| jonpry | Detule, maybe you try the SLUB allocator as well. its strange that vmalloc does what it does. i think it might be because of SLAB | 00:04 |
|---|---|---|
| jonpry | also there is strangeness in arch/arm/mach-msm/include/mach/msm_iomap-7x00.h | 00:08 |
| jonpry | .27 allocates at 0xF0000000 for peripherals, not sure about 35, but we are located at 0xE*. this probably removes 256MB of virtual space from the kernel. | 00:10 |
| Detule | jonpry, so this call is a kmalloc call, are you sure that somehow defaults to vmalloc | 00:24 |
| Detule | i mean in ppp_deflate | 00:24 |
| Detule | i just tried applying the patch, i was thinking about going w this for a few days | 00:26 |
| Detule | i guess both 27 and 35 allocate at 0xF8 | 00:34 |
| *** rpierce99 has quit IRC | 00:47 | |
| *** rpierce99 has joined #xdandroid | 00:48 | |
| *** rpierce99 has quit IRC | 00:49 | |
| *** Detule has quit IRC | 00:51 | |
| *** emwe has joined #xdandroid | 01:55 | |
| emwe | D3tul3: plz look in .35 ppp_deflate commits. i applied a patch from somewhere around the net which iirc is very (if not totally) similar to what jonpry posted. i had teh very same issues "months" ago on .35 | 01:57 |
| *** programmer8922 has quit IRC | 02:20 | |
| *** emwe has quit IRC | 02:23 | |
| *** emwe has joined #xdandroid | 02:25 | |
| hyc | wow, emwe is re-entrant | 02:26 |
| emwe | yap :) moin hyc | 02:41 |
| *** earider01 has joined #xdandroid | 04:33 | |
| *** earider01 has quit IRC | 04:43 | |
| *** ryannathans has quit IRC | 04:44 | |
| *** fgj has joined #xdandroid | 04:44 | |
| *** fgj has left #xdandroid | 04:51 | |
| *** fgj has joined #xdandroid | 04:51 | |
| *** fgj has left #xdandroid | 04:52 | |
| *** turbolapin has joined #xdandroid | 05:29 | |
| *** turbolapin is now known as lapin{GR} | 05:29 | |
| lapin{GR} | goodmorning all! | 05:30 |
| *** ryannathans has joined #xdandroid | 06:03 | |
| *** turbolapin has joined #xdandroid | 06:22 | |
| *** turbolapin has left #xdandroid | 06:22 | |
| *** lapin{GR} has quit IRC | 06:23 | |
| *** hardwalker has quit IRC | 06:23 | |
| *** lapin{GR} has joined #xdandroid | 06:24 | |
| *** xda-Squirrels has quit IRC | 06:25 | |
| *** lapin{GR} has quit IRC | 06:45 | |
| *** Detule has joined #xdandroid | 07:24 | |
| Detule | emwe: thanks, here's a stupid question how do you look at a particular file's commit history on gitorious | 07:26 |
| emwe | Detule: git log <path-to-file> | 07:38 |
| emwe | git log -p ... for the diff as well | 07:38 |
| emwe | git log <dir> works as well. | 07:38 |
| emwe | if you want to look at a log from another ref ... git log <path> <ref> where <ref> is a remote/branch e.g. | 07:39 |
| Detule | yeah i was wondering if it can be done on gitorious rather than locally....if not, they should make it their next feature :) | 07:39 |
| emwe | or was it the other way around. don't remember ;) | 07:39 |
| emwe | if you add a remote and fetch a branch or all branches from it, you can browse it's log just fine | 07:40 |
| Detule | do you recall what were the symptoms on your phone whenever deflate would blow up | 07:40 |
| emwe | symptoms was stacktraces at least. what does git log show what i had written? ;) | 07:40 |
| emwe | dunno if data (smd7) crashed in the sense it stopped working. | 07:41 |
| emwe | too long ago. was one of the first things i had to "find a fix for" when data was working. | 07:41 |
| emwe | shit, gitorious doesn't allow the history to be shown... object too large. thanks. | 07:43 |
| emwe | Detule: https://gitorious.org/linux-on-qualcomm-s-msm/linux-msm/commit/5d73b77ab2dd023c6c7d55902d8783f232b423ff | 07:45 |
| emwe | Detule: so the comment leaves room for interpretation ;) | 07:45 |
