**** BEGIN LOGGING AT Sat Oct 27 03:00:02 2012 Oct 27 03:00:59 robbiethe1st: Oct 27 03:01:03 ? Oct 27 03:01:08 next time you update backupmenu Oct 27 03:01:18 Yes? Oct 27 03:01:21 please make untar ignore timestamp issues Oct 27 03:01:27 It does. Oct 27 03:01:41 But it shows up as errors anyway, doesn't do any harm. Oct 27 03:01:44 then why did it spew out thousends of errors on me and break my rootfs? Oct 27 03:01:54 *broke Oct 27 03:02:24 It will spew out errors - that's just a listing of stderr. But it's set to ignore most all errors - go check the source for the exact tar options used Oct 27 03:02:40 Why do you think it broke your rootfs? Oct 27 03:02:55 Did you stop it before it was complete? Oct 27 03:03:04 because it made a fine bootable device unbootable Oct 27 03:03:07 not at all Oct 27 03:03:16 OK, and what kernel were you using? Oct 27 03:03:17 it said that it was complete in big nice green text Oct 27 03:03:24 powerkernel Oct 27 03:04:06 Did the backup you were restoring have the same kernel as currently flashed to the device? Oct 27 03:04:14 hmmmmmm Oct 27 03:04:18 let me think Oct 27 03:04:33 If you've got a different kernel running - flashed to the kernel partition - the backup won't have kernel modules for it Oct 27 03:04:33 no Oct 27 03:04:36 therefor, no booting Oct 27 03:04:54 One way to check, use flasher to flash /only the stock kernel/ Oct 27 03:04:57 ok that's understandable Oct 27 03:05:01 If it boots, great. Oct 27 03:05:13 i can't flash, (barely) broken microUSB Oct 27 03:05:22 i'll go to a repair shop and solve that problem Oct 27 03:05:29 then I'll be able to do stuff with it Oct 27 03:05:34 Yeah. Oct 27 03:05:59 But if it's not the kernel, that's weird - I /generally/ have no issues, and rarely see complaints Oct 27 03:06:02 however the problem came up as very erratic Oct 27 03:06:20 like half of the things still work, most doesn't Oct 27 03:06:33 Did you flash the optfs as well? Oct 27 03:06:45 hm Oct 27 03:06:48 Because all of your app's files are stored on there Oct 27 03:06:54 I don't *think* so, but I'm not sure Oct 27 03:07:17 rootfs has main OS, and some binaries, plus symlinks to most apps Oct 27 03:07:24 which are on the opt/home fs Oct 27 03:07:28 yeah I know what they're used for pretty much Oct 27 03:07:49 but I can't see what difference flashing optfs would make in my case Oct 27 03:08:05 bootmenu crashes after 5s of idle (probably watchdog issue), maemo doesn't boot at all, backupmenu crashes unless I press keys during startup (then it gives a "watchdog not found" error), and menu options mostly doesn't work (though they respond fine) Oct 27 03:08:12 It'd only be a real issue if your current apps are different from the old apps Oct 27 03:08:31 Well, that's interesting Oct 27 03:08:38 I wonder if that's a kernel issue... Oct 27 03:08:55 I tried to re-restore the thing, but I cannot mount eMMC properly now Oct 27 03:09:19 it told me to run fsck (pretty sure it wouldn't change things, but it didn't work anyway - "fsck.ext2: No such file or directory") Oct 27 03:09:41 So you've got no fsck on the rootfs Oct 27 03:09:55 wow, could the backup have gotten corrupted/been incomplete? Oct 27 03:09:58 root shell just gave me a black screen, though it *always* did (even when things were working fine) Oct 27 03:10:25 i'm pretty sure that's not the case, unless "Completed successfully!" (or whatever the