**** BEGIN LOGGING AT Mon Nov 14 02:59:57 2011 Nov 14 13:51:38 oneiric firefox, when being compiled on a machine with linux 3.0 kernel, will fail to compile with: Nov 14 13:51:42 ../coreconf/config.mk:71: ../coreconf/Linux3.0.mk: No such file or directory Nov 14 13:51:50 fun :) Nov 14 13:52:10 oh, could you file a bug, our buildds all use old kernels Nov 14 13:52:47 (yet) Nov 14 13:52:50 looking at it right now Nov 14 13:53:18 we have some pandas but that are currently all assigned to armhf porting iirc Nov 14 13:53:40 so trhey havent seen any higher level builds like desktop packages yet Nov 14 13:53:55 it is expected that we drop all but panda soon though Nov 14 14:12:09 Hi Nov 14 14:16:46 How can "remove" a jump to a tag in a macro ??I to change this code --> http://pastebin.com/CrkXZZJK (tag already defined obvius) Nov 14 14:21:39 punxos: what are you trying to do exactly? I don't understand from your code Nov 14 14:22:38 this code is inside a macro, so I can't create a tag because the compiler fail due to tag is redefined Nov 14 14:22:48 is it still correct to dd the 11.10 pandaboard image to the SD card? or is a different method used now? Nov 14 14:24:49 brendand, neither the images nor the method to write tham changed Nov 14 14:24:53 *them Nov 14 14:25:01 its the same as 11.04 Nov 14 14:25:43 ogra_ - ok, thanks Nov 14 14:26:10 punxos: so, where you say "tag", you mean "label Nov 14 14:27:33 punxos: Is this as assembler macro, or are you working with C inline assembler? Nov 14 14:29:05 assembler macro Nov 14 14:29:42 dmart: I found the solution: add $ to label Nov 14 14:29:49 thx anyway Nov 14 14:31:43 punxos: "dollar local labels" don't work on arm -- though if you have some code where they do appear to work, I'd like to see it Nov 14 14:31:56 1$ Nov 14 14:31:56 -> Nov 14 14:32:02 Error: junk at end of line, first unrecognized character is `1' Nov 14 14:32:22 dmart: beq $label Nov 14 14:32:28 $label: some code Nov 14 14:33:56 dmart: sorry not work :( Nov 14 14:33:59 $ is allowed in symbol names on arm -- you haven't really solved the problem there; you just have a different label name when you ass $ Nov 14 14:34:09 I dint' do enough checks Nov 14 14:34:21 yes yes Nov 14 14:34:23 sorry Nov 14 14:34:35 .oO( arm is a british company, try a pound sign instead of a dollar one ;) ) Nov 14 14:34:44 xD Nov 14 14:34:55 dmart: this example code-> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.kui0100a/armasm_cegceijg.htm Nov 14 14:35:53 * dmart doesn't know if UTF-8 is allows in symbol names for £ :P Nov 14 14:36:00 hehe Nov 14 14:36:26 :D Nov 14 14:36:32 punxos: are you working with the ARM RVCT assembler (armasm) or GNU as? The syntax for macros is completely different; the example is for armasm Nov 14 14:37:04 arm-none-eabi- Nov 14 14:37:19 OK, so GNU as then Nov 14 14:41:09 punxos: porting that example to GNU as might look something like this: http://pastebin.ubuntu.com/738238/ Nov 14 14:41:52 dmart: thanks Nov 14 14:42:01 But I suspect that doesn't quite solve your problem. Can you paste your actual source code so I can see what you're trying to do? Nov 14 14:42:38 yes 1 sec Nov 14 14:44:14 dmart: Nov 14 14:44:14 http://pastebin.ubuntu.com/738241/ Nov 14 14:44:28 just update a register if r0 is 0 Nov 14 14:48:56 dmart: but this macro is inside other macro due to that your example not work Nov 14 14:51:49 hello ubuntu guys! Nov 14 14:52:11 does anyone know of a particular brand/model of SD card that actually supports erase/secure erase? Nov 14 14:54:54 yes, a drill will securely erase any sd-card Nov 14 14:55:02 the bigger the drillbit the better Nov 14 14:59:59 not useful :D Nov 14 15:05:03 punxos: in this case, you could avoid the branch using conditional assembly Nov 14 15:05:33 mm I dont' know how Nov 14 15:05:51 a short sample please ? Nov 14 15:09:26 punxos: here are some examples of things you can do: Nov 14 15:09:27 http://pastebin.ubuntu.com/738263/ Nov 14 15:10:09 The GNU assembler documentation is a bit vague about some acpects of macros, but it is worth reading Nov 14 15:10:53 http://sourceware.org/binutils/docs-2.21/as/Macro.html#Macro Nov 14 15:11:06 dmart: oooo thanks very much ! great job Nov 14 15:13:36 hope it helps Nov 14 15:13:50 yes it's perfect :) Nov 14 15:14:13 works perfectly Nov 14 15:14:41 which approach did you choose? Nov 14 15:15:03 beq 1f Nov 14 15:15:03 1: Nov 14 15:15:19 it's more visual for me Nov 14 15:16:55 That's a good simple solution, and works in a lot of cases. If your code is performance-critical, you might still want to use conditional instructions; though it really depends on how reliably that branch can be predicted (and the sophistication of branch prediction on the CPU(s) you're targeting) Nov 14 15:24:41 note that if you are building thumb2 (or you want to support thumb2) then see IT instruction instead.. Nov 14 15:40:55 Hi Nov 14 15:41:45 I have a beagleboard, can I install ubuntu on it ? Nov 14 15:42:27 well i guess so :) Nov 14 15:44:11 antoniodariush, depends what version of beagle and which flavour of ubuntu Nov 14 15:45:14 I have a project but running on nanonote so i need to cross compile and use toolchains everytime to compile the software Nov 14 15:45:17 (you cant easily run ubuntu-desktop on a beagle C4 ... not enough ram, but it will run ubuntu-server fine for example) Nov 14 15:46:21 as long I can avoid cross compilation I am fine with any flavour of ubuntu Nov 14 15:46:44 right, use -server for this Nov 14 15:48:03 I have a beagleboard-xM Nov 14 15:48:15 ok i'll have a look thanks Nov 14 15:49:28 even with Xm you want -server unless you actually want to run and use desktop apps ... if its for compiling, -server is the best chouce to not waste resources Nov 14 15:49:40 *choice Nov 14 15:52:16 yeh that's true Nov 14 15:52:35 dmart: thanks for the info Nov 14 15:54:57 punxos: no problem -- glad it was helpful Nov 14 17:47:22 If a irq is raised when the user is in destination instruction jump. if you do "subs pc.lr,#4" maybe is not true (if the las jump was done). Any can explain me that ? Nov 14 17:58:58 ogra_, NCommander: https://blueprints.launchpad.net/ubuntu/+spec/ubuntu-arm-p-optional-initrd had work items, and now it doesn't - what's the story there? :) Nov 14 17:59:27 slangasek, NCommander just blindly pasted notes in, while i'm working on the spec atm Nov 14 17:59:57 (I'm the drafter, ask him why he blindly pastes stuff without talking to me) Nov 14 18:00:42 slangasek, apw, tim and me had another long session in the hallway later, actually looking at code ... the kernel side changed a bit Nov 14 18:01:06 so i'm rewriting the stuff atm Nov 14 18:02:58 hmm Nov 14 18:03:14 slangasek, i'll finish it by tomorrow, do you want any specific person assigned to the foundations WIs ? Nov 14 18:03:25 ro should i just put the team in and leave it to you Nov 14 18:03:47 *or Nov 14 18:04:10 the latter is fine Nov 14 18:04:17 many of these are probably taken by me personally Nov 14 18:04:31 k Nov 14 18:12:38 * ogra_ smiles seeing libpam-sshauth in the archive Nov 14 18:31:06 Hi Nov 14 19:12:25 Was there an answer to the OMAP3 question? Was it dropped for 12.04? Nov 14 19:12:39 no, it wasnt Nov 14 19:12:52 (what was the question though ?) Nov 14 19:18:54 int_ua: The only real question is whether or not to declare no-support for ubuntu-desktop on omap3 