| Detule | nice, k i am currently running a kernel with that patch applied | 07:45 |
| Detule | i am hoping this was the source of the bug i was seeing on 39 | 07:46 |
| emwe | smd7 stopped working? | 07:47 |
| Detule | no the bug where the timer (and certain functions of the phone) would slow down | 07:48 |
| Detule | looking at my logs i invariable see a smd_close(), then smd trying to open channel 7, and right after that a pppd page allocation failure | 07:48 |
| emwe | was it only you who had this or also jonpry? | 07:49 |
| emwe | i am very very very seldomly seeing this | 07:49 |
| emwe | i think mostly reboot, and sleep a while | 07:50 |
| emwe | then clock is off. but so seldom. | 07:50 |
| Detule | not sure, i think jonpry may have seen the bug on his device but i don't know if he is running 39 full time....i am, and this happens every couple of days or so on average....i only have two logs saved, and the above behavior is in both | 07:51 |
| Detule | i think jonpry is onto something with the allocations in arch/arm/mach-msm/include/mach/msm_iomap-7x00.h, looks like in 39 they are off by a significant amount leaving less space for the kernel | 07:52 |
| Detule | if that's the case, maybe this is why this bug crops up more oftren on 39, kmalloc starts struggling earlier to find contiguous allocations....dunno my understanding of how the slab woks is next to nothing | 07:55 |
| emwe | i think .27 .35 and .39 iomap layouts are differently last time i looked. | 08:03 |
| emwe | at least in little pieces. | 08:03 |
| emwe | am also quite ignorant about it. ;) | 08:04 |
| emwe | you had been running debian not any android-flavour, right? | 08:04 |
| Detule | right | 08:04 |
| Detule | i was just enjoying a nice python script timestamping my dmesg output ;) | 08:05 |
| Detule | emwe: to be clear, running debian in my chroot | 08:06 |
| *** lapin{GR} has joined #xdandroid | 08:07 | |
| lapin{GR} | anyway here who can help me with a simple git question? | 08:08 |
| emwe | Detule: sorry for asking again... you didn't have any iowait issues, or did you with .39? | 08:09 |
| Detule | how would they manifest? | 08:09 |
| emwe | start a long-not-used app which is not preloaded and do your daily nap to wait until it's ready. | 08:11 |
| emwe | i see loop0, loop1 and mmcqd goign crazy when this happens. | 08:11 |
| Detule | no, but running partitioned build with NILFS for my data partition | 08:11 |
| emwe | i think part of the problem is that the dalvik cache is on data partition and this *might* cause a lot of io | 08:12 |
| emwe | why this manifests so hard i dunno. | 08:12 |
| Detule | imo, partitions are much better than looped builds...and then nilfs2 has treated me very nicely | 08:13 |
| hyc | agreed, looped builds suck | 08:14 |
| emwe | lol | 08:16 |
| emwe | i am still eager to find out what causes these issues for me as .27 has never been that bad on looped builds iirc | 08:16 |
| emwe | first i am going to try to pull 2.6.35.7 to longterm 2.6.35.14 this evening. | 08:18 |
| emwe | in the hope their's some improvement. early 2.6.35 where said to be really painful regarding io. htc also upgraded to 2.6.35.10 .. perhaps there's some good stuff pending. | 08:20 |
| emwe | s/their's/there's | 08:21 |
| *** xda-Squirrels has joined #xdandroid | 08:58 | |
| *** fishhead2567 has joined #xdandroid | 09:05 | |
| *** gsommer has joined #xdandroid | 09:34 | |
| *** emwe has quit IRC | 09:41 | |
| *** rpierce99 has joined #xdandroid | 09:43 | |
| *** Detule has quit IRC | 09:46 | |
| D3tul3 | micropklt_set_color_led_state(1) -> solid green led, led_state(2) -> solid amber led, led_state(3 or 4) -> blinking green led.....is there a parameter for blinking amber led? | 09:48 |
| *** emwe has joined #xdandroid | 09:51 | |