text is) isn't certain Oct 27 03:10:39 also wouldn't that also make restore impossible? Oct 27 03:10:56 Well, what that knows is that tar completed successfully Oct 27 03:11:12 whether the backup is complete or is what you actually want... no real way to tell Oct 27 03:11:21 DocScrutinizer05 thought it was because of a bug in which backupmenu didn't tar without ignoring timestamp, which means that everything that didn't have a time of 0 failed Oct 27 03:11:21 all it's doing is running simple shell commands, basically Oct 27 03:11:42 s/time/last changed time/ Oct 27 03:11:42 FIQ meant: DocScrutinizer05 thought it was because of a bug in which backupmenu didn't tar without ignoring last changed timestamp, which means that everything that didn't have a time of 0 failed Oct 27 03:11:47 It shouldn't. Oct 27 03:11:54 Check the source actually Oct 27 03:12:15 the file you're looking for is /usr/share/BackupMenu/BackupMenu.item - shell script Oct 27 03:12:20 look for 'gtar' lines. Oct 27 03:12:51 Hm, can you mount anything via USB? Oct 27 03:13:36 Dinnertime. Be back soon. Oct 27 03:14:27 sorry, but when I check the source, I see no trace of a "ignore timestamp", though my knowledge of gtar might just be limited Oct 27 03:14:53 gtar p"$backupRootCompress"cf $backupRootLocation ./ --ignore-failed-read --numeric-owner --exclude=./tmp/* 2> /tmp/tarstatus & pid=$! <-- is the line Oct 27 03:16:23 no... Oct 27 03:16:35 I checked manpage quickly, the options given seems to be unrelated to that Oct 27 03:18:32 possible that --ignore-failed-read ignores mtime though, so you might be correct Oct 27 03:19:52 oh well, I can't really diagnose anything at the moment due to microUSB problems, as said Oct 27 03:20:30 Ok, I can see that a modified kernel would make Maemo unbootable. Oct 27 03:20:41 er, different kernel between restore and pre-restore Oct 27 03:21:17 I can *not* however see why that would broke backupmenu as horrible as it did (or bootmenu for that matter, which doesn't even use the kernel pre-boot) Oct 27 03:25:33 Also, I though that the kernel *were* in the rootfs, i.e. didn't think about it much when I flashed a new kernel in first place Oct 27 04:04:03 FIQ, I can't give you too much help with regards to the screwed-up restore; only thing I can suggest is doing a diff of your backup image(if you have a copy elsewhere) against a previous backup or another person's. Oct 27 04:04:53 i'll see what I can do once i can reflash and boot the device (as it's all on eMMC etc) Oct 27 04:06:04 Yeah Oct 27 04:06:20 Um, check the thread. There's an old 'bootstrap' rootfs image you can try Oct 27 04:06:23 flashable Oct 27 11:31:07 While this was running, /sys/class/power_supply/bq24150-0/status said "Charging". I'd guess that this is not quite true. The N900 was receiving power but not enough to run the N900 and the battery was emptying. /sys/class/power_supply/bq24150-0/charge_current says "1750". What does that mean? I doubt that it means that the N900 is receiving 1750 mA. Oct 27 11:31:10 haha wat Oct 27 11:31:19 DocScrutinizer05: http://talk.maemo.org/showpost.php?p=1286272&postcount=115 - explain Oct 27 11:40:52 from memory, there is no way to measure charging current Oct 27 11:41:06 only battery current Oct 27 11:41:45 'charging' means have vbus, from the charger perspective Oct 27 11:53:42 wtf, where's the download link for apkenv? :s Oct 27 