| 01:01.29 | *** join/#htc-linux mitsutaka (~mitsutaka@123.116.125.0) |
| 01:27.10 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:fd2f:5462:fff6:41c1) |
| 01:29.41 | *** join/#htc-linux dunedude (~Lemon@c-98-225-120-28.hsd1.az.comcast.net) |
| 01:29.58 | *** part/#htc-linux dunedude (~Lemon@c-98-225-120-28.hsd1.az.comcast.net) |
| 01:54.35 | *** join/#htc-linux ahigerd (~ahigerd@173.218.156.102) |
| 01:54.35 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
| 02:11.25 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:fd2f:5462:fff6:41c1) |
| 02:20.34 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:fd2f:5462:fff6:41c1) |
| 02:36.05 | ahigerd | jonpry: D: |
| 02:36.14 | ahigerd | Just had the issue recur |
| 02:38.25 | jonpry | oh yeah |
| 02:38.41 | ahigerd | Pulling logs now |
| 02:38.46 | ahigerd | You want everything or just the radio? |
| 02:39.01 | jonpry | dmesg maybe |
| 02:39.20 | jonpry | i'm not really good with this radio stuff |
| 02:39.37 | ahigerd | http://pastebin.com/6xnnbis1 |
| 02:39.47 | ahigerd | There's kernel, system, and radio logcats |
| 02:41.00 | ahigerd | So dmesg is near the bottom |
| 02:41.37 | jonpry | looks like you ran out of memory |
| 02:41.59 | ahigerd | And lowmem killed the dialer? O.o |
| 02:43.23 | ahigerd | No, that OOM was a while ago |
| 02:43.25 | ahigerd | I knew about that one |
| 02:43.43 | jonpry | so what exactly happened |
| 02:45.15 | *** join/#htc-linux mitsutaka (~mitsutaka@219.143.36.82) |
| 02:45.27 | ahigerd | Open dialer, input number, place call, put phone to ear, no sound. Phone has dropped to launcher and the dialer doesn't acknowledge that there's a call. |
| 02:45.35 | ahigerd | I tried to place the call again and both calls came up. |
| 02:46.12 | ahigerd | Look at line 2204 |
| 02:46.28 | ahigerd | This is the classic representation of the issue: AT+CLCC request gets responded to with "WCDMA", which is utterly meaningless |
| 02:46.54 | jonpry | 03-30 21:34:56.273 D/AT ( 1280): AT> AT+CLCC |
| 02:46.54 | jonpry | 03-30 21:34:57.063 D/AT ( 1280): AT< WCDMA |
| 02:47.00 | ahigerd | Yup. That's it. |
| 02:47.04 | ahigerd | It's SUPPOSED to respond with a list of current calls. |
| 02:47.15 | jonpry | at least it isn't a no response |
| 02:49.49 | ahigerd | detule hacked up the RIL to catch this failure mode and retry the request, but I took his hack out to test this kernel. |
| 02:50.04 | ahigerd | The retry is less than optimal, it's slow -- can take up to 20 seconds to get things back in sync |
| 02:50.12 | ahigerd | But at least it doesn't make the OS think the call was dropped |
| 02:50.29 | jonpry | at+clcc takes 20 seconds? |
| 02:50.46 | ahigerd | The retry cycle can |
| 02:50.50 | ahigerd | If it fails enough |
| 02:51.18 | jonpry | so its common to get continuous failure? |
| 02:51.50 | ahigerd | Semi-common |
| 02:52.06 | ahigerd | It doesn't help that the retry is delayed a few seconds |
| 02:52.12 | ahigerd | detule could tell you more; I don't know what's holding it up |
| 02:54.06 | ahigerd | As an example of what the response is SUPPOSED to look like, scroll up to line 2194 |
| 02:54.37 | ahigerd | This case is actually a bit odd in that it didn't crap out in placing the call (so I got a couple rings) but when the call state changed (presumedly to "answered") the re-query failed |
| 02:55.03 | ahigerd | The call is actually still going, but since the OS doesn't know about it the audio routing is turned off; I've heard from people I've called that they can still hear me though |
| 02:56.24 | ahigerd | We've had some speculation about the proximity sensor being related to the problem, since this didn't happen in FRX, FRX doesn't have prox, and we've seen some state change storms in the logs now and then. |
| 02:57.53 | jonpry | yeah i would get rid of the delay |
| 02:57.54 | ahigerd | is pondering putting detule's hacked RIL back in place just to keep the phone usable until the next attempt |
| 02:58.02 | ahigerd | There's no intentional delay |
| 02:58.21 | ahigerd | There's just a span of time between retries for no reason we can tell for sure |
| 02:58.25 | jonpry | <PROTECTED> |
| 02:58.38 | ahigerd | It's not INTENTIONAL; there's nothing we can do to just "get rid" of it |
| 02:58.50 | jonpry | ic |
| 02:58.54 | ahigerd | detule was thinking about making one of the threads sleep to get other chatter out of the picture |
| 02:59.16 | ahigerd | Not sure exactly what he had in mind there |
