:45] <kliegs> yes
[15:41:52] <seanpaul> petermayo: I'll start it, but hopefully we'll figure it out before we have an answer
[15:42:30] <petermayo> make_chroot isn't that long is it?
[15:46:30] <petermayo> seanpaul: which package is it looking for for panel?
[15:46:53] <seanpaul> petermayo: libpanel is owned by sys-libs/ncurses
[15:55:43] *** ers has left #chromium-os
[15:56:02] *** ers has joined #chromium-os
[15:56:02] *** ChanServ sets mode: +v ers
[16:02:20] *** rbyers_ has joined #chromium-os
[16:36:10] <petermayo> seanpaul: which path had the alsa portage pacakge configure script?
[16:36:52] <seanpaul> petermayo: tmp/portage/media-sound/alsa-utils-1.0.21-r2/work/alsa-utils-1.0.21/configure
[16:36:57] <petermayo> <second topic> : arthur looks like it needs a configuration script to move from u-boot to u-boot-next
[16:47:04] <petermayo> seanpaul: are there any changes around parallelization ?
[16:50:00] *** cchristiansen has quit IRC
[16:51:07] <petermayo> seanpaul:  Im' thinking about trying a clobber on one or more of these bots ...
[16:51:11] *** jujugre has quit IRC
[16:51:29] <seanpaul> petermayo: go for it, i'm still wading through CLs
[16:53:14] *** BladeFreak has joined #chromium-os
[16:53:47] <petermayo> khim: Is your CL likely to turn any builders green?
[16:57:27] <petermayo> Unfortunately, someone's CL kicked the builders off without a clobber seconds before my clobber request made it in.
[16:59:30] <seanpaul> petermayo: Would you like to revert?
[17:00:09] *** vmil86 has quit IRC
[17:00:53] <petermayo> Not unless we have a good choice, or a complete list for the no - need CLs I think.
[17:03:05] *** vmil86 has joined #chromium-os
[17:09:19] *** jujugre has joined #chromium-os
[17:19:45] *** Adys has joined #chromium-os
[17:31:48] *** jongpil has joined #chromium-os
[17:33:22] <adlr> dumb question: where is the source to autotests? (specifically login_CryptohomeMounted)
[17:33:42] * adlr is the sheriff today
[17:35:30] <adlr> found it: third_party/autotest/files/client/site_tests/login_CryptohomeMounted
[17:35:49] <petermayo> adlr: welcome to the nightmare of figuring out why ncurses is not supplying new_panel to the alsa-utils.
[17:36:14] <adlr> awesome. i love nightmares...
[17:37:42] *** CrashTest_ has left #chromium-os
[17:37:46] *** Adys has quit IRC
[17:38:47] *** Adys has joined #chromium-os
[17:42:44] *** jongpil has left #chromium-os
[17:43:01] <seanpaul> So as far as I can tell, the alsa issue was introduced between 18:20 and 21:25 (PST) on Thursday
[17:43:33] <seanpaul> I've been going through the merged CLs between those two times, trying to find the culprit
[17:45:20] <seanpaul> I'm going to try to repro on my personal chroot, and then start reverting suspicious CLs locally until I can get a clean build
[17:49:08] <adlr> cool
[17:53:14] <adlr> does anyone know what 404s in the login_CryptohomeMounted test would mean? is that a spurious failure?
[17:53:38] <cmasone> adlr: coming by
[17:53:45] <adlr> cmasone: thanks
[17:56:40] *** rbyers_ has quit IRC
[17:59:14] *** rbyers_ has joined #chromium-os
[18:00:46] *** JasonO has joined #chromium-os
[18:00:46] *** JasonO has joined #chromium-os
[18:01:02] *** jujugre has quit IRC
[18:03:59] <adlr> cmasone: thanks for helping resolve the login_* tests
[18:04:08] <adlr> now, ARM is the main issue
[18:04:17] *** jrbarnette has joined #chromium-os
[18:05:38] <seanpaul> Was anyone looking into this alsa-utils failure on Friday?
[18:08:01] <cmasone> seanpaul: not that I know of.  dgarrett was looking at ARM stuff.
