**** BEGIN LOGGING AT Wed Sep 26 02:59:58 2012 Sep 26 04:31:55 ds2, you mean pocket-sphinx? Sep 26 05:19:56 ka6sox: TI released some free package... donno what it is called Sep 26 05:20:17 Hi all. I purchased a BeagleBoard when it first came out. I played with it a bit, got to know and understand the double/triple bootloader stuff (Xloader I think? and uBoot) I think I tossed Debian on there too, over top of ArchLinuxArm (I think it was...). then it sat in the drawer until now. Has much changed since then regarding this board? Sep 26 05:22:07 koen ping? Sep 26 05:29:04 ugh...need to bring down narcissus Sep 26 05:29:27 before europe kills it...permanently Sep 26 05:49:59 ka6sox: pong Sep 26 05:53:19 koen, we lost a HD on Cheli...so I need to halt emerald as it is severely impacting the rebuild of the array. Sep 26 05:54:21 I have halted the other builders on that machine. Sep 26 06:07:31 ka6sox: ok Sep 26 06:09:41 I turned off everything but Opal and its git :D Sep 26 06:53:23 I'm having trouble getting my beagle to boot. When I hold the user button, i can see that the MLO on the EMMC is being loaded, but the board keeps resetting Sep 26 06:53:43 I'm powering it via USB; perhaps that's an issue? Sep 26 06:54:12 It can load X-Loader just fine, and try to boot via nand, and it doesn't seem to reboot Sep 26 06:54:19 but of course there is nothing in nand, so it hangs Sep 26 06:54:49 actually, nand goes into boot loop Sep 26 06:58:04 figured it out; shitty USB extension cable. Sep 26 06:58:19 figured Sep 26 07:00:49 Russ: thanks for the help... Sep 26 07:01:04 any time Sep 26 07:01:14 I feel like I know you from this channel before Sep 26 07:03:54 ya, I've been here for some time Sep 26 07:19:08 Russ: I can create a network connection via the usb connection, right? Sep 26 07:19:18 I need to setup a bridge on the host, right? Sep 26 07:25:37 that's what I do Sep 26 07:25:53 setup a rule to add usbX to the bridge interface Sep 26 07:26:51 heh, anytime you have an order on frys.com with more than 3 items with free shipping, it charges shipping Sep 26 07:27:07 solution left as an exercise to the reader Sep 26 07:59:56 yay -- Linux desmas-e 2.6.35.4-x1 #1 PREEMPT Fri Aug 27 13:19:09 UTC 2010 armv7l GNU/Linux Sep 26 08:13:16 <_av500_> 'jour Sep 26 08:13:56 bon! Sep 26 08:14:49 <_av500_> we'll see for that part Sep 26 10:06:09 is it just me or is feeds.angstrom-distribution.org down? Sep 26 11:40:09 gm all Sep 26 11:40:22 anyone go to the Florida Linux User Exchange? Sep 26 11:47:15 flu exchange? Sep 26 11:51:03 <_av500_> european flu is much better Sep 26 12:04:40 koen: is feeds.angstrom-distribution.org down? Sep 26 12:12:30 * mdp is reminded that the Ohio linuxfest is this weekend..ugh Sep 26 13:26:46 koen: got a remedy for this build error http://arago-project.org/testresults/linux/mainline/2012-03-09_21-27-11/beagleboard/build-log.txt? Happens when I compile with a 3.2 kernel. Sep 26 13:27:51 no errors with 3.2 kernels here Sep 26 13:29:28 koen: hm. I grabbed a 3.2 from sakoman, so that could maybe cause it. Sep 26 13:31:20 I'll try linux-mainline_3.2 instead Sep 26 14:01:11 tasslehoff: most likely the patch 0006-Revert-OMAP-DSS2-remove-update_mode-from-omapdss-v3.2.patch is not applied in your recipe Sep 26 14:10:31 <_av500_> crazy shit Sep 26 14:10:34 <_av500_> there, I said it Sep 26 14:15:48 sp00k: ah. thanks. I'll still try mainline though. it's just that I've had more success with pm when using linux-sakoman until now, so I started there :) Sep 26 14:16:35 * tasslehoff tries to upgrade away from his current musb sorrows Sep 26 14:17:14 tasslehoff: switching to freescale? Sep 26 14:27:49 koen, I'm implementing pinconf_dbg_show() now. Sep 26 14:28:20 bah Sep 26 14:32:13 <_av500_> tasslehoff: you can run from