| *** IINSPECTOR-0011 has left #xdandroid | 00:11 | |
| *** IINSPECTOR-0011 has joined #xdandroid | 00:11 | |
| *** NeoMatrixJR has joined #xdandroid | 00:11 | |
| *** serren has quit IRC | 00:12 | |
| *** rpierce99 has quit IRC | 00:28 | |
| *** mes has quit IRC | 00:35 | |
| *** Fudge has quit IRC | 00:41 | |
| *** NeoMatrixJR has quit IRC | 00:55 | |
| *** IINSPECTOR-0011 has quit IRC | 01:21 | |
| *** programmer8922 has joined #xdandroid | 01:23 | |
| *** mes has joined #xdandroid | 01:24 | |
| *** programmer89221 has joined #xdandroid | 01:27 | |
| *** programmer8922 has quit IRC | 01:27 | |
| *** Fudge has joined #xdandroid | 02:17 | |
| *** programmer89221 has quit IRC | 02:46 | |
| *** programmer8922 has joined #xdandroid | 02:49 | |
| smiley- | wee.. the phone survived the night.. 72% battery and almost 12 hours deep sleep | 03:20 |
|---|---|---|
| *** programmer8922 has quit IRC | 03:24 | |
| *** programmer8922 has joined #xdandroid | 03:25 | |
| *** bzo has quit IRC | 03:30 | |
| *** GlemSom has joined #xdandroid | 03:34 | |
| *** phh has quit IRC | 04:00 | |
| *** phh has joined #xdandroid | 04:01 | |
| *** GPFerror has joined #xdandroid | 04:37 | |
| *** emwe has joined #xdandroid | 05:27 | |
| emwe | stinebd: hyc: the power.c patch is broken? tested the whole merge request out with raph and stock .27. wen't fine. | 05:37 |
| hyc | yes it's broken | 05:52 |
| hyc | it has the logic reversed | 05:52 |
| hyc | whch makes me wonder why it works for you | 05:52 |
| emwe | .27 + frx06+ on raph | 05:59 |
| emwe | not seen an issue. actually the raph just runs with that lying around here sleeping | 05:59 |
| emwe | i didn't test on rhod and topa as i am fed up switching userlands all the time | 05:59 |
| emwe | hyc: allthough i should have been whondering why there was a compilation error mentioned by alex. he compiled under a gb tree if i recall his comments right. | 06:01 |
| emwe | all was building fine. | 06:01 |
| *** programmer8922 has quit IRC | 06:04 | |
| *** valyk has joined #xdandroid | 06:21 | |
| valyk | heya anyone | 06:21 |
| valyk | how can i edit the ext2 file on windows 7 | 06:22 |
| valyk | ? | 06:22 |
| valyk | everyone* i mean:) | 06:22 |
| hyc | emwe: yeah, the whole circumstances of the patch are suspect. | 06:24 |
| hyc | but compared to the previous code, the logic is definitely backwards | 06:25 |
| emwe | i should have paid more attention. figured all was fine after testing with the raph. | 06:27 |
| emwe | hyc: you ever seen "msm_sleep(): clk_set_rate 128000000 failed" in dmesg? | 06:29 |
| emwe | ah... you might not. can be wis dropped that ramp-up fix in his pm changes again | 06:30 |
| hyc | nope, definitely have not seen that | 06:31 |
| *** valyk has left #xdandroid | 06:32 | |
| *** valyk has joined #xdandroid | 06:33 | |
| emwe | will look your combo branch again then | 06:37 |
| emwe | it fails for me on rhod400. likely 128mhz is not the lowest stepping frequency. like jb has also mentioned for his diam | 06:38 |
| emwe | might be it fails and stais at 19.2mhz during resume making the wakes so ultra long here.. | 06:38 |
| hyc | well, 128000 isn't a multiple of 19200 | 06:41 |
| hyc | shou;d be 115200 I'd guess. dunno | 06:41 |
| hyc | or 122803200 | 06:47 |
| emwe | 122800 | 06:49 |
| emwe | the CA acpuclock selects different stepping tables according to the different plls | 06:50 |
| emwe | topa is fine with 128mhz | 06:50 |
| emwe | other devices seem to as well | 06:50 |
| emwe | ok, wake seems also fixed for rhod400 with that ramp-up fix commented. then the theory with resume @19.2mhz and ultra slow wakes makes sense. | 06:53 |
| emwe | hyc: you noticed some rebuild problems after syncing new code and it didn't get picked up by the android buildsystem? | 06:59 |
| emwe | i am just rebuilding my whole froyo tree just to verify if that power.c merge got built at all yesterday. not that I have fooled myself by believing it got pulled into the compile... | 06:59 |