[18:09:19] <seanpaul> cmasone: thanks... I just wanted to make sure I wasn't duplicating effort
[18:19:43] *** behdad has joined #chromium-os
[18:21:06] *** Adys has quit IRC
[18:21:21] *** unreal_ is now known as unreal
[18:26:57] *** jrbarnette_ has joined #chromium-os
[18:30:41] *** jrbarnette has quit IRC
[18:30:42] *** jrbarnette_ is now known as jrbarnette
[18:35:51] *** rbyers_ has quit IRC
[18:37:06] *** rbyers_ has joined #chromium-os
[18:41:03] *** lipsinV2 has quit IRC
[18:51:21] *** satorux_ has quit IRC
[18:52:20] *** satorux_ has joined #chromium-os
[18:52:20] *** ChanServ sets mode: +v satorux_
[18:53:09] *** flackr has quit IRC
[18:53:09] *** sosa has quit IRC
[18:53:09] *** rbyers has quit IRC
[18:53:09] *** vpalatin has quit IRC
[18:53:09] *** mmocny has quit IRC
[18:53:09] *** cwolfe has quit IRC
[18:53:09] *** kliegs has quit IRC
[18:53:09] *** rbyers_ is now known as rbyers
[18:53:34] *** sosa has joined #chromium-os
[18:53:34] *** ChanServ sets mode: +v sosa
[18:53:45] *** vpalatin has joined #chromium-os
[18:53:45] *** ChanServ sets mode: +v vpalatin
[18:53:46] *** cwolfe has joined #chromium-os
[18:53:46] *** ChanServ sets mode: +v cwolfe
[18:57:01] *** chocobo__ has joined #chromium-os
[18:57:02] *** ChanServ sets mode: +v chocobo__
[18:57:18] *** chocobo__ has quit IRC
[19:04:33] *** mascondante has quit IRC
[19:06:20] *** kliegs has joined #chromium-os
[19:06:21] *** ChanServ sets mode: +v kliegs
[19:06:31] *** mmocny has joined #chromium-os
[19:06:31] *** ChanServ sets mode: +v mmocny
[19:06:34] *** rbyers_ has joined #chromium-os
[19:06:34] *** ChanServ sets mode: +v rbyers_
[19:06:48] *** flackr has joined #chromium-os
[19:06:48] *** ChanServ sets mode: +v flackr
[19:16:46] *** wfrichar has joined #chromium-os
[19:16:46] *** ChanServ sets mode: +v wfrichar
[19:27:21] <seanpaul> alsa's configure script is failing with "/lib64/libgcc_s.so.1: file not recognized: File format not recognized"
[19:27:25] <seanpaul> kliegs: ^
[19:30:08] *** perezd has joined #chromium-os
[19:31:47] <kliegs> looking at that the cross compiler has -L/lib64 as a command line flag.  This seems wrong for the cross compiler
[19:31:57] <kliegs> Anyone know of any tools/configuration changes recently that could have caused this?
[19:32:32] <cmasone> kliegs: maybe ping raymes/asharif on email?
[19:33:01] <cmasone> kliegs: I can't find them on IRC
[19:33:04] <kliegs> cmasone: thanks.
[19:34:43] *** cwolfe has quit IRC
[19:37:28] *** cwolfe has joined #chromium-os
[19:37:28] *** ChanServ sets mode: +v cwolfe
[19:39:58] *** mmocny has quit IRC
[19:41:28] *** jshin has joined #chromium-os
[19:41:28] *** ChanServ sets mode: +v jshin
[19:43:12] *** sergiu has joined #chromium-os
[19:43:12] *** sergiu has joined #chromium-os
[19:45:48] *** powderluv has joined #chromium-os
[19:45:48] *** ChanServ sets mode: +v powderluv
[19:48:48] <seanpaul> adlr: ping
[19:50:37] <seanpaul> I believe the CL that broke alsa config (or rather exposed the bug in alsa config) is http://gerrit.chromium.org/gerrit/#change,3557
[19:51:04] <seanpaul> Does anyone know about this? I'd like to revert it, but I'm hesitant since there are other CLs from the same committer
[19:51:22] <seanpaul> I'm not sure what dependencies exist between them (if any)
[19:51:37] <adlr> seanpaul: pong. looking
[19:51:38] <cmasone> seanpaul: that change has already been revereted
[19:51:51] <adlr> i see
[19:52:03] <cmasone> seanpaul: also...why not reach out to asharif?
[19:52:19] <seanpaul> cmasone: I have, no response
[19:52:30] <seanpaul> cmasone: The CL was reverted, but it's been reapplied
[19:53:21] <cmasone> seanpaul: then perhaps anush?
[19:53:34] <cmasone> powderluv: ping ^^^
[19:53:53] <seanpaul> adlr: Can you sneakernet ping asharif?