| *** gsommer has quit IRC | 10:01 | |
| *** lapin{GR} has quit IRC | 10:36 | |
| jonpry | D3tul3, to change that stuff all you really need to do is replace 0xE000 with 0xF000 and change ./include/mach/vmalloc.h:#define VMALLOC_END 0xe0000000UL | 11:13 |
| D3tul3 | jonpry: just to make sure, need to change all 0xE0 values in iomap-7x00.h with 0xF8 (this would make it the same w 35 and 27) | 11:31 |
| jonpry | D3tul3, in that case SPL_BASE would be 0xF9 | 11:33 |
| emwe | jonpry: this increases the vmalloc reserve by 256mb, is this right? or o am i totall off/wrong? | 11:33 |
| jonpry | just add 0x18** to all the _BASE numbers | 11:34 |
| jonpry | emwe it increases the virtual range used by the vmalloc allocator | 11:34 |
| emwe | jonpry: this is vmalloctotal in /proc/meminfo i assume? | 11:35 |
| jonpry | should be | 11:35 |
| emwe | jonpry: i have 573440 kB on .35. what you got? | 11:36 |
| jonpry | i don't have my phone | 11:36 |
| emwe | heh | 11:36 |
| jonpry | i'm guessing its 256 | 11:36 |
| jonpry | mb | 11:36 |
| D3tul3 | 180224 | 11:36 |
| jonpry | lol | 11:36 |
| D3tul3 | whew | 11:36 |
| emwe | which kjernel D3tul3 ? | 11:37 |
| D3tul3 | 39 | 11:37 |
| D3tul3 | k jonpry got it thanks | 11:38 |
| *** krisse has joined #xdandroid | 11:49 | |
| krisse | arrrghhh: hi | 11:50 |
| krisse | Tried running tiwlan but it blockes at the same location... | 11:51 |
| arrrghhh | hrm | 11:51 |
| arrrghhh | do you have logs? | 11:51 |
| krisse | Can u find some design info on this kernel... Just to see what it does next | 11:52 |
| arrrghhh | not sure what you mean | 11:52 |
| krisse | I have logcat and dmesg | 11:52 |
| *** sigma6 has quit IRC | 11:52 | |
| krisse | I added logging | 11:53 |
| arrrghhh | pastebin pls | 11:53 |
| krisse | Its blocking after irq return | 11:53 |
| krisse | I think i posted the error yesterday | 11:54 |
| krisse | I have it not in front | 11:54 |
| krisse | Currwntly mobile | 11:54 |
| krisse | Ok i pastbin it when i'm home | 11:55 |
| krisse | You wil be here for some time? | 11:55 |
| arrrghhh | i'll see if i have the old stuff | 11:55 |
| arrrghhh | not sure, i'm at work | 11:55 |
| krisse | :) ok | 11:56 |
| krisse | Its logged no? | 11:56 |
| xda-Squirrels | im at work too lolz | 11:57 |
| xda-Squirrels | name of the game today is crappy old ibm informix database and pfs: first choice database | 11:58 |
| *** krisse has quit IRC | 12:06 | |
| xda-Squirrels | why are people so technologically dense? | 12:11 |
| xda-Squirrels | clinic is afraid that me copying an MS SQL .bak backup file is going to screw up their system and are making me wait, despite our scheduled time to get data, until they're out of the system. | 12:12 |
| xda-Squirrels | */rageface* | 12:12 |
| *** programmer8922 has joined #xdandroid | 12:12 | |
| xda-Squirrels | which puts me right into the middle of another, more important and expensive, contract's appointment time. | 12:13 |
| *** krisse has joined #xdandroid | 12:13 | |
| arrrghhh | xda-Squirrels, tell them they're not paying you enough | 12:17 |
| xda-Squirrels | lol | 12:18 |
| xda-Squirrels | i pushed back their appointment with me to get data to after 1pm | 12:19 |
| xda-Squirrels | they can deal with it | 12:19 |
| stinebd | tell them they're paying you too much | 12:19 |
| xda-Squirrels | f that ;o! | 12:19 |
| xda-Squirrels | they're getting paid too much if they dont see that copying a simple sql backup file isn't going to trash their system | 12:20 |
| stinebd | sweet, that ghibli movie i never saw is on on demand | 12:26 |
| xda-Squirrels | Word. | 12:31 |
| *** GlemSom has joined #xdandroid | 12:49 | |
| D3tul3 | emwe: what back color led? :) | 12:51 |