| 03:00.19 | ahigerd | Anyway, that's the crash course on what's going on and what we know so far. |
| 03:00.55 | ahigerd | hyc's been fighting with this for a long time, too |
| 03:01.21 | jonpry | i wonder if its some kind of unsolicited notification |
| 03:02.35 | ahigerd | We've pondered that but my question is if it's a race condition with an unsolicited notification where's the REAL response? Shouldn't that at least process through? |
| 03:03.02 | ahigerd | Oh, I forgot to mention, this happens with outgoing SMS too; the SMS will appear to fail but the recipient will get it anyway |
| 03:03.21 | ahigerd | BRB |
| 03:04.14 | jonpry | the new kernel fixes something |
| 03:05.24 | jonpry | maybe the amss does not approve of getting requests when its in CREG |
| 03:05.48 | jonpry | also WCDMA==UMTS==3G GSM |
| 03:18.13 | *** join/#htc-linux detule (~detule@unaffiliated/d3tul3) |
| 03:19.22 | jonpry | AT$WCDMA? |
| 03:21.06 | detule | oh looks like i caught the party |
| 03:22.02 | jonpry | of one |
| 03:22.47 | ahigerd | I'm back |
| 03:22.51 | ahigerd | And yeah, http://pastebin.com/6xnnbis1 |
| 03:22.55 | ahigerd | Problem recurred |
| 03:23.01 | ahigerd | Using stock RIL, not your patched one |
| 03:23.20 | detule | jonpry, fixed 3.3? |
| 03:24.21 | jonpry | no haven't done anymore tracing |
| 03:25.12 | jonpry | does the ril properly handle creg's with network type? |
| 03:26.04 | detule | which creg's ? |
| 03:27.23 | jonpry | case 4: /* +CGREG: <n>, <stat>, <lac>, <cid>, <networkType> */ |
| 03:28.02 | ahigerd | Line 1578 |
| 03:28.11 | ahigerd | er, nm |
| 03:28.22 | ahigerd | That's just the one-parameter version |
| 03:30.19 | jonpry | need that dumping of smd channel stuff |
| 03:31.31 | detule | these messages bug me AT< [WCDMA] WAKEUP PDA: U(),T(),R(DS) |
| 03:32.41 | jonpry | maybe disabling wcdma would help |
| 03:32.50 | ahigerd | It doesn't |
| 03:33.04 | ahigerd | It just does the same thing except saying "GSM" instead |
| 03:33.53 | ahigerd | Believe me, I tried that >.> |
| 03:36.03 | detule | how did you disable it? |
| 03:36.27 | ahigerd | Turned off dat |
| 03:36.27 | ahigerd | a |
| 03:36.37 | detule | the strange thing is, i get those WCDMA prefixed wakeup messages as well -> on spring |
| 03:36.39 | detule | sprint |
| 03:36.42 | detule | how did you turn it off |
| 03:36.49 | ahigerd | Through the network settings panel |
| 03:36.52 | jonpry | plain AT or a channel open and close might cause that message |
| 03:36.59 | jonpry | so naive |
| 03:37.03 | jonpry | :p |
| 03:37.21 | ahigerd | Oh I'm sure. We've seen this spurious message show up in response to other requests to the RIL |
| 03:37.30 | ahigerd | It's just most obvious when it makes a call appear to fail or keeps a call from hanging up |
| 03:37.40 | jonpry | you have that at command injector thing? |
| 03:37.46 | ahigerd | Me? No |
| 03:38.15 | jonpry | me neither |
| 03:38.18 | jonpry | http://ibot.rikers.org/%23htc-linux/20090817.html.gz |
| 03:38.26 | jonpry | seems to have been going on for a bit |
| 03:38.50 | ahigerd | 2009? o.o |
| 03:40.20 | detule | yeah open/close but why the hell am i getting a continuous stream of those |
| 03:41.22 | jonpry | maybe its opening and closing |
| 03:42.19 | ahigerd | researches a bit |
| 03:43.30 | ahigerd | I wonder if it'd be possible for us to land docs on the radio |
| 03:45.04 | ahigerd | What radio do we have anyway? |
| 03:45.28 | jonpry | not really. there is a guy who has the amss source but he won't give it up |
| 03:46.11 | ahigerd | Blah. |
| 03:46.42 | ahigerd | (What's "amss" anyway?) |
| 03:46.48 | *** join/#htc-linux DuperMa (~Duper@85-250-200-172.bb.netvision.net.il) |
| 03:47.22 | jonpry | radio more or less. advanced mobile subscriber software stack |
| 03:49.11 | ahigerd | Ah. |
| 03:50.55 | detule | as bzo succintly explained it to me once, 'amss is the os running on the radio' |
| 03:51.21 | ahigerd | Ah, k. |
| 03:51.34 | ahigerd | A quick search for the acronym expansion seems to indicate this is a Qualcomm specific thing? |
| 03:51.59 | jonpry | indeed |
| 03:52.08 | detule | who has the amss source? |
| 03:52.18 | jonpry | wistilt2 |
| 03:52.25 | detule | damn |
| 03:54.25 | jonpry | well i think if that bs is actually coming in on the smd channel then there is probably nothing that can be done |
| 03:54.42 | jonpry | btw did you notice that out of bounds compile warning on smd.c? |
| 03:54.50 | ahigerd | It didn't happen in FRX though, so something's triggering it |
