00:03.40 | *** join/#htc-linux WisTilt2 (~wisgreg@wireless248.wirelesstcp.net) |
00:18.37 | jonpry | heya WisTilt2 |
00:19.09 | WisTilt2 | hi jonpry whats up |
00:19.39 | jonpry | guess d3tul3 is breaking 30fps in neocore |
00:20.20 | d3tul3 | er |
00:20.25 | jonpry | no particular reason why. with 60hz panel |
00:20.28 | WisTilt2 | cool, what did he do? arrrghhh got 36+ at one point but that was pretty oc'd |
00:21.31 | WisTilt2 | over 30 with 60hz is not really possible i'd think unless something is upping vsync with whatever he's doing |
00:21.32 | d3tul3 | did nothing, happened twice this morning, can't reproduce it....it crossed my mind it might be a weird interaction with some timer issue |
00:22.03 | WisTilt2 | d3tul3 is this on .39 or 3.x? |
00:22.10 | d3tul3 | 3.x |
00:22.12 | jonpry | i got the compositor going. still need moar speed though |
00:22.40 | d3tul3 | oh wait it happened again 33.3 ok now i have my wife here so i can take a photo with her phone |
00:23.17 | WisTilt2 | i took out the fake vsync crap and msm_fb seems happier but without oc'ing and setting vsync to 70hz i have no idea how fps would go up. could be some timer jacking with the calculations for fps in neocore maybe? |
00:23.20 | jonpry | gotta make sure your not hallucinating |
00:24.08 | jonpry | i still do not understand the origin of the refresh/2 limit |
00:25.31 | WisTilt2 | jonpry what kind of frame rate you getting now with compositor working? |
00:26.02 | jonpry | i can't run any of the benchmarks so its all subjective |
00:26.16 | jonpry | i could probably put in something to measure it i guess |
00:26.29 | WisTilt2 | did you ever try pumping up gpu another 50mhz or so? |
00:26.46 | jonpry | its still not quite as fast as the no compositor setup |
00:26.53 | jonpry | no i'm on stock clocks |
00:27.28 | WisTilt2 | simple change in clock-wince if you want to see if it makes any visual difference? |
00:27.33 | *** join/#htc-linux stroughtonsmith (~steven@86-43-208-204-dynamic.b-ras2.bbh.dublin.eircom.net) |
00:28.03 | jonpry | well i think the big killer right now is that surfaces are in non cached memory, where as it used to be cached |
00:28.20 | jonpry | also i don't have copybit |
00:28.37 | WisTilt2 | jonpry: change these 2 lines in clock-wince in grp rail and try it -> |
00:28.39 | WisTilt2 | REG_AND(MSM_GRP_NS_REG, 0x7fff); |
00:28.39 | WisTilt2 | REG_OR(MSM_GRP_NS_REG, 0x1a); |
00:29.06 | WisTilt2 | those values will run gpu at PLL2/4 which should be 1056 if you're not oc'd |
00:29.43 | jonpry | right now gpu is at 196? |
00:29.47 | WisTilt2 | yep |
00:29.57 | WisTilt2 | no |
00:30.12 | WisTilt2 | yes, sorry, thinking memory. |
00:30.49 | jonpry | k. i will try. be a bit. trying to build a 7.1 image |
00:30.52 | WisTilt2 | those changes just set the gpu registers to use PLL2/4. |
00:31.17 | *** join/#htc-linux Cass (~Cass@188-220-34-222.zone11.bethere.co.uk) |
00:31.29 | WisTilt2 | im cleaning my wifes laptop. she seems to have clicked something she shouldn't have and has 16 different trojans on it now:( |
00:31.46 | WisTilt2 | stupid windows 7 crap |
00:31.52 | jonpry | i suspect that little time is spent in the gpu. all getting wasted by the things that render onto surfaces |
00:32.27 | jonpry | yeah gotta get rid of that windows |
00:33.26 | WisTilt2 | ive tried, she's hard headed about going to linux so has to live with all the MS problems |
00:34.28 | jonpry | i pulled the tech support plug. said its either linux or phone support |
00:34.55 | jonpry | i think she's buying a mac now. go figure |
00:36.13 | d3tul3 | http://dl.dropbox.com/u/38520332/IMG_20111210_192347.jpg |
00:36.22 | WisTilt2 | lol i even tried getting her into a mac but you know the old dog new tricks saying |
00:37.05 | WisTilt2 | d3tul3 does it get that fps all the time or is it hit or miss? |
00:37.47 | jonpry | does it have all the stuff? smoke and flames? |
00:38.00 | d3tul3 | yeah all textures |
00:38.07 | d3tul3 | hit or miss |
00:38.22 | d3tul3 | this is the 3rd time it's scored that today probably 20 runs |
00:38.32 | WisTilt2 | sound off also? |
00:38.37 | d3tul3 | yeah sound off |
00:39.16 | WisTilt2 | well 60hz +/- 10% so i guess 33fps would be right in that area. |
00:40.00 | WisTilt2 | just wonder what is letting it go that high. does it test down at 24-25fps at times with this same setup? |
00:40.13 | d3tul3 | 22.5~23 |
00:41.35 | WisTilt2 | d3tul3 what are your 4 PLLx @ lines currently? |
00:41.40 | jonpry | i think it gets stuck in bins |
00:41.51 | jonpry | so refresh/2 or /3 |
00:42.53 | d3tul3 | WisTilt2, sorry rebooting but I am sure the PLLs are stock |
00:43.02 | *** join/#htc-linux mitsutaka (~mitsutaka@123.112.97.210) |
00:49.58 | WisTilt2 | do you guys know if GB actually is setup correctly for enabling BT? i've been back working on that and everything is working on the kernel side afaict but userland never enables it. |
00:53.49 | jonpry | hci0 goes up? |
00:55.39 | jonpry | android/libhardware more or less expect hciattach to never return and for a device names hci0 to go into the UP state within 10 seconds |
00:57.02 | WisTilt2 | yes every up but acts like when userland tries turning it on it never sees that its up |
00:57.23 | jonpry | netcfg shows it in up state? |
00:58.01 | WisTilt2 | using hcitools i can see BT devices so its up i assume |
00:58.48 | jonpry | maybe the dev name is wrong or something? |
00:59.21 | WisTilt2 | yeah thats what im wondering. dont know what name userland expects and i dont have the source for that to look. |
00:59.36 | jonpry | hci0 |
01:00.55 | WisTilt2 | thats what it is. whats puzzling me is the device isnt there until i check it on in userland so its working at that point. |
01:01.25 | jonpry | so userland unchecks it for you eventually? |
01:01.44 | *** join/#htc-linux toer (~tore@179.81-166-86.customer.lyse.net) |
01:02.05 | WisTilt2 | it never get the check mark, says enabling then times out but the kernel does startup the hci0 device when you check it. |
01:03.39 | jonpry | permissions? |
01:03.42 | jonpry | uventd.rc? |
01:04.35 | jonpry | this cm7.1 build came out pretty nice |
01:04.51 | jonpry | its like a few frames away from feeling like a galaxy |
01:05.09 | WisTilt2 | nice |
01:05.35 | WisTilt2 | i guess uventd.rc is not there after boot so need to mount system image? |
01:06.00 | jonpry | it should be |
01:07.06 | WisTilt2 | im adb'd in and its nowhere to be found on the running system |
01:07.15 | WisTilt2 | uventd.rc |
01:08.15 | WisTilt2 | ueventd.rc? |
01:09.29 | WisTilt2 | there is no /dev/hci* in there at all |
01:09.50 | jonpry | even after the device comes up? |
01:11.05 | WisTilt2 | nope, just checked it, userland said turning on bluetooth, the hci0 is there in /dev but ueventd.rc doesn't have it |
01:11.44 | WisTilt2 | looks like it has everything else including camera which i havent used this boot |
01:12.07 | jonpry | what are the perms on /dev/hci0? |
01:12.12 | WisTilt2 | 666 |
01:12.29 | WisTilt2 | i changed those, they were 640 i think |
