Friday, 2011-08-05

*** Xofrats has quit IRC00:27
*** arrrghhh has joined #xdandroid01:20
arrrghhhI'm deeply bothered that the DIAM is not considered a Legacy / Low Activity device.01:20
arrrghhhon xda-devs01:20
*** arrrghhh has quit IRC01:30
*** emwe has joined #xdandroid01:56
hycheh02:08
hycOK, so it looks like nobody else is gonna write a new smd_tty patch huh02:09
hycI'll take a look at bringing jonpry's original patch in, and see if that helps the smd_tty buffer mismatch errors02:09
hychttp://highlandsun.com/hyc/FRX06+/zImage-0133602:32
hycand modules-2.6.27.46-01336-g019636e.tar.gz02:32
hycwould appreciate some DIAM/RAPH user testing it02:32
emwehyc: hah, nobody has time and at least me not the expertise ;)03:03
emwehyc: did you mod that specific smd for the msm7500 or what have you done?03:04
*** programmer8922 has quit IRC03:06
hycI just reverted back to jonpry's original version of the smd_tty patch03:12
hychttps://gitorious.org/~hyc/linux-on-qualcomm-s-msm/hycs-linux-msm/commit/019636e4db74df96edd97870e284da90312790aa03:14
hycI'm not sure it will make any difference03:14
hycjust figured it's worth a try03:14
hycprobably the smart thing to do would've been to add the actual values to the error message so we can see what the mismatch amounts to03:16
hycdoh03:16
emweok, and as i see in logs the RAPH800 has SMD7?03:18
hycyeah03:18
emweshit03:18
emwethat means i need to pull htc_hw over to .3503:18
hycwell, looking at the smd_tty code, it doesn't have any device-specific init.03:18
hycso every model gets smd703:18
emwehm, we could reverse that reroute hack which is in place in smd_tty_open03:19
emwei think we could have left it as is i think then ;)03:20
emweopen smd1 and reroute for topa, raph and some others.03:20
emwerhod03:20
emwedon't have the code at hand03:20
hyceh, whatever03:20
emweso if i am missing htc_hw it defaults to CDMA?03:22
hycI guess so03:22
emwehmpfs03:23
emwehas the smd_tty patch been tested yet on the diam500/raph800?03:26
emwehyc: and what i was talking about in smd_tty_open: https://gitorious.org/linux-on-qualcomm-s-msm/hycs-linux-msm/blobs/019636e4db74df96edd97870e284da90312790aa/arch/arm/mach-msm/smd_tty.c#line13903:27
hycthe current autobuild kernels are all dying on raph800 with smd_tty buffer mismatch03:27
emweyes i know they die. just trying to figure if your reversal patch can be pushed to autobuild03:28
hycI just wrote it just now03:28
hycso no, nobody has tested it03:28
emweah heh. then let's wait some hours. ;)03:28
emwesorry03:28
*** rpierce99 has quit IRC03:29
hycand if it solves the problem, I don't really know why03:29
hycsince most of the code is functionally the same03:29
hycunless it's a big difference between work and delayed_work callbacks03:30
hycheck, I'm going to extend that error message while we're waiting03:31
emweif their's a diff between their usages then it sounds like buffers aren't filled up fast enough? ah heck. i am too ignorant in all that smd hacks :)03:31
hycok, rebuilt kernel with better error message03:33
hycsame zImage-01336, but now http://highlandsun.com/hyc/FRX06+/modules-2.6.27.46-01336-g019636e-dirty.tar.gz03:34
emwei will post up your links to arrrghhh for public testing03:38
emwehyc: now tell me... did color-led notifications from userland always work? like receiving a mail, red led, charging red, on full batt, switch to green led?03:43
hycnot that I know of03:43
emwelol, ok, that works fine here haha03:43
emweit must be the fixed brightness tables then ;)03:44
hyccool03:44
emweor just a correct microp-led impl on .3503:44
hycbut that also makes no sense :P03:44
hycok, that makes more sense03:44
emwethey should not retrieve any led brightness for color leds from the tables.. but who knows... really :)03:44
hycyeah, not me :P03:48
*** GPFerror has joined #xdandroid04:03
emwexdandroid later tell Xofrats can you please get back to me regarding the iowait stuff you observed? thanks.04:08
xdandroidemwe: The operation succeeded.04:08
hychave you tried running with higher sdcc clock?04:09
emwe.35 has it to 50MHz but the only clock we have around in clock-wince is 40MHz04:10
emweit doesn#t solve my class10 issues04:10
hycok04:10
hycthat's the only answer that matters ;)04:10
emwei really have high iowaits and system get#s unbearably slow04:10
emweon class6 rhod, though04:11
emwei reduced read ahead back to 256kb and it makes it better04:11
emwei enabled vm block_dumps and the max it reads is 256 sectors at once04:11
emwedunno if that can be tuned to read more at once.04:11
emweor is there some loop-mount options perhaps for tuning loop reads?04:12
hycdunno04:12
hycand probably reading more than 256 sectors at once would just continue to slow things down04:12
emwebut that is just 128kb if i am right, am i?04:13
hycyes04:13
hycthe point is that readahead is only valuable when you've got a lot of sequential reads04:14
emweunsure how dalvik+jit has an impact here...04:14
hycand that really only applies when loading code04:14
emweyeah, loading apps is sick04:14
emwetages ages and cpu spikes up for ages04:14
emwes/tages/takes04:14
emweneed someone handing me over powertop or iotop for android04:15
hycI'm guessing the lowmemory killer settings also affect that04:15
hyccpu spike is probably more due to paging out04:15
hycdunno04:15
emwelowmemkiller being to aggressive?04:15
hycno, not aggressive enough04:15
hyckeeping too many apps loaded04:15
emwecpu spikes come from mmcqd, loop0, loop104:16
hycinstead of leaving RAM to be used by cache04:16
hyceh ok04:16
emwewe have vfs_cache_pressure to 200 allowing kernel do drop some cached dirents and inodes earlier04:16
emwedefault is 100 which is said to be "fair".04:16
*** kalemas has joined #xdandroid04:17
emwehyc: btw, coincidence has hit. rhod died tonight. apart from it dying during the day as well while driving around. it still reacted to plugs/unplugs (charging led toggling fine) just couldn't get panel to wake. might be an msm_ts driver thingy actually.04:20
hycthat would make more sense04:21
emwestrange thing is though, one can't adb into the device04:21
emwewhich sounds like... userland is fscked up04:21
hycyeah04:21
emwekernel is up fine it seems.04:22
emwei need a permalogger thingy it seems.04:23
emweyou once told me the stuff you did/do. do you have that written down somewhere?04:23
emwegscripts something?04:23
hycyeah04:23
hycbut I didn't have a good logger for kernel msgs04:24
hycwas thinking about adding that to the logcat cmd04:24
emweredirect dmesg to logcat? yeah. i had that idea once too.04:25
emwecool thing would be you have kernel and logcat in one file with the high probability that they are in time.04:25
hycwell, my patched dmesg fixes the timestamps so they match logcat04:26
emweeven with kernel being in UTC?04:27
hycyep04:27
hychere's my gscript cd /sdcard; mv logg.txt logg.0.txt; mv logr.txt logr.0.txt04:27
hycnohup logcat -v time > logg.txt & nohup logcat -v time -b radio > logr.txt & nohup cat /proc/kmsg > klog.txt &04:27
emwecould this also be stuffed in <build>.user.conf?04:29
hycmebbe. I seem to recall someone tried that, and starting scbs would interfere for some reason04:29
emwegoing to give it a try, thanks.04:29
*** kalemas has left #xdandroid05:11
smiley-oops  im in queue buying the new xperia05:15
emwetststs05:28
*** GlemSom has joined #xdandroid05:54
*** lotjuh has joined #xdandroid06:06
*** hardwalker has quit IRC06:18
smiley-the kovsky celebrate with rebooting twice06:24
lotjuhcan anyone tell me how I can fix the problem that my htc doesn't want to run the haret.exe file cause it doesn't have the right certificates?07:53
ryannathansuse a new winmo ROM07:58
ryannathansredownload haret or the package, maybe corrupt.07:59
lotjuhhm07:59
*** rpierce99 has joined #xdandroid08:03
emwehm, i think there where some nocert patches for winmo roms08:26
lotjuhah, I'll look for that08:26
emwesome .cab to install if your rom is not certificate-less08:26
lotjuhcause I'm pretty sure my files are not corrupted08:26
emwenocert.cab or similar. don't recall exactl.y08:26
lotjuhkk, thx (-:08:26
emwenp08:26
lotjuhhm, only found 1 nocert.cab file, and my phone can't install it08:54
emwelook harder. sorry, can't help right now.09:20
lotjuhhehe, np, I'm still looking (-:09:22
lotjuhtrying to find a different windows 6.1 rom09:22
emwelotjuh: See this post for the nocert cab file (http://forum.xda-developers.com/showpost.php?p=1286970&postcount=157)09:59
emwedunno if you tried this.09:59
emwefound this text in a xda forum09:59
emwegtg. bye.09:59
*** emwe has quit IRC09:59
*** ORi| has joined #xdandroid10:08
*** ORi| is now known as ORi|ON10:09
*** |Jeroen| has joined #xdandroid10:29
SpecRwell, FRX07 is working correctly in raph800 with the library copied and the conf file modified11:40
rpierce99data too?11:41
SpecRwhat do you mean by data?11:43
rpierce993g11:43
rpierce99evdo11:43
rpierce991x11:43
SpecRi use the raph only as a wi-fi device11:43
SpecRhere we have no cdma network11:43
rpierce99oh, yeah arrrghhh tested that11:44
*** Detule has quit IRC11:57
*** sigma6 has quit IRC11:58
*** D3tul3 has joined #xdandroid12:13
*** arrrghhh has joined #xdandroid12:16
arrrghhhSpecR: like i said before, if you don't use service then all the fantastic benefits of the new RIL are moot for you.12:16
arrrghhhwould be nice to have someone with a RAPH800 that actually has service confirm the rest of the crap from the new RIL12:19
arrrghhhMMS, netloc, data...12:19
*** D3tul3 is now known as Detule12:24
smiley-mm  new xperia is grrreat12:32