| 03:55.01 | detule | yeah that's been there for a while |
| 03:55.11 | detule | it's not failing in early_alloc |
| 03:56.06 | jonpry | good to know |
| 04:02.03 | jonpry | i'm guessing create mapping is the culprit. it has all kinds of rules its trying enforce |
| 04:02.33 | detule | well it certainly manages through at least one full cycle through that loop |
| 04:02.40 | detule | including a create_mapping call |
| 04:03.10 | jonpry | interesting |
| 04:10.59 | detule | it makes it through the loop only once, it fails soon after the second create_mapping call |
| 04:15.14 | jonpry | soon after like during vm_area_add_early? |
| 04:17.40 | detule | well here's my iotable_init http://pastebin.com/HdHz1FnB |
| 04:18.21 | detule | and my ramconsole output is "rrrrrr" |
| 04:18.54 | detule | (with some other stuff before that) |
| 04:19.28 | jonpry | your not incrementing i |
| 04:19.35 | jonpry | and it should be i+3 |
| 04:19.39 | jonpry | er i*3 |
| 04:20.08 | detule | ugh |
| 04:20.48 | *** join/#htc-linux DuperMan (~Duper@85-250-200-172.bb.netvision.net.il) |
| 04:21.03 | jonpry | imho it might be easier to read with a little more entropy |
| 04:30.02 | detule | rrrrrrrrrrrrrrrrrrrrrrrrr |
| 04:30.48 | detule | 25 by my count |
| 04:31.32 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
| 04:31.41 | jonpry | and 2 are from before the loop? |
| 04:31.57 | detule | yeah |
| 04:32.38 | detule | you know i was wondering about this |
| 04:32.47 | detule | how does it handle csr and gpt |
| 04:32.52 | detule | when they i think have the same base |
| 04:33.13 | jonpry | there are just multiple virtual addresses that point to the same phys |
| 04:33.56 | jonpry | so it dies on the 8th entry? |
| 04:33.58 | detule | so it's failing on CSR right |
| 04:34.12 | detule | right |
| 04:34.47 | jonpry | adsp |
| 04:35.16 | detule | we must be looking at different places |
| 04:35.38 | detule | msm_io_desc[] in mach-msm/io.c |
| 04:51.21 | *** join/#htc-linux jonpry (~jon@adsl-98-85-196-230.mco.bellsouth.net) |
| 04:51.53 | *** join/#htc-linux detule (~detule@unaffiliated/d3tul3) |
| 04:51.58 | jonpry | yeah can't have both CSR and GPT in MSM_DEVICE |
| 04:52.12 | jonpry | i have it dying on 17 |
| 04:52.15 | jonpry | after create mapping |
| 04:52.24 | jonpry | pretty sure its ad5 |
| 04:53.43 | detule | his after you took out csr? |
| 04:53.55 | jonpry | yes |
| 04:55.44 | jonpry | there are lots of bug_on's in vm_area_add_early |
| 04:57.02 | jonpry | hmm. got ram console with ad5 removed |
| 04:59.17 | detule | sweet |
| 04:59.25 | detule | wonder why it doesn't like ad5 |
| 05:01.02 | jonpry | i dunno. probably best to figure that out now |
| 05:01.34 | detule | we should probably try adding these in order by virt address |
| 05:02.01 | jonpry | this debug hack will need to be removed. and its the only way to figure this out |
| 05:03.09 | jonpry | i think its saying the region overlaps another |
| 05:03.38 | jonpry | which it totally does |
| 05:03.45 | jonpry | wtf |
| 05:04.23 | detule | with what |
| 05:07.11 | jonpry | all kinds of stuff |
| 05:08.22 | detule | we also have ad5_base defined to be something else in the header file |
| 05:08.34 | jonpry | whole thing was a disaster and was blowing up dex |
| 05:10.49 | jonpry | bleh usb controller crashed again |
| 05:10.58 | detule | i don't see it ad5 starts at F83 right |
| 05:11.05 | detule | + 13m that's F9 ? |
| 05:11.10 | detule | what's in between? |
| 05:15.39 | jonpry | well first there is mdp overflowing into mdc |
| 05:16.20 | jonpry | beyond that i'm not sure anymore |
| 05:18.35 | jonpry | ah nm |
| 05:18.51 | jonpry | it was never ad5. it's msm_shared_ram_base |
| 05:19.04 | detule | yeah MDP is overflowing into shared ram base |
| 05:19.25 | detule | mdp base + size = F811 |
| 05:20.15 | detule | so er move shared_ram base up? |
| 05:20.50 | jonpry | i rearranged that file so it is more readable |
| 05:20.52 | detule | uh but then have to move other things as well |
| 05:22.27 | jonpry | pushed |
| 05:23.16 | jonpry | i think that mother will boot |
| 05:23.29 | jonpry | but i blew up my usb controller again :( |
| 05:25.41 | detule | why the whole from F84 to F87 |
| 05:26.00 | jonpry | there is an 86 |
| 05:26.22 | jonpry | and its not supposed to be there |
| 05:27.01 | jonpry | meant to do spl@85 ad5 @86 |
| 05:27.15 | detule | yeah that 'll work nicely |
| 05:29.11 | jonpry | i hope so |