| emwe | D3tul3: there's that button for conference speaker. the little micropphone sign is the "back led" | 12:51 |
| emwe | on the back. | 12:52 |
| emwe | i've been debugging the led cmds a bit. the last of the four bytes we send is for controlling blink duration when a "blink led programm" is selected. | 12:52 |
| D3tul3 | by the way i am sure you know this but for the record buff[2]=5 is amber blink buff[2]=4 is green blink | 12:53 |
| emwe | the first two somehow select the led programm to run. it's not that first is back and second byte is front led.. | 12:53 |
| emwe | isn't it in the enum color_led just like that? | 12:53 |
| D3tul3 | i didn't think those were defined in color_led | 12:54 |
| D3tul3 | but i am probably wrong | 12:54 |
| emwe | ah no right. i had them in in between. i know it's these. just reverted that stuff i had here | 12:54 |
| emwe | because the assumption first byte is back and second is for front led is wrong it seems | 12:55 |
| D3tul3 | who needs frequency then :) | 12:55 |
| emwe | you can have it blink n-seconds only | 12:55 |
| emwe | instead blink forever | 12:55 |
| D3tul3 | oh that's a nice feature | 12:55 |
| emwe | fourth byte (not counting cmd byte) controls that | 12:55 |
| emwe | so the last 0xff is infinity .. set it to 0x01 and you get it blinking once | 12:55 |
| emwe | only if blink led is on, though. | 12:56 |
| D3tul3 | makes sense | 12:56 |
| emwe | there's some magic i haven't figured entirely. | 12:56 |
| emwe | i can trigger led-disco here with dbgfs and a little shell script | 12:56 |
| D3tul3 | cool, i have high hopes for this vmalloc and ppp_deflate patched kernel that i am on right now | 12:57 |
| emwe | that fix solved it for me. :) | 12:57 |
| D3tul3 | if it hopes to be the cure for whatever ails 39 i am doing away with the sleep debug led notifications all together | 12:57 |
| D3tul3 | s/hopes/proves/ | 12:57 |
| emwe | i am just merging stable and longterm .35 in my tree. just running 2.6.35.9 ... still a way to go till 2.6.25.14 | 12:58 |
| emwe | 2.6.35.14 i meant... | 12:58 |
| D3tul3 | once i get no interference from the sleep debug notifications i'll look into writing a blink callback though what i had in mind was a lot simpler (and more hack-ish) it seems than what you are doing | 12:59 |
| emwe | i am way from what i have in mind regarind leds. but i am trying to keep pragmatic, yet. | 13:02 |
| *** knchris has joined #xdandroid | 13:02 | |
| D3tul3 | you have the right idea | 13:02 |
| emwe | or microp management itself. | 13:02 |
| stinebd | ok guys | 13:02 |
| stinebd | i got a good one for you | 13:02 |
| emwe | merge acoustic? or plan your japan trip? | 13:02 |
| stinebd | the japan trip was back in may | 13:02 |
| stinebd | this is even better | 13:03 |
| stinebd | #1 0xb00037b4 in call_array (ctor=0xafd40004, count=4, reverse=<value optimized out>) at bionic/linker/linker.c:1615 | 13:03 |
| stinebd | when running an aosp image i built for atrix | 13:03 |
| * emwe hides | 13:03 | |
| stinebd | i've been working on this for 20 hours straight and got nothing | 13:04 |
| stinebd | CM built and runs just fine | 13:04 |
| stinebd | taking bionic from nvidia's repo has the same crasher | 13:05 |
| stinebd | i'm honestly lost | 13:05 |
| stinebd | btw i can't get the whole nvidia tree because it goes 30KB/s | 13:06 |
| arrrghhh | can't wait a few weeks? :P | 13:12 |
| arrrghhh | D3tul3, what about GPS killing sleep? i don't know if there's any other sleep blockers...? | 13:12 |
| *** krisse has quit IRC | 13:13 | |
| *** krisse has joined #xdandroid | 13:15 | |
| arrrghhh | krisse, i don't see any pastebins in the logs | 13:16 |
| *** knchris has left #xdandroid | 13:16 | |
| arrrghhh | just mentions of what's going on | 13:17 |
| krisse | :) | 13:17 |