arrrghhhmeh12:33
arrrghhhhyc: i haz bootani on my RAPH80012:34
arrrghhhi never ran a kernel with the smd_tty patch that broke things, so i'd assume this is a good sign :P12:34
arrrghhhseems to be workin great12:36
arrrghhhi guess i'll boot a few more times just to make sure...12:36
*** touchpro has joined #xdandroid12:44
touchprohyc, initial test on raph800 that zimage and modules lets the system boot up, but something is hanging the system up to a point where its barely usable, thanks for looking into this for us cdma punks12:45
arrrghhhtouchpro: works great on my RAPH80012:45
arrrghhhyou also have the new RIL?12:45
touchproi havnt shoehorned that in yet, in previous tests i had to let the data.img be created without having the raph plugged in (for adb) so im just restarting now and im gonna copy the new ril over12:47
arrrghhhplease do12:47
arrrghhhunless you're on FRX06 with this new kernel12:47
arrrghhhif you're running the RIL from FRX07, then that explains the slowness.12:47
touchproi dont have service at the moment to test out data but i should in a few days, and im on frx07 right now12:48
arrrghhhdamnit jim12:49
arrrghhhnew RIL12:49
arrrghhhhave you seen mah post?12:49
touchproarrrghhh do you still have problems with adb?12:49
arrrghhhhyc: are you planning on committing this, or should i add your test kernel to my post?12:50
arrrghhhtouchpro: yea, i never get any notification on the phone or on the PC.  it charges via USB, but there's no other indication that USB is actually connected.12:50
touchproare you plugged in to computer when you boot frx? thats the only way i have ever gotten adb to work on raph80012:50
arrrghhhah12:53
arrrghhhi asked about that12:53
arrrghhhsmiley- made me sound like i'm crazy and it 'just works' on his crapintosh.12:54
touchpronah its been a known bug for a while, i think we may be the only device that still has to do that12:54
*** Ivo has joined #xdandroid12:55
arrrghhhheh12:55
arrrghhhkovsky still has some oddities like that12:55
arrrghhhalrighty, that would explain my lack of ADB12:55
*** fishhead2567 has quit IRC12:55
arrrghhhi do remember that now, you have to boot with activesync enabled specifically, right?12:55
arrrghhhlike you can't boot in usb mass storage mode...12:55
touchproyeah lol im logcatting right now as im booting frx07 now, i dont know about the activesync i dont even have it installed on my lappy, i just plug up and boot12:56
arrrghhhhum12:56
arrrghhhok12:56
touchprois bootani slow for you?12:57
arrrghhheh12:59
arrrghhhi would say "no"12:59
arrrghhhmainly because it's always been a little choppy12:59
arrrghhhwith the RIL that shipped with FRX07, it was painfully slow12:59
arrrghhhand would outright freeze at times12:59
touchprousually on mine its choppy for a few seconds, then corrects itself, but it doesnt seem to, its probably the ril12:59
touchprodamn adb cut out halfway through the boot process, something killed it so im gonna have to bindmount the ril13:00
*** emwe has joined #xdandroid13:03
arrrghhhdidn't you do that..?13:04
arrrghhhFFFFFFFFFFUUUUUUUUUUUUU Phone.apk13:04
arrrghhhugh13:04
emwemeep13:04
arrrghhhemwe: really getting tired of missing calls :/13:04
emwe.27?13:05
arrrghhhyes13:05
emwepm.[idle_]sleep_mode=1 i assume?13:05
arrrghhhpretty sure it happened on .35/.39, but i haven't run those for more than a day or two at a time.  no extended testing.13:05
arrrghhhsleep_mode=1 yes.  i don't have any idle_sleep_mode setting in my startup..13:05
emweok, just making sure.13:06
emwei get so less phone calls i can hardly judge i miss any ;)13:06
emwei only know it takes ages bringing up the phone app13:06
emwebut i am no got verfication candiate as .35 is on my phones.13:06
arrrghhhhum13:06
arrrghhhyea13:06
arrrghhhtaking ages when there's an incoming call is killing me.13:07
arrrghhhi thought i read something about G1 guys placing Phone.apk permanently in RAM13:07
arrrghhhsounds dangerous, but if it allows me to answer calls i'm all for it.  what do you think?13:07
emwenah, that's not gonna work out.13:07
emwewe only have that refresh thread in rootfs13:07
emweit's all only half baked solutions iirc13:07
arrrghhhdamnit13:08
arrrghhhshoehorn a snapdragon into mah RHOD?  :D13:08
emweeven cyano has done some stuff, but in the last ressort, nothing can keep it in ram i think if the system wants it away13:08
arrrghhhmight as well see if i can cram some more RAM in while i'm at it.13:08
emweoh yeah, i am getting pissed by the slowness ;)13:08
arrrghhhhum13:08
arrrghhhi really do like this hardware platform.13:08
arrrghhhcurious if NAND would assist in this at all.  i have to think it would help *some813:09
arrrghhher13:09
arrrghhh*some*13:09
emweyeah, the sd is burried with io. if that wouldn't be, i think we can get a nicer feel for everything.13:09
arrrghhhindeed13:10
arrrghhhfull NAND ahead :P13:10
arrrghhhjust gotta convince hyc it's a good idea.13:10
arrrghhhperhaps some roofies will convince him...13:10
Detuleimo a partitioned build is a nice intermediary between looped and nand13:10
touchprovogue runs great on nand lmao im gonna run away now13:10
Detulei've experienced less i/o mucking13:10
arrrghhhDetule: i noticed performance was WAY worse on dedicated partitions13:10
arrrghhhperhaps my SD card is on its way out... i don't know.13:10
emwearrrghhh: btw, i got userland led notifications up. that is neato. charge led and mail noti and so on.13:11
arrrghhhdon't tell the users taht13:11
emweplays nicely with sleep indicator led as well13:11
arrrghhh:P13:12
arrrghhhorly?13:12
emweit's just a matter of implementing correctly in .27 i think13:12
emwesomething's missing there13:12
Detuleemwe: you have this somewhere up on gitorious?:)13:12
Detuleoh not finished on the kernel side, got it13:12
emweDetule: it's just a matter of actually setting the leds in the hooks13:12
emwethat#s my assumption. really dunno why that was never tracked13:12
emwei just had some stubbed color led hook and just implemented it for rhod/topa on .3513:13
emwewhich is a copy of .27 micropklt code that is.13:13
emwenow i am gonna see if i need htc_hw... damn raph800 :P13:14
emweaccording to the re-established radio mode detection ril code i have to...13:14
arrrghhhhaha13:15
emweand if that doesn't cost me a shitload of time, debug h2w headset malfunction13:15
emwewe need machine variant detection from htc_hw for 35mm and rhodw panel stuff anyway13:15
arrrghhhyea13:16
emwebtw, there's still one rhod400 gsm user who complains still to loose signal over night13:16
emwehe always reports on the autobuild comments.13:16
emwei can't complain at all13:16
emweDetule: you've been around yesterday where Xofrats talked about some IO findings?13:17
emweeek, does this sound less-perf? Running ENG build: no pre-dexopt!13:18
Detulei saw the post on xda13:18
emweDetule: do you by chance have a link handy? otherwise which forum?13:18
arrrghhhhttp://forum.xda-developers.com/showthread.php?t=120363813:19
arrrghhhemwe: ^^13:19
Detuleemwe: so it's these calls micropklt_set_led_states that set the led to amber and green -> i would love to change those13:19
Detulearrrghhh beat me to it13:19
Detulealso the one about changing sdcc manually13:20
emweDetule: micropklt_set_color_led_state13:21
emweDetule: that one is currently only called on suspend resume... on .35 i have it green on suspend, and wake nothing. which i really do favour. i don't wan't any blinking or amber during sleep because that would drive my gf crazy when that thing os on over night :)13:23
touchprogrrr whats the best way to edit system.ext2 so i can just put this ril in manually without bindmounts?13:23
*** |Jeroen| has quit IRC13:23
arrrghhhtouchpro: loop mount on ubuntu box?13:23
arrrghhhbindmounts are fine..13:23
arrrghhhi should say loop mount on linux box ;)13:24
emweDetule: ahem... as i just see, there's no green/amber led sysfs devices publised?!13:24
touchproarrrghhh when you bindmounted the ril did it correctly detect cdma phone? mine still says gsm phone detected on haret boot13:24
arrrghhhi didn't look at dmesg13:24
emweDetule: can you do a adb shell ls -la /sys/class/leds on .27 please?13:24
arrrghhhbut i could make a call & i had service...13:24
touchprok did you see the cdma menus instead of gsm in system settings etc?13:25
emwemake sure you are bind mounting to /system/lib/ since frx0713:25
touchproemwe ill have to check, thanks for the advice13:25
emweor easier, remount /system rw and adb push to /system/lib13:25
emweway easier than this bindmounting foobar13:26
emwe08-05 19:18:38.860 D/RIL     ( 1306): Using CDMA Phone13:26
emwe^^ i hate you13:26
touchprolol i wanted to do the remount and adb push but adb dies for me halfway through boot for some reason, never did before13:26
arrrghhhtouchpro: definitely in CDMA mode13:27
Detuleemwe: would .39 do13:27
arrrghhhi have system select, "change the cdma roaming mode"13:27
emweDetule: pardon?13:27
arrrghhhobviously i can't swtich to GSM, this isn't a worldphone13:27
touchprok, emwe's advice about mounting to /system/lib was a correct assumption, i was following the original ril post mounting to /lib13:28
arrrghhhtouchpro: that's NOT what my post said.13:28
arrrghhhmy post directs you to bindmount to /system/lib...13:28
emwetouchpro: rild is setup to look in /system/lib since frx0713:28
touchprooh sorry i hadnt seen your post arrrghhh13:28
arrrghhh...13:28
emweDetule: you meant if .39 has led notifications? if so then no, because the code got copied from .27 verbatim more or less13:28