| 05:29.39 | jonpry | blasted 3.3 kernel finds all your memory problems |
| 05:29.57 | detule | i wonder what kind of issues that mdp overlap could be causing on 3.2 |
| 05:30.26 | detule | i have 1% battery life on my laptop left charger at work, don't think this will compile through |
| 05:30.52 | jonpry | down to the wire |
| 05:32.08 | detule | /home/oliver/Documents/Android/Kernels/linux-3.3/arch/arm/mach-msm/board-htcrhodium.c:1399: undefined reference to `arch_ioremap_caller' |
| 05:32.43 | jonpry | yeah there are some commits |
| 05:32.54 | detule | i just pull whatever is on git |
| 05:33.59 | jonpry | pushed |
| 05:40.37 | *** join/#htc-linux DuperMa (~Duper@46-116-60-26.bb.netvision.net.il) |
| 05:42.22 | *** join/#htc-linux detule (~detule@unaffiliated/d3tul3) |
| 05:42.54 | detule | jonpry, it boots but you probably know this by now:0 |
| 05:43.29 | detule | got a kernel panic from somewhere in acoustic but that's a battle for another day |
| 05:43.37 | jonpry | no my usb is broke for now |
| 05:43.39 | jonpry | awesome |
| 05:43.44 | jonpry | video and all? |
| 05:44.10 | detule | no it panics in the flying text when it turns on acoustic |
| 05:44.21 | detule | did you take mach-msm from 3.2? |
| 05:44.28 | jonpry | yes |
| 05:44.56 | detule | alright should be easy to hunt down ... i am out for the night, thanks for this ... as usual :) |
| 05:45.13 | jonpry | i'm not sure we have i2c yet. a1010 and tps whatever should be in though |
| 05:45.22 | jonpry | yeah i need sleep too. good night |
| 05:45.31 | detule | good night |
| 06:52.54 | *** join/#htc-linux rob_w (~bob@ppp-88-217-81-39.dynamic.mnet-online.de) |
| 06:52.57 | *** join/#htc-linux rob_w (~bob@unaffiliated/rob-w/x-1112029) |
| 07:04.35 | *** join/#htc-linux LargePrime (~LargePrim@184.244.158.129) |
| 07:27.38 | *** join/#htc-linux DuperMan (~Duper@46-116-60-26.bb.netvision.net.il) |
| 07:35.28 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-107-19.dhcp.stcd.mn.charter.com) |
| 07:51.31 | *** join/#htc-linux jonpry (~jon@adsl-98-85-196-230.mco.bellsouth.net) |
| 08:19.46 | *** join/#htc-linux G4y (~heaheahah@160.8.202.84.customer.cdi.no) |
| 08:21.02 | *** join/#htc-linux kiozen (~kiozen@ppp-88-217-31-28.dynamic.mnet-online.de) |
| 08:26.12 | *** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net) |
| 09:12.49 | *** join/#htc-linux MacDrunk (~marper@201.165.119.167) |
| 09:13.00 | *** part/#htc-linux MacDrunk (~marper@201.165.119.167) |
| 09:16.27 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:fd2f:5462:fff6:41c1) |
| 09:32.01 | *** join/#htc-linux swc|666 (~goldbond@Aircrack-NG/Friend) |
| 09:40.56 | *** join/#htc-linux GNUtoo (~gnutoo@host149-73-dynamic.56-82-r.retail.telecomitalia.it) |
| 09:44.35 | *** join/#htc-linux ccube (ccube@nx.ccube.de) |
| 09:59.36 | *** join/#htc-linux ccube (ccube@nx.ccube.de) |
| 10:43.04 | *** join/#htc-linux LordDeath (~LordDeath@cable-81-173-164-174.netcologne.de) |
| 10:54.58 | *** join/#htc-linux mitsutaka (~mitsutaka@123.116.125.0) |
| 11:06.06 | *** join/#htc-linux pbaxter (~pbaxter@host172-113-dynamic.6-79-r.retail.telecomitalia.it) |
| 11:16.12 | *** join/#htc-linux kiozen (~kiozen@p578a42db.dip0.t-ipconnect.de) |
| 11:18.17 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
| 11:59.30 | *** join/#htc-linux LargePrime (~LargePrim@173-149-66-3.pools.spcsdns.net) |
| 12:13.06 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
| 12:37.54 | *** join/#htc-linux helicopter88 (~helicopte@host223-118-dynamic.37-79-r.retail.telecomitalia.it) |
| 12:50.53 | *** join/#htc-linux detule (~wircer@unaffiliated/d3tul3) |
| 13:11.14 | *** join/#htc-linux LargePrime (~LargePrim@108.100.105.12) |
| 13:15.06 | *** join/#htc-linux yuuki (~yuuki@dsbg-d9bb1f15.pool.mediaWays.net) |
| 13:22.17 | *** part/#htc-linux detule (~wircer@unaffiliated/d3tul3) |
| 13:22.25 | *** join/#htc-linux detule (~wircer@unaffiliated/d3tul3) |
| 13:24.50 | *** join/#htc-linux Alex[sp3dev] (~alexander@178.76.204.11) |
| 13:59.16 | *** join/#htc-linux DuperMan (~Duper@46-116-60-26.bb.netvision.net.il) |
| 14:11.13 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:fd2f:5462:fff6:41c1) |
| 14:14.22 | Alex[sp3dev] | wtf is modern linux 'desktop' written by some dope morons. tons of stupid abstractions (*Kit, udisks, upower, systemd) that not only do not work but are constantly rewritten (udisks2, ConsoleKit is dropped), and newer versions feature the absence of command line interface |