[19:54:07] <cmasone> seanpaul: he doesn't sit near us
[19:54:35] <adlr> anush isn't at his desk
[19:54:39] <adlr> yeah, i'm not too close to asharif
[19:54:47] <cmasone> anush is on chat, though
[19:56:01] <adlr> trying to get him on here...
[19:56:03] <powderluv> hi
[19:56:41] * powderluv (catching up on the context)
[19:58:12] <powderluv> seanpaul: you suspect 3557 for some breakage ?
[19:58:34] <seanpaul> powderluv: yes
[19:58:46] <seanpaul> powderluv: It looks like Gerrit is out-of-sync with the git tree
[19:58:51] <powderluv> why ?
[19:58:54] <powderluv> ohh i see
[19:58:59] <seanpaul> powderluv: I think git commit 9b69ac75545a0dfdb8f38d40a71b17d64f16a7f9 is the offender
[19:59:15] <powderluv> if you go to the gitweb on gerrit it is different from what you see on git.chromium.org ?
[19:59:38] <powderluv> in chromiumos-overlay ?
[19:59:43] <seanpaul> powderluv: I'm just looking at the Gerrit CL
[19:59:52] <seanpaul> powderluv: I will look at gitweb through Gerrit
[20:00:17] <powderluv> if you suspect something being out of sync that is a good place to start
[20:01:03] <seanpaul> powderluv: I see the reapplication of the patch in Gerrit's gitweb
[20:01:55] <kliegs> Just to focus on the tree breakage
[20:02:08] <kliegs> Is there any issues with seanpaul reverting 3557 to fix the arm compiling issues?
[20:02:10] <powderluv> http://gerrit.chromium.org/gerrit/#change,3769
[20:02:24] <cmasone> kliegs: other than the fact that that CL has already been reverted?
[20:02:40] <kliegs> cmasone: it hasn't been reverted
[20:02:50] <powderluv> is there some log that is pointing to that change ?
[20:02:59] <kliegs> cmasone: Go look at the actual git log in either Gerrit or gitweb and you'll see no revert
[20:03:00] <seanpaul> cmasone: http://git.chromium.org/gitweb/?p=chromiumos/overlays/chromiumos-overlay.git;a=history;f=chromeos/scripts/setup_board;h=c0c820557f0d3d658bbca885608ab3978a75144c;hb=master
[20:03:16] <seanpaul> powderluv: that CL is abandoned
[20:03:39] <cmasone> kliegs: it was reverted and reapplied in a different CL
[20:04:29] <powderluv> so where is the buildfailure pointing to this CL ?
[20:04:44] *** magn3ts has joined #chromium-os
[20:05:14] <kliegs> powderluv: http://chromegw.corp.google.com/i/chromiumos/builders/arm%20tegra2%20full/builds/3362/steps/BuildTarget/logs/stdio
[20:05:18] <powderluv> i see the chromeos-bootimage failure
[20:05:47] <kliegs> If you look on the machine at the configure error logs for alsa-utils, you'll see its failing to build its test binary because the cross compiler for arm is linking in libgcc from /lib64 which is a 64-bit x86 host library
[20:06:02] <kliegs> This error is only showing up on the full builders because the alsa-utils package was not changed
[20:06:09] <kliegs> So the binary builders would not notice a problem
[20:06:27] <powderluv> ok so the chromeos-u-boot failure is something else
[20:06:52] <kliegs> yes. and its believed to be fixed. i think either dianders or hung-te resolved it
[20:07:52] <powderluv> ok getting the toolchain folk on irc
[20:08:02] *** djkurtz has quit IRC
[20:08:19] *** JasonO has quit IRC
[20:08:24] *** saintlou has joined #chromium-os
[20:08:24] *** ChanServ sets mode: +v saintlou
[20:08:44] *** raymes has joined #chromium-os
[20:08:44] *** ChanServ sets mode: +v raymes
[20:09:21] <petermayo> seanpaul:  which ncurses was new_panel added for?  I thought you looked at a page that said 5.8 this AM?
[20:09:52] <seanpaul> petermayo: ncurses/libpanel is a red herring
[20:10:19] <petermayo> b/c?
[20:10:37] <petermayo> of the arch list switch in the toolchain?
[20:10:53] <powderluv> fyi: raymes is looking into it
[20:10:58] <seanpaul> petermayo: the issue is that alsa's configure script is failing on a gcc cross compiler command that is linking against a lib64 library
[20:11:05] <seanpaul> powderluv: thanks!
[20:11:12] <powderluv> he is on here too
[20:12:33] <raymes> hi
[20:12:54] *** jrbarnette_ has joined #chromium-os
[20:13:05] <raymes> kliegs: do you suspect this is gcc-config related?