emwearrrghhh: adb shell ls -la /sys/class/leds if you can13:29
Detulepstebin.com/JqXAPYXP13:29
emweoh ;)13:29
Detuleer13:30
emweDetule: haha... no amber/red or green led devices publised13:30
emweDetule: if these are present liblights uses them and voila, led notifications work13:30
Detulewell we need those devices then13:31
emweDetule: https://gitorious.org/linux-on-qualcomm-s-msm/linux-msm/blobs/08e25bc0cc6389587af19c0b4f636f36cf1581c1/arch/arm/mach-msm/board-htctopaz-led.c (currently used for topa and rhod, dirty hack, needs refactor)13:31
emweDetule: are you gonna look into it then? ;)13:31
Detulenot if there are any expectations attached:)13:32
emwelol13:32
emwebtw, there's a "stupid" copy of the function we talked earlier. micropklt_speaker_led() is basically the same code just also sets back-led for some rhods. should be made on function having 2 args, color for front and back13:33
emwejust didn't do that yet for .3513:33
emwearrrghhh: that back led is not on our rhod400, no? (magically hidden? never seen it)13:34
arrrghhhit works in winmo13:35
arrrghhhand wistilt2 made it turn on when doing voice command, etc13:35
emweuh?! we have that too?13:35
emwerhod400?13:35
arrrghhhyes13:35
arrrghhhon the mute button13:35
emweis it shining through the button?13:35
arrrghhhi've only seen it green or red13:35
emweaha13:35
arrrghhhbut i think it can do blue too13:35
arrrghhhyes13:35
emwehmhmhm13:35
emweman, i wanted htc_hw and h2w not led...13:35
emweok, convinced. i wanna see that.13:36
arrrghhhcam13:37
arrrghhhcan't seem to get adb working on this PC13:37
arrrghhhoh well13:37
Detuleemwe: thanks for the pointers, first thing i want to change the sleep/wake led pattern on my rhod, and i gather i can do that with that set_color_led_state function13:37
emweDetule: yap, exactly. you can also look at the little bit cleaner code i linked for you.13:38
Detulethanks again13:39
emweDetule: microp_led_set_color_led() is your friend. might get some love later on ;)13:39
emweDetule: your welcome. i am always glad to point and help if i am able to.13:39
arrrghhhtraining about over, cya guys later.13:41
*** arrrghhh has quit IRC13:41
*** Ivo has quit IRC13:54
*** Ivo has joined #xdandroid13:54
*** touchpro has quit IRC14:30
*** rpierce99 has quit IRC14:32
*** rpierce99 has joined #xdandroid14:33
hyci don't have commit access to the kernel tree, i don't think. and i still don't know what the functional difference is between the two patches.14:34
hycwould like to understand that....14:34
emwelol, i think i found teh color led testmode14:35
emwefront led can do yellow as well14:35
rpierce99yellow + red = orange?14:36
emweyeah, it's a dirty yellow. call it orange, yes14:36
rpierce99oh, my orange is orange14:37
emweback can do red orange and gree14:37
emwe(if that thing is really the testmode... looks like)14:37
*** fishhead2567 has joined #xdandroid14:49
hychmmm. the only difference between work and delayed_work is being able to specify a time delay14:53
hycand jonpry's patch uses a time delay of zero14:53
hycduh14:53
hycok, need to clean up that patch some more14:53
*** arrrghhh has joined #xdandroid15:11
lotjuhemwe: I got it to work to upgrade to windows 6.5 first15:11
emwelotjuh: ok.15:11
emwearrrghhh: you sure wis said he got blue also for back?15:11
emwearrrghhh: i got quite some combinations now, but not blue.15:11
arrrghhhemwe, it wasn't wis who said that15:12
emweah sorry15:12
arrrghhhi was thinking there was more than just green/red15:12
arrrghhhi've definitely seen green/red15:12
arrrghhhi thought there was some other color tho...15:12
emwewe got orange as well ;)15:12
arrrghhhah15:12
arrrghhhpretty easy to do, just like the front LED15:12
emweand amber+green flash 2s15:12
arrrghhhit can do orange, but i've never seen it in WinMo?15:12
emweback can do green red and orange15:12
arrrghhhi guess it doesn't matter what winmo can do15:13
arrrghhhjust what the hw is capable of ;)15:13
hycone more test kernel for you15:13
arrrghhhmoi?15:13
hycsame filename for zImage15:13
hycyes15:13
arrrghhhaw you shouldn't have.15:13
emwehyc: does your first reversal patch skip that smd_readx() or is that still in place?15:13
hycplus modules-2.6.27.46-01336-g85e1331.tar.gz15:13
hycemwe: that's gone15:14
emweok15:14
hycand this current patch deletes it15:14
hychttps://gitorious.org/~hyc/linux-on-qualcomm-s-msm/hycs-linux-msm/commit/85e1331f97f1a886e351b6a784f24ff527dad6f115:14
hycif it runs fine for arrrghhh then I think you should merge this commit15:14
emwehyc: i am ready more or less ;)15:15
hyccool15:16
arrrghhhi'll try to get adb mojo flowin too15:18
arrrghhhw00t, adb15:20
arrrghhhemwe, you just wanted to see the beginning of the boot process, yes?15:21
arrrghhhhttp://pastebin.com/W28D4ukE15:21
emwearrrghhh: just the SMD setup :)15:21
arrrghhhtell me if that's sufficient15:21
arrrghhhhyc, so far so good.  anything specific i should do/try?15:21
hycnope. if it boots up and there's no smd_tty buffer mismatch errors, it's good.15:26
hyctesting data would be best but we obviously have to leave that to someone else15:26
arrrghhhheh15:26
hychm, tho you could also run it on your rhod15:26
arrrghhhyea, someone mentioned that data didn't work on their DIAM50015:26
rpierce99yeah fixing the raph and breaking the rhod would be no bueno15:27
hyclol15:27
hycpretty sure it will be fine, jonpry must've tested this originally on his rhod15:27
hycbut yeah, nice to have confirmation15:27
arrrghhhthat's all he haz15:27
arrrghhhis RHOD15:27
arrrghhhalright.  i need to get this new RIL on my RHOD15:28
hycwell, give it a try on your rhod too15:28
arrrghhhso i'll just get it all on thar15:28
hycok15:28
arrrghhhsure, 1 sec15:28
arrrghhhemwe, looks like SMD setup is in that paste, let me know if you need moar15:28
emwearrrghhh: nah, that was enough. keeping local copies of these outputs. that's why i asked :)15:29
arrrghhhhaahaha15:29
arrrghhhnice15:29
arrrghhhRAPH800 boot was a gap eh?  :P15:29
emwediam500 and raph800 both have the MSM7500 that's why diam500 should suffer equally well15:29
arrrghhhindeed15:30
Detuleemwe: yikes microp-ng15:37
Detuleyou could have warned me15:37
hycoh. I suppose I still need to update the ril to use RSINFO for the baseband version15:38
emweDetule: why? doesn't matter which "backend" you use. -ng for me just looks like handier registration, nothing more, nothing less. anything other technical is up to alex to explain ;)15:39
emweit's just the write call which is different15:39
hycarrrghhh: did you ever get radiooptions 11 to show you AT@v ?15:40
Detulek, i guess i shouldn't try to hard figuring out how it differs maybe i'll just bring it all together in my tree15:40
arrrghhhperhaps i'm doing it wrong15:40
arrrghhhhyc, is it radiooptions 11 <enter> then AT@v15:40
arrrghhhor all one line...15:40
arrrghhh?15:40
hycall one line15:40
hycradiooptions 11 "AT@v"15:41
arrrghhhk, that's how i was passing it15:41
arrrghhhand it didn't work...15:41
arrrghhhhrm i think i might've used a single quote15:41
hycthen it sounds like you don't have the patched rild15:41
arrrghhhlet me try again15:41
hycor libril.so actually15:41
rpierce99on mine it didn't error as long as i was root, but i didn't see a response either15:41
arrrghhhi bind mounted 'em15:41
hycweird15:41
arrrghhhand i checked the md5sum's15:41
arrrghhhthey matched...15:41
hycand no response in the logcat at all?15:42
arrrghhhi saw some errors that i thought might've been related15:43
arrrghhhbut i didn't see the command itself in the log, if that's expected.15:44
arrrghhhi can answer a call with radiooptions?15:46
arrrghhhmy problem is solved, i just always have to be adb'd into my phone lol15:46
arrrghhhhyc, it worked15:47
arrrghhhi'm using adb this time, maybe it makes a difference.15:47
hyccool15:47
arrrghhhthe answer it returned isn't the same as what's in rsinfo tho15:47
hycwell, you need to be root obviously15:47
hycno, it should be different15:47
arrrghhhoh ok15:47
arrrghhhand i was root before...15:47
hycok15:47
arrrghhhit threw an error that it couldn't open it before15:47
hycjust gimme the output15:47
arrrghhhmeh15:47
arrrghhhlol sorry15:48
arrrghhh08-05 13:47:01.520 D/AT      ( 1365): AT< 2.00.0015:48
arrrghhhgood, or do you want the whole log?15:48
hycI think that's fine15:49
arrrghhhAT+CLAC returned... nada it seems15:50
hycjust wanted to make sure the rest looks sane15:50
arrrghhhAT< 4 lol15:50
hycok15:50
arrrghhhhere i'll pastebin the whole thing15:50
*** GlemSom has quit IRC15:50
hycis rsinfo in your pastebin too?15:50
hycI guess I have an older one here somewhere15:50
arrrghhhhave at it15:50
arrrghhhyou can see the CLAC and @v at the bottom15:51
hyccool, got it15:52
arrrghhhexcellent15:52
arrrghhhuhhh15:53
arrrghhhso far RHOD is not happy15:53
hycbummer15:56
hycby the way new ril2-test.zip15:56
hycthis should get the right version info15:56
hycso it should show up in settins/about phone15:56
arrrghhhfuck15:56
arrrghhhi'm retarded15:56
arrrghhhbooting again...15:57
hycwhat happened? wrong modules?15:57
arrrghhhnagh15:57
arrrghhhi forgot to rename the rootfs... d'oh.15:57
arrrghhhi was doing many things all at once, i know that's the best way to do it.15:57
arrrghhh:D15:57
hyclol15:57
arrrghhhso... what should show up in settings/about phone?15:58
arrrghhhyou didn't fix baseband yet15:58
arrrghhhoh15:58
arrrghhhyou did15:59
arrrghhhand you're telling me to download a new test RIL15:59
hycyes15:59