| krisse | I'm just here | 13:17 |
| arrrghhh | that's fine | 13:17 |
| arrrghhh | i was just sayin :P | 13:17 |
| krisse | never used pastbins before | 13:19 |
| krisse | I will upload zip with 2 files in there | 13:19 |
| krisse | 1 logcat | 13:19 |
| arrrghhh | heh, it's not hard | 13:19 |
| krisse | 1 dmesg | 13:19 |
| krisse | or kmsg | 13:19 |
| arrrghhh | hum | 13:19 |
| arrrghhh | if it's multiupload or some such site i won't be able to get it. | 13:19 |
| krisse | to activate the wifi only thing I do is load the wlan.ko | 13:20 |
| krisse | that generates already the error | 13:20 |
| krisse | now the freeze | 13:20 |
| arrrghhh | ok | 13:21 |
| krisse | I only need to load wlan.ko right? | 13:22 |
| D3tul3 | arrrghhh: yeah i am not suggesting killing sleep dbg notifications for the mainline kernel, just in my tree.....i use gps only sporadically and then i am mindful to turn it on/off...whereas full time notifications i think would be an everyday benefit | 13:23 |
| arrrghhh | i think so | 13:23 |
| arrrghhh | far from positive tho | 13:23 |
| arrrghhh | D3tul3, damn you | 13:23 |
| arrrghhh | :P | 13:23 |
| arrrghhh | yea, i want to get rid of the sleep notification | 13:24 |
| arrrghhh | people don't get it, still. | 13:24 |
| krisse | strange | 13:24 |
| krisse | have a different result now | 13:24 |
| krisse | hmm | 13:24 |
| rpierce99 | i'm still not convinced that my suggested change to the leds won't work for mainline | 13:24 |
| D3tul3 | it will | 13:25 |
| stinebd | it won't | 13:25 |
| stinebd | trust me | 13:25 |
| arrrghhh | it can? | 13:25 |
| rpierce99 | it does? | 13:25 |
| D3tul3 | oh you mean with users | 13:25 |
| arrrghhh | lol | 13:25 |
| stinebd | wait what was the change? | 13:25 |
| arrrghhh | we need to resolve all the blockers of sleep before eliminating the sleep LED | 13:25 |
| arrrghhh | AFAIK, GPS is the only one... no? | 13:25 |
| D3tul3 | amber light only with cpu on AND screen off | 13:25 |
| arrrghhh | other than stupid apps at random. | 13:25 |
| stinebd | won't work | 13:26 |
| rpierce99 | amber light would just indicate a sleep blocker, would go off instead of green sleep, and would go off with screen on | 13:26 |
| D3tul3 | won't work in what sense | 13:26 |
| stinebd | in the sense that phh would never accept the patch | 13:26 |
| rpierce99 | because phh is an unreasonable SOB or because there is something wrong with it | 13:27 |
| stinebd | mostly the 2nd one | 13:27 |
| D3tul3 | come on let's not tease it out of you, what's the larger issue | 13:27 |
| stinebd | the larger issue is why you want to do screen logic in the core of the kernel outside of its driver | 13:28 |
| D3tul3 | it's one call outside of the criver | 13:28 |
| D3tul3 | driver | 13:28 |
| arrrghhh | cuz they want to have their cake and eat it too | 13:28 |
| stinebd | i don't care if it's one call or 800 trillion jillion calls | 13:29 |
| D3tul3 | shrug | 13:29 |
| stinebd | anyway better to do it in userland | 13:29 |
| stinebd | and i accept patches for anything | 13:30 |
| arrrghhh | why can't the blockers on sleep get fixed, and then sleep LED eliminated entirely? | 13:30 |
| rpierce99 | so it's not because the logic isn't sound, it's because implementation would be hackish, i can accept that | 13:30 |
| arrrghhh | emwe fixed the GPS issue in GB. I'm sure it can be ported back to FroYo | 13:30 |
| emwe | arrrghhh: remember i said i fixed nothing in that respect? it just works. likely some hooks need impl. dunno. haven't looked at froyo. and i'd rather like to look at the hd2 gps library worked on by tytung :) | 13:31 |
| D3tul3 | stinebd: i am actually not even sure any calls outside of the driver are needed | 13:31 |
| arrrghhh | emwe, meh. you fixed it, deal with it :P | 13:31 |