01:12.50 | WisTilt2 | or 644 they were |
01:13.55 | jonpry | you can't really manually change it. thats what ueventd is for |
01:14.47 | d3tul3 | WisTilt2, apart from enabling BT_LPM are you doing much beyond that in the kernerl |
01:14.48 | WisTilt2 | dont remember where i changed that but it was 644 i believe. either way its showing 666 in /dev |
01:15.02 | d3tul3 | i would like to give it a go but don't want to start reinventing the wheel |
01:15.36 | *** join/#htc-linux asnos_ausente (~Tob@unaffiliated/asnos) |
01:15.44 | WisTilt2 | i enabled several other things BT related in config also. need to diff from autobuild to see which ones, been awhile since i put this aside |
01:15.59 | jonpry | i dunno that /dev/hci0 is actually used. probably just the dbus socket |
01:20.33 | *** join/#htc-linux toastcfh (~toastcfh@41-32.200-68.tampabay.res.rr.com) |
01:20.59 | WisTilt2 | d3tul3: looks like these are the other 2 changes in config -> |
01:21.01 | *** join/#htc-linux toastcfh (~toastcfh@unaffiliated/toastcfh) |
01:21.02 | WisTilt2 | CONFIG_BT_HCIBTSDIO=y |
01:21.02 | WisTilt2 | CONFIG_SERIAL_BCM_BT_LPM=y |
01:22.27 | d3tul3 | alright, you also said you changed the permissions on that bdaddr hook right? |
01:22.34 | WisTilt2 | i have to look at the source and see what else i changed. i did something else but like i said its been awhile. |
01:22.40 | *** join/#htc-linux mastermerlin1 (~Adium@p4FEE4831.dip.t-dialin.net) |
01:22.44 | WisTilt2 | yes bdaddr was what i changed from 644 to 666 |
01:23.03 | WisTilt2 | thanks for reminding me, knew some perms were changed and that was it |
01:23.45 | WisTilt2 | i think you were the one that actually suggested that |
01:27.02 | *** join/#htc-linux toer (~tore@179.81-166-86.customer.lyse.net) |
01:28.49 | WisTilt2 | actually perms were 400 initially |
01:32.18 | jonpry | d3tul3, definitely some serious trouble on the 210 with 3.1 |
01:32.48 | d3tul3 | jonpry, still with the g-sensor? |
01:33.02 | jonpry | don't think so |
01:33.29 | jonpry | i went back to "Disable g-sensor: could be causing...." and its still broke |
01:33.45 | d3tul3 | what's the issue? |
01:34.06 | d3tul3 | i thought you had tried it and it was working fine for you |
01:34.24 | jonpry | the x axis is either all left or all right |
01:35.20 | jonpry | i've gotten it in the middle occasionally |
01:36.46 | WisTilt2 | d3tul3: other thing i changed, which may not matter dont know, is CONFIG_SERIAL_MSM_HS=y |
01:38.47 | d3tul3 | WisTilt2, thanks, bcm_bt_lpm is missing completely in 3.1 |
01:44.37 | d3tul3 | jonpry, no idea, no need to revert all the way, just disabling CONFIG_BMA should not compile the driver at all....didn't 3.1 work for you at one point in time? |
01:45.57 | jonpry | yeah it was fine |
01:48.15 | jonpry | the broken kernels are being built on a different machine |
01:48.37 | jonpry | still doesn't work at upgrade to 3.1.4 |
01:49.07 | jonpry | before i run calibration, there are both x and y coord. but after i only get y's |
01:50.11 | d3tul3 | different toolchain? |
01:50.18 | jonpry | yes |
01:50.37 | d3tul3 | the new codesourcery toolchains do something funny with software alignment |
01:50.42 | d3tul3 | dur |
01:50.44 | d3tul3 | memory alignment |
01:50.52 | jonpry | just sounds like such a strange thing to happen |
01:51.16 | d3tul3 | need a config patch, though it's something to explore for the future, like we can have some sort of hardware alignment |
01:51.30 | d3tul3 | alignment_Trap i think the config otpion is called |
01:51.41 | jonpry | hrm. and that will fix it? |
01:51.42 | d3tul3 | supposedly reduces overhread |
01:51.56 | *** join/#htc-linux snadge (~snadge@unaffiliated/snadge) |
01:52.14 | d3tul3 | well i tried the patch once and built with the same toolchain and had some strange things happen to my home screen |
01:52.23 | d3tul3 | same=new |
01:52.56 | snadge | im working on a vision ICS build, and ive merged the EGL workarounds and using the msm7k libgralloc |
01:53.28 | snadge | but i cant get it to boot unless i push a libhwcomposer from another build.. i cant figure out what builds hardware qcom display libhwcomposer.. |
01:54.47 | jonpry | d3tul3, that sound like what is happening to me. only i don't have the config option |
01:55.50 | jonpry | snadge, i had to use the null composer thing in libhardware. just built it with mm. make didn't seem to do it automatically |
01:56.01 | snadge | Android.mk has ifeq BOARD_USES_QCOM_HARDWARE .. which is defined in BoardConfig.mk |
01:56.15 | snadge | ahh |
01:56.29 | jonpry | hwcomposer.default.so is not from msm7k |
01:56.39 | d3tul3 | jonpry, let me find you the chat log i had with emwe |
01:57.29 | snadge | so i just want the null composer that makes since.. since i dont want to be using it yet, im using hax / |
01:57.31 | d3tul3 | jonpry, http://xdandroid.com/irclogs/%23xdandroid.20111006.log.html |
01:57.40 | jonpry | thanks |
02:00.18 | snadge | Install: out/target/product/vision/system/lib/hw/hwcomposer.default.so ... FFFFFuuuu ;) |
02:00.21 | snadge | thanks jonpry |
02:01.34 | snadge | so thats what the problem has been all along.. been on this for a few days now |
02:04.42 | *** join/#htc-linux asnos (~Tob@unaffiliated/asnos) |
02:13.30 | *** join/#htc-linux programmer8922 (~Evan@c-174-56-72-235.hsd1.nm.comcast.net) |
02:15.49 | snadge | you cant build libhardware until after you've build the .zip ? |
02:27.26 | jonpry | probably not |
02:33.42 | snadge | how do i hack hwcomposer.default.so into the build |
02:34.21 | snadge | so you just manually copy it in after its built? |
02:35.11 | jonpry | yeah i don't mess with those filesystem images |
02:35.20 | jonpry | and never to make clean |
02:35.27 | jonpry | er do |
02:50.35 | snadge | you can add to your device/vendor/model.mk |
02:51.29 | snadge | PRODUCT_PACHKES += libhwcomposer.default \ |
02:51.32 | snadge | <PROTECTED> |
02:51.51 | snadge | err PRODUCT_PACKAGES .. you get what i mean ;) |
02:53.25 | jonpry | ah. makes sense |
02:54.15 | jonpry | i dunno what source your working from but there is also a really nasty bug in that composer |
02:54.37 | snadge | i will check |
02:55.11 | snadge | https://github.com/Andromadus/android_hardware_libhardware |
02:55.36 | snadge | looks old |
02:56.06 | jonpry | https://github.com/Andromadus/android_hardware_libhardware/blob/ics/modules/hwcomposer/hwcomposer.cpp line 89 |
02:56.20 | jonpry | dpy and or sur can and will be null sometimes |
02:56.38 | jonpry | depending on the renderer, this will send it sky high |
02:57.12 | snadge | ok so test for null and not call swap buffers? |
02:57.45 | jonpry | right |
03:03.05 | snadge | <PROTECTED> |
03:03.06 | snadge | <PROTECTED> |
03:03.44 | snadge | something like that? |
03:03.54 | jonpry | that'll do it |
03:04.18 | jonpry | some people are working on a copybit based thing for msm8k |
03:04.34 | jonpry | not sure if 7230 has c2d |
03:05.44 | snadge | im working on vision atm.. other people have semi working builds with egl workarounds |
03:06.06 | snadge | the repo im working with is cm based and public tho |