arrrghhhi think i need to go back to bed and try today again.15:59
hycheh.15:59
arrrghhhRHOD seems fine.16:00
hycok, good16:00
hycemwe: I think that patch is ready to go16:00
arrrghhhlol, sleep LED freaked me out then i remembered this if from your local tree :P16:01
hycheh yeah, probably better not to hand this one out if people are gonna freak.16:01
hycjust let it get into the autobuild and forget about it16:02
arrrghhhmeh, the fix is really for RAPH/DIAM folk16:02
arrrghhhwhich have a completely different sleep indicator16:02
hycok16:02
arrrghhhbut yea, since this patch seems good then commit & forget.16:02
arrrghhhhrm16:03
arrrghhhwhat's interesting is ADB works just as expected assuming i boot with USB in16:03
arrrghhhso the kernel just isn't initing it on its own16:03
emwenaaaaaaah right in this moment my mother in law calls with a computer issue16:08
arrrghhhhaha16:09
arrrghhhteamviewer16:09
emweexactly. just need to install here on openSUSE16:09
emwe.27 is just building.16:10
hycI believe, tho I have no proof, that in winmo the arm9 clock is faster, so MPEG4 video decode is faster16:18
arrrghhhhum16:18
arrrghhhOC ARM9?  :P16:18
hycbut my attempts to muck with the clocks never got any results16:18
arrrghhhbummer16:19
arrrghhhi know that alex figured out how to OC the gfx proc on kovsky, BLAC and I think RAPH/DIAM as well16:19
hycI think we should get this thread closed/deleted http://forum.xda-developers.com/showthread.php?t=120457616:25
hycsince he says his work is based on tiad816:25
hycmost likely that means there's no source code for anything he "fixed" ...16:25
emwearrrghhh: what's been teh results with that wistilt occassional incall audio fix?16:26
arrrghhhhaven't seen anything overwhelmingly positive16:28
arrrghhhbut it doesn't seem to break anything either16:28
arrrghhhand wistilt2 said it fixed all 15 of his RHOD300's.16:28
hyccouldn't hurt, sure16:29
emweasking because i got that also in the commits. didn't notice even ;)16:29
emweif it doesn't hurt but fixes 15 rhod300 then fine.16:30
emwedone.16:31
arrrghhhk16:32
emweso unfortunately now those poor guys need a fixed up ril, too, right?16:32
arrrghhhyea16:33
arrrghhhi made a post on that16:33
arrrghhhfor FRX08 i guess...16:33
hycyou tested the latest ril2-test already? got baseband version reported now?16:34
arrrghhhyup16:35
arrrghhh1.12.34f_2.00.0016:35
hycok cool16:35
arrrghhhthanks for the reminder, i got distracted there :P16:35
hycthen anything else that's still unsupported can just rot16:36
hycpushed16:37
lotjuhhm, wireless on rhod100 with frx06 should work right? (-:16:37
emwelotjuh: why the hell are you using an old release?16:39
hycok, post updated with new ril2 binary16:39
lotjuhI tried frx07, but it was hanging in the boot16:39
lotjuhit kept looping16:39
arrrghhhlotjuh, where16:39
arrrghhhmany RHOD100 users are on 0716:39
lotjuhsec, I'll start it up again16:40
emwearrrghhh: i was just about to note the ts-calibration file might be missing, but i will stop trying to take your role. you can do that better.16:40
lotjuhthe kernel packed with it just rebooted my phone during the boot16:40
arrrghhhemwe, lol16:41
arrrghhhlotjuh, did you try updating?16:41
arrrghhhand how far in the boot process did you get?16:41
lotjuhso I got the latest bzimage and initrd16:41
lotjuhand that one was looping16:41
arrrghhhwhere16:41
lotjuhI'm checking (-;16:41
arrrghhhok16:42
lotjuhit's rebooting16:42
arrrghhhsure16:42
xdandroidNew kernel available (20110805_223228) at http://zimages.googlecode.com/files/htc-msm-linux-20110805_223228-package.tar.bz216:42
arrrghhhemwe, thanks for the push, i shall update my thread ;)16:42
hyccool, that should settle everything then16:43
lotjuhI get a lot of Ignored duplicate definition messages16:44
emwearrrghhh: what you could do is put a link to the xda-forum post for the updated/required ril in the autobuild comment section in TIAD8'ish capitalization16:44
arrrghhhemwe, meh16:44
lotjuhin /init.rc it seems16:44
hycignore those16:45
arrrghhhlotjuh, what do you mean you get messages in /init.rc...16:45
arrrghhhlotjuh, do you ever see the boot animation?16:45
hycin /init.rc there are duplicate service definitions16:45
hycor /init.android thinks there are16:45
hyclike I said, ignore it16:45
lotjuhwell it keeps looping in it16:45
lotjuhI get the same dupplicate messges over and over again16:46
lotjuhand I see snd_ioct comming by every now and then as well16:46
lotjuhit's the same message block I keep seeing16:46
lotjuhso it's not going any further16:46
hycthen possibly a corrupted binary16:47
arrrghhhlotjuh, redownload the full bundle, format the card with the hp tool, try again from step 1.16:47
lotjuhI'll try copying it again, I use the same for the frx06 and that one does work16:47
arrrghhhuse the same what16:47
lotjuhkernel / initrd16:47
arrrghhhok, that shouldn't matter.16:48
hycsure, same kernel / initrd should work regardless16:48
lotjuhyeah, that's what I thought16:48
lotjuhbut my wireless doesn't work on them16:48
lotjuhanyway, I'll recopy it16:48
hycbut rootfs / system / data ... nope16:48
lotjuhmaybe it did get corrupted16:48
arrrghhhwireless...16:48
arrrghhhi thought you couldn't even boot -_-16:48
lotjuhI can't boot in 0716:48
lotjuhI can on 0616:48
*** vahid2222 has joined #xdandroid16:55
arrrghhhhi vahid222216:56
arrrghhhyou can talk in the channel, it's ok :P16:56
lotjuhhehe16:56
vahid2222lol16:56
arrrghhhso no data eh?16:56
arrrghhhyou said you're on Sprint?16:56
vahid2222no, nothing16:56
arrrghhhdid you check the APNs?16:56
vahid2222no, i'm on bell16:56
arrrghhhoh right16:57
arrrghhhyou're that bell guy16:57
vahid2222yep, not working16:57
vahid2222apparently i am16:57
arrrghhhso leaving the default values for MNC/MCC or whatever it still disappears?16:57
vahid2222no, if i leave them at default, the apn could be saved16:57
vahid2222but no data16:57
arrrghhhMCC?MNC16:57
vahid2222both of them16:58
arrrghhhok, and the APN is "selected"?16:58
arrrghhhby that i mean the radio button is green16:58
vahid2222yes,16:58
arrrghhhnot gray16:58
vahid2222selected16:58
arrrghhhhum16:58
vahid2222it is green16:58
arrrghhhwell i guess logs are going to be necessary and hopefully hyc can take a peek16:58
arrrghhhunless anyone can think of another reason why data wouldn't work...?16:58
arrrghhhi'm thinkin logs will be good16:58
vahid2222how do i get a log?16:58
arrrghhhthere's this fantastic app16:59
arrrghhhgetlogs16:59
arrrghhhyou can install it in Android if you have a file explorer app installed16:59
arrrghhhor if you're in winmo, move the apk to AndroidApps16:59
arrrghhhthe apk is in AndroidApps\Other FYI16:59
vahid2222what's apk16:59
arrrghhhit's like a cab for Android17:00
arrrghhhall app's have an APK to install them17:00
vahid2222k, lemme see what i can do17:00
arrrghhhsure17:00
lotjuhfrx07 with the original kernel etc loops at the same spot17:01
arrrghhhlotjuh, so you never see the boot animation?17:01
lotjuhnope17:01
arrrghhhdo you ever see A N D R O I D in the terminal?17:01
lotjuhyeah, after that shows the dupplicate messages right?17:02
arrrghhhor is it before that point...17:02
arrrghhhi'm not sure tbh17:02
arrrghhhobviously you have a ts-calibration file17:02
arrrghhhuhm what else17:02
lotjuhyeah, it asked me to touch the 5 points so it created the file17:02
arrrghhhyou'd get a rootfs error if you had everything in a folder and a bad/missing rel_path setting17:03
arrrghhhyou were able to do that on FRX07?17:03
vahid2222arrrghhh, i don't have data, how do i get this getlogs on my phone, can't find it online17:03
lotjuhidd, had the wrong rel_path earlier, so that's fixed now17:03
lotjuhyeah I was on FRX0717:03
arrrghhhvahid2222, uhm.. the apk is in AndroidApps\Other17:03
lotjuhbut it asks for the points while you're still in the console17:03
arrrghhhlotjuh, yea but it usually knocks them off the page before you can hit all of 'em17:03
arrrghhhthe kernel vomit messes it up17:04
lotjuhah, well I was able to get past that point at least (-:17:04
arrrghhhhence why i suggest using an old kernel in the steps17:04
arrrghhhhrm17:04
vahid2222got it17:04
hycdoes that still happen? I removed a bunch of those idiot kernel messages17:04
arrrghhhlotjuh, do you have adb?17:04
hycit should be perfectly fine now17:04
arrrghhhhyc, on autobuild?17:04
hycyep17:04
lotjuhI don't know what it is, so I don't think so17:04
arrrghhhorly17:04
arrrghhhdamnit i don't test these things frequently :P17:05
hycwell apparently it was fine for lotjuh17:05
arrrghhhapparently17:05
arrrghhh:P17:05
lotjuh(-=17:05
arrrghhhit's been a while since ive tested that17:05
lotjuhI could try the kernel packed with FRX0617:05
arrrghhhhyc, when did you do that17:05
arrrghhhlotjuh, that's old17:05
lotjuhI know17:05
lotjuhbut the current one doesn't seem to want to boot d-:17:05
arrrghhhso... have you tried any other kernels on FRX07?17:06
arrrghhhi'm honestly not sure what would bungle it up17:06
arrrghhhdo you have adb?17:06
lotjuhalthough it could be the initrd though17:06
lotjuhsince it seems to fail there17:06
arrrghhhhasn't changed since april17:06
lotjuhhm, true17:06
lotjuhand no, as I said I don't know what adb is (_=17:06
arrrghhhalthough FRX06 predates it17:06
lotjuhso I don't think I have it (-;17:06
lotjuhah!17:06
hycarrrghhh: commit 57d2faf May 3017:06
lotjuhso it could still be the initrd17:06
arrrghhhhyc, damn.17:07
lotjuhshall I try the FRX06 one then?17:07
arrrghhhlotjuh, you should be on FRX07-everything17:07
hycinitrd is rock solid17:07