| hyc | mm, dunno. I very rarely do a full top-down build | 07:00 |
| emwe | ... and just grabbed the system.img and tested... | 07:00 |
| emwe | i always do for just pushing the system.img via adb | 07:00 |
| hyc | since I'm usually patching specific things I usually only recompile that one thing. | 07:00 |
| emwe | only been doing that for the ril mostly. otherwise i mostly push the whole image. | 07:01 |
| hyc | and also a bit lazy. Since I have stuff like the video codecs that have to be added to the build separately, I don't like to blow away my system.img | 07:01 |
| emwe | ;) | 07:01 |
| emwe | 3 out of 6 commits from your combo made it to autobuild branch just now. | 07:02 |
| hyc | Oh? will have to take a look | 07:02 |
| emwe | just small ones. prox irq, mdp and clk late init for kexec | 07:03 |
| hyc | cool | 07:03 |
| hyc | kexec is a good one. .39 kernel dies on kexec too | 07:04 |
| emwe | raph also spits out like 4 unbalanced irq traces on boot for raph-navi | 07:04 |
| emwe | but i am not really motivated getting into that. | 07:04 |
| hyc | so you've mostly fixed the GSM radio problem on sleep? | 07:04 |
| emwe | it's working here now. power drain is really not noticeable. | 07:04 |
| hyc | Nice. that's going to make a lot of people happy | 07:05 |
| hyc | they don't have to be stuck on 1253 any more | 07:05 |
| emwe | would even dare to say it got better... might be just subjective feeling. | 07:05 |
| hyc | heh | 07:05 |
| emwe | happy bzo popped up and we could talk about that issue | 07:05 |
| hyc | the mdp_dma squelch should also make it easier to do TS calibration now | 07:05 |
| hyc | instead of having everything scroll off the screen... | 07:06 |
| emwe | there was a time everything stopped when calibration up was running iirc | 07:06 |
| emwe | or was it just it did something else that long with no output? | 07:06 |
| hyc | something else I'm sure. at any rate, the log spam made it impossible to see the calibration stuff | 07:06 |
| emwe | ah | 07:07 |
| smiley- | my phone was struck by the Alexander bug | 07:07 |
| emwe | smiley-: ? | 07:07 |
| smiley- | = my son throwing in the floor so the battery disconnected ;) | 07:07 |
| hyc | oh, screen wouldn't turn off? | 07:07 |
| hyc | lol | 07:07 |
| emwe | bah, gitorious is a pita in it's commit activities not showing the original authors but just the commiters. | 07:08 |
| xdandroid | New kernel available (20110618_130410) at http://zimages.googlecode.com/files/htc-msm-linux-20110618_130410-package.tar.bz2 | 07:12 |
| hyc | woohoo | 07:12 |
| smiley- | hmm.. I wonder if the X1 can boot on that one | 07:12 |
| hyc | I'll have to rebase on that and pull in JB's new audio stuff | 07:12 |
| *** Detule has joined #xdandroid | 07:14 | |
| phh | [03:24:58] <hyc> where did your busybox build come from? I wonder if I should merge my dmesg patches back into it | 07:15 |
| phh | [03:25:30] <stinebd> phh probably | 07:15 |
| phh | dzo probably :D | 07:15 |
| hyc | ;) | 07:15 |
| hyc | dzo? hangs around still? doesn't seem like it | 07:16 |
| hyc | emwe: you mentioned you've got a working patch for instant cable status? | 07:22 |
| emwe | hyc: it's just a minor thing and i am not sure it's right, but it works on raph/rhod/topa. | 07:22 |
| hyc | I'd like to have that... | 07:23 |
| emwe | i am using the initial/old smem semaphore id from JBs initial work on that which never made it into official patches | 07:23 |
| hyc | ah | 07:23 |
| emwe | i am tempted to push that one of | 07:23 |
| emwe | use 0x91 instead 0x11 as the smem semaphore id | 07:23 |
| hyc | Ah, I saw you pushed that in .35 | 07:23 |
| hyc | yes? | 07:23 |
| emwe | i've beeen using his initial work for a while on .35 and all fine. and whondered why it screwed up with the official patches | 07:23 |
| emwe | yes, .35 has it | 07:23 |