musb, but you cannot hide Sep 26 14:34:03 _av500_: I'm still gonna try. Sep 26 14:37:13 I managed to hide as long as I used oe-classic, but the layers of oe-core were too thin to hide behind. Sep 26 14:39:04 heh Sep 26 14:45:46 indeed I even tried to disable musb on beaglebone-3.6 but it clings to life to break the build Sep 26 14:45:55 tough little zombie Sep 26 17:22:23 jkridner_: what are your thoughts on adding the register address/offset for the pinmux to the bone json? Sep 26 17:23:32 If that is what is needed to create the pincntl entries, then I agree I should added. Sep 26 17:24:24 right now I need it to parse the current mux, setting mux at runtime is another thing Sep 26 17:24:37 by which I mean "completely impossible" Sep 26 17:25:11 but babysteps, reading it works now Sep 26 17:25:23 root@bone-mainline:/sys/kernel/debug/pinctrl/44e10800.pinmux# grep 8a0 pins Sep 26 17:25:24 pin 40 (44e108a0) 00000027 pinctrl-single Sep 26 17:25:43 I've been running your mux-dt.js... introducing a simple bug-fix in bonescript to not give the 'setup not defined' error. Sep 26 17:25:53 jkridner_: and I have node 0.6 for not having a decent way to print hex numbers Sep 26 17:26:04 hate* Sep 26 17:26:51 koen, you'll take read-only and be happy with it Sep 26 17:27:07 koen, oh, and thanks for distracting me from real work yet again :P Sep 26 17:27:18 * koen aims to please Sep 26 17:27:43 I'm sure someone will rant about me enabling uio_pruss instead of what I'm supposed to be doing ;) Sep 26 17:28:40 jkridner_: first bone related patch for uboot sent upstream for review Sep 26 17:28:41 koen: so, is that file exactly a dt file? I never got over the hump to figure out what I needed to output. Sep 26 17:29:03 mdp shut down irc for real work Sep 26 17:29:13 koen: by joelagnel? Sep 26 17:29:18 woglinde: that sounds a bit drastic Sep 26 17:29:35 jkridner_: no, by me Sep 26 17:29:35 no Sep 26 17:29:37 koen, I don't like your 1 second delay either Sep 26 17:29:53 k, because joelagnel sent one for DDR3. Sep 26 17:30:05 jkridner_: https://github.com/koenkooi/linux/blob/3.6-for-panto-rebase/arch/arm/boot/dts/am335x-bone.dts Sep 26 17:30:06 mdp you will find out that you wilkl finished your projects on time Sep 26 17:30:14 jkridner_: that's how a full dt looks like Sep 26 17:30:18 s/how/what/ Sep 26 17:30:28 someone rung? Sep 26 17:30:40 woglinde: likely, but unacceptable Sep 26 17:30:40 jkridner_: https://github.com/koenkooi/linux/blob/3.6-for-panto-rebase/arch/arm/boot/dts/am335x-bone.dts#L27 shows the pinctrl nodes, the js just generates the content Sep 26 17:30:47 should I "Acked-by" both patches? Sep 26 17:31:06 mdp *g* your porblem Sep 26 17:31:12 ;) Sep 26 17:31:33 btw. living scrum at work is really nice Sep 26 17:31:41 * koen wonders if he can automatically generate the json Sep 26 17:31:45 till later diner now Sep 26 17:31:47 joelagnel seems to have used the 'lt' name. doesn't seem like any reason to hold it up by discussing that. Sep 26 17:32:06 woglinde, the distraction was provided on linux-omap, gotta shut that down too :( Sep 26 17:32:28 jkridner_, i noticed that in his hdmi branch Sep 26 17:32:56 would it be fine to "Acked-by" it based on a reading of the code? Sep 26 17:36:12 koen: patch link you sent is for the kernel. what was the uboot patch? Sep 26 17:36:38 last u-boot patch I have for you was fixing the tty02 bug. Sep 26 17:38:40 jkridner_: decrease bootdelay and mount rfs RO Sep 26 17:39:29 does the rfs read-only mount improve boottime? Sep 26 17:40:10 anyway, I'll comment on it when I see it, but it hasn't come across u-boot@lists.denx... for me yet Sep 26 17:43:08 jkridner_: try doing fsck on ext4 when it's mounted Sep 26 17:43:23 hint: it's not like xfs :) Sep 26 17:52:21 koen: you just need the first