| 14:14.22 | *** join/#htc-linux kiozen (~kiozen@ppp-88-217-31-28.dynamic.mnet-online.de) |
| 14:16.06 | *** join/#htc-linux whitekidney (~wkl0l@141.0.136.22) |
| 14:22.55 | ahigerd | systemd works just fine >.> |
| 14:23.02 | ahigerd | I'm a fan |
| 14:25.20 | Alex[sp3dev] | okay, it does. as long as you're using fedora. what i don't like about these new things are. 1. they are non-portable, bound to linux 2. xml-based configuration (ck, dbus) is poorly documented and hard to setup and 3. these things get rewritten every year, and compatibility breaks (new interfaces) and shits like NM don't even have CLI out of the box |
| 14:25.34 | ahigerd | I'm not using Fedora >.> |
| 14:26.17 | Alex[sp3dev] | gentoo? |
| 14:26.37 | detule | no command line? |
| 14:26.41 | ahigerd | For my desktop, yeah, though that's running init-ng; the box I have running systemd is Debian |
| 14:27.03 | *** join/#htc-linux Guest30608 (~heaheahah@160.8.202.84.customer.cdi.no) |
| 14:27.24 | ahigerd | I agree about NM though. Linux wifi configuration sucks. |
| 14:27.49 | Alex[sp3dev] | i was sick of wifi |
| 14:28.11 | Alex[sp3dev] | so now i'm just running supplicant daemon and manually writing networks to wpa_supplicant.conf |
| 14:28.19 | ahigerd | Honestly I just stick with wpa_supplicant and try not to worry about making it friendly for people that aren't me |
| 14:28.20 | Alex[sp3dev] | and manually running dhclient |
| 14:28.40 | ahigerd | Gentoo's wpa_supplicant integration is fine, I just modify the conf and /etc/init.d/net.wlan0 restart |
| 14:29.11 | ahigerd | The Debian box is actually running master mode so I can't say much there |
| 14:29.27 | ahigerd | (It's a router/server) |
| 14:30.16 | Alex[sp3dev] | i'm thinking of moving to gentoo. binary distros are a bit inflexible (can't turn off some bells-and-whistles or compile what i need) and i'm sick of debian packaging, it's so complicated compared to archlinux or gentoo |
| 14:30.32 | ahigerd | That's why I use Gentoo *nods* |
| 14:30.39 | ahigerd | I've got the patience to deal with compiles |
| 14:30.53 | ahigerd | And with systems getting faster even that's not as big of a deal anymore |
| 14:30.58 | ahigerd | My first Gentoo machine spent five weeks compiling. XD |
| 14:31.09 | ahigerd | (Pentium II 266MHz) |
| 14:31.52 | ahigerd | (Actually that box was up and running in about two days, but then I built OpenOffice -- 2 weeks -- and GNOME -- 3 weeks.) |
| 14:32.04 | detule | i like fedora but it's too damn hard to keep up with the cycle |
| 14:33.15 | helicopter88 | why is eclipse on gentoo too much outdated? |
| 14:33.27 | Alex[sp3dev] | i first tried gentoo on my 'old' pc. (amd 3000+, 1gig ram, ancient hdd). building world took about a day or two which was not fun. recently i tried on my laptop (6gig ram, portage tmpdir in ram and ssd) - blazingly fast |
| 14:33.41 | ahigerd | I dunno, but I hate Eclipse anyway >.> |
| 14:34.17 | Alex[sp3dev] | i hate java anyway |
| 14:34.46 | helicopter88 | who doesn't hate java? |
| 14:34.46 | ahigerd | Likewise |
| 14:34.47 | *** join/#htc-linux kiozen (~kiozen@ppp-88-217-31-28.dynamic.mnet-online.de) |
| 14:36.38 | *** join/#htc-linux Rajko (~rajkosto@2001:470:d76b:da7a:fd2f:5462:fff6:41c1) |
| 14:42.51 | jonpry | i like java |
| 14:48.05 | *** join/#htc-linux kiozen (~kiozen@ppp-88-217-31-28.dynamic.mnet-online.de) |
| 14:56.49 | *** part/#htc-linux detule (~wircer@unaffiliated/d3tul3) |
| 15:09.50 | *** join/#htc-linux detule (~detule@unaffiliated/d3tul3) |
| 15:12.45 | detule | what was the adb wait for device syntax |
| 15:13.14 | GNUtoo | adb wait-for-device |
| 15:13.27 | detule | thanks |
| 15:13.27 | jonpry | detule, you get that thing going? |
| 15:13.38 | detule | i am just sitting down trying to see the stacktrace |
| 15:13.46 | detule | adb comes up for a second |
| 15:14.20 | jonpry | did you remove the tracing patch? |
| 15:14.29 | detule | yeah this is straight from git |
| 15:15.02 | detule | i guess i can ramconsole just as easy |
| 15:15.30 | jonpry | adb is a good sign |
| 15:27.32 | detule | pretty sure we're not probing any of the i2c devices |
| 15:29.45 | detule | i guess we need msm_i2c |
| 15:31.48 | jonpry | sounds right |
| 15:33.42 | *** join/#htc-linux Hadrian (Hadrian@89.28.124.7) |
| 15:33.43 | Hadrian | i'm searching for a girl... |
| 15:33.58 | jonpry | named sandoz? |
| 15:35.08 | detule | jonpry, liftoff bootani and all |
| 15:35.27 | jonpry | oh cool |