11:56:39 thp must've taken it down. Oct 27 11:56:44 http://talk.maemo.org/showpost.php?p=1284909&postcount=326 Oct 27 11:56:49 meh Oct 27 11:56:49 Found here: http://wiki.maemo.org/Apkenv Oct 27 11:56:59 (.zip containing a .deb) Oct 27 11:58:49 thankfully the wayback machine exists Oct 27 11:59:03 aww, it's not archived :( Oct 27 12:00:31 Uh Oct 27 12:00:33 I just linked it. Oct 27 12:00:36 I'll try again. Oct 27 12:00:40 http://talk.maemo.org/showpost.php?p=1284909&postcount=326 Oct 27 12:00:59 >harmattan Oct 27 12:02:13 Ah Oct 27 12:07:38 kerio: no /sys/class/power_supply/bq24150-0/status without pali's bq24150.ko, so I have NFC what does "charging" mean Oct 27 12:10:47 kerio: http://talk.maemo.org/showpost.php?p=1241530&postcount=1 so please ask the OP of thread, not me Oct 27 12:11:29 who's got a Nexus 7? Native ubuntu on it since yesterday, with hildon on it could be almost maemo. Oct 27 12:29:59 How to set maemo's clock from cli? Oct 27 12:30:22 gui is so slow Oct 27 12:30:38 I want to set time format and time zone from cli Oct 27 12:30:40 date & hwclock? Oct 27 12:30:49 ooh Oct 27 12:30:59 DocScrutinizer05, hwclock? Oct 27 12:31:08 silly UI wants you to swipe/scroll all the way Oct 27 12:33:43 LaoLang_cool: nevermind, for setting TZ and format you need other tools, no idea what's the maemo way to edit them Oct 27 12:34:22 tzset? Oct 27 12:34:32 LaoLang_cool: setting clock is date(1) and hwclock(8) Oct 27 12:34:59 hello Oct 27 12:35:04 hello Oct 27 12:39:08 I tried to use gconftool-2 -s /apps/clock/time-format --type=boolean ture, but it doesn't work, after I launch clock, the time format will be changed to default 12hr again Oct 27 12:39:28 I got confused Oct 27 12:42:45 * pressed Oct 27 12:43:15 xray7225: eh? Oct 27 12:44:34 LaoLang_cool: I assume the 'ture' is a typo you introduced here? Oct 27 12:45:30 LaoLang_cool: check by >> gconftool-2 -g /apps/clock/time-format<< Oct 27 12:46:27 DocScrutinizer05, it's typo ;p Oct 27 12:46:39 sure, but where? Oct 27 12:46:54 as long as it's a typo here, it doesn't affect cli Oct 27 12:47:03 I have set it to true, but after launch clock, the value will be restored to 'false' Oct 27 12:47:17 I don't know why Oct 27 12:47:26 in cli however it won't throw an error (I just checked) but I doubt it will result in what you intended Oct 27 12:49:42 LaoLang_cool: check out tzconfig, tzset, and locale Oct 27 12:50:06 s/set/select/ Oct 27 12:50:07 DocScrutinizer05 meant: LaoLang_cool: check out tzconfig, tzselect, and locale Oct 27 12:50:39 After running gconftool-2, time showing on status bar will be changed to 24hr format, but after running clock, it will be restored back to 12hr Oct 27 12:50:56 H iave no tzset cmd here in maemo ... Oct 27 12:51:41 tzseelct Oct 27 12:51:50 tzselect grrrr Oct 27 12:52:06 -sh: tzselect: not found Oct 27 12:52:19 IroN900:/home/user# ll /usr/bin/tzselect Oct 27 12:52:21 -rwxr-xr-x 1 root root 6894 2010-06-08 14:19 /usr/bin/tzselect Oct 27 12:52:35 but tz can give tzselect as a completion candidate, weired... Oct 27 12:52:41 nfc which pkg brought it Oct 27 12:52:45 I'm running as root now Oct 27 13:08:11 oh, tzselect use /bin/bash as shell, I haven't it, so it says not found Oct 27 13:09:08 DocScrutinizer05: libc6, apparently Oct 27 13:10:42 kerio: ?? tzselect? Oct 27 13:11:16 yep Oct 27 13:11:19 kerio: dafq a lib comes with cmdline binaries? Oct 27 13:11:21 dpkg -S `which tzselect` Oct 27 13:11:51 wow Oct 27 13:12:05 kerio: thanks Oct 27 13:12:32 kerio: you think you can stop trolling now that your stand-in is here again? :-D Oct 27 13:12:39 nope Oct 27 13:12:47 way too useful... ;-P Oct 27 13:13:01 i feel that the mixture of trolling and helping is what really upsets people Oct 27 13:13:08 hehehehe Oct 27 13:13:14 might be about right Oct 27 13:14:20 and in the meantime, i help people :D Oct 27 13:15:22 I'm however afraid that - if this tz* stuff comes with libc6 - it might be of no effect on maemo Oct 27 13:16:38 prolly a strace of the time&language picker in maemo settings might reveal what's primary location to tweak for such TZ and format stuff Oct 27 13:16:46 in maemo Oct 27 13:17:27 or just dbus-monitor Oct 27 13:17:57 alternatively do a ls -lR / >before-ls; gconftool -R / >before-gc Oct 27 13:18:16 the change settings-tz, then compare before* to after* Oct 27 13:18:50 then* change... Oct 27 13:20:44 (dbus) I'm not sure this class of settings is done via dbus Oct 27 13:38:40 trolling? you dont say... Oct 27 13:39:23 i do say Oct 27 13:43:58 precisely Oct 27 16:38:51 Pali: what's the state of uboot as a full NOLO replacement? Oct 27 16:39:14 uboot has no support for n900 onenand Oct 27 16:39:17 Pali: ROMBL tries to read xloader from the uSD apparently, so it would be neat if we had a way to run an emergency emergency system from there Oct 27 16:39:27 kerio, but in qemu it show me hello world to serial console :-) Oct 27 16:39:42 so I think uboot spl is working for n900 Oct 27 16:40:01 Pali: what about the hw initialization? Oct 27 16:40:17 kerio, I do not know if omap n900 can boot xloader from sd card... Oct 27 16:40:33 Pali: it can Oct 27 16:40:46 kerio, at least xloader+uboot spl doing all what is needed for qemu Oct 27 16:40:59 look at the end of the TRM for details Oct 27 16:41:23 freemangordon, it needs to configure to change boot sequence Oct 27 16:41:37 and I do not know if it is possible on n900 without serial console... Oct 27 16:42:00 at least there *is* command for "cold-flashing" for it Oct 27 16:42:09 to boot from mmc Oct 27 16:42:40 Pali: the boot order depends on some DWORD in RAM iirc Oct 27 16:42:57 but I do not know exact format of xloader on non gp devices Oct 27 16:43:08 in trm is only MLO file for gp device Oct 27 16:43:13 I see Oct 27 16:43:38 I already looked at it when I implemented cold flashing... Oct 27 17:20:12 i'm eager to test MLO on uSD, what's been the exact requirements? VFAT partition, MLO.*? on \ ? Oct 27 17:20:54 I mean, dd if=dev/NOLO of=MyDocs/MLO.bin would be a start Oct 27 17:21:06 I'd expect boot to fail and freeze on that Oct 27 17:21:23 Pali: freemangordon: ^^^ Oct 27 17:24:26 DocScrutinizer05, see http://omappedia.org/wiki/Bootloader_Project Oct 27 17:25:34 see also: http://linux.fjfi.cvut.cz/~zub/omap34xx-boot-order/ Oct 27 17:25:44 but I have no idea if this working on HS devices Oct 27 17:29:00 DocScrutinizer05: i'd guess MBR and the first partition formatted as vfat Oct 27 17:29:20 kerio: nope, a file called MLO Oct 27 17:29:29 DocScrutinizer05: no, i mean Oct 27 17:29:37 MLO.bin in the first vfat partition of the MBR Oct 27 17:30:16 sure I guess it will look for MBR to get partiton table to learn what's first partition Oct 27 17:30:34 and the file is "MLO" in root, not "MLO.bin" Oct 27 17:31:16 i see Oct 27 17:34:59 Oh LOL Oct 27 17:35:02 