[20:13:40] <kliegs> raymes: I'm not sure of the root cause.  But for some reason the cross compiler is getting -L/lib64 and -L/usr/lib appended to its command line
[20:13:51] <kliegs> I don't know enough about the toolchain to know which piece causes that
[20:14:12] <raymes> ok
[20:15:21] *** jrbarnette has quit IRC
[20:15:50] <raymes> do we know when this started to happen?
[20:16:30] <seanpaul> raymes: It was introduced Thursday night between 18:20 and 21:25 PST
[20:16:42] <raymes> thanks
[20:17:14] *** jrbarnette_ has quit IRC
[20:19:01] <seanpaul> raymes: Gerrit change 3557 says it's reverted, but there's no corresponding revert in the git log
[20:19:03] <kliegs> raymes: http://gerrit.chromium.org/gerrit/3557 is suspected to be the cause.  Gerrit says in the CL history that it is reverted but looking at the actual git logs we can't find the revert
[20:19:20] <seanpaul> yeah... what kliegs said :)
[20:19:34] * kliegs ponders that great minds think alike
[20:19:45] <seanpaul> I was thinking the same thing!
[20:20:19] <raymes> kliegs: it was repushed as http://gerrit.chromium.org/gerrit/#change,3770 i'm guessing that is the cause
[20:20:44] <kliegs> raymes: I mean if you look at the timestamp in 3557 it never shows up in the logs
[20:21:18] <kliegs> however you know the toolchain stuff better so I'll trust you on this
[20:21:27] <raymes> oh i see
[20:21:47] <raymes> either way the change is currently in though
[20:22:00] <powderluv> just submit a new CL to revert it
[20:22:17] <seanpaul> powderluv: Ok, coming
[20:22:40] <powderluv> and we can get chrome-infra-team to investigate the commit not showing up
[20:22:55] *** bfreed has joined #chromium-os
[20:22:55] *** ChanServ sets mode: +v bfreed
[20:22:57] <powderluv> seanpaul: please get raymes/asharif on the review
[20:23:20] *** JasonO has joined #chromium-os
[20:23:29] <raymes> thanks
[20:23:29] <powderluv> and we can open a bug at crbug.com under build infra for the missing commit
[20:28:54] *** rharrison_chrome has quit IRC
[20:29:19] *** nick761_ has quit IRC
[20:29:36] *** nick761 has joined #chromium-os
[20:30:07] <seanpaul> raymes: The CL is up for your review, PTAL
[20:30:23] *** davidjames has joined #chromium-os
[20:30:23] *** ChanServ sets mode: +v davidjames
[20:33:04] <seanpaul> powderluv: http://code.google.com/p/chromium/issues/detail?id=88950
[20:33:34] <powderluv> cool thanks.
[20:37:16] *** leavittx has joined #chromium-os
[20:42:13] <petermayo> powderluv: kliegs: seanpaul: That message in gerrit means that a revert CL was created by pushing the revert button:  the revert CL has to be approved and pushed the same as any other afterward.
[20:42:34] <kliegs> petermayo: thanks. that makes sense
[20:43:29] <seanpaul> dianders: ping
[20:43:32] <petermayo> I found it the hard way too, when reverting a CL in my last sheriff rotation.
[20:43:40] <dianders> seanpaul: pong
[20:44:26] <seanpaul> dianders: Do you have the know-how to increase the size allowance for bootstub.bin? It looks like tegra2 binary is still failing with that error
[20:44:38] <seanpaul> dianders: and that's the only thing stopping me from opening the tree
[20:46:34] <dianders> seanpaul: Ick.  I can give it a quick shot.  There's another bug I'm tracking down right now (crosbug.com/17437) that's also pretty important.  ...but I can give it a shot.  Can you send me any details you have?
[20:46:38] * kliegs is waiting for the tree to open so he can close it
[20:47:07] *** dianders has left #chromium-os
[20:47:27] *** dianders has joined #chromium-os
[20:47:28] *** ChanServ sets mode: +v dianders
[20:48:16] <dianders> ...sorry, accidentally closed IRC window.  :(  Hope I didn't miss ur response...
[20:48:54] <seanpaul> dianders: I was hoping you had the details :) I can look into where allowances are made for bootstub
[20:49:06] <dianders> Can you point me to the exact failure?
[20:49:26] <dianders> This is the arm generic full?
[20:49:51] <seanpaul> dianders: "__main__.PackError: blob too large: bootstub.bin: 657408 > 557056"
[20:50:00] <seanpaul> dianders: in pack_firmware_image
[20:50:02] <dianders> Which build?