| emwe | ok. | 13:31 |
| arrrghhh | and yes, i would like you to look at that HD2 GPS lib. that would be HUGE | 13:31 |
| D3tul3 | probably everything could be done inside microp-klt.c much like it's done with the current debug notifications | 13:32 |
| emwe | just add a module param and be done with it. | 13:32 |
| emwe | and have it default to off if you like. | 13:32 |
| emwe | man, when i read the linux commit logs somehow cold showers run over back... | 13:35 |
| arrrghhh | haha | 13:37 |
| arrrghhh | shivers eh | 13:37 |
| stinebd | shiver he timbers | 13:37 |
| krisse | Hi all, does somebody now where the parameters are configured in the kernel for NAND. I mean like IRQ's and so one? or are they dynamically asigned? | 13:39 |
| arrrghhh | emwe, i don't know what he means | 13:42 |
| arrrghhh | do you? | 13:42 |
| stinebd | arch/arm/mach-msm board files | 13:42 |
| arrrghhh | thx | 13:43 |
| emwe | i think that answered it. | 13:44 |
| stinebd | was i right? | 13:44 |
| emwe | i wouldn't have said sth different :) | 13:44 |
| arrrghhh | hehe | 13:45 |
| krisse | hi all | 14:04 |
| krisse | I have a problem with insmod wlan.ko | 14:04 |
| krisse | after doing that the insmod freezes | 14:04 |
| krisse | this is the pastbin | 14:04 |
| krisse | http://pastebin.com/B05vcDMs (dmesg) | 14:04 |
| krisse | http://pastebin.com/GXYscQiJ (logcat) | 14:05 |
| krisse | can somebody help? | 14:05 |
| krisse | I added some logging to the kernel. | 14:05 |
| krisse | but didn't find how to go further then msm_sdcc.c | 14:05 |
| *** rpierce99 has quit IRC | 14:32 | |
| *** rpierce99 has joined #xdandroid | 14:32 | |
| *** rpierce99 has quit IRC | 14:33 | |
| *** rpierce99 has joined #xdandroid | 14:35 | |
| *** Detule has joined #xdandroid | 14:44 | |
| *** programmer8922 has quit IRC | 14:53 | |
| *** xda-Squrls has joined #xdandroid | 15:23 | |
| *** xda-Squirrels has quit IRC | 15:24 | |
| *** xda-Squrls is now known as xda-squirrels | 15:24 | |
| stinebd | haha http://www.engadget.com/2011/08/12/t-mobile-overages-on-200mb-data-plan-in-throttling-out/ | 15:39 |
| stinebd | tmo and att's plan is brilliant | 15:39 |
| stinebd | have tmo become a twin of att so they can't even tell them apart, for them to block the sale | 15:40 |
| stinebd | the fcc will be powerless to stop them | 15:40 |
| stinebd | tomorrow sprint will start charging $2/MB after 2GB in hopes of being bought by verizon | 15:41 |
| arrrghhh | krisse, these aren't complete logs | 15:47 |
| krisse | yes | 15:47 |
| krisse | I know that | 15:47 |
| arrrghhh | uhm | 15:47 |
| arrrghhh | ok | 15:47 |
| krisse | its the part from when I started the wifi | 15:47 |
| krisse | not enough? | 15:47 |
| arrrghhh | i see an untracked PID at the end | 15:47 |
| arrrghhh | i don't see where the driver goes back down in the dmesg | 15:47 |
| arrrghhh | the logcat is pretty much useless | 15:48 |
| arrrghhh | E/WifiService( 156): Error configuring interface tiwlan0, :java.lang.IllegalStateException: Expected code 213, but got 600 | 15:48 |
| arrrghhh | that's all that really is about wifi in it | 15:48 |
| arrrghhh | it'd probably be best to provide all logs | 15:48 |
| krisse | that last untracked pid is I think from unplugging my usb cable | 15:51 |
| krisse | ok | 15:51 |
| krisse | do that | 15:51 |
| stinebd | woooooo linker fixed | 15:52 |
| stinebd | I/SurfaceFlinger( 1391): SurfaceFlinger's main thread ready to run. Initializing graphics H/W... | 15:52 |
| stinebd | D/ ( 1391): 2s elapsed in InitFifo; is daemon started? | 15:52 |
| stinebd | teehee | 15:53 |
| stinebd | silly zygote, if you have to ask that question, the answer is no | 15:53 |
| *** xda-squirrels has quit IRC | 16:01 | |