column (pinRegOffset) in bone.js, not pinRegMode, correct? Sep 26 17:53:06 well Sep 26 17:53:16 I need offset in the DT Sep 26 17:53:32 but, it makes sense to put it in bone.js, right? Sep 26 17:53:33 but I need offset + 0x800 to parse debugfs Sep 26 17:53:43 yes, it makes sense to put it there Sep 26 17:53:59 is the get_maintainers.pl script not giving all maintainers to whom you shall submit your patches ? Sep 26 17:54:02 but thanks to nodejs' inability to print hexnumbers we need to do some hacks Sep 26 17:54:19 bizulk: yes, but only use that on pure mainline based kernels Sep 26 17:54:38 bizulk: as in: base your patch on latest mainline and run the script on that Sep 26 17:55:22 koen: I did it on the linux-omap master (after pull). Sep 26 17:56:15 It gave me Greg K. email. But he is not a linux-omap maintainer, and I clearly disturbed / confused him. Sep 26 17:56:31 it can print hex numbers Sep 26 17:57:42 jkridner_: not without hacks Sep 26 17:57:57 jkridner_: if you parse it as a number, you have lost Sep 26 17:58:07 unless you do buffer magic, which doesn't work in 0.6.x Sep 26 17:58:17 0.8.x claims to have fixed that Sep 26 17:58:38 are you familiar with util.format? Sep 26 17:59:09 guess node-sprintf is an option Sep 26 17:59:10 jkridner_: that didn't work when I tried it :) Sep 26 18:08:28 koen: https://github.com/jadonk/bonescript/commit/2d2e68894241475859fec504da229711ea91f5e4 Sep 26 18:08:33 Finally what is the purpose of the linux-omap tree compared to the other trees ? Sep 26 18:08:48 bizulk: confuse people Sep 26 18:09:13 koen: I'll use JSON.stringify(bone, null, 4) to produce bone in the future. Sep 26 18:09:18 koen: You mean : we should not use that one ? Sep 26 18:09:33 jkridner_: awesome Sep 26 18:09:38 bizulk, omap specific stuff lives there until the next merge window Sep 26 18:10:02 bizulk: it generally best to base patches on mainline or linux-next, unless told otherwise after the first try Sep 26 18:10:36 bizulk, and get_maintainers.pl is just a guideline...it's sometimes wrong or at least missing people that need cced Sep 26 18:10:38 koen: I'm still not sure what I need to *OUTPUT* for DT files. Sep 26 18:11:02 It may be nothing but what you've already done in mux-dt.js Sep 26 18:11:09 right Sep 26 18:11:20 jkridner_: have a look at https://github.com/beagleboard/kernel/blob/beaglebone-3.6/hacks/beboprcape-dt.diff Sep 26 18:11:39 jkridner_: that shows where I put the output of mux-dt.js into the dt Sep 26 18:11:51 jkridner_: and why I added the comments :) Sep 26 18:11:59 mdp: okay. but so how I do know to who "--to" & "--cc" ? Sep 26 18:12:20 k. I just think bonescript should be able to produce output that is directly consumable in some way, rather than cut-and-paste. Sep 26 18:12:37 yes Sep 26 18:12:41 my motivation ran out Sep 26 18:12:45 bizulk, use that and then you will be corrected if necessary Sep 26 18:12:56 It seems we need to put the DT information into the EEPROMs or something and I really don't know what that something is. Sep 26 18:13:12 mdp: If the script does not return correct result. how can I know the difference between someone that "responsible for part of Sep 26 18:13:12 the kernel tree" with "someone responsible for this specific git tree" Sep 26 18:13:40 bizulk, it's only relevant to mainline submissions Sep 26 18:13:53 mdp: I did do that. But know Greg L. Ask me something like that : "what are you doing you stupid fool" Sep 26 18:14:08 jkridner_: up for a short phonecall? Sep 26 18:14:11 how the individual maintainer manages their queue of stuff and where it lives is a separate issue Sep 26 18:14:13 sure. Sep 26 18:14:18 bizulk, link? Sep 26 18:14:21 I have 15 minutes Sep 26 18:14:50 mdp: link on what ? email ? they are directly sent to me Sep 26 18:15:23 