[20:50:09] <seanpaul> tegra2 binary
[20:50:24] <dianders> OK, thanks!  I'll look at it.
[20:50:30] <dianders> Bug #?
[20:52:51] <petermayo> seanpaul: you creating the bug,or shall I?
[20:53:12] *** JasonO has quit IRC
[20:53:46] <seanpaul> dianders: http://code.google.com/p/chromium-os/issues/detail?id=17440
[20:54:04] <seanpaul> dianders: Sorry, I was busy creating the bug :p
[20:55:29] <dianders> seanpaul: Thanks.  Setting things up for tegra2_dev-board.
[20:56:38] <seanpaul> dianders: awesome!
[20:58:04] <kliegs> FYI:  Doing a sync, setup_board --force and a manual emerge of alsa-utils after the revert is working for me so far
[21:02:36] *** powderluv has quit IRC
[21:07:48] *** JasonO has joined #chromium-os
[21:10:36] *** xiyuan has quit IRC
[21:11:58] *** patcito has joined #chromium-os
[21:14:47] *** jglasgow has joined #chromium-os
[21:14:47] *** ChanServ sets mode: +v jglasgow
[21:22:05] *** cros_ has joined #chromium-os
[21:23:47] *** jrbarnette has joined #chromium-os
[21:23:48] *** ChanServ sets mode: +v jrbarnette
[21:23:59] <cros_> Can anyone help me grab a known booting version of the chromium OS source? If I repo init using the default manifest I will get frequent changes but I want something working and static to ease porting.
[21:24:33] <ellyjones> check out one of the branches
[21:24:36] <cros_> I'm interested in ARM btw.
[21:25:08] <cros_> ellyjones: I had a tough time figuring out what the branch naming convention was when I looked into that a while back.
[21:25:35] *** saintlou has quit IRC
[21:25:56] *** saintlou has joined #chromium-os
[21:25:56] *** ChanServ sets mode: +v saintlou
[21:28:26] <ellyjones> try 0.13.<largest>.B
[21:32:15] <cros_> ellyjones: This branch still changes though, correct? I was hoping there was some sort of tagged version.
[21:32:28] <ellyjones> haha
[21:32:29] <ellyjones> no
[21:34:44] <cros_> ellyjones: I don't know if that was directed at me. I don't see why people couldn't push changes to that branch.
[21:35:59] <ellyjones> yeah, it was directed at you :) I am making gentle fun of our own development process
[21:36:13] <ellyjones> the codebase is changing pretty fast right now so there is no real 'stable' branch
[21:37:56] *** Adys has joined #chromium-os
[21:38:33] <cros_> ellyjones: Thanks. I guess for someone working on a port to another ARM SoC I'm trying to find what code to start with. Right now when I run into problems I'm not sure if they are of my doing or if the base code is broken.
[21:41:33] <ellyjones> you could ask :)
[21:42:07] *** rharrison_chrome has joined #chromium-os
[21:42:08] *** ChanServ sets mode: +v rharrison_chrome
[21:42:34] <kliegs> cros_: Keep an eye on the tree status here too
[21:43:06] <kliegs> cros_: If you look at arm-generic or arm-tegra2 and see the same errors you're getting it will help you figure out if its a local issue or not
[21:53:30] *** rbyers has quit IRC
[21:53:51] <ellyjones> anyone know if I can use a mini-servo to power on/off a device? :)
[21:54:12] <cmasone> ellyjones: I am pretty sure yoiu can, but I have no idea how :-)
[21:54:19] <ellyjones> useless :P
[21:54:31] *** rbyers has joined #chromium-os
[21:54:32] <cros_> kliegs: I'm not talking of build errors. I'm talking of runtime problems which I'm not sure if the tree status helps with. I'm reading up on the buildbot stuff right now.
[21:57:19] <cmasone> cros_: the R13 and R12 branches are pretty stable in terms of changes, and unlikely to have new runtime issue cropping up on you.  That said, we haven't released on an ARm platform yet, so they might have problems that we will never fix if they only crop up on ARM
[22:00:05] <ellyjones> I want a driver for my host that allows it to act as a USB client
[22:00:16] <ellyjones> (specifically so it could pretend to be a USB disk to boot my dut)
[22:00:26] <ellyjones> does anyone know if this idea is batshit crazy?