| 15:45.54 | detule | jonpry, so we don't need this any more https://gitorious.org/linux-msm-rhod/linux-msm-rhod/commit/311eb9875c85dbd63122717fab33423c890ff13b |
| 15:50.41 | *** join/#htc-linux Gage_ (~Gage@68-191-233-59.static.kgpt.tn.charter.com) |
| 15:54.22 | *** join/#htc-linux mitsutaka (~mitsutaka@123.116.125.0) |
| 15:54.39 | jonpry | detule, we do |
| 15:55.42 | detule | i seem to remember not being able to get bootani |
| 15:55.45 | detule | without that patch |
| 15:56.37 | jonpry | one of the drivers uses ioremap instead of an MSM_IOMAP entry |
| 15:56.52 | jonpry | don't remember which one, probably hw3d |
| 15:57.15 | detule | yeah but bootani is going just fine without it |
| 15:57.39 | jonpry | something won't work |
| 15:59.09 | detule | oh you tucked it inside another commit |
| 16:00.41 | *** join/#htc-linux mitsutaka (~mitsutaka@123.116.125.0) |
| 16:03.37 | jonpry | touchscreen? |
| 16:05.08 | jonpry | damn there is a lot of patches in 3.2 |
| 16:06.52 | detule | i am about to add ts |
| 16:06.59 | detule | unless you got it queued up already |
| 16:07.29 | detule | trying to be more surgical than that input/toucschreen patch that's in 3.2 |
| 16:07.49 | jonpry | no i'm not working on it |
| 16:09.48 | jonpry | chances of hitting a deer in the next year in the us. 1:192 |
| 16:27.12 | detule | hm |
| 16:27.24 | detule | tssc_manager is registering interrupts |
| 16:27.43 | detule | but it doesn't actually do anything on the screen |
| 16:29.10 | jonpry | getevents say anything? |
| 16:29.33 | detule | yeah gets plenty |
| 16:30.11 | jonpry | probably locked then |
| 16:30.28 | jonpry | unlock with end key? |
| 16:31.14 | detule | hm unresponsive to all buttons but they all register with getevent |
| 16:31.55 | detule | here's some output of getevent http://pastebin.com/AAKwVgyu |
| 16:32.06 | jonpry | could the ui be frozen? |
| 16:32.16 | jonpry | does logcat talk about registering the input drivers? |
| 16:33.07 | jonpry | there was a patch i did in the older kernels to change the timesource used by evdev. i don't see it in 3.2 though |
| 16:34.04 | *** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring) |
| 16:34.07 | detule | I/InputManager-Callbacks( 1411): No input device calibration properties found for device tssc-manager. |
| 16:34.18 | detule | i can't recall if i see that in the regular logcat |
| 16:34.23 | detule | but it comes up for all input devices |
| 16:34.35 | detule | for gpio-keys and microp-keypad |
| 16:34.46 | jonpry | thats not normal |
| 16:35.08 | detule | binder? |
| 16:35.24 | detule | oh we're using staging/android from 3.2 |
| 16:36.12 | detule | hm |
| 16:36.17 | detule | i think we might need the wakelocks in evdev |
| 16:36.31 | detule | wait that won't make a difference |
| 16:41.51 | *** join/#htc-linux YOYOYOSF (698140c6@gateway/web/freenode/ip.105.129.64.198) |
| 16:45.37 | GNUtoo | detule, there is android stuff for 3.2? |
| 16:46.03 | detule | from google's tree |
| 16:47.26 | GNUtoo | where can I find that? |
| 16:47.30 | GNUtoo | paulk-desktop, ^^^ |
| 16:48.01 | paulk-desktop | detule, google kernels are 3.0 |
| 16:48.08 | detule | https://github.com/android/kernel_common/branches |
| 16:48.09 | paulk-desktop | I download kernel-omap this morning |
| 16:48.16 | detule | they have up to 3.3 in their tree |
| 16:48.36 | detule | you can forward port their staging/android from 3.0 if you wish ...but in 3.3 it's already mainlined |
| 16:49.07 | detule | (our staging/android from 3.2 is a forward port we've been carrying around perhaps since google's .39 tree) |
| 16:51.11 | *** join/#htc-linux jonpry (~jon@adsl-98-85-196-230.mco.bellsouth.net) |
| 16:53.25 | jonpry | do a diff between 3.2 and 3.3 evdev |
| 16:53.30 | detule | what are these idc files? |
| 16:53.31 | jonpry | see the < do_gettimeofday(&event.time); |
| 16:53.36 | jonpry | > getrawmonotonic(&ctime); |
| 16:53.43 | jonpry | need that or strange things happen |
| 16:53.45 | detule | i am missing these /system/usr/idc calibration files |
| 16:53.51 | paulk-desktop | ok |
| 16:53.51 | jonpry | no calibration for the button devices i think is a different problem though |
| 16:54.00 | jonpry | it looks like the only difference between the two versions is a line break |
| 16:54.05 | jonpry | probably just copy over 3.2 evdev |
| 16:54.12 | paulk-desktop | thanks |
| 16:54.15 | paulk-desktop | I'll do that |
| 16:54.35 | jonpry | detule, sorry i wrote bunch of undelivered messages |