| *** JesusFreak316 has joined #xdandroid | 16:21 | |
| *** Fudge has quit IRC | 16:28 | |
| *** Xofrats has joined #xdandroid | 16:51 | |
| Xofrats | I posted a patch to /init to support ext4 on xda, let me know if there are q's | 16:51 |
| emwe | Xofrats: where did you post it? | 16:52 |
| Xofrats | In the tp2 section, the thread about unclean shutdown | 16:54 |
| emwe | i should be reading more rhod forums it seems :) | 16:55 |
| emwe | what's your observations with ext4? | 16:55 |
| Xofrats | I've had 2 reboots so far, it hasn't eaten my data.img | 16:56 |
| Xofrats | I ended having to nuke mine yesterday anyway, so this was a good way to get it started | 16:56 |
| emwe | running ext4 system image looped i assume? | 16:57 |
| Xofrats | Yeah, but the patch supports partitioned ext4 now | 16:57 |
| Xofrats | ext4 _data_ image | 16:58 |
| emwe | ah | 16:59 |
| Xofrats | Now I just need to track major showstopper for .27 ext4 | 17:06 |
| Detule | ext4 might be more secure but i am not so sure it offers much better performance than ext2 | 17:07 |
| Detule | at least in terms of write speed | 17:07 |
| *** Xofrats has quit IRC | 17:10 | |
| *** emwe has quit IRC | 17:48 | |
| *** fishhead2567 has quit IRC | 18:03 | |
| *** JesusFreak316 has quit IRC | 18:03 | |
| *** Detule has quit IRC | 18:23 | |
| *** GlemSom has quit IRC | 18:26 | |
| *** programmer8922 has joined #xdandroid | 19:11 | |
| *** programmer8922 has quit IRC | 19:14 | |
| *** programmer8922 has joined #xdandroid | 19:37 | |
| *** JesusFreak316 has joined #xdandroid | 19:52 | |
| *** JesusFreak316 has quit IRC | 19:55 | |
| *** Fudge has joined #xdandroid | 19:55 | |
| *** JesusFreak316 has joined #xdandroid | 19:56 | |
| *** bluenote has joined #xdandroid | 19:56 | |
| *** Yusuke has joined #xdandroid | 19:57 | |
| *** Yusuke is now known as Yusuke14 | 19:57 | |
| bluenote | hey whenever I "power off" xdandroid, my phone reboots. is there a way I can cause it to actually just power off? | 19:57 |
| rpierce99 | that used to work, something in FRX07 stopped it from working, so unless you feel like debugging it yourself, no | 19:59 |
| JesusFreak316 | "Power off" then pull the battery. :P | 19:59 |
| rpierce99 | although even with FRX06 it always happened to me after using it for like 3 days | 19:59 |
| bluenote | it's always done that for me | 20:00 |
| bluenote | pre frx7 or not | 20:00 |
| rpierce99 | either way, it's a bug not a feature | 20:00 |
| bluenote | but unlikely to be fixed it sounds like? | 20:01 |
| rpierce99 | until someone cares enough to fix it | 20:01 |
| bluenote | understood | 20:01 |
| rpierce99 | but most of the devs don't reboot into winmo that often, so i doubt it | 20:01 |
| *** programmer8922 has quit IRC | 20:16 | |
| stinebd | [19:59:26] <rpierce99> that used to work, something in FRX07 stopped it from working, so unless you feel like debugging it yourself, no | 20:17 |
| stinebd | sounds like arm9 crashing | 20:17 |
| rpierce99 | so just make arm9 not crash during shutdown? Done! glad to get that one marked off the list | 20:18 |
| *** programmer8922 has joined #xdandroid | 20:22 | |
| stinebd | heh | 20:23 |
| stinebd | listen | 20:23 |
| stinebd | people keep clamoring for a reboot option | 20:23 |
| stinebd | that gives it to them :P | 20:23 |
| bluenote | heh | 20:53 |
| bluenote | there's an app for rebooting, theres no shutdown app that works :/ | 20:54 |
| arrrghhh | you want to actually shutdown the phone? | 20:59 |
| arrrghhh | pulling the battery sounds like the app for that :D | 21:00 |
| bluenote | free, too! | 21:03 |
| *** rpierce99_ has joined #xdandroid | 21:08 | |
| *** rpierce99 has quit IRC | 21:08 | |
| *** rpierce99_ is now known as rpierce99 | 21:08 | |