03:07.08 | jonpry | vision is 7230 right? |
03:08.20 | jonpry | i've been using straight cm9 repos. but we have lots of trouble with ics on these old 7201's |
03:19.06 | snadge | yeah i just thought i'd help out, out of curiosity.. im learning a LOT as im doing this |
03:19.20 | snadge | but because of that, unfortunately im not a huge help |
03:19.49 | snadge | just assisting with others efforts really.. trying to get something semi useable and buildable.. so other people can look and fix bugs, without having to port half the device first |
03:23.41 | jonpry | ics is a tough one. lots of strange changes |
03:24.07 | jonpry | just had his phone enter hyperspeed |
03:24.25 | jonpry | might as well rename it the touch pro 3 |
03:24.48 | rpierce99 | liking your new mod a little better now? |
03:25.13 | jonpry | yep added another patch. and now it puts the old stuff to shame |
03:25.25 | jonpry | cm7 way faster than gbx0a |
03:25.54 | rpierce99 | because of your magic or just in general? |
03:26.04 | jonpry | magic |
03:26.10 | jonpry | and more magic |
03:26.14 | jonpry | laced with magic |
03:26.25 | rpierce99 | with magic sprinkles |
03:26.32 | jonpry | that too |
03:31.12 | *** join/#htc-linux rzk (~rzk@93-80-178-77.broadband.corbina.ru) |
03:31.39 | *** join/#htc-linux mitsutak_ (~mitsutaka@123.112.97.210) |
03:49.15 | snadge | how do i get it to rebuilt the image files ? |
03:49.35 | snadge | never mind it worked :) |
03:49.45 | snadge | delete system.img |
04:06.26 | *** join/#htc-linux raymonddull (~raymonddu@c-24-11-68-98.hsd1.mi.comcast.net) |
04:19.45 | snadge | hmm.. my gralloc.msm7x30 is half the size of the one which works |
04:19.53 | snadge | i just kanged it from CM repo |
04:25.42 | *** join/#htc-linux Rob2222 (~Miranda@p4FFF2D20.dip.t-dialin.net) |
04:31.02 | snadge | jonpry: are you using a modded hardware/msm7k/libgralloc ? |
04:48.20 | *** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net) |
04:53.39 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
04:58.10 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-62.client.mchsi.com) |
04:58.15 | *** join/#htc-linux BHSPitMonkey (~stephen@unaffiliated/bhspitmonkey) |
05:55.39 | arrrghhh | WisTilt2, you thar? |
05:56.07 | WisTilt2 | hi arrrghhh, just watching a dumb movie with the wife |
05:56.12 | arrrghhh | lol |
05:56.21 | arrrghhh | sounds... fun? |
05:56.42 | WisTilt2 | oh yeah. something about ice tornadoes |
05:56.55 | arrrghhh | just got back from our company holiday party. skipped out just before the b-52's went on stage. |
05:57.04 | arrrghhh | dunno who the hell voted for them, but they can suck it. |
05:57.15 | WisTilt2 | lol |
05:57.36 | arrrghhh | well if you're busy i won't bug ya. |
05:57.54 | arrrghhh | just wanted to see where you were at and add some context to BT if you needs it |
05:58.25 | WisTilt2 | no fire away. movie is over thank god |
05:58.32 | arrrghhh | heh |
05:58.41 | arrrghhh | AFAIK BT has worked for a while in GB |
05:58.53 | arrrghhh | but audio routing was kinda flaky w/o acoustic |
05:59.11 | arrrghhh | emwe enabled BT in .35 and ran into an interesting bug where the phone wouldn't sleep unless BT was on. |
05:59.28 | arrrghhh | i think he said his headset would pair & work fine, but he couldn't xfer any files. |
05:59.42 | arrrghhh | i had a similar experience. i could pair with another data device, but when it came down to xfer it wouldn't budge. |
06:00.22 | arrrghhh | there's also some odd ways WinMo bleeds over to Android with BT |
06:00.33 | arrrghhh | like if you leave BT on in WinMo, it freaks out Android... not sure if you were aware of that bug. |
06:00.49 | arrrghhh | i think we discussed it. can't remember who i've talked with what about anymore :P |
06:01.13 | WisTilt2 | i never had the prob with sleeping. i had it working on .39 with froyo but no audio of course due to acoustic in the kernel but in GB it wont enable so i put it aside for the night. |
06:01.23 | arrrghhh | hum |
06:01.40 | arrrghhh | i'm sure there were changes between Froyo and GB. |
06:01.58 | arrrghhh | what, I'm not sure. emwe would be able to answer that one better than I. |
06:02.30 | WisTilt2 | its something simple im sure and has to be kernel related, just need to go through it again and find why userland isnt seeing it come up. |
06:02.39 | arrrghhh | hrm |
06:02.50 | arrrghhh | well autobuild .35 BT works I'm almost 99.9% sure. |
06:03.02 | arrrghhh | just that odd bug I mentioned earlier is still outstanding. |
06:03.40 | WisTilt2 | yeah i tried alex's tree and it works, cant remember when i tried that if it was with froyo or gb though, that was couple months ago. |
06:03.50 | arrrghhh | heh |
06:03.53 | arrrghhh | it's all a wash. |
06:04.12 | arrrghhh | it does work on GB tho. emwe cranked on that for a little while. |
06:04.25 | WisTilt2 | yeah we'll get it working. there's not a whole lot in .39 that isnt working so coming down the home stretch. |
06:04.38 | arrrghhh | i'm pretty sure he said it works really well with his headset. just not with another phone/PC for xfer. |
06:04.46 | arrrghhh | yea that's sweet :D |
06:05.20 | arrrghhh | i wish jb had time to come back and play with this stuff |
06:05.30 | WisTilt2 | i didnt much care about BT before but my new truck has it and its pretty nice but i have to use *cough* winmo to use it now |
06:06.13 | arrrghhh | HEATHEN! |
06:06.15 | arrrghhh | :P |
06:07.13 | WisTilt2 | that brings back memories of the old show Sanford & Son lol |
06:07.35 | arrrghhh | can't say i've seen many episodes of that show... if any. |
06:08.08 | WisTilt2 | yeah many decades ago, back when tv was simple |
06:08.38 | arrrghhh | hehe. i don't even really have tv service, but i watch tv shows... |
06:08.53 | arrrghhh | i get some of the OTA channels piped thru with my cable interwebs. |
06:09.47 | arrrghhh | so what's new with .39? |
06:09.48 | WisTilt2 | directv here for 18+ yrs. cant live without espn and a few other channels |
06:10.20 | arrrghhh | seems like that mystery has gotten more convoluted... |
06:10.24 | jonpry | snadge, mine is way different. i think the usual strategy is to use gingerbread gralloc except for framebuffer.cpp |
06:10.34 | arrrghhh | i wish you could buy channels a la carte |
06:11.01 | arrrghhh | retarded i have to buy a package with 200 channels when i maybe watch 5 main channels. might order 15-20 at most lol. |
06:11.17 | WisTilt2 | me too. pay for hundreds of channels you never watch |
06:11.19 | jonpry | WisTilt2, i got hyperspeed |
06:11.25 | arrrghhh | jonpry, do tell |
06:11.29 | WisTilt2 | ? ludicrous speed? |
06:11.32 | jonpry | its like touchpad |
06:11.33 | arrrghhh | lol |
06:11.38 | jonpry | for sure |
06:11.38 | arrrghhh | you tease |
06:11.43 | jonpry | touch pro 3 |
06:11.52 | jonpry | no joke |
06:11.57 | WisTilt2 | tp3 i like that. what'd you do? |
06:12.12 | arrrghhh | jonpry, they made that device. t'was a flop. |
06:12.20 | jonpry | my compositor is just that good |
06:12.40 | WisTilt2 | is this on GB? |