based systems with <512M memory. Currently, running ubuntu-desktop on a Beagle C4 is extremely painful due to swappiness. Nov 14 19:19:54 It takes me over an hour to do any release testing on it alone (other platforms can be smoke tested in 20 minutes). Nov 14 19:28:08 GrueMaster: I'm running Ubuntu Core on the N900 from class 6 microSD and I wouldn't say that swappiness is a serious problem. Nov 14 19:28:39 Ubuntu-core is not Ubuntu-Desktop (with unity-2d, etc). Nov 14 19:29:11 GrueMaster: are there daily releases of the Core? Nov 14 19:29:34 Only when core packages change afaik. Nov 14 19:30:05 We may have scaled back image building to 3/week due to armHF bring up. Nov 14 19:30:34 Where are they published? There are no ARM: http://cdimage.ubuntu.com/ubuntu-core/daily/current/ Nov 14 19:31:48 looks like daily/20111112 is the most recent for armel. We'll need to make sure that they don't get deleted. Nov 14 19:31:52 GrueMaster: I don't really care getting it _daily_. I care about their existing :) Nov 14 19:32:04 Yes, agreed. Nov 14 19:32:18 infinity: ^^^ Nov 14 19:32:25 Any comments? Nov 14 19:33:26 Hrm. Curious. Nov 14 19:33:34 May be an accidental oops from the cdimage cutover. Nov 14 19:33:36 Or something. Nov 14 19:33:39 P.S. I've posted a question also: http://askubuntu.com/questions/78875/was-omap3-support-dropped-for-12-04 Nov 14 19:33:57 Although, I see armel on the 12th. Nov 14 19:34:01 int_ua: omap3 will not be dropped. Nov 14 19:34:09 http://cdimage.ubuntu.com/ubuntu-core/daily/20111112/ Nov 14 19:34:12 20111112 is the last daily armel on every palce Nov 14 19:34:15 *place Nov 14 19:34:17 We went over the question of omap3. Nov 14 19:35:02 GrueMaster: Thanks :) Nov 14 19:35:03 I will still be actively testing on beagleXM (512M), and am looking to get ubuntu-desktop running on the Nook Color this cycle. Nov 14 19:35:37 int_ua: So it is. We need to hunt that down, thanks. Nov 14 19:36:05 I would guess possible pool-churn (again). Nov 14 19:38:17 GrueMaster: No. Nov 14 19:38:24 GrueMaster: There aren't even failed build attempts. Nov 14 19:38:39 GrueMaster: I'll look into it. Nov 14 19:38:45 hrm. That is odd. Nov 14 19:40:58 And more curiously, there are builds on the livefs builders... Nov 14 19:41:30 cdimage full (again)? Nov 14 19:43:26 no. Nov 14 19:43:31 New machine. Nov 14 19:43:35 Also, ubuntu-server built. Nov 14 19:44:01 And ubuntu. Nov 14 19:44:08 Nevermind my claims that nothing was building. Nov 14 19:44:50 Looks like only core is missing. And logs are missing, but that's a different issue entirely. Nov 14 19:45:22 Oh! Nov 14 19:45:32 annonaceae is dead. Nov 14 19:45:34 I think. Nov 14 19:45:48 infinity: kubuntu-netbook and kubuntu-mobile are missing too Nov 14 19:46:17 int_ua: kubuntu-netbook is dead. Nov 14 19:46:26 Hrm. That is a beagleXM buildd. Nov 14 19:46:28 int_ua: kubuntu-mobile is half there. Nov 14 19:46:51 int_ua: (basically, all omap3, mx5, and the -core images are missing, because their buildd went kaput) Nov 14 19:46:55 infinity: s/kubuntu-netbook/kubuntu/ Nov 14 19:46:56 All the *ceae buildds are beagleXMs from linaro iirc. Nov 14 19:46:56 * infinity chases up with IS now. Nov 14 19:53:21 Okay, annonaceaeaaeaaeee fixed. Dailies should look more normal tomorrow. Nov 14 20:02:19 heh Nov 14 20:04:08 Thanks a lot :) Nov 14 20:06:04 Is Annonaceae a server name?) Nov 14 20:06:42 No, it is one of many buildds in our pool. It is a beagleXM platform. Nov 14 20:07:11 We build the entire armel pool (arch=armel/armhf) on actual hardware. Nov 14 20:24:56 int_ua: "server" is a pretty liberal use of the word, but it's a hostname, yes. **** ENDING LOGGING AT Tue Nov 15 02:59:57 2011