bizulk, if you don't properly chop your patches up into logical changes then get_maintainers.pl will return garbage. garbage in -> garbage out Sep 26 18:15:28 jkridner_: can you PM me your number? I don;t have TI LDAP anymore :) Sep 26 18:15:41 bizulk, if you posted your patch publicly provide the link here Sep 26 18:15:46 koen, LOL Sep 26 18:16:55 mdp: I sent it with "git send-email". Sep 26 18:17:01 yes, I'm made of fail today Sep 26 18:17:39 mdp: that's not the patch but why I did that : http://comments.gmane.org/gmane.linux.ports.arm.omap/85183 Sep 26 18:17:48 bizulk, as _av500_ says, "pretend I'm not looking over your shoulder" Sep 26 18:18:14 bizulk, pastebin or provide links to the cmdline you used and/or relevant posting on the public list I can see Sep 26 18:18:35 mdp: okay. Just a second Sep 26 18:22:16 mdp: http://pastebin.com/EeZSTNvG Sep 26 18:26:17 bizulk, first think is you didn't cc any of the mentioned lists. that's guaranteed to piss off greg Sep 26 18:26:22 s/think/thing/ Sep 26 18:27:58 and it's sufficient to cc him as the overall staging maintainer and tony as a large commit signer, but you really need to cc lkml otherwise people are unable to publicly review it Sep 26 18:28:08 mdp: "s/think/thing/" ? Yes I've just understood I did this mistake (just because I directly used google interface that just ... well you known). Sep 26 18:28:52 you've got a bunch of commit signers in that driver, I'd cc them all if you are new to who has an interest in what Sep 26 18:29:35 I'd ignore those that have the occasional sign as it's normally somebody that pops up and contributes one patch..and they'll catch it on lkml Sep 26 18:29:44 anyway, there's some fuzzy logic to this stuff Sep 26 18:29:55 no "exact steps" like we all yearn so much for Sep 26 18:31:59 bizulk, oh, I recommend you always use --dry-run first on git send-email...inspect your to/cc list before it's too late Sep 26 18:33:27 mdp: I'll see about this option Sep 26 18:33:31 <_zANDER_> Hey all. I just installed Debian on my beaglebone and found a problem with serial console. Where can I post it and the solution? Sep 26 18:33:48 mdp: now he directly wrote to me whithout cc', that's annoying and confusing. Shall I put these now by myself. Sep 26 18:34:23 mdp: many thanks for your comments Sep 26 18:34:46 tell him you'll repost publicly due to send-email error..he'll understand ;) Sep 26 18:35:33 np, good luck Sep 26 18:35:53 I usually get a reply from his mail-robot Sep 26 18:36:11 "my scripts detected that you did something stupid, here's how to fix it" Sep 26 18:39:14 koen: hehe Sep 26 18:39:49 still looking on the git.kernel.org where is written the purpose of each repository but I don't find it. Sep 26 18:41:56 koen, I was hoping for something like, "what are you doing hacking on parport?" from that submission but just a bland accept notice ;) Sep 26 18:42:04 from his auto scripts Sep 26 18:42:27 <_zANDER_> I just finished installing Debian on my beaglebone based on this site http://elinux.org/BeagleBoardDebian. Sep 26 18:42:48 <_zANDER_> Found a problem with the console login. Looks like it spawns 2 getty's. Sep 26 18:43:15 <_zANDER_> Had to edit /etc/inittab to take out second modem getty in order to login. Sep 26 18:52:16 zander write a debian bugreport Sep 26 18:52:24 or fix it in elinux wiki Sep 26 18:53:08 <_zANDER_> woglinde: so I can use the standard debian bugreporter for beaglebone arm version then. Sep 26 18:58:20 woglinde: hi. I was wondering about your musb comment the other day. have you experienced cases where userspace has managed to fudge up musb? Sep 26 18:58:36 non-musb-related userspace, I mean Sep 26 19:01:41 tasslehoff sorry was a misunderstanding Sep 26 19:01:51 I did not look at the musb code at all Sep 26 19:02:29 