[22:03:07] <vpalatin> ellyjones: you need a host supporting USB-device mode. You won't find such a thing on x86. By the way I did that with a nexus one, i was working pretty well (I reworked the N1 kernel mass storage gadget driver)
[22:03:10] <cros_> cmasone: Appreciate the information. I haven't been able to get the UI up yet so I am concerned with major issues like X not starting successfully. Based on what you're saying I guess that shouldn't be a concern :).
[22:03:54] <rochberg> ellyjones:  a)  I hear rumors about people doing this over NFS.  b)You'll need the right USB controller
[22:04:04] *** Macer has joined #chromium-os
[22:04:05] <rochberg> they are difficult to find, but I have [redacted]s
[22:04:08] <ellyjones> vpalatin: aw :( I want a device that I can write to from the host and then boot from the target
[22:04:15] <Macer> hm
[22:04:29] <kliegs> ellyjones: I kept thinking in my head I should write what you say
[22:04:41] <vpalatin> ellyjones: by the way, as mentionned by rochberg you can do it with NFS, it is working for ARM
[22:04:48] <kliegs> if I ever make some 20% time it might be worth it. although the NFS work seems close enough to what you want
[22:04:59] <vpalatin> I will be supported on x86
[22:05:03] <Macer> is chromium os booting on typical x86 hw yet?
[22:05:10] <vpalatin> I got the main pieces in place last week for it
[22:05:14] <Macer> i was going to see about making a build for my via based artigo
[22:05:21] <ellyjones> vpalatin: oh *really*? Interesting
[22:05:42] * ellyjones is _this_ close to being able to work remotely :)
[22:06:23] <vpalatin> ellyjones: you also need your servo mini , but they have arrived here in MTV
[22:06:29] <ellyjones> I have one
[22:07:54] <seanpaul> dianders: I think I've got a firmware image layout that will accommodate the larger bootstub
[22:08:00] <seanpaul> dianders: I'm just testing it now
[22:08:48] <dianders> seanpaul: OK.  Would be interested in seeing the change and what you did.
[22:08:58] <dianders> sealpaul: Also check w/ robotboy, since he's looking into various solutions, too.
[22:09:30] <seanpaul> dianders: sure.
[22:09:45] <seanpaul> dianders: If anton is back, he might be the best guy for the job
[22:10:13] <dianders> seanpaul: he's back.  We've been chatting about the problem and various issues...
[22:10:46] <seanpaul> dianders: I just moved the offsets around to accommodate it, I haven't tried to strip anything out to save space -- I was optimizing for time :)
[22:11:07] <dianders> Ah, cool.  That might be good enough...
[22:11:32] *** petermayo has quit IRC
[22:11:52] *** petermayo has joined #chromium-os
[22:16:33] <seanpaul> dianders: I've assigned you as a reviewer on the patch
[22:16:44] <dianders> seanpaul: Thx.  Looking for it.
[22:17:25] <adlr> I kicked off ARM builds to get seanpaul's cl: Revert "Removed copying of libgcc files from gcc-config."
[22:17:29] <adlr> would I need to clobber those?
[22:17:56] <seanpaul> adlr: I don't think so
[22:18:17] <adlr> okay, cool, wanted to check
[22:18:56] *** powderluv has joined #chromium-os
[22:18:56] *** ChanServ sets mode: +v powderluv
[22:19:41] <seanpaul> adlr: Thanks for kicking those... it looks like alsa-utils built on the tegra2 full builder, so we should be past that error
[22:19:51] <adlr> great
[22:20:06] <adlr> so the u-boot issue remains for sure
[22:21:14] <seanpaul> adlr: yup :(
[22:22:05] *** flackr has quit IRC
[22:26:14] *** petermayo has quit IRC
[22:26:26] <dianders> seanpaul: Talked to Anton about your CL.  Neither of us have great ideas about implications of moving things around in flash.  Gonna check w/ Randall, though it looks like CL / Tom are the real owners of the flash map...
[22:33:59] *** Adys has quit IRC
[22:34:43] *** dmazzoni has joined #chromium-os
[22:35:03] <ers> Does the u-boot problem need to block opening the tree? It sounds like it's understood, and that addressing it won't be affected by unrelated commits.
[22:38:06] <adlr> i'm also curious how folks feel about opening the tree
[22:38:34] *** rbyers has quit IRC
[22:39:30] *** rbyers has joined #chromium-os
[22:39:51] <cmasone> adlr: righteously indignant?
[22:45:55] <gauravsh> Has the tree really been closed all weekend?
[22:46:17] <ers> yup.
[22:47:56] <ers> Because I missed a couple of windows of openness on Friday, I haven't been able to submit anything since Wednesday.