06:13.06 | jonpry | yeah cm7.1 |
06:13.10 | arrrghhh | oO |
06:13.15 | arrrghhh | you haz 7.1 singing? |
06:13.21 | arrrghhh | gimmie |
06:13.23 | arrrghhh | :D |
06:13.30 | jonpry | i think i can get a little more speed by getting rid of the debug messages |
06:13.48 | jonpry | and i'm running with no OC of any kind |
06:14.31 | jonpry | i'll image it. need to make a change to the init |
06:15.12 | arrrghhh | cool |
06:16.07 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
06:16.37 | WisTilt2 | what kernel you running? |
06:16.43 | jonpry | 3.1 |
06:16.53 | WisTilt2 | damn |
06:17.21 | jonpry | why's that? |
06:17.41 | WisTilt2 | we're going to have to retire .39:( |
06:17.44 | arrrghhh | bleeding edge ftw |
06:17.46 | arrrghhh | hahahaha |
06:17.52 | arrrghhh | WisTilt2, it will live on |
06:18.22 | jonpry | well. i think 3.1 is a continuation of that effort |
06:18.31 | arrrghhh | jonpry, seriously if you got CM7.1 to be fast, I'm sure other stripped builds are going to scream. |
06:18.37 | arrrghhh | indeed |
06:19.05 | jonpry | latest kernel without simultaneous upgrade of random drivers |
06:19.34 | jonpry | that was the goal. and we need to keep it alive. as soon as its possible to abandon 39. we should do it |
06:20.18 | WisTilt2 | is 3.1 fairly up to date with all the .39 changes? i know detule has been working it over quite a bit |
06:20.43 | jonpry | i'm not 100% convinced it is stable. but detule has been doing a good job of keeping it up to date |
06:21.52 | jonpry | arrrghhh, who needs stripped builds. we have touch pro 3 |
06:21.55 | WisTilt2 | tomorrow ill boot it up and give it a whirl. |
06:22.33 | arrrghhh | jonpry, lol |
06:22.44 | arrrghhh | you are a tease. |
06:22.59 | jonpry | image in a few |
06:23.04 | arrrghhh | :D |
06:23.05 | arrrghhh | awesome |
06:23.14 | WisTilt2 | btw, ill have a rhod400 in my hands monday |
06:24.32 | WisTilt2 | not that i wont be sending kernels your way arrrghhh, just will make it easier to work out some of these differences quicker. |
06:24.46 | arrrghhh | haha |
06:24.54 | arrrghhh | np, i'm sure it'll make it a lot easier on you. |
06:25.16 | arrrghhh | i know when i was trying 'help' jonpry with building ICS i was just severely slowing him down lol |
06:25.45 | jonpry | still gotta get this compositor going on ics |
06:26.20 | snadge | im still trying to buypass the compositor ;) |
06:26.33 | snadge | and you were right.. it is libgralloc-qsd8k that it needs to be linked to |
06:27.53 | jonpry | even if you bypass the hwcomposer thing. you still have a way better compositor than we do |
06:28.14 | arrrghhh | ICS seems pretty gfx intensive to put it lightly. |
06:29.23 | jonpry | and floating point |
06:29.34 | arrrghhh | heh |
06:30.03 | jonpry | i think zerofree is the key to images |
06:42.20 | jonpry | uploading... |
06:43.26 | arrrghhh | sweet |
06:49.08 | jonpry | http://www.megaupload.com/?d=M23I8OK7 |
06:49.14 | jonpry | decompress and dd to card |
06:49.31 | jonpry | don't let it sleep |
06:49.49 | arrrghhh | don't let it sleep |
06:49.50 | arrrghhh | lol |
06:50.06 | jonpry | sometimes screen does not refresh properly |
06:50.18 | arrrghhh | so it just dies in sleep? |
06:50.31 | jonpry | it will restart flinger |
06:50.35 | arrrghhh | ah |
06:50.39 | WisTilt2 | so this needs to go on sd card by itself? |
06:50.41 | arrrghhh | minor details. |
06:50.48 | jonpry | yeah |
06:51.02 | arrrghhh | dd it to a fat32 partition...? |
06:51.09 | arrrghhh | er |
06:51.13 | arrrghhh | dd it to an entire card |
06:51.17 | jonpry | yup |
06:51.21 | arrrghhh | i see |
06:51.26 | jonpry | its 2 partitions |
06:51.30 | arrrghhh | ah right |
06:51.32 | arrrghhh | like ICS |
06:51.42 | WisTilt2 | i dont have my other cards here so will have to try the magic tomorrow |
06:51.44 | jonpry | yup |
06:51.46 | arrrghhh | i can use a 512 mb card? or is that too small |
06:51.50 | jonpry | 1gb |
06:52.06 | arrrghhh | drat. i have a 2gb layin around too somewheres... |
06:52.09 | WisTilt2 | i guess i can copy my current one and clear it. this one is 8gb |
06:52.24 | jonpry | it would fit on 512 but things would need to get resized |
06:52.30 | arrrghhh | np |
06:53.01 | jonpry | 8gb is fine. my card is 8gb |
06:53.36 | WisTilt2 | so just format fat32 the dd it thats it? |
06:53.43 | WisTilt2 | *then |
06:54.18 | jonpry | no formatting |
06:54.37 | jonpry | dd to the root device. image includes the partition table |
06:55.08 | WisTilt2 | k. need to copy my stuff off first... |
06:55.41 | jonpry | i went for maximum easy install |
06:56.18 | jonpry | there are a lot of bugs. i think most of them are not real big problems |
06:56.33 | jonpry | will take some work to fix though |
07:15.35 | arrrghhh | crap |
07:15.44 | arrrghhh | i did it to /dev/sdf1 instead of /dev/sdf. |
07:15.50 | arrrghhh | let's try that again... |
07:15.53 | WisTilt2 | lol noob |
07:15.56 | WisTilt2 | :) |
07:16.01 | arrrghhh | no joke :P |
07:16.06 | jonpry | durp |
07:16.22 | arrrghhh | at least i was able to identify my failure haha |
07:16.30 | WisTilt2 | i just started dd a couple mins ago, took forever to copy off my 6.4gb of stuff on the card |
07:16.34 | arrrghhh | about my only saving grace usually |
07:16.54 | arrrghhh | it was done then i thought about it, and looked at fdisk again and facepalmed. |
07:17.06 | jonpry | not double i hope |
07:17.42 | arrrghhh | eh, i'm not dd'ing stuff daily. i wish i got to work more with @ work. |
07:17.50 | arrrghhh | jeez |
07:17.53 | arrrghhh | i missed 'linux' |
07:18.01 | arrrghhh | i wish i got to work more with linux @ work. yeesh. |
07:18.59 | jonpry | i thought you meant dd |
07:19.05 | arrrghhh | hah |
07:19.11 | jonpry | you are familiar with this? http://www.allamericanblogger.com/wp-content/uploads/double-facepalm1.jpg |
07:19.16 | arrrghhh | nah, we only have a few linux servers in my dept |
07:19.32 | arrrghhh | most are "appliances" which means no root access. |
07:19.44 | arrrghhh | lol, yea i am aware of the double facepalm :P |
07:21.19 | WisTilt2 | jonpry did you stick a SSD in that new laptop? |
07:21.58 | jonpry | no. i'm not sure if this one has a slot. other one did |
07:22.08 | jonpry | got 7200 rpm drive though |
07:23.13 | WisTilt2 | i have a 256gb ssd along with the 500gb mechanical for mass storage. linux is so fast on a ssd with an i7 |
07:23.25 | arrrghhh | jonpry, the first partition is FAT16? |
07:23.55 | jonpry | could be |
07:24.00 | arrrghhh | lol |
07:24.01 | jonpry | like 68mb or something |
07:24.07 | arrrghhh | oh well. giving it a whirl |
07:24.16 | WisTilt2 | so we just hit haret and thats it? |
07:24.34 | jonpry | yup |
07:24.41 | WisTilt2 | wow that haret boot was way faster than normal |
07:25.13 | jonpry | no boot animation |
07:26.09 | jonpry | so long wait. then you gotta hurry up and unlock it. but things are real sketchy. because it hasn't settled yet. and i think it switched orientation to landscape and then back again for no particular reason |
07:26.24 | WisTilt2 | i mean haret itself. bar went by so fast then scrolling text |