| emwe | i don't care there too much to break stuff because i am the only "user". | 07:24 |
| hyc | ah | 07:24 |
| emwe | well, break is the wrong word. | 07:24 |
| hyc | that .35 still isn't usable on rhod? | 07:24 |
| emwe | you can boot and use a bit. no led and clamshell stuff | 07:25 |
| emwe | no wifi for anything yet. | 07:25 |
| hyc | ah | 07:25 |
| emwe | no bt either. i am lagging forver. | 07:25 |
| hyc | sheesh | 07:25 |
| hyc | not sure whether to spend time on it, or just pull .39 and see where that goes | 07:26 |
| emwe | but it's one thing just working on your own with private/cloned kernel and userland or trying to stick to stock | 07:26 |
| hyc | stock .27 is really out of the question. | 07:26 |
| emwe | yeah, but as you see, it still gets support. | 07:26 |
| hyc | yes... | 07:27 |
| emwe | and jb is still striving for acoustic integration there. | 07:27 |
| hyc | true | 07:27 |
| emwe | totally understandable. nothing works best as .27 still | 07:27 |
| hyc | just wondering how much effort is actually needed to get .35 to equal functionality | 07:27 |
| emwe | depends on how "dirty" you wan't it | 07:27 |
| hyc | .39 seems to be still too fresh | 07:28 |
| emwe | i'd like to have an led abstraction layer/driver still | 07:28 |
| emwe | pita to copy&paste 99% management code for every board. | 07:28 |
| hyc | yeah, the LED code is messy | 07:28 |
| emwe | but micro-klt is not an option | 07:28 |
| emwe | microp | 07:28 |
| hyc | why not? | 07:28 |
| emwe | ten variations and whatnot for the different machine variants | 07:29 |
| hyc | yeah | 07:29 |
| emwe | perhaps i am too easily side trackable and get nothing done by that. | 07:29 |
| hyc | true. it bugs me to see that it registers all that stuff in /sys for LEDs that don't exist on my phone | 07:29 |
| emwe | yap. making userland eventually use the wrong one. | 07:30 |
| hyc | heh | 07:30 |
| emwe | oh man .. froyo still building. | 07:30 |
| emwe | c2d@2.8ghz in this thinkpad used to be fast in the past... | 07:31 |
| hyc | lol | 07:31 |
| smiley- | booted the above kernel on the x1.. everything seems to be working so far ;) | 07:31 |
| hyc | I think that's about what my laptop has | 07:31 |
| hyc | 2.53ghz here :P | 07:32 |
| hyc | nothing in those latest changes would have affected X1 regardless | 07:33 |
| emwe | ok, so wis' pm doesn't employ that ramp-up fix anymore. so neither you nor any .39 user will see it again. going to undo that as it fixes at least rhod400 wake here. | 07:39 |
| hyc | cool | 07:40 |
| emwe | it was initially introduced for some rare devices having issues with a clock change from 19.2 to 528mhz where the latter was the clock on entering suspend. | 07:41 |
| *** helicopter88 has joined #xdandroid | 07:43 | |
| emwe | but can this really happen at all? bah. reverting. one could introduce sth like acpuclk_get_lowest_stepping_frequency() iterating over the boot-constructed stepping table. | 07:43 |
| hyc | seems unlikey yeah | 07:55 |
| hyc | btw, that smd fix I emailed seems to be working well | 07:56 |
| *** valyk has left #xdandroid | 07:56 | |
| emwe | i could pull from your combo? | 07:57 |
| hyc | sure | 07:57 |
| hyc | meantime I'm going to post a note on the forums that the latest autobuild fixes the rhodw gsm sleep bug | 07:59 |
| smiley- | only thing that doesnt seem to work is the battery charge | 07:59 |
| hyc | none of that has been touched | 08:02 |
| xdandroid | New kernel available (20110618_135046) at http://zimages.googlecode.com/files/htc-msm-linux-20110618_135046-package.tar.bz2 | 08:02 |
| hyc | it was the same problem before right? | 08:02 |
| smiley- | I was running another kernel that sp3dev had done some battery charge fixes for | 08:03 |
| hyc | well tell Alex to post his fixes back to the main tree | 08:04 |
| hyc | emwe: did you try that pm ramp-up patch using 122800 instead of 128000 ? | 08:05 |