woglinde: ah. that was actually kinda good to hear Sep 26 19:02:32 I only experienced problems with it when you are power it up via musb pon desktop pc Sep 26 19:02:52 when the desktop pc has gadet stuff loaded Sep 26 19:03:16 tasslehoff I thought you have debugged it now Sep 26 19:03:59 woglinde: just started looking at it today. have had a few other things to do (as always). Sep 26 19:04:09 hm ah Sep 26 19:06:39 trying to upgrade to 3.2, to see if it just goes away :) Sep 26 19:14:31 hehe Sep 26 19:14:41 upgrade to 3.6 haha Sep 26 19:15:06 also, it's a bug that occurs twice a week Sep 26 19:15:10 :) Sep 26 19:17:39 twice a week is to much Sep 26 19:32:35 more than one nasty bug hiding under the fridge? Sep 26 20:05:15 Tartarus, patches for extra uarts on bones sent to u-boot list Sep 26 20:05:34 I'm not sure how to do what's in my 4/6 patch, any guidance would be gladly accepted Sep 26 20:06:33 for anyone else who wants to try them out, see -> https://github.com/bradfa/u-boot/commits/bone-uart-patches-for-mainline Sep 26 20:09:06 * mdp tries to remember what is special about uart3 Sep 26 20:09:13 mdp, not pinned out on bone Sep 26 20:09:18 at least not rx/tx Sep 26 20:09:23 rts/cts are though! Sep 26 20:09:37 bradfa: Are you a subscriber to u-boot@lists.denx.de? Sep 26 20:09:40 hrm Sep 26 20:09:44 If not they need to be approved by the moderator Sep 26 20:09:45 Tartarus, yes, as of about 2 days ago Sep 26 20:09:49 ah, ok Sep 26 20:09:51 kicking gmail Sep 26 20:10:42 or your mail queue hasn't flushed out? Sep 26 20:10:53 They got all cc'ed back to me already Sep 26 20:11:23 I don't see them in the denx.de list archive, though Sep 26 20:11:29 bradfa, I guess I would prefer seeing the define for COM4...without an _evm_uart3 target...for the inevitable custom board that uses uart3.. Sep 26 20:12:14 mdp, OK, seems reasonable Sep 26 20:12:16 Tartarus, curious what you think there Sep 26 20:12:28 bradfa, what he thinks is far more important though ;) Sep 26 20:12:34 mdp, I need to think about it. And it's nap time for the kids right now, so afk a bit Sep 26 20:12:40 ok, cya Sep 26 20:12:48 it's head home time for me Sep 26 20:13:36 cya2 Sep 26 20:13:39 if my patches don't come through the list, please let me know, it's my first time with git send-email Sep 26 20:13:54 I sent them to myself first and everything seemed peachy Sep 26 20:14:11 check out --dry-run Sep 26 20:15:06 mdp, just tried sending to another email, that worked. Sent through my fastmail.fm smtp and got to my gmail inbox Sep 26 20:15:22 so everything seems to work Sep 26 20:15:27 ok Sep 26 20:15:40 there's some good tutorials on setting up for using gmail Sep 26 20:15:52 I have to use gmail due to inadequate TI mail service Sep 26 20:16:32 I use personal smtp since otherwise I'm on exchange and it marks anything I send through it as spam because it's plain text and not html... Sep 26 20:17:16 but regardless, now really time to go home, ping me if the patches don't make it through by morning and I'll send again Sep 26 20:18:29 ah, I got u-boot "message awaits approval" emails now Sep 26 20:18:36 so, smtp FTW! Sep 26 20:18:39 ta ta Sep 26 20:30:03 could using a speaker cable into a board blow an audio codec? Sep 26 20:30:30 or at least damage it Sep 26 21:51:40 <_av500_> back Sep 26 22:22:26 front Sep 26 22:23:18 <_av500_> +1 Sep 26 22:37:12 prpplague: you there? Sep 26 23:09:41 fester: sorry busy atm Sep 26 23:17:43 ah ok Sep 26 23:17:44 np Sep 26 23:56:51 prpplague: i was just wondering if you thought this might work with the beagle, since they're both linux: https://github.com/Sh4d/LPD8806 Sep 26 23:57:26 fester: it's userspace, it should be fine Sep 27 00:07:41 thanks Sep 27 00:07:53 but he's gone **** ENDING LOGGING AT Thu Sep 27 02:59:57 2012