| *** Xofrats has joined #xdandroid | 21:15 | |
| Xofrats | You know my back cover wore off from taking it off too many times | 21:16 |
| ryannathans | i bet your mad. | 21:17 |
| *** bluenote has quit IRC | 21:18 | |
| *** krisse has quit IRC | 21:19 | |
| *** Xofrats has quit IRC | 21:24 | |
| *** ryannathans has quit IRC | 21:41 | |
| *** programmer8922 has quit IRC | 22:06 | |
| *** xda-Squirrels has joined #xdandroid | 22:23 | |
| xda-Squirrels | om nom nom | 22:27 |
| *** xda-Squirrels has left #xdandroid | 22:27 | |
| *** xda-Squirrels has joined #xdandroid | 22:27 | |
| *** ryannathans has joined #xdandroid | 22:29 | |
| *** f00bar80 has joined #xdandroid | 22:41 | |
| f00bar80 | do you support epad devices ? | 22:42 |
| JesusFreak316 | Lin?k | 22:53 |
| JesusFreak316 | Link? | 22:53 |
| JesusFreak316 | It's only for msm7201 based windows mobile handsets. | 22:53 |
| JesusFreak316 | Yeah, no; you'd be better off just googling it. | 22:54 |
| *** TheSeven has quit IRC | 22:56 | |
| *** [7] has joined #xdandroid | 22:56 | |
| xda-Squirrels | storms are a-brewin' | 22:57 |
| *** programmer8922 has joined #xdandroid | 23:01 | |
| *** Xofrats has joined #xdandroid | 23:02 | |
| *** JesusFreak316_ has joined #xdandroid | 23:02 | |
| * Xofrats kicks the squirrel in the nutz | 23:02 | |
| *** JesusFreak316 has quit IRC | 23:04 | |
| xda-Squirrels | :(! | 23:05 |
| *** f00bar80 has quit IRC | 23:07 | |
| Xofrats | Who maintains the rootfs git repo? | 23:07 |
| arrrghhh | who else | 23:08 |
| arrrghhh | stinebd | 23:08 |
| Xofrats | Hmm, need to talk to him about incorporating the ext4 changes | 23:09 |
| arrrghhh | what about the so-called 'brain-dead' way the modules are handled? | 23:12 |
| Xofrats | That too | 23:12 |
| Xofrats | Mine is a hack, because it's dealing with /lib/modules after the hack | 23:13 |
| Xofrats | fact... | 23:13 |
| Xofrats | If I were to do it, I'd check if find -type l matched uname -r, then nuke the dir before ungzipping | 23:14 |
| arrrghhh | meh | 23:14 |
| Xofrats | That way no stray modules | 23:14 |
| arrrghhh | so long as it's fixed | 23:14 |
| arrrghhh | need some way to detect | 23:14 |
| arrrghhh | no need to do that every boot y'know | 23:14 |
| Xofrats | Mine wouldn't | 23:15 |
| Xofrats | Combo of wc, find, and nuke-from-orbit | 23:15 |
| Xofrats | I could produce a patch I guess | 23:15 |
| JesusFreak316_ | Psst, arrrghhh, you should be proud; tomorrow I'm quitting Verizon and actually getting a T-Mobile sim for my HD2. | 23:18 |
| Xofrats | But first I need to get telnet working on this | 23:18 |
| Xofrats | adb ain't cutting it | 23:18 |
| Xofrats | Hmmm, usb0 has an interface w/telnetd running, but Win* doesn't see it | 23:20 |
| Xofrats | Do I have to enable tethering for usb0 to show up? | 23:21 |
| *** Xofrats has quit IRC | 23:27 | |
| arrrghhh | probably | 23:36 |
| *** Xofrats has joined #xdandroid | 23:40 | |
| Xofrats | Okay, a bit better.. Had to adb forward, and run telnetd -b 127.0.0.1 | 23:40 |
| Xofrats | Can't say it's the ideal way to do it, but oh well... | 23:41 |
| *** Xofrats has quit IRC | 23:45 | |
| *** JesusFreak316_ has quit IRC | 23:49 | |
| *** Xofrats has joined #xdandroid | 23:53 | |
| xda-Squirrels | meow | 23:54 |
| Xofrats | silly squirrel | 23:55 |
| xda-Squirrels | oh yeah, squirrels make a barking sound... */bark* | 23:55 |
| xda-Squirrels | chirpy bark, w/e ;p | 23:55 |
| Xofrats | Actually, they quack | 23:56 |
| Xofrats | I swear, this one squirrel... | 23:56 |
| xda-Squirrels | to finish converting one of my old vb.net (eww i know) codes to C# or be lazy... | 23:57 |
| xda-Squirrels | have most of the core code done, just need to tie it all into the UI | 23:57 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!