| 16:54.45 | jonpry | i wouldn't worry about the idc problem |
| 16:54.52 | jonpry | except on buttons |
| 16:55.12 | detule | are these suckers created on the kernel side? |
| 16:55.25 | jonpry | the need for idc's has been obviated in userland |
| 16:55.57 | jonpry | but userland determines if an input device is touch/button based on some heuristic that must be breaking |
| 16:56.22 | detule | it doesn't work with the evdev patch either jonpry |
| 16:56.23 | Alex[sp3dev] | heuristic is an euphemism for magic in this case |
| 16:56.29 | detule | and that evdev patch has getrawmonotonic |
| 16:57.18 | jonpry | ok |
| 16:57.24 | Alex[sp3dev] | paulk-desktop: hi btw |
| 16:57.24 | detule | want me to push this out to the tree even though it's not working |
| 16:57.30 | jonpry | sure |
| 16:57.46 | jonpry | maybe they redefined the evdev flags? |
| 16:57.54 | paulk-desktop | hi Alex[sp3dev] |
| 16:57.56 | Alex[sp3dev] | paulk-desktop: i'm getting 'bad address' error with ioctl on samsung modemctl. ever seen that? |
| 16:58.09 | paulk-desktop | nope |
| 16:58.11 | paulk-desktop | that is surprising |
| 16:58.17 | paulk-desktop | sgs2? |
| 16:58.28 | *** join/#htc-linux DuperMan (~Duper@46-116-60-26.bb.netvision.net.il) |
| 16:58.33 | paulk-desktop | look at aries device files in libsamsung-ipc |
| 16:58.35 | paulk-desktop | it should be similar |
| 16:58.48 | Alex[sp3dev] | paulk-desktop: http://pastebin.com/6B3fRHt6 maybe i'm doing it wrong |
| 16:58.56 | detule | jonpry, it's up there i probably messed up the tssc commit somehow |
| 16:59.20 | Alex[sp3dev] | paulk-desktop: thanks, will look at it. guess it won't be hard to implement sgs2 modem. in ics 3.0.15 kernel they're using modemctl similiar to nexus s 3.0 kernel |
| 17:00.12 | paulk-desktop | Alex[sp3dev], also there is already software for sgs2 (userland side): https://github.com/grindars/freeril-i9100 |
| 17:00.23 | paulk-desktop | bbl |
| 17:03.57 | detule | looking at this getevent output it's interpreting touches correctly it seems "BTN_TOUCH" |
| 17:04.09 | detule | i like this "getevent -l" from ics |
| 17:08.02 | jonpry | there doesn't really seem to be a diff between 3.2/3.3 drivers/input |
| 17:09.46 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-107-19.dhcp.stcd.mn.charter.com) |
| 17:10.16 | jonpry | you don't have like a pegged system server or anything? |
| 17:12.36 | detule | no |
| 17:13.15 | jonpry | well i think the trouble has nothing to do with drivers/input |
| 17:15.12 | jonpry | which is unfortunate |
| 17:23.02 | jonpry | hrm i don't think i patched kernel/power/main.c |
| 17:23.14 | jonpry | for that matter hole kernel/power is suspect |
| 17:24.04 | jonpry | yup fubar'd that one |
| 17:28.09 | jonpry | pushed |
| 17:28.50 | *** join/#htc-linux DuperMan (~Duper@46-116-60-26.bb.netvision.net.il) |
| 17:29.33 | *** join/#htc-linux solos1nce (592a2459@gateway/web/freenode/ip.89.42.36.89) |
| 17:30.09 | solos1nce | hello |
| 17:30.14 | solos1nce | anybody here ? |
| 17:31.52 | detule | cool jonpry that did it |
| 17:32.19 | jonpry | awesome |
| 17:48.58 | jonpry | all systems go? |
| 17:49.09 | *** join/#htc-linux simpson (77eb360e@gateway/web/freenode/ip.119.235.54.14) |
| 17:49.36 | simpson | Alex[sp3dev], ping |
| 17:50.10 | detule | jonpry, sure i am cherry-picking my way through 3.3 i just noticed |
| 17:50.15 | detule | that i've been kexecing on it |
| 17:50.19 | detule | without the kexec patch from 3.2 |
| 17:50.39 | jonpry | lol |
| 17:50.47 | jonpry | they fixed something for us for once |
| 17:50.55 | detule | pretty amazing |
| 17:52.16 | Alex[sp3dev] | simpson: pong |
| 17:52.36 | simpson | can you check this for me ? http://pastebin.com/hMu9NW6M should work ? |
| 17:52.50 | CptAJ | detule, the 2010q3 toolchain you mention is this one right? https://sourcery.mentor.com/sgpp/lite/arm/portal/release1600 |
| 17:53.04 | CptAJ | I dont see one specifically labeled "q3" |
| 17:53.21 | simpson | last release with that label was 2010q1 |
| 17:53.30 | detule | CptAJ, that's the one |
| 17:54.02 | *** join/#htc-linux G4y (~heaheahah@160.8.202.84.customer.cdi.no) |
| 17:54.02 | CptAJ | alrighty, gonna give it a shot |
| 17:54.29 | Alex[sp3dev] | simpson: looks ok |
| 17:54.31 | simpson | call me paranoid, but i stick to 2010q1, anything more recent makes bad binaries for me |
| 17:54.37 | Alex[sp3dev] | simpson: got problems with that code? |
| 17:55.17 | simpson | yeah, it it just hangs, doesn't crash in my code probably because lk's exception handlers don't dump registers |