arrrghhhforget FRX0617:07
arrrghhhdelete it17:07
lotjuhwell yeah, but as long as we don't know what exactly is failing...17:07
arrrghhhFRX07 has a full bundle just like 06 did17:07
lotjuhI downloaded the full FRX07 bundle17:07
arrrghhhalthough there's been a bunch of fixes since17:07
arrrghhhok17:07
arrrghhhand you didn't mix/match anything just ran it?17:08
lotjuhidd17:08
arrrghhhcrap17:08
arrrghhhwell you tried latest rootfs/kernel and still fail?17:08
lotjuhI even used the hp tool to format my disk etc17:08
lotjuhidd17:08
lotjuhhm17:08
lotjuhnot the latest rootfs for 0717:08
lotjuhonly the latest kernel17:08
arrrghhhok17:08
arrrghhhgrab the latest rootfs17:08
arrrghhhand there's another new kernel17:09
hycyeah, need latest rootfs17:09
arrrghhhas of a few minutes ago17:09
lotjuhkk17:09
arrrghhhshouldn't matter much for your phone17:09
arrrghhhbut still17:09
arrrghhhgrab it :P17:09
lotjuhI'll see if this laptop understands the phone as well (linux)17:10
lotjuhmy bf took mine with windows on it to play a game -.-17:10
arrrghhhlinux is better for most things17:12
arrrghhh'cept well, gaming.17:12
lotjuhah, got it mounted (-:17:12
lotjuhhehe, I know (-;17:12
lotjuhhence why I want it on my phone as well d-:17:13
arrrghhhhaha17:13
lotjuhlatest rootfs is from 27-07?17:15
arrrghhhyup17:15
lotjuhkk17:15
lotjuhcamping wireless is slow (-=17:17
arrrghhhcamping wireless..??17:17
lotjuhyeah, I'm on the camping atm17:17
lotjuhin our tent17:17
*** dadaR6 has joined #xdandroid17:17
arrrghhhwhen i go camping17:17
arrrghhhno laptops are ever allowed17:18
lotjuhgoing to CCC next week17:18
lotjuhhaha17:18
lotjuhI can't go camping without my laptop (-;17:18
arrrghhhthen you're not camping!17:18
arrrghhhi usually go places where normal cars can't get to17:18
arrrghhhgotta go 'wheelin17:18
dadaR6hello17:18
arrrghhhhowdy17:18
dadaR6I have the issue that most people have about service drop after receiving a call17:19
arrrghhhtried new RIL?17:19
dadaR6are you able to reproduce it or you need some of my logs?17:19
dadaR6yes, i just did17:19
arrrghhhi can't reproduce it, no.17:19
arrrghhhso with the new RIL, it still happens?17:19
arrrghhhand you verified with logs that you're on the new RIL...?17:19
dadaR6[HTC Vogue Community RIL hyc e787acd 2011-08-05 13:36:38 -0700]17:20
arrrghhhk17:20
arrrghhhcan you provide hyc with full logs...?17:20
arrrghhhofc after the issue has occurred17:20
arrrghhhso recreate the problem, pull logs, submit pastebin of said logs :D17:20
dadaR6yes, i just need to clean up some personal info i guess17:20
arrrghhhnp17:20
arrrghhhgetlogs does most of that for you17:21
arrrghhhif not all of it... depending on what you care about17:21
dadaR6yes, i just used it, but with files instead of pastebin17:21
arrrghhhdoes the same thing17:21
arrrghhhthis is so odd...17:21
arrrghhhyou're VZ?17:22
arrrghhhor what carrier17:22
dadaR6sprint17:22
arrrghhhorly17:22
arrrghhhdang...17:22
arrrghhhi definitely do not have that problem.  "old" RIL or not.17:22
dadaR6:-(17:22
dadaR6can i PM you the logs on xda?17:23
arrrghhh..sure?17:23
arrrghhhi'm going to give them to hyc17:23
dadaR6ok, give me 1min17:23
arrrghhhassuming that's OK at least17:23
arrrghhhcuz i can look, but he fixes :P17:23
lotjuhgah, same thing17:24
emwearrrghhh: force_cdma can be dropped and shouldn't be used anymore, right?17:24
arrrghhhcorrect17:24
emwethanks17:24
arrrghhhit's ignored... if you think that's causing an issue?17:24
arrrghhhalthough RAPH folk never used it17:24
emwenono, just pulling over .27 htc_hw17:24
arrrghhhnever needed to, they're not... ok17:24
arrrghhhsorry17:24
emweand dropping a shitload of stuff from the file17:24
Detuleemwe: compiling 39 with that spiffy led code17:24
arrrghhhmy brain just freakin runs when i try to troubleshoot this crap sometimes17:24
lotjuhright after the A N D R O I D message it start doing stuff with init and then it says something with "untracked pid exited" and it starts again with the init stuff and dupplicates17:25
arrrghhhDetule, nice.  is it configurable, or is it just how it is...?17:25
emweDetule: i do still have to do some "debuggin" in the sense of trial and error to figure other led combinations. need to get orange somehow. dunno the values, yet :)17:25
arrrghhhlotjuh, that's... bad.  so you're on linux?  is adb installed on that rig?17:25
arrrghhhemwe, wistilt2 had all that crap sorted out...17:26
arrrghhhi don't know if the logs will cut it17:26
arrrghhhcrap i was supposed to find that section of the logs too17:26
emweirclogs?17:26
arrrghhhhe gave me some test kernels17:26
arrrghhhthat basically ran thru all the possible combinations17:26
lotjuharrrghhh: nope, but I could fix that I guess17:27
Detulearrrghhh: dont know probably need to sit down and explore....i just hope whatever zImage comes out actually works17:27
arrrghhhlol17:27
Detuleemwe: that board-led code does look nicer than whatever is in microp-klt17:27
Detuleand by nicer i mean i can halfway understand what's happening17:28
emweDetule: yes, but i am already facing some issues. that code currently deals only with one microp client. but for button backlight, rhod needs the "ksc" one.17:28
vahid2222arrgghh, I installed getlogs, I can gather the logs and email them, is that all?17:28
emweDetule: and... rhodium audio microp code has another client to klt. i somehow need to get both clients in there. haven't looked deeper yet.17:28
arrrghhhvahid2222, there's a pastebin button that will upload 'em directly to pastebin17:29
arrrghhhor to a file on your sd card17:29
vahid2222do i neet to change the apn or anything?17:29
vahid2222before i gather the logs17:29
arrrghhhno17:29
Detuleok, you're thinking ahead, that's good....i'll just be happy if i can gain a measure of control over teh stupid led17:29
arrrghhhit should've tried to connect17:29
arrrghhhemwe, <WisTilt2> hey, you know where the blue led might be controlled? it is stuck on from whatever lights change i did and can't find it to save my life!17:30
arrrghhhlol17:30
arrrghhhthere's a blue LED somewhere17:30
Detulei think rpierce99 had the best suggestion in the forums, something like light ON only if ARM is working AND screen off17:30
Detules/light/LED/17:31
Detulek i gotta jet later fellas thanks again emwe17:32
emweDetule: have fun17:32
emwenp17:32
emwearrrghhh: lol, ok. looking for blue somehow. haha...17:32
emwearrrghhh: but there's no extra leds on some rhod variants, no? ;)17:32
*** touchpro has joined #xdandroid17:32
arrrghhhi... don't think so17:33
arrrghhhbut i can't say i know for sure17:33
arrrghhhthese bastards and their RHOD100's.  front cam's and what not.17:33
touchprohey arrrghhh can you package up the bundle your using and send it my way please? im having trouble getting things to work now, keep getting a rootfs not found issue17:33
arrrghhhtouchpro, uh17:33
arrrghhhthat's a simple fix17:33
touchpro?17:33
arrrghhhyou have a rootfs.img, right?17:33
arrrghhhnamed just that ^^17:33
touchproyep17:33
arrrghhhdo you have Android in a folder17:34
touchprochecked it 5 times17:34
arrrghhhharet.exe17:34
touchpronope on root17:34
arrrghhhrootfs.img17:34
touchproyes17:34
touchproyes17:34
arrrghhhit's all a the root?17:34
arrrghhhok17:34
touchproyes17:34
arrrghhhin startup.txt, do you have a rel_path...?17:34
touchpronope17:34
arrrghhhyou're sure.17:34
touchpropositive17:34
arrrghhh...17:34
arrrghhhthose are literally the only reasons17:34
arrrghhhyou don't have file extensions hidden or something17:34
arrrghhhand have rootfs.img.img17:34
arrrghhhsomeone managed to pull that the other day17:35
lotjuhhehe17:35
arrrghhhmade me want to stab myself in the face17:35
touchprolol no rootfs.img.img17:35
lotjuhfirst thing I always do in windows, turn off the hidden extensions option (-=17:35
arrrghhhheh17:35
arrrghhhtouchpro, then your rootfs is messed up...?17:36
arrrghhhyou didn't rename the zip did you?  lol17:36
touchprono, but im pretty sure damn vogue linux put a damn folder named andboot on my sdcard, so ill delete that and try again17:36
arrrghhhseriously tho, those are the only reasons for that error.  some issue with the rootfs, wrong location of stuff, rel_path set wrong, something along those lines.17:37
arrrghhhuhm17:37
arrrghhhAndroid puts an 'android' folder17:37
touchprorel_path isnt set in my startup.txt so im not sure that would be an issue though17:37
dadaR6arrrghhh: here are my logs: http://pastebin.com/XdY4y9fW17:37
touchproandboot not android17:37
arrrghhhtouchpro, XDAndroid does not create any such folder.17:37
dadaR6I leave them on for a bit, but will remove them17:37
touchproyes i know, vogue droid does, and i cant find my other memory card to test the raph with so im using the one out of my vogue17:38
arrrghhhdadaR6, you said service drops to 0?  x on the bars?17:38
dadaR6I see a X in the signal bar17:39
touchprolmao wow how random, the andboot folder created by my vogue stops it from booting, must be a remnant somewhere left over in the initrd (my guess)17:39
dadaR6and I cannot make any calls17:40
arrrghhhweird.. i thought that was signal strength -> AT< $HTC_CSQ: 6,417:40
dadaR6Searching for service in the lock screen17:40
arrrghhhdadaR6, did you try making a call?17:40
arrrghhhin this log17:40
dadaR6yes17:40
arrrghhhk thx17:40
arrrghhhhyc, you see this?17:40
dadaR6i get error saying no service available17:40
dadaR6I am in good service area, I usually get full bars17:40
lotjuhinstall sdk platform-tools17:41
dadaR6Service just came back now (dropped when i first talked to you)17:41