| hyc | I seem to recall trying both but I don't remember it making any difference | 08:05 |
| hyc | perhaps it didn't affect rhod210 | 08:06 |
| emwe | hyc: if you tell it set a rate it doesn't know about it freaks out. i haven't tested specifically the rhod with 122800. but as topa has 128mhz in it's table and it works there i just "assume" it to be fine. | 08:07 |
| hyc | ahhh | 08:07 |
| hyc | I thought it just refused to set it if it didn't match a known freq | 08:07 |
| emwe | sry yeah, it refuses yes. what i meant | 08:07 |
| emwe | and stays at the freq it is in. which on resume is 19.2mhz | 08:08 |
| hyc | ok | 08:08 |
| hyc | strange that you have 128000 | 08:08 |
| hyc | even G1 has 122800 | 08:08 |
| emwe | topa has 128 | 08:08 |
| emwe | rhodw here has 122800 apart from others. | 08:08 |
| emwe | <6>[ 0.267028] 19200 -1 1 19200 1 30720 -1 4 | 08:09 |
| emwe | <6>[ 0.267089] 120000 1 8 60000 2 61440 -1 5 | 08:09 |
| emwe | <6>[ 0.267150] 122880 0 2 61440 2 61440 -1 4 | 08:09 |
| emwe | <6>[ 0.267211] 176000 2 6 88000 2 61440 -1 6 | 08:09 |
| emwe | and now the RAPH100: | 08:09 |
| emwe | <6>[ 0.277801] 19200 -1 1 19200 1 30720 -1 5 | 08:09 |
| emwe | <6>[ 0.277862] 122880 0 2 61440 2 61440 -1 4 | 08:09 |
| emwe | <6>[ 0.277923] 128000 1 6 64000 2 61440 -1 7 | 08:09 |
| emwe | <6>[ 0.277984] 176000 2 6 88000 2 61440 -1 6 | 08:09 |
| hyc | interesting | 08:10 |
| emwe | so we could actually make 122880 the common denominator | 08:10 |
| emwe | (if it's selected as a stepping frequency on all devices) | 08:10 |
| hyc | right | 08:10 |
| emwe | but let's see how many complaints rise up again. i mean, we've been running without this patch months before it got introduced. | 08:10 |
| emwe | let's just see. | 08:11 |
| hyc | true | 08:11 |
| hyc | my rhod210 matches your rhodw | 08:11 |
| hyc | ok, it's 5am here. going to bed | 08:12 |
| hyc | 'night | 08:12 |
| emwe | night hyc. | 08:12 |
| emwe | ok, that fresh froyo build with power.c present doesn't exhibit the bad behaviour here. | 08:12 |
| emwe | screen going down fine and sleeping. | 08:13 |
| emwe | oh, nice to know your rhod210 matches rhod400 in regards to stepping frequencies. | 08:15 |
| smiley- | hyc: yeah.. seems to be this one.. it's not in the main tree :( https://gitorious.org/~ast/linux-on-qualcomm-s-msm/alex-linux-xperia/blobs/htc-msm-2.6.27/drivers/power/ds2746_battery.c | 08:18 |
| *** Detule has quit IRC | 08:52 | |
| xdandroid | New kernel available (20110618_144657) at http://zimages.googlecode.com/files/htc-msm-linux-20110618_144657-package.tar.bz2 | 08:57 |
| *** rpierce99 has joined #xdandroid | 09:43 | |
| *** rpierce99 has quit IRC | 10:01 | |
| *** Kraln has quit IRC | 10:13 | |
| *** programmer8922 has joined #xdandroid | 10:15 | |
| *** Kraln has joined #xdandroid | 10:17 | |
| *** GPFerror has quit IRC | 10:36 | |
| *** helicopter88 has quit IRC | 11:38 | |
| *** helicopter88 has joined #xdandroid | 12:08 | |
| *** helicopter88 has quit IRC | 13:32 | |
| *** mes has quit IRC | 13:33 | |
| *** helicopter88 has joined #xdandroid | 14:31 | |
| *** Detule has joined #xdandroid | 14:38 | |
| *** helicopter88 has quit IRC | 15:03 | |
| *** JesusFreak316 has joined #xdandroid | 15:47 | |
| *** programmer8922 has quit IRC | 16:09 | |
| *** emwe has quit IRC | 16:36 | |
| *** Detule has quit IRC | 16:45 | |
| *** JesusFreak316 has quit IRC | 17:05 | |
| *** GlemSom has quit IRC | 17:13 | |
| *** Miga_ has joined #xdandroid | 17:44 | |
| *** Miga has quit IRC | 17:47 | |
| stinebd | hyc wake up | 18:42 |
| hyc | no | 18:43 |
| stinebd | you see my whining about tiwlan? | 18:44 |
| stinebd | the wpa driver chokes with the interface renamed | 18:44 |
| hyc | hmm, don't think so, what happened? | 18:44 |
| hyc | oh | 18:44 |
| stinebd | i think we're better off just keeping ti named tiwlan0 | 18:44 |
| hyc | bummer. I didn't see your posts | 18:45 |
| stinebd | you said this worked on a g1? or was it just that you were able to rename the device? | 18:46 |