07:26.45 | jonpry | hmm |
07:26.53 | jonpry | i find the speed is related to the card |
07:27.16 | WisTilt2 | same card. seemed about 4 times quicker |
07:27.44 | jonpry | it might be best to let it crash and then unlock it on the second time. then it won't do that crazy orientation stuff |
07:28.36 | WisTilt2 | whoa |
07:29.44 | WisTilt2 | so this is in landscape only or did i mess it up? |
07:30.06 | jonpry | just hit back button |
07:30.19 | jonpry | it takes a few seconds to switch then |
07:30.34 | WisTilt2 | ah, that did it. ok this is nice. definitely ludicrous speed hehe |
07:31.07 | jonpry | coming apart at the seems |
07:31.11 | arrrghhh | lol |
07:31.44 | WisTilt2 | opps i hit end key for home by habit and i probably need to pull battery now |
07:31.54 | arrrghhh | aw |
07:32.13 | arrrghhh | that's a pretty crippling bug, sad. |
07:32.45 | jonpry | its just all kinds of serious bugs |
07:32.59 | jonpry | need to find out if they are worth fixing |
07:33.14 | arrrghhh | heh |
07:33.37 | jonpry | i think its pretty sweet |
07:33.39 | arrrghhh | ludicrous speed sounds worth fixing. |
07:34.07 | jonpry | you get yours going arrrghhh? |
07:34.11 | WisTilt2 | its peppy ill say that. rebooting and making note not to hit end this time |
07:34.30 | arrrghhh | jonpry, literally just clicked haret. had to flash back to winmo lol :P |
07:34.55 | WisTilt2 | jonpry haret it actually around 10 times faster than normal |
07:34.57 | jonpry | ugh nand |
07:35.13 | WisTilt2 | blue bar goes by so fast you almost dont see it |
07:35.35 | jonpry | must be my super awesome formatting |
07:35.51 | arrrghhh | jonpry, you love nand |
07:35.51 | WisTilt2 | yeah its like the sd card is on steroids now |
07:35.57 | arrrghhh | what DPI is this? seems tiny |
07:36.21 | jonpry | ldpi |
07:37.24 | arrrghhh | woah |
07:37.29 | arrrghhh | lots of wacky artifacts for sure. |
07:37.33 | arrrghhh | if that's what you call 'em. |
07:38.04 | jonpry | depends on what you are doing. there are like 3 issues i know about |
07:38.49 | arrrghhh | well notification bar kinda freaked out |
07:39.05 | jonpry | yeah. it uses a special method i have disabled |
07:39.47 | arrrghhh | swiping screens seems to create visual artifacts too |
07:39.57 | jonpry | you can bring it down. key is to press on it for 5 seconds or so |
07:40.02 | jonpry | or i don't have that |
07:40.08 | jonpry | er oh |
07:40.51 | jonpry | the status bar flashes while swiping. but that is just the normal problem with status bar |
07:41.04 | arrrghhh | lol |
07:41.16 | arrrghhh | this RIL/build seems to be GSM only as well. |
07:41.25 | jonpry | talking about artifacts elseware while swiping? |
07:41.26 | arrrghhh | i guess that shouldn't surprise me :P |
07:41.33 | arrrghhh | hrm |
07:41.35 | jonpry | i think it has no ril |
07:41.40 | arrrghhh | they were in the status bar and on the desktop as well. |
07:42.03 | jonpry | i don't have any except in the status bar |
07:42.05 | WisTilt2 | yeah no service here on gsm but everything is very smooth. i have no artifacts yet here |
07:42.07 | arrrghhh | yea you can see these lines flickering in the top dots |
07:42.17 | arrrghhh | no artifacts? |
07:42.18 | arrrghhh | hum. |
07:42.23 | snadge | jonpry: what version of libgralloc-qsd8k are you using? |
07:42.27 | jonpry | yeah top dots |
07:42.34 | snadge | the ones im using come up with an error about 3 buffers or something.. and 0hz |
07:42.46 | snadge | from CM |
07:42.47 | jonpry | snadge we don't have snapdragon |
07:43.06 | jonpry | plain libgralloc |
07:43.20 | arrrghhh | WisTilt2, can you try to move an icon/widget around on the desktop? |
07:43.25 | arrrghhh | it's... let's just say broken for me. |
07:43.32 | jonpry | that will make stuff happen |
07:43.45 | snadge | oh okay.. therse a port for Desire S that has a gralloc.msm7x30 that works |
07:43.50 | snadge | but its 3kb smaller than the one im building |
07:43.55 | jonpry | its just the erase thing is not quite the right size |
07:43.59 | snadge | i'd *love* to know where they got that from |
07:44.22 | snadge | i can just overwrite my non working version with that one.. but thats not the point ;) |
07:44.23 | jonpry | probably cm7.2 |
07:44.25 | jonpry | er 7.2 |
07:44.33 | jonpry | 7.1 |
07:44.55 | snadge | which wont build against an ICS tree right? |
07:45.05 | snadge | eg.. copying that lib in there is unlikely to just work |
07:45.14 | arrrghhh | jonpry, this says it's 7.2 RC0 |
07:45.20 | arrrghhh | from ze Hero |
07:45.52 | jonpry | hmm |
07:46.16 | arrrghhh | in about phone |
07:46.31 | jonpry | ic |
07:46.41 | arrrghhh | nbd, still a cool preview. thx jonpry, gonna hit the sack. talk to you guys later. |
07:46.44 | jonpry | i pulled today. they must be up to something :p |
07:46.49 | arrrghhh | hah |
07:47.10 | jonpry | thanks arrrghhh . good night |
07:47.24 | jonpry | snadge, it will build fine |
07:47.47 | jonpry | just have to patch framebuffer.cpp because ics passes refresh rate = 0 |
07:47.59 | jonpry | which causes a divide by zero in unpatched gralloc |
07:48.08 | WisTilt2 | this is pretty nice. i did get the lines at the top when dragging icons but everything else is looking good. |
07:48.22 | snadge | ahh cheers.. i'll try that :) |
07:49.22 | jonpry | WisTilt2, thanks. the fixes shouldn't really change performance. just need to make sure this is actually going to be fast enough before trying to polish it all up |
07:49.51 | WisTilt2 | its fast for sure. i like this resolution also, looks real clean |
07:49.58 | jonpry | don't need general pmem pool |
07:50.34 | snadge | wtf does this code mean |
07:50.44 | snadge | int refreshRate = 1000000000000000LLU / |
07:50.56 | snadge | ( ... whole heap of crap ... ); |
07:50.57 | snadge | huh ? |
07:51.11 | jonpry | look at how they do it in hardware/libhardware |
07:51.20 | jonpry | there is a new variable. refreshRateQuotient |
07:52.19 | snadge | pixclock must be 0? |
07:52.45 | jonpry | probably |
07:53.13 | snadge | i'll just comment that out |
07:53.39 | jonpry | WisTilt2, ldpi fan? |
07:54.06 | WisTilt2 | yes. nice seeing more stuff on things like settings, folders, etc |
07:54.53 | jonpry | isn't that how froyo used to be? |
07:55.10 | WisTilt2 | not like this |
07:55.45 | jonpry | like how you can how fast its rendering in the status bar? |
07:56.11 | WisTilt2 | by the flickering? |
07:56.17 | snadge | jonpry: what about the error 3 buffers thing ? |
07:56.17 | jonpry | yeah |
07:56.43 | jonpry | not sure what that means |
07:56.59 | snadge | error: 'GRALLOC_MODULE_PERFORM_CREATE_HANDLE_FROM_BUFFER' |
07:57.11 | snadge | so thats why cms one was forked ;) |
07:57.40 | jonpry | haven't seen that error before |
07:58.02 | jonpry | dequeue native buffer and black screen is common |
08:00.25 | jonpry | i gotta sleep. good night |
08:00.47 | WisTilt2 | jonpry: nice job. me too, ttyl |
08:09.35 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-91-189.dynamic.mnet-online.de) |
08:30.38 | *** join/#htc-linux LjL (~ljl@unaffiliated/ljl) |