dadaR6so it eventually come back, ahaha17:41
lotjuhinstalling*17:41
lotjuhshould have adb then17:41
dadaR6without doing anything17:41
arrrghhhdadaR6, can you pull another set to compare?17:42
dadaR6radio only?17:42
arrrghhhuhm17:42
arrrghhhi usually pull all of them17:42
dadaR6k17:42
arrrghhhnot sure if hyc needs all of 'em17:42
touchproif [ "$CARD_PATH" = "" ];then CARD_PATH="andboot" fi (straight from init in initrd.gz) i dont know how to patch but can we change that? it probably only affects me17:43
arrrghhhlogcat & radio would probably be good17:43
arrrghhhnot sure if dmesg is needed or not...17:43
dadaR6let me get them all17:43
arrrghhhtouchpro, if you're using andboot, then use andboot17:43
lotjuhmust be an andboot initrd.gz then17:43
touchproandboot on vogue is used to hold the androidinstall.tar and androidupdate, on our builds it isnt used unless you partition (just a guess)17:44
arrrghhhno it's used17:45
touchprodo you know for what?17:45
arrrghhhuhm17:45
arrrghhhit basically says if you don't have a rel_path17:45
arrrghhhbut andboot exists17:45
arrrghhhthen rel_path is andboot...17:45
lotjuharrrghhh: so now I have adb, what now? (-=17:46
touchprooh, andboot, androot, tomato tomahto lmao17:46
arrrghhhlotjuh, logssss17:46
arrrghhhlotjuh, i'd guess it's something that dmesg or logcat will expose.  doubt it's a radio issue.17:46
lotjuharrrghhh: how does it work? I got the adb on the laptop17:47
arrrghhhis the device showing up?17:47
arrrghhhadb devices17:47
dadaR6arrrghhh: second logs after service recovers: http://pastebin.com/58vsEcGj17:48
lotjuhwell, the device is currently running windows since I can't boot the FRX07 d-:17:48
arrrghhhFRX06 works you said?17:49
arrrghhhdadaR6, thx17:49
lotjuhyeah, I'll start that then, wasn't sure if that would work17:49
arrrghhheh?17:49
arrrghhhwait17:50
arrrghhhlotjuh, i'm just curious if FRX06 works17:50
lotjuhit works17:50
arrrghhhok17:50
arrrghhhhave you updated FRX06 at all?17:50
dadaR6arrrghhh: is that ok to remove those pastebin now that you got them? I dont like to leave things around (I will forget about it)17:50
arrrghhhyea that's fine17:51
lotjuhyes, new kernel and new rootfs for FRX0617:51
lotjuhold initrd though17:51
arrrghhhlotjuh, so newest stuff boots on FRX06?17:51
lotjuhyeah17:52
arrrghhhwth...17:52
lotjuhthat's what I said from the start (-=17:52
lotjuhonly difference between FRX06 and FRX07 is the initrd I think17:52
lotjuhwant me to see if it will show up in adb while it's looping during the startup of 07?17:54
lotjuhI'm not sure what exactly is needed for adb to recognize my device17:55
arrrghhhwell there's plenty of differences in the system image17:57
lotjuhgah, this windows 6.5 rom keeps enabling my data connection while I don't want it to17:57
lotjuharrrghhh: yeah, that's true17:57
arrrghhhso what i want you to do is make sure this works - FRX06 with the newest everything17:57
arrrghhhinitrd, rootfs, & kernel.17:57
lotjuhok, 07 has the latest initrd right? (-:17:58
*** dadaR6 has quit IRC17:59
arrrghhhyes18:01
*** vahid2222 has quit IRC18:02
*** dadaR6 has joined #xdandroid18:03
lotjuhis there a seperate page where I can get the latest rootfs for 06 or should that already be the latest one in the zip file?18:03
*** touchpro has quit IRC18:03
dadaR6arrrghhh: about the lost service18:03
dadaR6it seems this is only happening when missing a call18:04
dadaR6if i pick the call, or reject it, it is all fine18:04
dadaR6only when i missed it18:04
dadaR6(person hangs up before going to voicemail or i pick it up)18:05
arrrghhhdadaR6, odd... thx18:07
lotjuharrrghhh: 06 boots with the 07 initrd and latest kernel18:07
lotjuhjust not sure if I got the latest bootfs for it18:07
lotjuhrootfs*18:07
arrrghhhhrm18:08
arrrghhhwhy do you do these things to me...18:08
lotjuhhaha, yeah sorry18:09
lotjuhI wish it would just work as well (-;18:09
arrrghhhthen use 06 i guess?18:10
lotjuh06 doesn't work completely tough18:10
lotjuhwhen I attach it to my laptop and choose USB connect it keeps searching18:10
arrrghhhthat'll never work on looped mounts18:10
lotjuhand my wireless doesn't want to start18:10
arrrghhhso... not fixed in 0718:11
lotjuhah ok18:11
arrrghhh06 & 07 should be same for wifi18:11
arrrghhhas far as core functionality goes18:11
lotjuhwonder why my wifi says error then18:11
arrrghhhthere was some stuff changed, so it should work better18:11
lotjuhI know it used to work18:11
arrrghhhhave you read the FAQ?18:11
lotjuhon 04 orso18:11
arrrghhhit did used to work18:11
arrrghhhhrm18:11
arrrghhhthen you probably don't have to do the wifi-nvram trick18:11
arrrghhhit's not wep?18:11
lotjuhit doesn't even want to start the wireless device (-:18:12
arrrghhhOo18:12
arrrghhhif you swapped kernels18:12
arrrghhhdid you also swap modules?18:12
arrrghhhthose go hand-in-hand.. they must match18:12
dadaR6arrrghhh: a ssimple guess would be that the phone thinks that it lost signal because the call dropped, and signal a service lost18:12
lotjuhI got a new modules file18:12
lotjuhdidn't download it myself though18:12
dadaR6when in fact, the service was not lost18:12
lotjuhit created it itself18:12
arrrghhhlotjuh, that's not possble.18:13
arrrghhhpossible*18:13
lotjuhit said on the wiki that when you download a new kernel you should remove the old one because of the modules18:13
lotjuhbut it doesn't say you have to get new modules yourself18:14
arrrghhhwow18:14
lotjuhand as I said, I got 2 mocules files now18:14
arrrghhhyou really sure that's what it says?18:14
lotjuh modules*18:14
lotjuhsec (-:18:14
lotjuhEdit the startup.txt to change the line set KERNEL zImage to set KERNEL zImage_VERSION, OR18:15
lotjuh Delete the original zImage file and rename the new zImage_VERSION file to zImage. Delete the original modules-VERSION.tar.gz file to avoid any confusion as to which modules go to which kernel. The kernel must match the modules, or several advanced features will not work - the main one being wifi.18:15
arrrghhhdelete the original modules18:15
lotjuhI did the first18:15
lotjuhit said OR (_=18:15
arrrghhhuhm18:15
lotjuhanyway, where did that new file come from then o.O18:16
arrrghhhthat OR is either in the wrong place18:16
arrrghhhor18:16
arrrghhhi don't know18:16
lotjuhhehe18:16
arrrghhhhrm18:16
arrrghhhthat is kinda confusing now that i'm re-reading that... shit.18:16
lotjuhand there's no download link for modules on that Incremental Updates page either18:16
arrrghhhthe zimage and modules come together18:17
arrrghhhin that .tar.gz file18:17
arrrghhhand the way that is worded is poor, i apologize18:17
lotjuhahh, well I guess that's where that file came from then (-;18:17
arrrghhhgonna check if someone changed it or if i'm that retarded18:17
lotjuhI will remove the old one to make sure it uses the new modules file18:18
lotjuhhehe18:18
arrrghhhhrm18:19
arrrghhhy'know... i guess that would work18:19
arrrghhhyou don't have to delete 'em if you edit the startup.txt18:19
arrrghhhi don't think i wrote that part either, someone much smarter than me did probably.18:20
arrrghhhcuz i never mess w/the startup.txt's reference to the kernel.  i always just leave it zImage heh18:20
*** Xofrats has joined #xdandroid18:21
XofratsI is here18:21
arrrghhhah crap18:21
lotjuhI'm just gonna remove the old ones completely and rename it to zImage as well (-:18:21
arrrghhhthere goes the neighborhood18:21
arrrghhhlotjuh, heh ok.  just make sure startup.txt matches ;)18:21
XofratsNah, the neighborhood was already gone by the time you got here18:21
arrrghhh...zing?18:22
Xofratsbtw the wince clock for msmsdcc is id=2118:22
XofratsSo I assume setting fmax does increase the clock18:22
emweXofrats: it looks for the next lower18:23
emweXofrats: btw, it's "mweirauch" here :)18:23
XofratsDoes it? How can I test it?18:23
*** dadaR6 has quit IRC18:23
emweXofrats: just set clock_wince.debug_mask=15 as i wrote and watch in dmesg18:23
emweyou'll see what it actually sets.18:23
* Xofrats found that class4 nonHS card and the 16GB card output weird18:24
emwethe code is in clock-wince.c if you wan't to look at it.18:24
XofratsOK, I'll do it once I get ntop compiled18:24
hycI didn't see dadaR6's pastes before he deleted 'em18:24
hycarrrghhh: you have copies of them?18:25
arrrghhhhyc, yes.  pm comin18:25
XofratsThe only other thing of interest is in core.c18:25
emwethat is?18:29
Xofratssomething about wait_for_completion or something18:31
XofratsWhich is where I sometime see oops18:32
*** dadaR6 has joined #xdandroid18:33
XofratsBut I suspect that's from the sdio side18:33
lotjuhok, I could turn Wi-Fi on now, then it started scanning, but it turned itself off, and when I tried to turn it back on it says 'Error'18:34
*** ringer has joined #xdandroid18:35
arrrghhhlotjuh, this is FRX06?18:36
lotjuhyeah18:36
arrrghhhcan't support that18:36
lotjuhwith the kernel that was just pushed18:36
stinebdfrx06 blows18:36
arrrghhhlol18:36
lotjuhwell at least it boots d-:18:36
lotjuhcan't say that from frx07 yet (-=18:36
arrrghhhstinebd, for some reason she can't boot FRX07.  trying to sort that out (i thought)18:36
stinebdare you aware your smilies are upside-down?18:36
lotjuhyeah (-;18:37
arrrghhhLOL18:37
arrrghhhcocaine's one hell of a drug18:37
XofratsMy smiles are right side up ^^)18:37
stinebd;)18:37
lotjuhthat's what you get when you learn 'chatting' in irc when there was no msn yet d-:18:37
stinebdwe have a lot of people with trouble booting frx07 for some reason18:37
arrrghhhstinebd, a lot?18:37