[22:49:25] <adlr> i think the question at this point is, if we know all the ARM issues, and have people working on them, do we really need to keep the tree closed?
[22:49:39] *** xiyuan has joined #chromium-os
[22:49:40] *** ChanServ sets mode: +v xiyuan
[22:50:56] *** CrashTest_ has joined #chromium-os
[22:51:05] *** Styx has quit IRC
[22:51:42] <ers> I believe that opening the tree now would be consistent with past practice.
[22:52:50] *** Styx has joined #chromium-os
[22:54:42] <jglasgow> I feel that we should open the tree.  There are CLs stacking up waiting to get it.  The longer it is closed, the more risk that the flood of CLs breaks it again.
[22:54:52] *** saintlou has quit IRC
[22:56:55] *** Styx has quit IRC
[22:57:12] <jrbarnette> isn't it the case that the flood of CLs could break ARM further without anyone being able to see it?
[22:57:15] <adlr> okay, i just want to look into a couple other issues dianders just mentioned
[22:57:34] *** rharrison_chrome has quit IRC
[22:57:36] <adlr> are we allowed to mention board codenames in here?
[22:57:41] *** saintlou has joined #chromium-os
[22:57:42] *** ChanServ sets mode: +v saintlou
[22:57:47] <adlr> for unreleased things
[22:58:19] <gauravsh> you are but then you will be fired. :)
[22:58:25] <gauravsh> j/k. I am not really sure.
[22:59:19] <adlr> okay, so the internal x86-[upcoming acer model] binary R13 is failing
[22:59:49] <kliegs> Slightly late to the discussion but given there have been multiple arm issues, I feel strongly the tree should remain closed until the arm issues are resolved
[22:59:52] <srao> I thought we were allowed to as long as we don't give away what HW they are associated
[23:00:03] <seanpaul> I'd really like to keep the tree closed
[23:00:17] <kliegs> There have been multiple issues masking others, opening now and allowing checkins will likely just cascade the problem
[23:00:19] <marcheu> can we plan those as days off maybe next time?
[23:00:21] <seanpaul> We have people working on the issues that closed it
[23:01:03] <dianders> kliegs: Arthur issues are understood (and I'm OK with them not blocking tree opening).
[23:01:31] <dianders> kliegs: tegra2-devboard problems are also understood.  I don't feel like these need to keep the tree closed, but am OK if others disagree.
[23:01:48] <dianders> seanpaul: Are you still around?
[23:02:23] <seanpaul> dianders: yep
[23:02:27] <seanpaul> I'm here
[23:02:36] <kliegs> dianders: I'm fine with Arthur being special as its not a generic arm platform and we try to prevent internal bots from closing external tree
[23:03:02] <dianders> Cool.  Randall made some comments on your CL.  I think he was suggesting giving a little wiggle room and also making things a little rounter numbers...
[23:03:15] <dianders> ....err, rounder
[23:03:33] <seanpaul> dianders: I just responded to them
[23:03:34] <dianders> Can RO section be 0x00110000?
[23:03:38] <dianders> Ah, cool.  :)
[23:04:23] <kliegs> I still think the tree needs to remain closed until tegra2-binary goes green AND the tegra2 full builder passes the point it was failing at before
[23:04:46] <dianders> kliegs: OK, fair enough.  We're working our best on that as 1st priority...
[23:04:54] <marcheu> and until I get a pony
[23:05:06] <kliegs> dianders: glad to hear that.  I know people are working on it so I'm sure it will get opened
[23:05:18] <kliegs> marcheu: For small enough values of pony, sure
[23:05:43] <marcheu> kliegs: I think you're aiming for a catastrophe if we keep the tree closed, if you look at the number of changes in gerrit
[23:05:55] <kliegs> marcheu: maybe. but 1 hour either way won't change that
[23:06:02] <marcheu> oh if it's 1h sure
[23:06:23] <seanpaul> dianders: Sure, the RO section could be 110000, but I'm not sure it really improves readability since all of the other numbers are odd-balls :)
[23:06:34] <kliegs> i'm not sure the exact timeline - I was only helping earlier on the alsa issue and haven't been involved in the u-boot issues
[23:07:00] <kliegs> However this should be a strong reminder to everyone why to avoid major changes Friday to the code trees :)
[23:07:58] <kliegs> And also why its important for some changes to make sure people do full builds on their own machines if they think their issue could affect other packages.