08:56.54 | *** join/#htc-linux rzk (~rzk@93-80-178-77.broadband.corbina.ru) |
09:34.14 | *** join/#htc-linux bukington (~bukington@fac34-2-82-228-151-145.fbx.proxad.net) |
09:56.04 | *** join/#htc-linux rob_w (~bob@ppp-93-104-166-218.dynamic.mnet-online.de) |
09:56.04 | *** join/#htc-linux rob_w (~bob@unaffiliated/rob-w/x-1112029) |
10:11.04 | *** join/#htc-linux helicopter88 (~helicopte@95.234.35.222) |
10:36.47 | *** join/#htc-linux GNUtoo (~gnutoo@host1-132-dynamic.51-79-r.retail.telecomitalia.it) |
10:50.27 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-91-189.dynamic.mnet-online.de) |
11:03.45 | *** join/#htc-linux friehmaen (~fm@fm.xers.de) |
11:39.48 | *** join/#htc-linux ElFinLazz (~elfinlazz@112.154.189.103) |
12:15.23 | *** join/#htc-linux mastermerlin (~Adium@p4FEE4831.dip.t-dialin.net) |
12:19.41 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:5c31:34c:50f9:c8b1) |
12:34.03 | *** join/#htc-linux LordDeath (~LordDeath@cable-81-173-164-253.netcologne.de) |
12:49.46 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-91-189.dynamic.mnet-online.de) |
14:19.13 | *** join/#htc-linux asnos_ausente (~AndChat@unaffiliated/asnos) |
14:20.28 | *** join/#htc-linux iamjw (~pure@219.134.169.85) |
14:50.29 | *** join/#htc-linux asnos (~Tob@unaffiliated/asnos) |
15:16.16 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com) |
15:38.06 | *** join/#htc-linux d3tul3 (~oliver@unaffiliated/d3tul3) |
15:50.07 | *** join/#htc-linux bukington (~bukington@fac34-2-82-228-151-145.fbx.proxad.net) |
16:09.44 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
16:18.39 | *** join/#htc-linux Echo31 (~olivier@lan31-9-88-177-158-48.fbx.proxad.net) |
16:33.13 | *** join/#htc-linux jonpry (~jon@c-24-17-200-206.hsd1.wa.comcast.net) |
16:34.33 | *** join/#htc-linux ftoz (~root@cst-prg-168-12.cust.vodafone.cz) |
16:54.22 | *** join/#htc-linux d3tul3 (~oliver@unaffiliated/d3tul3) |
16:54.30 | *** join/#htc-linux oliver__ (~oliver@pool-96-234-141-27.bltmmd.east.verizon.net) |
16:57.00 | *** join/#htc-linux raymonddull (~raymonddu@c-24-11-68-98.hsd1.mi.comcast.net) |
17:45.53 | *** join/#htc-linux asnos_ausente (~Tob@unaffiliated/asnos) |
17:54.55 | *** join/#htc-linux rob_w (~bob@ppp-93-104-166-218.dynamic.mnet-online.de) |
17:54.56 | *** join/#htc-linux rob_w (~bob@unaffiliated/rob-w/x-1112029) |
18:01.09 | *** join/#htc-linux LjL (~ljl@unaffiliated/ljl) |
18:03.14 | *** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net) |
18:07.46 | d3tul3 | jonpry you have the new toolchain going? |
18:14.41 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-91-189.dynamic.mnet-online.de) |
18:21.39 | *** join/#htc-linux LjL (~ljl@unaffiliated/ljl) |
18:32.56 | *** join/#htc-linux stroughtonsmith (~steven@86-43-208-204-dynamic.b-ras2.bbh.dublin.eircom.net) |
18:42.45 | jonpry | d3tul3, no i switched to 2010q3 and everything works again |
18:48.13 | *** join/#htc-linux [acl] (~abel@cpe-69-203-141-229.si.res.rr.com) |
18:48.26 | [acl] | who is this animal who emailed about 3.5mm |
18:48.54 | [acl] | damn.. i see how its gonna be ... |
18:49.02 | jonpry | hi [acl] |
18:49.04 | d3tul3 | [acl] hey |
18:49.07 | [acl] | hey dood |
18:49.09 | rpierce99 | i wondered the same thing |
18:49.23 | d3tul3 | you try that camera patch? |
18:49.26 | [acl] | jonpry: dood quicik question. Is this scbs part of .35 ? i never understood the whole thing |
18:49.36 | jonpry | must be the 3rd user of nand |
18:49.49 | [acl] | jonpry: doubt it since that patch isnt on nand |
18:50.00 | d3tul3 | scbs is not in .35 |
18:50.04 | [acl] | ahh ok |
18:50.05 | jonpry | [acl], no. there is patches for 27. and i ported it to 39/3.x |
18:50.09 | [acl] | thanks for clearing it up |
18:50.28 | [acl] | detule: patch for cam ? sorry i havent had time. |
18:50.34 | jonpry | i got hyperspeed cm7.1 |
18:50.40 | [acl] | sexy |
18:50.44 | [acl] | cuz of the GL ? |
18:50.47 | rpierce99 | d3tul3: did you ever look into the scbs freeze when plugging in while sleeping thing? |
18:50.50 | jonpry | yeah |
18:51.01 | [acl] | yeah i saw the patch but didnt realy look |
18:51.02 | [acl] | good work |
18:51.29 | jonpry | lots of bugs. not really ready for primetime |
18:51.47 | d3tul3 | rpierce99 haven't had a chance |
18:52.07 | rpierce99 | is that only on rhodw? |
18:52.44 | [acl] | i'm confused a bit by thie egl patch that was email tho |
18:52.54 | [acl] | all i see is some LogD's and a null returned |
18:53.03 | [acl] | guess so we can report back whats available ? |
18:53.31 | d3tul3 | yeah it was so that we can see which extensions are supported by the vendor blob |
18:53.40 | [acl] | fucking qcom |
18:53.53 | [acl] | the winmo driver has more extensions i think |
18:53.59 | jonpry | the return null is what makes it go |
18:54.01 | [acl] | actually im sure it does |
18:54.17 | jonpry | vial? |
18:54.22 | [acl] | i always supported the idea |
18:54.27 | [acl] | plus vial the damn ril too :-p |
18:54.35 | [acl] | haha |
18:54.48 | [acl] | this is just sad. we were gayed out of some good functions |
18:54.48 | jonpry | lol |
18:54.55 | jonpry | what does it have? |
18:55.29 | [acl] | forgot. but i remember being able to use the driver directly to draw on screen. while on android you cant do some shit. this was from early work back in the good ol days when 3d didnt work |
18:55.39 | jonpry | d3tul3 pointed out the other day that our blob actually does support frame buffer objects |
18:55.48 | jonpry | but they have ARB on the end of the function names |
18:56.10 | [acl] | really |
18:56.12 | [acl] | hmm |
18:56.15 | jonpry | we can directly draw on screen |
18:56.42 | jonpry | i'm doing that with compositor. not quite finished yet |
18:56.55 | [acl] | damn with FBOs all those small hacks would natively work |
18:57.42 | [acl] | well hope it all goes well. and im shocked to be honest |
18:57.52 | [acl] | no other msm7k device actually dove into this |
18:58.15 | jonpry | blazing trails |
18:58.22 | phh | rhod is top android msm7k device ! |
18:58.24 | phh | wait what. |
18:58.38 | [acl] | phh: sup dood |
18:58.57 | [acl] | detule: hey dood this cam patch. work for you good ? |
18:59.06 | d3tul3 | works fine |
18:59.13 | jonpry | hey phh |
18:59.28 | [acl] | detule: damn so my dreams of bringing open source jpeg encoding die here |
18:59.42 | [acl] | detule: but it made sense |
18:59.53 | d3tul3 | well not sure if you were having the same JPEGTASK crap out that we were having...try it out |
19:00.20 | [acl] | detule: yeah that was the issue for a while, but ill try it out. that was bugging everyone |
19:00.59 | [acl] | jonpry: have you tried to call the FB0 function alredy ? |
19:01.05 | [acl] | jonpry: damn im exited for this |
19:01.30 | jonpry | no. but its working well enough that i could try to turn on electron beam |
19:01.32 | d3tul3 | i would like to try getting those patches in xdandroid rather than replacing whole repos |
19:01.51 | d3tul3 | i mean get them in my xdandroid tree |
19:02.53 | jonpry | why use an old frameworks_base? |