| 17:57.08 | Alex[sp3dev] | simpson: i'm pretty sure that by default it is not loaded to 0x0, but to some other address. or are you using some wince header like tinboot? |
| 17:57.49 | Alex[sp3dev] | besides i would not be sure memset doesn't trash r0 |
| 17:58.41 | simpson | that code is from a small sub project in lk, i am trying to execute some arm binaries through lk |
| 18:04.05 | Alex[sp3dev] | simpson: http://pastebin.com/p2VyCqQX when playing with LK, i just modified tinboot's address to match the one linux is loaded to and just loaded the wrapped lk binary as if it were a zImage |
| 18:04.35 | simpson | cool, thanks |
| 18:04.48 | Alex[sp3dev] | simpson: it is based on tinboot, but wince header may be broken, i have not flashed it, only ram-booted |
| 18:05.02 | Alex[sp3dev] | if you need flashing, there's code for photon in git |
| 18:05.11 | simpson | i have the wince rom header from the leo lk |
| 18:31.01 | *** join/#htc-linux GNUtoo (~gnutoo@host149-73-dynamic.56-82-r.retail.telecomitalia.it) |
| 18:31.36 | *** join/#htc-linux Alex[sp3dev] (~alexander@178.76.204.11) |
| 18:35.43 | jonpry | why are rhod's always the first phone to run a kernel? |
| 18:46.10 | detule | camera wifi works |
| 18:46.43 | ahigerd | I dunno, maybe we just have a well-understood arch? |
| 18:48.58 | *** join/#htc-linux helicopter88 (~helicopte@host223-118-dynamic.37-79-r.retail.telecomitalia.it) |
| 19:04.01 | detule | i had +40MB worth of modules-*.tar.gz in /sdcard |
| 19:04.12 | ahigerd | Ehe, I should probably clean that up too |
| 19:21.14 | *** join/#htc-linux skynet (SkyNet-200@gateway/shell/trekweb.org/x-uvkxhbmlxptvupaa) |
| 19:21.28 | *** join/#htc-linux skynet-2000 (SkyNet-200@unaffiliated/skynet2000) |
| 19:21.28 | *** join/#htc-linux skynet-2000 (SkyNet-200@gateway/shell/trekweb.org/x-uvkxhbmlxptvupaa) |
| 19:40.47 | detule | jonpry, good time to push to the tree? |
| 19:41.02 | jonpry | sure |
| 19:44.42 | jonpry | how far did you get? |
| 19:45.16 | detule | hard to go by the 3.2 tree commit log because a lot of it was done in the initial patch-set you pushed |
| 19:45.59 | jonpry | everything works? |
| 19:46.15 | detule | probably not bluetooth but i stopped after the gsensor |
| 19:46.27 | detule | scores +23 in neocore so that's pretty good :) |
| 19:46.33 | detule | my one true test |
| 19:47.07 | jonpry | cool :) |
| 19:47.22 | jonpry | need autobuild |
| 19:50.23 | jonpry | it was a rocky start but we really whipped this one out |
| 19:50.27 | detule | we should probably backport that io table to 3.2 |
| 19:50.55 | jonpry | indeed |
| 19:51.00 | detule | not sure where the "we" happened...it was all you...sigh...i'll shine when 3.4 comes around |
| 19:51.38 | jonpry | i dunno you made most of the peripherals work |
| 19:51.54 | detule | but i have to say that tracing method makes these ports less intimidating |
| 19:52.23 | jonpry | we will have to use that again some time |
| 19:52.24 | detule | at some point we should checkout staging/android from the initial commit |
| 19:52.47 | jonpry | yes. and upgrade usb |
| 19:54.17 | jonpry | probably 2 months until 3.4 drops |
| 20:02.21 | detule | i am out later |
| 20:17.34 | *** join/#htc-linux mastermerlin (~Adium@p4FEE4E3F.dip.t-dialin.net) |
| 20:22.29 | *** join/#htc-linux swc|666 (~goldbond@Aircrack-NG/Friend) |
| 20:33.07 | *** join/#htc-linux LTxda (ltxda@c-98-194-107-10.hsd1.tx.comcast.net) |
| 20:33.07 | *** join/#htc-linux LTxda (ltxda@unaffiliated/ltxda) |
| 20:33.21 | *** join/#htc-linux LTxda (ltxda@c-98-194-107-10.hsd1.tx.comcast.net) |
| 20:33.21 | *** join/#htc-linux LTxda (ltxda@unaffiliated/ltxda) |
| 20:46.19 | *** join/#htc-linux simpson (77eb360e@gateway/web/freenode/ip.119.235.54.14) |
| 21:40.05 | *** join/#htc-linux mastermerlin (~Adium@p4FEE4E3F.dip.t-dialin.net) |
| 22:45.36 | CptAJ | Whitestone touchscreen is up and running =) |
| 22:45.55 | CptAJ | it WAS the toolchain -_- |
| 22:46.41 | CptAJ | pretty unreliable compiler action there >_> |
| 22:46.58 | CptAJ | thank god you guys mentioned that. I would've been chasing my tail forever |
| 22:47.06 | CptAJ | cheers to you! |
| 22:59.46 | *** join/#htc-linux GNUtoo (~gnutoo@host149-73-dynamic.56-82-r.retail.telecomitalia.it) |
| 23:20.03 | *** join/#htc-linux swc|666 (~goldbond@Aircrack-NG/Friend) |
| 23:23.21 | *** join/#htc-linux detule (~detule@pool-96-234-128-143.bltmmd.east.verizon.net) |