XofratsI predate icq and msn too18:37
stinebdwell not a lot by your standards18:37
stinebdbut by mine18:38
arrrghhhi don't know of a specific issue on RHOD18:38
arrrghhhsome devices need an updated kernel18:38
emwenon-keyboarders ^^18:38
arrrghhhand the CDMA RAPH/DIAM issue is pretty much licked, thanks to hyc18:38
XofratsMy frx07 had trouble until I removed all the arrrghhh stuck in it18:38
stinebdto save me from backlogging, what's the model in question here?18:38
stinebd400?18:38
lotjuhRHOD10018:38
stinebdooh18:39
stinebda strange one perhaps18:39
lotjuhdutch18:39
arrrghhhyup18:39
arrrghhhbut i don't know of any RHOD100 users that have trouble booting18:39
stinebdwhat does it do on frx07? ever see any scrolling text?18:39
lotjuhI guess I'm gonna start the process all over again tomorrow, format the sd, just get FRX07 on it and see if it helps18:39
lotjuhyeah, it gets past the android message18:39
arrrghhhyea18:39
XofratsThere was a patch on the msm aosp that changed wait_for_completion to completion_io18:39
arrrghhhdoesn't hit bootani tho18:39
stinebdsweet, i'd love the logcat18:40
XofratsBut that just seems cosmetic18:40
lotjuhand then it starts looping in the duplicate messages18:40
stinebdah18:40
arrrghhhstinebd, i tried to get her up on adb18:40
arrrghhhhaven't finished that yet18:40
arrrghhhand i'm not sure how well that will work on 07 ..18:40
stinebdlogcat works fine without su18:40
stinebdanyway if you need it i can whip up a current rootfs with secure off18:41
stinebdonly 5 more consecutive days of work and then i might be useful again18:41
XofratsCan't you setprop via local.prop?18:41
stinebdnot that18:41
XofratsWihout baking a new rootfs18:41
XofratsThat or froyo.user.conf18:42
stinebdunderstandably android is quite picky about ro.secure18:42
stinebdonce it's set, that's it for the session18:42
XofratsPpl underestimate how powerful *.user.conf is18:43
stinebdi'm not people18:43
stinebdif you wanna go around remounting and sed'ing in user.conf that's fine, but i don't trust those things18:43
XofratsI forgot, you're an android ^^)'18:44
XofratsActually, during userinit everything is rw18:44
XofratsAs I found out18:44
stinebdarrrghhh: you've got mail18:44
XofratsMan ntop compiles slow...18:45
stinebdand it's funny18:45
arrrghhhk18:45
stinebdnot ha-ha funny18:45
stinebdmore like stab-stab funny18:45
lotjuhlol18:45
stinebddo a search for "like i found in the build" and you'll see what i mean18:46
arrrghhhoh yea18:47
arrrghhhthat dude was in htc-linux18:47
stinebdk18:47
stinebdmaybe i'll reply18:47
stinebdi feel especially stabby18:47
hycarrrghhh: new ril2-test.zip should fix the BSINFO problem18:48
arrrghhhok18:48
arrrghhhi... can't reproduce this isse18:48
arrrghhhissue*18:48
stinebdthen consider it fixed18:48
hycyeah, need to feed it to dadaR6 or someone else with it18:48
arrrghhhmind if i download it, host it and pass it out like candy?18:48
arrrghhhyea18:48
hycgo ahead18:48
arrrghhhthx18:48
dadaR6hyc: I m here18:48
dadaR6anything i can try?18:49
arrrghhhstinebd, i couldn't reproduce the issue in the first place18:49
arrrghhhdadaR6, yes, new RIL :P18:49
stinebdthen it's fixed18:49
dadaR6where to get it?18:49
XofratsIs riltest for raph/diam onry?18:49
arrrghhhXofrats, no18:49
hycI'll let arrrghhh post it for dadaR618:49
stinebdrong time18:49
stinebdarrrghhh will end up putting it on some file sharing site18:50
arrrghhhdropbox18:50
Xofratsrilwarez18:50
stinebdmore like virusbox18:50
arrrghhhlol18:50
arrrghhhstinebd, multiupload18:50
arrrghhhyour fave18:50
Xofrats0day ril18:50
stinebdyes, the porn banners are a nice touch18:50
arrrghhhall the respectable sites have porn banners18:51
XofratsNot if you have privoxy18:51
XofratsWhich I am half tempted to run on my rhod18:51
arrrghhhdadaR6, http://dl.dropbox.com/u/12691845/ril2-test.zip18:51
arrrghhhor http://db.tt/bDHkp8918:52
arrrghhhy'know, whatever18:52
arrrghhhthe second one has a higher chance of giving you cancer tho18:52
stinebdfirst one has a higher chance of giving you aids18:52
Xofratsstd or terminal illness18:52
arrrghhhchoose wisely18:52
dadaR6arrrghhh, thanks, rebooting with it18:53
arrrghhhnp18:53
XofratsDamn thing needs to compile already...18:54
XofratsAt least one good thing about this compile: Phone hasn't segfaulted even at 691mhz18:58
stinebdarrrghhh: question19:00
dadaR6hyc, looking good so far19:00
dadaR6I was not able to reproduce it anymore19:00
hycok19:00
hycplease send new log19:00
hycthere's still something else funky in the last logs you sent19:00
dadaR6I see the data goes to 3g, 1x, triangle(?), then back19:01
stinebduntil a certain unknown and inconsistent point, my atrix will stop waking up very often for processing and will become very battery efficient. why? and how come it isn't always very battery efficient?19:01
Xofratstriangle is roaming19:01
stinebdnot sure if i should blame motorola or nvidia19:01
dadaR6ah, i never saw that one before :-P19:01
arrrghhhstinebd, as i learned in my training class today, blame God.19:02
XofratsNo, blame yourself, or God19:02
stinebdi prefer to blame things that cannot shove lightning bolts up my bum19:02
arrrghhhlol19:02
arrrghhhseriously, the instructor kept saying "we'll let God take care of it.." etc19:03
dadaR6hyc, getting the logs for you19:03
arrrghhhmeaning, let the system take care of it automagically.  dumb.19:03
XofratsOK, system is running out of memory, brb19:03
*** Xofrats has quit IRC19:04
stinebdheh19:05
stinebdpoor guy19:05
stinebdi assume19:05
stinebdhard to keep track of genders on irc19:05
dadaR6hyc, http://pastebin.com/6XLNedUq19:08
hycok got it19:08
dadaR6up for 10 mins19:09
dadaR6i forgot to boot with air plane mode, I hope it did not screw up anything...19:10
hyclooks fine19:11
hycyour log shows that the modem still screwed up the same as before, and that the patch handled it as expected.19:11
hycso that's all good19:11
dadaR6ok, nice19:12
dadaR6also, I dont know if WinMo are tweaking how they display their signal strength19:12
dadaR6but usually I get 'weaker' signal reported on XDAndroid19:12
dadaR6I know it does not mean anything, just want to mention it...19:12
arrrghhhi think  my winmo ROM has more bars19:13
stinebdi know of airports with many bars19:13
hycyeah, at this point it doesn't mean much19:14
stinebdis the ril reporting snr correctly at least?19:14
stinebdor whatever that thing is that means stuff19:15
dadaR6yep, it does not matter as long as we can keep service :-)19:15
hycthe ril reports whatever the modem reports19:15
hychere I think the modem reports strength 0-719:15
hycthe ril has a hardcoded lookup table to turn that into dBm19:15
stinebdhah19:16
hycthe table values were defined by me, using a dartboard19:16
hycso, whatever dude19:16
stinebd<3 htc radios19:16
stinebd0-7 is pretty freaking random19:16
arrrghhhhahahah19:17
dadaR6easy then, just report a +1 value ahaha19:17
dadaR6that's what Apple did with their iPhone, isnt it? :-P19:17
dadaR6it makes user feel better19:17
stinebdi was able to get their leaked signal algorithm19:17
stinebdif (radioSignalPercent < 99) radioSignalPercent = 99;19:18
dadaR6sounds about right :-)19:18
stinebdplus tax19:19
hycanyway, it's more complicated than that. the modem reports 2 values - CDMA strength and EVDO strength19:19
stinebdwhat about gsm?19:19
hycon dadaR6's last log, the CDMA strength is always higher than EVDO19:19
hycyeah, GSM/WCDMA have two values to19:20
stinebdmakes sense19:20
hycanyway, Android will report the one you're currently using. so if it says you're on EVDO19:20
hyc(and current report is CDMA 6 / EVDO 3)19:20
hycyou will see a low number of bars19:20
hyceven though the CDMA signal is very strong19:20
dadaR6yes, this is the same as WinMo19:21
dadaR6i usually get 4 bars or so in EDVO19:21
dadaR6then i make a call, and jump to 7-8 bars19:21
arrrghhhyea, i've noticed higher signal strength when in 1x19:22
dadaR6which is what you describe19:22
arrrghhhthen it drops when in ev19:22
dadaR6yep19:22
*** programmer8922 has joined #xdandroid19:22
stinebdis that attributable to the higher error tolerance in that modulation?19:22
stinebdbuzzwords19:22
hycno idea19:23
dadaR6but i still think they cheat on their EDVO reported signal19:23
hycit's always possible19:23
hycand really, like I said, I came up with these table values without any guide whatsoever19:24
hycso the fact that the EVDO bars are anywhere close is just random chance19:24
hycmebbe I'll start using rand() in the next revision19:24
stinebdsounds good19:28
stinebda nice gingerbread feature19:28
hycheh.19:29
hycnew ril2-test.zip. one more try at smoothing out the call status notifications19:29
*** Xofrats has joined #xdandroid19:29
hychttp://highlandsun.com/hyc/ril2-test.zip19:29
hycdadaR6: feel free to try this one too19:29
XofratsMeh, so why would clock-wince lie about the freq19:29
hycto help us nail folks like you who fall prey to placebo effect19:30
*** programmer8922 has quit IRC19:30
*** ringer has quit IRC19:31
dadaR6hyc, ok, trying...19:31
emweXofrats: it does not lie. it sets to the best possible it knows. and those clock register setups have been REd and no better are know. ok, on .35 there is a 40MHz one. try that on .27 and see if it improves it for you.19:33
XofratsWell, the cmdline referenced msmsdcc_fmax19:34
Xofratsemwe: and how would I do that?19:35
emweXofrats: look at .35 clock-wince, find the difference in the table, copy it to .27, compile, go. ;)19:35
XofratsYeah, turtle and goliath problem... git wants me to pull 700MB before I can compile19:36