| hyc | I just renamed the device | 18:46 |
| hyc | lemme boot that up again | 18:46 |
| stinebd | are you selling that g1? | 18:48 |
| hyc | hadn't planned to, no | 18:48 |
| hyc | hmm, I can't get the wlan_loader to run manually on my G1 | 19:07 |
| hyc | and wpa_supplicant startup is all done in libhardware_legacy.so, I can't grab it from init.rc on the G1 | 19:07 |
| stinebd | screw it then | 19:13 |
| stinebd | i wish we could symlink devices | 19:13 |
| hyc | yeah kinda sucks | 19:14 |
| stinebd | you have clout in the kernel world, make it happen | 19:14 |
| hyc | lol | 19:14 |
| hyc | Well, we can still avoid editing init.rc by putting the -Dtiwlan0 -itiwlan0 in wifi.c | 19:14 |
| stinebd | yeah | 19:15 |
| stinebd | dhcpcd needs to be fixed i guess | 19:15 |
| hyc | same solution | 19:15 |
| stinebd | that actually might be automagic | 19:15 |
| stinebd | i forget who spawns that | 19:15 |
| stinebd | netd hopefully | 19:16 |
| stinebd | on the positive side, we don't need to depend on the iputils clone | 19:17 |
| hyc | oh I was looking in the wrong init.rc | 19:17 |
| hyc | cyanogenmod build has init.rc, init.goldfish.rc, init.trout.rc and init.sapphire.rc | 19:17 |
| hyc | perhaps we just need to figure out how to chain init.rc files like they do | 19:18 |
| stinebd | it involves different board names | 19:18 |
| stinebd | on boot init will load init.<hardware>.rc | 19:18 |
| hyc | ah | 19:19 |
| stinebd | goldfish being the emulator, trout g1 i guess, sapphire mytouch or whatever it is | 19:19 |
| hyc | so what are you getting? I get wpa_supplicant exiting with no messages at all | 19:24 |
| stinebd | after successfully renaming tiwlan0, i get Set_key Invalid key or something | 19:25 |
| stinebd | 3 or 4 times | 19:25 |
| stinebd | and then the standard wpa_supplicant driver error message | 19:25 |
| stinebd | Set_key is a function in the tiwlan0 wpa driver | 19:25 |
| hyc | ok | 19:26 |
| stinebd | wpa_printf(MSG_ERROR,"Set_key: Wrong Key\n"); | 19:26 |
| stinebd | that's what i get | 19:26 |
| stinebd | now, that only happens on the first try after booting | 19:26 |
| stinebd | if i try again, it'll go straight to the standard driver error message, without any Set_key error | 19:26 |
| *** Miga__ has joined #xdandroid | 19:37 | |
| *** Miga__ is now known as Miga | 19:37 | |
| *** Miga_ has quit IRC | 19:40 | |
| hyc | oh, I can't edit the G1 rootfs init.rc files. I think it's a squashfs or some other read-only type of filesystem | 19:42 |
| hyc | (I can edit, but on reboot it's just reverted...) | 19:43 |
| *** Detule has joined #xdandroid | 20:12 | |
| *** mes has joined #xdandroid | 20:28 | |
| *** Yusuke has joined #xdandroid | 20:29 | |
| *** Yusuke is now known as Yusuke14 | 20:29 | |
| *** Detule has quit IRC | 20:56 | |
| *** Detule has joined #xdandroid | 21:08 | |
| *** programmer8922 has joined #xdandroid | 21:20 | |
| *** Starfox has joined #xdandroid | 21:45 | |
| *** Starfox is now known as Guest38734 | 21:46 | |
| Guest38734 | Meh, stupid nickserv... | 21:47 |
| *** Guest38734 is now known as xofrats | 21:47 | |
| xofrats | I guess I could pull epic and run it via android... | 21:48 |
| *** xofrats has quit IRC | 21:51 | |
| *** Xofrats has joined #xdandroid | 21:55 | |
| Xofrats | Okay, a bit better | 21:55 |
| *** Xofrats has quit IRC | 22:05 | |
| *** Xofrats has joined #xdandroid | 22:13 | |
| *** [7] has quit IRC | 22:18 | |
| *** TheSeven has joined #xdandroid | 22:19 | |
| *** Xofrats has quit IRC | 22:20 | |
| *** Fudge has quit IRC | 22:23 | |
| *** Fudge has joined #xdandroid | 23:18 | |
| *** Detule has quit IRC | 23:21 | |
| *** Xofrats has joined #xdandroid | 23:34 | |
| *** programmer8922 has quit IRC | 23:34 | |
| *** GPFerror has joined #xdandroid | 23:42 | |
| *** Detule has joined #xdandroid | 23:53 | |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!