IroN900:~# ./omap34xx-boot-order Oct 27 17:35:04 sys_boot[5:0]: 0x10 Oct 27 17:35:05 Boot order: OneNAND USB UART3 MMC1 -- Oct 27 17:35:47 Pali: mio. thanks Oct 27 17:39:10 so if that's true and I'm not mistaken, you *can't* coldflash a working xloader+NOLO - or could you? Oct 27 17:39:49 USB is coldflash? Oct 27 17:39:56 well, xloader still might check USB before chainloading existing NOLO Oct 27 17:40:05 DocScrutinizer05: remember the nandtester Oct 27 17:40:08 he had no xloader Oct 27 17:40:53 kerio: yep, that's pretty easily the reason why he could coldflash Oct 27 17:41:36 ROMBL didn't find a valid xloader in NAND, so listened to ZSB for coldflashing Oct 27 17:41:41 USB* Oct 27 17:41:47 hm Oct 27 17:42:02 DocScrutinizer05: so we *are* set, if we find a way to boot from uSD Oct 27 17:42:15 hmm? Oct 27 17:42:27 oh no wait, there's still a problem Oct 27 17:42:32 Boot order: OneNAND USB UART3 MMC1 -- Oct 27 17:42:52 it only goes to mmc1 if it can't load the nand xloader Oct 27 17:43:02 yep, exactly Oct 27 17:43:49 and what xloader does, this is a completely different story then Oct 27 17:44:36 http://omappedia.org/wiki/Bootloader_Project#Second_Stage_Boot Oct 27 17:48:08 cp $whatever /media/mmc1/u-boot.bin Oct 27 17:48:18 might be worth a try Oct 27 17:48:49 >>In the case of loading the u-boot via the SD card, the SD card x-loader looks for a FAT32 partition on the first MMC controller and scans the top level directory for a file named "u-boot.bin". It then transfers the file into main memory and transfers control to it.<< Oct 27 17:49:07 ...huh, i have an u-boot.bin on my MyDocs Oct 27 17:49:34 so I guess 'our' xloader doesn't do that Oct 27 17:50:54 it might look at USB for somebody trying to coldflash NOLO, and load 2nd NOLO from there into RAM, but it probably won't look at MMC1 for u-boot.bin Oct 27 17:51:13 :( Oct 27 17:51:48 DocScrutinizer05, colflashing is done via USB peripherla booting Oct 27 17:51:58 and usb peripheral booting is done *before* omap booting Oct 27 17:52:09 sorry? Oct 27 17:52:19 so you can cold flash Oct 27 17:52:26 what's omap booting? Oct 27 17:52:38 I mean booting from above your list Oct 27 17:52:43 what's usb peripheral booting Oct 27 17:53:16 the above list tells where ROMBL looks for xloader/MLO Oct 27 17:53:36 what's MLO? Oct 27 17:53:53 once xloader got loaded, it again has some list (of possibly only one source) where to look for NOLO Oct 27 17:54:06 kerio: another name for xloader Oct 27 17:54:16 DocScrutinizer05, bootrom first check if there is omap asic message sent from usb Oct 27 17:54:29 if yes then it wait for another message via usb Oct 27 17:54:38 if no then bootrom booting from list Oct 27 17:54:45 oooh Oct 27 17:55:03 and one message is "wait for boot image via usb" and boot it Oct 27 17:55:07 wtf is this specified now? Oct 27 17:55:36 USB in that boot order is booting from usb device Oct 27 17:55:41 via host mode Oct 27 17:55:48 WUT? Oct 27 17:55:55 (but this is not working on n900) Oct 27 17:55:59 that's not what I think ROMBL can do Oct 27 17:56:13 yes, omap boot rom can do it Oct 27 17:56:28 boot xloader from assrage device??? Oct 27 17:56:57 omap bootrom can boot xloader from nand, onenand, mmc, usb flash, ... Oct 27 17:57:01 look into trm Oct 27 17:57:10 usb FLASH? Oct 27 17:57:12 duh Oct 27 17:57:26 usb flash disk Oct 27 17:58:13 DocScrutinizer05, see: https://github.com/radare/0xFFFF/blob/master/src/cold-flash.c#L82 Oct 27 17:58:34 all possible boot messages Oct 27 18:04:35 Pali: does ROMBL try multiple times to load the xloader? Oct 27 18:05:08 kerio, I do not know what omap rom doing Oct 27 18:05:13 try look at trm Oct 27 18:05:26 TRM "26.4 Device Initialization by ROM Code" doesn't mention mass storage mode for USB, only "download" from host Oct 27 18:07:54 hm, there was booting file from FAT partition on usb device (via usb host mode) Oct 27 18:09:55 don't see it anywhere in this region of TRM Oct 27 18:10:07 have to check "Figure 26-8. Overall Booting Sequence" for a while Oct 27 18:10:33 (p.3455 of my TRM) Oct 27 18:22:55 Pali: I only see "/* Omap HS USB boot message */" in this 0xffff source you linked me to, and that's what I see as well in e.g. "26.4.5 Peripheral Booting" in TRM. AIUI it's waiting for image to get sent via USB "plaintext", not mounting a USB mass storage device and load some file from there Oct 27 18:23:50 that's pretty much what we expect to see for coldflashing Oct 27 18:26:27 Pali: "Figure 26-9. Device List Setup" says "power-on reset? YES: take all devices from sys_boot [pins]" Oct 27 18:28:29 Pali: so in the end I don't see why we could make ROMBL load xloader from USB - with "Boot order: OneNAND USB UART3 MMC1 -- " - when xloader in NAND is succeeding for "try to load" Oct 27 18:30:04 DocScrutinizer05, ok Oct 27 18:31:32 I'd be happy to learn I missed something, or that maybe output of ./omap34xx-boot-order was incorrect Oct 27 18:32:36 for now it seems to me there's no way to override a working xloader in OneNAND - which otoh makes quite some sense Oct 27 18:33:38 you can't nuke xloader without booting device, and you can't boot device in a way not meant to be valid by xloader Oct 27 18:34:11 so in that regard the device is tamper-proof Oct 27 18:47:35 DocScrutinizer05: tamper-proof as long as you don't have a computer and a usb cable handy Oct 27 18:47:57 kerio: exactly NOT Oct 27 18:48:54 since you can't stop ROMBL from loading and executing OneNAND xloader by any means, other than destructing the OneNAND based xloader, on a power-on-reset boot Oct 27 18:49:25 i thought coldflashing happened before the onenand xloader was loaded Oct 27 18:49:37 only if xloader fails Oct 27 18:49:52 oic Oct 27 18:50:40 see: "Boot order: OneNAND USB UART3 MMC1 -- " Oct 27 18:51:03 ~nexus6 Oct 27 18:51:07 ~nexus Oct 27 18:51:15 btw time window for polling USB is 300ms, if anybody is interested Oct 27 18:51:48 DocScrutinizer05: so what happens if xloader is fine but NOLO is borked? Oct 27 18:51:59 that depends on xloader Oct 27 18:52:05 the current xloader Oct 27 18:52:08 the one signed by nokia Oct 27 18:52:13 yep Oct 27 18:52:34 no, i mean, what does it do if NOLO is borked? Oct 27 18:52:46 that depends on xloader Oct 27 18:53:29 DocScrutinizer05: so you don't know? Oct 27 18:53:33 since "it" == xloader in your question Oct 27 18:54:13 no, I don't know since I haven't disassembled xloader yet Oct 27 18:54:24 i thought somebody had tried it by now Oct 27 18:54:30 ...but ofc if there's no way to recover... Oct 27 18:55:18 you probably could tell from your PCs syslog, watching for xloader ENUM after ROMBL ENUM Oct 27 18:56:03 I still have to investigate and sync the data from syslog with my above reasoning Oct 27 18:57:49 actually for poor _lena this aren't exactly good news Oct 27 18:58:06 