emweXofrats: btw, i was wrong... its 48MHz clock in there...19:36
emweMSM_CLOCK_REG(48000000,   1, 0x04, 0x02, 0, 2, 4, 1, 245760000), /* UART_HS, 48MHz */19:36
emwehmhmhm now i whonder what is taken.19:36
emwewas it you who compiles on his phone?19:38
XofratsYes19:38
emweget a computer.19:38
Xofrats[    6.442871] clock-wince: set_mdns_host_clock: 21, freq=32000000 calc_freq=32000000 pll1=960000000 expected pll =76800000019:38
Xofrats[   99.149230] clock-wince: pc_clk_set_rate: id=19 rate=4800000019:38
Xofrats[   99.149261] clock-wince: set mdns: 19, 48000000; bitidx=0, offset=a419:38
Xofrats[   99.149353] clock-wince: set_mdns_host_clock: 19, freq=32000000 calc_freq=32000000 pll1=960000000 expected pll =76800000019:38
XofratsAny particular reason why msmsdcc is capped at 32mhz on .27?19:39
XofratsAnd if the freq is set, how is it supposed to do variable clocking that it supposedly does?19:40
emwe[    3.270690] clock-wince: a11_clk_set_rate: id=21 rate=4915200019:40
emwe[    3.270721] clock-wince: set_mdns_host_clock(21, 49152000); bitidx=0, offset=ac19:40
emwe[    3.270751] clock-wince: a11_clk_is_enabled not implemented for clock: id=2119:40
emwe[    3.270782] clock-wince: set_mdns_host_clock: 21, freq=48000000 calc_freq=61440000 pll4=0 expected pll =24576000019:40
arrrghhhrboot brb19:40
emweno idea why. because it was enough for low end cards it seems19:40
emweit always selects the next lower clock because these reverse engineered clock tables aren't complete.19:41
emweso its 32MHz on .27 that's it. alex added the 48MHz one.19:42
emwei think i should leave now. really late.19:42
emwegnight.19:42
*** arrrghhh has quit IRC19:45
*** emwe has quit IRC19:47
*** Xofrats has quit IRC19:51
hychmmmm. when you have data active, and a voice call comes in, and then the voice call hangs up, the ril also hangs up the data session.19:51
hycstupid POS19:51
hycI hate finding out that I created a bug...19:52
rpierce99well i for one hate you for that, good lord hyc can you do anything right? lol19:56
rpierce99i guess since maybe you fixed one or two other things we might be able to forgive you for this one, but don't let it happen again19:56
hycwhimmper whimper19:57
*** Xofrats has joined #xdandroid19:59
XofratsOK now gitorious's web interface is being evil...19:59
XofratsI'm in the .35 branch, on clock-wince, and when I try to raw blob, gives me .27...20:00
* Xofrats wonders if it's opera mini...20:01
*** Xofrats has quit IRC20:02
hycon my rhod210 when a call is dropped, I get an @HTCDIS message for voice calls and a +PCD message for data calls20:05
hycit doesn't look like CDMA reports these messages20:05
hyccany anyone confirm?20:05
hycI guess it's not a huge deal...20:06
rpierce99on cdma we don't drop calls20:07
dadaR6hyc, the new ril you sent works too20:07
hyc:P20:07
hycdaraR6 ok cool thanks20:07
dadaR6however, I am often stuck in roaming now...20:07
hycwell, that's a side-effect of this patch20:07
dadaR6after a call drop (hangs up before i pick up)20:07
hycthe modem returns a result that has zero information20:07
hycso until it regains its senses, it has no idea what network it's talking to20:08
dadaR6your previous RIL did not do that I think20:08
hycno, it would simply fail the request, which made the Phone.apk think you had no service20:08
dadaR6no no, not that 'old' one20:08
hycoh the one immediately before?20:09
dadaR6the one from earlier today (3:40pm?)20:09
dadaR6yes20:09
hycthat seems unlikely, since the patches were in completely different areas20:09
dadaR6i saw the roaming, but it disappear quickly20:09
dadaR6now it stays20:09
dadaR6until i place a call20:10
hyceh.20:10
hycsend new logs if you want me to investigate20:11
hycotherwise nothing20:11
rpierce99hyc have you figured out what it is about these phones that make them do this, I have no problem, obviously20:11
dadaR6ok, let me gather them20:11
hycnot a clue.20:11
rpierce99radio version?20:11
hycI just see what's in the radiologs. the modem stops providing useful info.20:11
hyccould be.20:11
rpierce99could be easier to just up/down grade everyones radio that runs into this20:12
hycwhat radio version are you running? rpierce99 and dadaR6 ?20:12
rpierce99is that baseband version?20:12
hycyep20:12
dadaR6how to get the radio version?20:13
hycSettings / About Phone20:13
rpierce9961.27W.25.23U_2.32.00WF20:13
hycactually it's in the getlogs output20:13
hycI see that dadaR6 has 61.27W.25.23U_1.96.00WF20:13
hychmmm20:14
rpierce99same phone?20:14
*** Xofrats has joined #xdandroid20:14
hycI dunno what models you guys have20:14
rpierce99rhod40020:14
dadaR6same20:14
rpierce99so yeah, maybe if dadar6 upgrades his radio the issue will go away20:14
hycthen perhaps it's all just due to that radio version20:15
XofratsOK is it me or does the .35 clock-wince not define the 48mhz for the cdma pll?20:15
rpierce99http://forum.ppcgeeks.com/cdma-tp2-development/90342-radio-collection-update-06-17-10-rhodium_w-radio-roms-mirrors-included.html20:15
rpierce99SPCS 2.32.00WF20:16
dadaR6logs:20:17
hychm, mebbe you should be running 2.32.00WU so you can use your GSM20:17
dadaR6http://pastebin.com/0TiRd8QN20:17
hycbut anyway...20:17
XofratsYeah, the 6.5 sprint rom is 2.3220:17
hycan upgrade may be a good idea20:17
dadaR6it is possible that upgrading will help20:17
XofratsYes20:17
dadaR6I am using an old version still on WinMo 6.120:18
Xofrats2.32 is pretty decent20:18
rpierce99you should just upgrade to 6.5 then while your'e at it20:18
XofratsAgreed, although 6.x is dead regardless20:18
dadaR6not really, i dont need it, no benefit for me20:18
rpierce99yeah i suppose, some people still bounce into winmo on occasion20:18
XofratsIf I were to download the msm kernel tarball, what else would I need to compile a working kernel+module, without going into git and the 700mb madness?20:19
dadaR6it's not broken, so I am not going to fix it :-)20:19
XofratsYour radio is broken though20:20
dadaR6which might broke it more20:20
XofratsShould be able to just find the radio ruu20:20
rpierce99it's linked in that post20:20
rpierce99well it's an exe, so should be even easier20:20
Xofratsit's a self-extracting zip20:21
hycok whatever. I see sometimes the info is there, sometimes it's missing in your log20:21
rpierce99oh, haha20:21
hycnothing I can do about that.20:21
dadaR6well, this is fixed as far i am concerned20:21
dadaR6i dont care much about that triangle :-P20:21
hyccool20:21
hycI certainly don't care20:22
dadaR6I still get service, so this is all i care20:22
rpierce99as long as it's not actually bouncing you over to a verizon tower for some reason20:22
dadaR6even if it is, then what?20:22
rpierce99then you pay roaming charges...20:23
dadaR6roaming is included for me as most of plans?20:23
XofratsExcessive roaming will get you booted20:23
rpierce99no idea, i have 2 sprint phones, one is a company phone, no idea what's included, the other is an old 3rd part employee plan i'm grandfathered into20:23
XofratsSupposedly more than 50% usage or 800min/month20:23
dadaR6as soon as i dial in, roaming goes away20:23
dadaR6and I am way below 800 min/month20:24
dadaR6like 10 times lower ahaha20:24
dadaR6for all my calls20:24
Xofratshyc: Do I need anything other than the tarball from gitorious to compile a working kernel+module?20:25
rpierce99we should have arrrghhh post to the thread though that for those who have this issue they should see if their radio is up to date20:25
hycyou need the compiler20:25
XofratsWell, he posted it on the raph forum20:25
hyctools20:25
hycblah blah bla20:25
Xofrats... I know that, I meant source-wise20:25
dadaR6hyc, thanks a lot20:26
hycn/p20:26
hycno, just the source20:26
rpierce99the cross compiler, right?20:26
Xofratsthis is a rhod-specific issue, rhod40020:26
XofratsNo, I compile on ze rhodium20:26
rpierce99haha oh20:26
XofratsLinux localhost 2.6.27.46-01342-gb9e00e1 #30 PREEMPT Mon Jul 18 22:34:22 CEST 2011 armv6l GNU/Linux20:26
Xofratshyc: Think you could look at the .35 clockwince and see what needs to be banged up for cdma 48mhz?20:27
hycnope20:28
hycI've had zero luck with the clock code20:28
XofratsHmmm, k20:28
XofratsI guess I need the ti.git too?20:32
hycdunno. you have ti wlan?20:32
Xofratsrhod does no?20:34
hycrhod is broadcom20:34
hycso no20:34
XofratsOK, so I'm good to go20:34
dadaR6hyc, your fixes are ok to submit btw?20:35
*** Xofrats has quit IRC20:35
hycyes, I've already pushed them to xdandroid repo20:36
dadaR6cool20:36
dadaR6byee20:37
*** dadaR6 has quit IRC20:37
hycbye20:37
*** D3tul3 has joined #xdandroid20:37
stinebdhyc: you should work on your important project20:39
hycyeah20:40
hycdon't remind me :P20:40
hyc ? have you been looking at my other projects?20:44
stinebdno20:44
stinebdyou mentioned something about it after the release20:44
hycyeah, it's been slow going20:45
hycfigure I need to get to feature complete by this month so I can start into testing next month20:45
*** ORi|ON has quit IRC21:03
*** ORi| has joined #xdandroid21:04
rpierce99hyc, don't worry about it, follow the advice of the most interesting man in the world: http://tinyurl.com/3ve9qbr21:12
*** Entropy512 has joined #xdandroid21:23
*** programmer8922 has joined #xdandroid21:26
*** programmer8922 has quit IRC21:27
*** rpierce99 has quit IRC21:30
*** programmer8922 has joined #xdandroid21:39
*** whiteE has joined #xdandroid22:15
*** [7] has quit IRC22:38
*** TheSeven has joined #xdandroid22:38

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!