19:03.09 | jonpry | patch may apply clean as its against 2.3.7 now |
19:03.31 | jonpry | complete overwrite of msm7k/libgralloc would be best |
19:03.39 | d3tul3 | the xdandroid frameworks_base has quite a few patches in |
19:03.40 | [acl] | damn |
19:03.41 | [acl] | ballsy |
19:03.50 | jonpry | and libhardware/include/gralloc stuff |
19:04.04 | jonpry | i'm guessing most of those patches are just cm kangs |
19:04.10 | [acl] | pretty much |
19:04.11 | d3tul3 | there's telephony stuff |
19:04.25 | phh | (still noone interested in hacking a D3 or ? :D) |
19:04.47 | phh | (and erm, its keyboard is really great) |
19:05.17 | jonpry | so just rip the whole services/surfaceflinger libs/ui include/ui |
19:05.43 | [acl] | damn |
19:05.51 | jonpry | phh my new touch pro3 is way better |
19:06.17 | jonpry | https://github.com/jonpry <<< source |
19:09.00 | d3tul3 | jonpry it's just the top commits in frameworks base right don't need to go digging in there |
19:09.23 | jonpry | umm no |
19:09.39 | jonpry | cause i worked on it for long time. then fast forwarded to .7 |
19:10.04 | jonpry | probably need to git diff against the cm-7.1.0 tag |
19:15.24 | jonpry | maybe not the tag. ad3d4bb0f5bb352516e74b56d7618befe27ddd91 |
19:18.02 | jonpry | d3tul3, http://pastebin.com/Hb6E7hcx |
19:20.05 | [acl] | jonpry: hmm.. i think effin qcom didnt have FBO calls in the original blob. This 1.1 does have more deliciousness. This is why it may have never matched winmo back in the day |
19:20.21 | [acl] | cock gobblers |
19:20.31 | [acl] | im too upset to work.. |
19:20.39 | jonpry | lol |
19:23.45 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-91-189.dynamic.mnet-online.de) |
19:26.35 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:5c31:34c:50f9:c8b1) |
19:47.24 | *** join/#htc-linux newbie1 (~helicopte@95.234.35.222) |
19:50.18 | *** join/#htc-linux avinashhm (~avinash-h@192.91.66.189) |
19:51.07 | *** join/#htc-linux avinashhm (~avinash-h@192.94.92.14) |
20:09.03 | *** join/#htc-linux ftoz (~root@214-151.gtt-net.cz) |
20:16.38 | *** join/#htc-linux WisTilt2 (~wisgreg@wireless248.wirelesstcp.net) |
20:44.22 | *** join/#htc-linux ftoz (~root@214-151.gtt-net.cz) |
20:48.57 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
20:50.01 | *** join/#htc-linux bukington (~bukington@fac34-2-82-228-151-145.fbx.proxad.net) |
20:50.06 | *** join/#htc-linux raymonddull (~raymonddu@c-24-11-68-98.hsd1.mi.comcast.net) |
21:02.22 | *** join/#htc-linux joshkoss (~joshua@c-76-19-180-159.hsd1.ct.comcast.net) |
21:15.37 | *** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net) |
21:16.12 | d3tul3 | jonpry, thanks for the diff |
21:21.56 | WisTilt2 | hi d3tul3. i just pushed the 70hz panel change to .39 if you want to stick it in 3.1. im cloning 3.1 and guess we'll do any further stuff on that kernel rather than have all the previous ones to keep in sync. |
21:24.30 | d3tul3 | oh cool yeah 3.1 is just that and the fake vsync commit behind |
21:25.05 | d3tul3 | btw, scbs is currently disabled in 3.1 once you clone it just add CONFIG_SCBS=y it will re-enable all scbs related code |
21:25.20 | WisTilt2 | that needs to get updated also. its been working great on all our rhods and gained another .5fps |
21:25.35 | WisTilt2 | ok on scbs thanks. cant live without that beauty:) |
21:26.07 | d3tul3 | scbs is what is causing the sod when plugging in during sleep btw, some interaction with the idle code |
21:26.19 | WisTilt2 | did you push that pmem check removal on 3.1 or we leaving it in for now? |
21:27.03 | d3tul3 | the pmem check is still there, you should try out the userland patch, i can shoot you a patched binary |
21:27.22 | d3tul3 | this way that check can remain in place for other resources that use pmem_adsp |
21:27.26 | WisTilt2 | userland patch fixed the overlap? |
21:27.53 | d3tul3 | i don;t think there ever was an overlap, it's the vendor blob that's misrepresenting the situation |
21:28.18 | d3tul3 | but yeah a userland patch can rectify the situation |
21:29.00 | WisTilt2 | i can live without it for now. i really dont use the camera anyway so its critical for me. |
21:29.41 | WisTilt2 | ill look into why scbs and idle are causing that. should be easy to trace if you are sure scbs is interacting with it to cause that sod condition. |
21:30.22 | jonpry | its probably quickwakeup |
21:30.36 | d3tul3 | yeah works fine without, also, i feel like I have to say this but since i've removed scbs, i have not had any timer issues whatsoever in 3.1 |
21:31.14 | WisTilt2 | hi jonpry. i can probably fix that in idle then. that whole idle code really isnt what it should be anyway since no one really used it. |
21:31.35 | jonpry | hi WisTilt2 |
21:31.53 | WisTilt2 | idle is probably putting things back in suspend before quickwakeup finishes |
21:31.55 | jonpry | i've been getting lots of i think usb related sod's w/o scbs on 3.1 |
21:32.33 | jonpry | the usb is so bad on my rhod. i think its might need surgery |
21:32.54 | d3tul3 | i have had a few too, these are different from before happen when device is not in suspend |
21:32.55 | jonpry | btw. Is msm7x27 pin compatible with 7201? |
21:33.15 | WisTilt2 | solder time? my main phone i use to do all this testing is getting close to need of repair on usb plug also |
21:33.50 | WisTilt2 | yes afaik they are pin for pin but there are several internal diffs with register offsets |
21:34.14 | jonpry | cause they just release a new 8k device that is drop in for 7x27 |
21:34.54 | jonpry | http://www.tomshardware.com/news/Qualcomm-Snapdragon-S4-class-Adreno-QRD,14236.html |
21:35.01 | *** join/#htc-linux joshkoss (~joshua@c-76-19-180-159.hsd1.ct.comcast.net) |
21:35.15 | WisTilt2 | we have some addendum's to our docs that has several of the other msm chip specs so ill look tomorrow |
21:35.33 | jonpry | serious solder time with that thing |
21:38.19 | WisTilt2 | btw, my autobuild is offline until sometime tomorrow. we had the lock re-keyed for that new closet and the keys the locksmith left are wrong ones. we had a 6hr power outage last night so that server and [acl]'s nand server are down. |
21:39.10 | d3tul3 | i think i got the patches in, it will be a minor miracle if this builds |
21:39.39 | jonpry | got the hardware stuff too? |
21:39.55 | d3tul3 | yeah that i just copied |
21:40.20 | jonpry | so not a clean patch? |
21:41.02 | d3tul3 | the frameworks_base was not clean, had to copy paste a few hunks over |
21:41.57 | jonpry | hmm |
21:42.44 | d3tul3 | what's this in your patch ifeq ($(BOARD_HAVE_BLUETOOTH),true) |
21:42.44 | d3tul3 | +khjakjsh |
21:43.07 | jonpry | was having build problems with cm7 |
21:43.13 | jonpry | just remove that |
21:43.27 | jonpry | trying to make sure it was not building with bluetooth |
21:45.15 | d3tul3 | wow it actually built surfaceflinger |
21:45.25 | d3tul3 | errored out in gralloc i must have messed something up |
21:45.39 | jonpry | which gralloc? |
21:45.48 | jonpry | mine also tried to build gralloc.default.so |
21:46.21 | d3tul3 | oh right |