who was that again? Oct 27 18:58:28 the girl which had a failing flashing-procedure Oct 27 18:58:35 oh right Oct 27 18:58:40 and even failing NOLO bat test Oct 27 18:58:54 i thought it was just a low battery Oct 27 18:59:13 unlikely on Vbat=4066mV or sth Oct 27 19:00:45 I'm just flabbergasted she said she swapped bupbat Oct 27 19:01:47 I think he actually did mean something else Oct 27 19:01:49 she Oct 27 19:02:04 not used to she @ irc :D Oct 27 19:04:53 DocScrutinizer05, do you know AT commands? or do you have some doc for it? Oct 27 19:05:14 I need to know what should "AT+CSRSF=0,0,0,0,0,7" do Oct 27 19:05:22 I know a few, but my good times are long gone ;-D Oct 27 19:05:58 when I connet my bluetooth headset into hfs mode, then blutooth daemon show this into syslog Oct 27 19:06:00 bluetoothd[1425]: Badly formated or unrecognized command: AT+CSRSF=0,0,0,0,0,7 Oct 27 19:06:11 also on desktop and on n900 Oct 27 19:06:30 DocScrutinizer05, so I think bluetooth daemon does not have implemented that code Oct 27 19:06:43 and I'd like to know what my bluetooth headset sending Oct 27 19:06:55 but I was not able to find anything about that AT command Oct 27 19:06:56 nfc Oct 27 19:07:06 never heard of it Oct 27 19:08:00 is there any documentation for AT? Oct 27 19:08:01 Pali: it is a CIA backdoor. Oct 27 19:08:04 and obviously pnatd doesn't know it either Oct 27 19:08:15 vi__, what? Oct 27 19:08:25 how? :D Oct 27 19:08:36 Pali: there's a lot of AT docs, but obviously nothing mentioning AT+CSRSF Oct 27 19:09:02 at least according to aunt google Oct 27 19:10:56 Pali: http://thebirdfree-personal.googlecode.com/svn-history/r13/trunk/Ontech/twitter/lib/aghfp/aghfp_parse.parse Oct 27 19:11:53 # CSR Extended Features Oct 27 19:11:54 # a CSR Supported Features Request Oct 27 19:11:56 { AT + CSRSF = %d:callerName , %d:rawText , %d:smsInd , %d:battLevel , %d:pwrSource , %d:codecs , %*:ignore \r } : aghfpHandleReponseCSRSupportedFeatures Oct 27 19:11:59 hm... Oct 27 19:12:46 codecs=7 Oct 27 19:20:10 Pali: http://forums.gentoo.org/viewtopic-t-910204-start-0.html Oct 27 19:21:36 there is no info about CSR... Oct 27 19:21:46 >>Feb 1 23:12:31 localhost bluetoothd[3065]: Badly formated or unrecognized command: AT+CSRSF=1,1,1,1,1,7 << Oct 27 19:21:51 what about CCR? Oct 27 19:21:56 but nobody explained what it is Oct 27 19:22:01 sure Oct 27 19:22:12 I found many logs on google with that AT Oct 27 19:22:14 there's really few in aunt google about it Oct 27 19:22:46 now I'm trying to understand that aghfp code from googlecode Oct 27 19:22:58 Pali: headset asking for codec support for a2dp? Oct 27 19:23:16 kerio, that AT command is for hfp Oct 27 19:23:22 oh Oct 27 19:23:27 Pali: headset asking for codec support for hfp? Oct 27 19:23:31 >:D Oct 27 19:23:37 but my headset has support for apt-x Oct 27 19:23:49 i thought apt-x was strictly for a2dp Oct 27 19:24:03 I thought too Oct 27 19:24:11 I need to know more about it Oct 27 19:24:12 for some very strange reason google doesn't yield any hits for "AT+CSRSF" Oct 27 19:25:45 If I had to guess, I'd suspect something along extended functions on headset, like number display, voice-dialing, whatnot Oct 27 19:26:09 well, probably that's what HFP means ;-P Oct 27 19:27:30 DocScrutinizer05, I have 4 buttons on headset and they working without problem (both n900 and notebook) Oct 27 19:28:53 Pali: so what did you say, who