[23:08:09] <dianders> seanpaul: I'd like thue R0 section to be a rounder number, though.  I don't know for sure, but sometimes flash parts have block sizes that are ~64K.  Since old RO size was 64K boundary I'd like to keep that if possible.
[23:08:25] <dianders> ...internal offsets can be more fickle--that's fine.
[23:08:26] <seanpaul> dianders: that makes sense
[23:08:35] <seanpaul> I will tweak and re-up
[23:08:40] <dianders> Cool, thanks.  :)
[23:11:41] <ellyjones> z/w 2
[23:11:42] <ellyjones> fail
[23:12:18] <seanpaul> dianders: Going to 110000 overflows the 2MB limit
[23:12:34] <dianders> seanpaul: Ick...
[23:12:44] <dianders> Looking...
[23:14:22] <dianders> Can you reduce the size of FW_MAIN_A?
[23:14:27] <dianders> and B?
[23:15:35] <dianders> There is roughly 47K of wiggle room there...
[23:15:41] <dianders> (I think)
[23:15:56] *** Ruetobas has quit IRC
[23:16:36] <seanpaul> dianders: That sounds great... Do you want to take this change over? It sounds like you know what you're doing here :)
[23:17:06] <seanpaul> dianders: (I'm happy to do it, but I don't want to step on your toes)
[23:17:27] <dianders> seanpaul: Sure, I'll give it a shot.  It may bit a bit of the blind leading the blind, but I'll give it a shot.  Anton is still working his angle of trying to reduce the overhead of the BCT.  ...and we have an email out to Jimmy, too.
[23:18:01] <seanpaul> dianders: excellent, thanks for your help!
[23:18:34] *** leavittx has quit IRC
[23:18:35] <dianders> seanpaul: Thank you, too.  I've not looked at the flashmap before, so you certainly pointed me in the right direction.  :)
[23:21:37] *** Ruetobas has joined #chromium-os
[23:22:02] <seanpaul> adlr, petermayo, bfreed: I'm heading out now. I'll keep an eye on the tree, but I won't be on IRC. Good luck keeping the peace!
[23:23:03] *** CrashTest_ has left #chromium-os
[23:23:14] <bfreed> seanpaul: Thanks for all the work.
[23:24:12] <bfreed> dianders: adlr kicked off some ARM builds that should not be affected by your work.  If they succeed, we will open the tree without you.
[23:24:42] <dianders> bfreed: Is kliegs OK with that?  He wanted to see the devboard pass...
[23:24:52] *** thieule has joined #chromium-os
[23:24:53] *** ChanServ sets mode: +v thieule
[23:25:24] <kliegs> bfreed, dianders: I think we should wait until we can see the full builder at minimum pass the previous failure points
[23:25:45] <kliegs> The ARM builds were having toolchain issues - until we know that's resolved opening it is risky and we could easily mask true problems.
[23:26:28] <kliegs> If tegra2 full passes build_packages and all the external binary builders go green that seems the minimum for reopening the tree
[23:26:46] *** cros_ has quit IRC
[23:27:14] <kliegs> The tegra2 full builder may likely fail in the build_images phase as it might have been started before the final changes to fix the firmware size were in, but if it reaches that point and fails and we've seen the binary builder complete that I feel we can be pretty confident that its in a good state
[23:28:43] <kliegs> Especially given the backlog of changes and the odds of one of those breaking things, making sure that the environment the changes into is working is critical.  Otherwise figuring out if its a new or existing issue can be painful, especially due to toolchain issues masking other problems (e.g. the alsa-utils issue was initially thought to be a ncurses or dependency issue until close examination)
[23:31:58] <bfreed> kliegs: Ok, I will keep an eye on tegra2_full and our two internal canaries.  tegra2_full was started just after sean's 3875 was checked in.  Is there some other firmware size issue other than what dianders is working on?
[23:32:53] <kliegs> bfreed: Sean's change was to fix the toolchain issues, not the firmware issue
[23:33:16] <kliegs> There were multiple firmware issues earlier that either dianders or hungte resolved (sorry for not remembering who did it exactly)
[23:34:03] <kliegs> bfreed: I think there's only the outstanding one, unfortunately it affects the builder that seanpaul picked so I'd expect it to go red in BuildImage.  But if it passes BuidlPackage it should be good enough
[23:36:15] *** rush2end_ has joined #chromium-os
[23:37:14] *** rush2end has quit IRC
[23:37:29] *** rush2end_ is now known as rush2end
[23:51:45] *** wfrichar has quit IRC
[23:56:30] *** Guest46629 has quit IRC
[23:58:39] *** rbyers has quit IRC





top