is spamming that AT+CSRSF string ? Oct 27 19:29:08 my bluetooth headset Oct 27 19:29:22 o.O Oct 27 19:29:29 I'd ignore that then Oct 27 19:29:33 everytime when I connect it Oct 27 19:29:54 yes I'm ignoring it, but I'd like to know what is my bluetootj headset trying to tell me... Oct 27 19:29:59 probably it's just checking some special function of e.g. cryPhone or whatever Oct 27 19:30:47 since the command returns error on PNATD it will fall back to more generic methods to achieve wahtever it needs to do Oct 27 19:32:34 maybe the headset would offer several codecs that the modem can digest natively, and thus avoid transcoding from e.g mp3 to pcm in SoC/APE Oct 27 19:33:09 Apple/whoever might have implemented such function to save some energy Oct 27 19:34:04 from "AT + CSRSF = %d:callerName , %d:rawText , %d:smsInd , %d:battLevel , %d:pwrSource , %d:codecs" I understand that my "AT+CSRSF=0,0,0,0,0,7" is name, text, sms, batt, pwr are unsupported (0) and codecs=7 Oct 27 19:34:26 :nod: Oct 27 19:34:47 so 7 supported codecs? Oct 27 19:34:53 :shrug: Oct 27 19:35:09 codec #7 ? Oct 27 19:35:31 nah, actually number of codecs Oct 27 19:35:46 or class of codecs, or whatever Oct 27 19:35:54 see header of that file: This file contains all AT commands as specified in the Headset and Hands Free specs. Oct 27 19:36:07 Copyright (C) Cambridge Silicon Radio Ltd. 2005-2009 Part of Audio-Adaptor-SDK 2009.R1 Oct 27 19:36:18 I noticed that, yeah Oct 27 19:36:44 whoever that might be Oct 27 19:37:01 Cambridge Silicon Radio Ltd Oct 27 19:37:37 CSR? Oct 27 19:37:44 seems it's rather a specific or proprietary cmd of them Oct 27 19:38:01 sure, CSR = Cambridge Silicon Radio Ltd Oct 27 19:38:43 other silicon bakers might have adopted the syntax Oct 27 19:39:05 like AT at large once been Hayes Oct 27 19:39:59 bottom line: a highly intriguing though negligible problem Oct 27 19:40:58 ergo exactly the stuff hackers spend 95% of their time when implementing a project, while the main heavy lifting is done in the remaining 5% of time, during several nightshifts Oct 27 19:41:08 ;-) Oct 27 19:44:59 http://www-personal.edfac.usyd.edu.au/staff/souters/Humour/Real.Programmer.Stories.html Oct 27 19:46:29 >> If there is not enough schedule pressure on the Real Programmer, he tends to make things more challenging by working on some small but interesting part of the problem for the first nine weeks, then finishing the rest in the last week, in two or three 50-hour marathons. << Oct 27 19:47:51 >>This not only inpresses the hell out of his manager, who was despairing of ever getting the project done on time, but creates a convenient excuse for not doing the documentation.<< Oct 27 19:49:54 Im sure that will result in good code! Oct 27 19:53:01 indeed, *real* code that nobody ever will be able to understand. that's how the "real programmer" likes his code Oct 27 19:53:22 50 hour marathons. ive been there... Oct 27 19:54:07 * DocScrutinizer05 too Oct 27 19:54:16 when I was younf and beautiful Oct 27 19:54:44 and I needed the money ;-P Oct 27 19:55:08 you're still beautiful docky! Oct 27 19:55:17 ;-P Oct 27 19:55:25 my sunglasses are, not me Oct 27 19:56:24 actually the only comment regarding my avatar so far been: "NOO!? For real???? I thought 'what a badass'!" Oct 27 19:58:00 DocScrutinizer05: thankgod the days where "real" programmers ruled the world are over Oct 27 19:58:21