21:46.25 | d3tul3 | the wrong gralloc is erroring out |
21:46.27 | jonpry | which doesn't work. so i changed hardware/libhwardware/Android.mk to not import the gralloc makefile |
21:47.51 | d3tul3 | i see it |
21:51.15 | jonpry | probably need kernel patch? |
21:52.14 | jonpry | http://pastebin.com/rSmQw8vf |
21:52.46 | jonpry | is it 8 minutes to touchpad firesale? |
21:53.03 | rpierce99 | no |
21:53.10 | rpierce99 | 2hr 8 min |
21:53.26 | rpierce99 | 6cst |
21:53.33 | jonpry | oh |
21:54.35 | jonpry | probably need some blobs too |
21:54.45 | jonpry | and there are 2 changes to ueventd.rc |
21:55.03 | d3tul3 | i need a sdcard reader |
21:55.19 | jonpry | phone not good enough? |
21:55.44 | d3tul3 | my usb port is in as much of a need of surgery as yours |
21:56.04 | jonpry | :) |
21:58.43 | jonpry | i think we should ditch rhodium fixup and go with cmd line like normal people |
21:59.00 | d3tul3 | api errors coz i had to butcher api/current.xml....isn't tehre some easy way to update that |
21:59.09 | jonpry | make update-api |
22:01.29 | jonpry | https://github.com/jonpry/gposit_blobs |
22:01.56 | jonpry | also i uploaded a new gralloc this morning |
22:02.09 | jonpry | hopefully you have the stuff with gpu1_cached |
22:02.26 | d3tul3 | i just kanged it 45mins ago |
22:04.11 | jonpry | should be fine then |
22:04.51 | jonpry | this is xdandroid high speed edition? |
22:05.35 | d3tul3 | :) don't know about that, likelihood of this build working as it should is probably pretty low |
22:05.38 | *** join/#htc-linux programmer8922 (~Evan@c-174-56-72-235.hsd1.nm.comcast.net) |
22:08.15 | d3tul3 | hey WisTilt2 |
22:08.27 | WisTilt2 | yo |
22:08.40 | d3tul3 | mentioned something about userland source re: bluetooth |
22:08.45 | d3tul3 | this is where it errors out for me https://gitorious.org/xdandroid/system_bluetooth/blobs/gingerbread/bluedroid/bluetooth.c#line166 |
22:09.35 | d3tul3 | jonpry, it built |
22:10.19 | WisTilt2 | d3tul3 is that where its erroring out after you enabled all the kernel config stuff? |
22:10.19 | jonpry | thats half the battle |
22:10.56 | d3tul3 | yeah, that ioctl must fail and it gives me a "Timeout" |
22:11.20 | WisTilt2 | so its getting to the timeout message? |
22:11.26 | d3tul3 | though i also brought in rfkill from Alex/emwe so my results might be off from yours but i thought you said something similar happening with you as well |
22:11.40 | d3tul3 | yeah i get the timeout message |
22:11.52 | WisTilt2 | looks like it thinks the device never comes up but it does |
22:12.31 | d3tul3 | do you get that Timeout message as well? |
22:12.33 | WisTilt2 | i did notice that there is also a dbgfs/dev/hci0 coming up when you enabled it via userland so no telling |
22:12.41 | WisTilt2 | no mine never made it that far |
22:12.52 | WisTilt2 | to the timeout message that is |
22:13.03 | d3tul3 | also what's proper output of "hciattach"? |
22:13.28 | WisTilt2 | proper output? dont follow |
22:14.06 | d3tul3 | when i type in hciattach in an adb prompt i get certain output, i wasn't sure what it's supposed to be outputting |
22:14.23 | jonpry | when it works it hangs |
22:14.30 | WisTilt2 | yep |
22:14.30 | d3tul3 | thanks it hung |
22:14.45 | WisTilt2 | what you get with hcitools dev? |
22:15.32 | WisTilt2 | i mean hcitool dev |
22:15.52 | d3tul3 | sorry didn't try that and don't have that kernel going...i did see what looked like a mac adress in the hciattach output and something about restarting brcm4325 firmware |
22:17.01 | WisTilt2 | you should be getting the mac addr. finally finished dl'ing 3.1 so going to build and get that loaded. |
22:21.05 | *** join/#htc-linux pigeon (~pigeon@eth5284.nsw.adsl.internode.on.net) |
22:21.06 | *** join/#htc-linux swc|666__ (~G.Gecko@irc.pentoo.ch) |
22:21.06 | *** join/#htc-linux cyanogen (shade@chemlab.org) |
22:21.07 | *** join/#htc-linux xawen (~Admin@pool-98-117-32-71.bltmmd.fios.verizon.net) |
22:21.16 | *** join/#htc-linux jason` (j@nitrology.com) |
22:21.35 | *** join/#htc-linux Mavy-bnc (mavfree@91.196.169.2) |
22:21.35 | *** join/#htc-linux joshkoss (~joshua@c-76-19-180-159.hsd1.ct.comcast.net) |
22:21.35 | *** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net) |
22:21.35 | *** join/#htc-linux raymonddull (~raymonddu@c-24-11-68-98.hsd1.mi.comcast.net) |
22:21.35 | *** join/#htc-linux avinashhm (~avinash-h@192.94.92.14) |
22:21.35 | *** join/#htc-linux Rob2222 (~Miranda@p4FFF2D20.dip.t-dialin.net) |
22:21.35 | *** join/#htc-linux DJWillis (djwillis@cpc1-bath5-2-0-cust122.aztw.cable.virginmedia.com) |
22:21.35 | *** join/#htc-linux khorben_ (~dont@port-83-236-217-161.static.qsc.de) |
22:21.35 | *** join/#htc-linux furtardo (~mks@nat/yahoo/x-qboneuhhhzudynjy) |
22:21.35 | *** join/#htc-linux T-Junk (~TechnoJun@chaosarmy.org) |
22:21.37 | *** join/#htc-linux programmer8922 (~Evan@c-174-56-72-235.hsd1.nm.comcast.net) |
22:21.37 | *** join/#htc-linux dcordes (~oedevel@unaffiliated/dcordes) |
22:21.37 | *** join/#htc-linux leviathan (~quassel@2001:470:26:484:6ef0:49ff:fee6:8dca) |
22:21.37 | *** join/#htc-linux sado1 (~sado1@v0ll.de) |
22:21.52 | *** join/#htc-linux Entropy512 (~quassel@cpe-69-205-145-233.stny.res.rr.com) |
22:21.52 | *** join/#htc-linux bukington (~bukington@fac34-2-82-228-151-145.fbx.proxad.net) |
22:21.52 | *** join/#htc-linux XirXes (~XirXes@67-2-1-110.slkc.qwest.net) |
22:21.52 | *** join/#htc-linux phh (~quassel@quassel.rezel.enst.fr) |
22:21.52 | *** join/#htc-linux balans (~server@82-170-217-205.ip.telfort.nl) |
22:21.57 | *** join/#htc-linux detule (~detule@hw-ma-6l13f61.mat.jhu.edu) |
22:21.59 | *** join/#htc-linux jonpry (~jon@c-24-17-200-206.hsd1.wa.comcast.net) |
22:21.59 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
22:21.59 | *** join/#htc-linux skodde (~skodde@unaffiliated/skodde) |
22:21.59 | *** join/#htc-linux bartman (~bart@2607:f2c0:a000:175:2e0:81ff:fe47:3d01) |
22:22.04 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com) |
22:22.04 | *** join/#htc-linux hasi (hasi@mausi.hasicon.org) |
22:22.09 | *** join/#htc-linux Bry8Star2g (Bry8Star2g@209.141.55.4) |
22:22.09 | *** join/#htc-linux ellisway (~ellis@188-220-43-90.zone11.bethere.co.uk) |
22:22.09 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
22:22.09 | *** join/#htc-linux ImCoKeMaN (~imcokeman@pool-96-249-151-84.hrbgpa.fios.verizon.net) |
22:22.09 | *** join/#htc-linux Blista (~nel@209.59.220.47) |
22:22.09 | *** join/#htc-linux Dinde (kayser@sur-internet.net) |
22:22.09 | *** join/#htc-linux Willd (willd@citu-202.citu.kth.se) |
22:22.09 | *** join/#htc-linux tomaw (tom@freenode/staff/tomaw) |
22:22.17 | *** join/#htc-linux friehmaen (~fm@2001:4d88:1ffa:3d:10::f) |
22:22.19 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
22:22.19 | *** join/#htc-linux Ondalf (~ondalf@unaffiliated/ondalf) |
22:22.19 | *** join/#htc-linux pw (~pw@coffee-addicts.org) |
22:22.23 | *** join/#htc-linux EdLin (~EdLin@securabit/listener/edlin) |
22:22.23 | d3tul3 | jonpry, changes to uevent |
22:22.23 | *** join/#htc-linux LordDeath (~LordDeath@cable-81-173-164-253.netcologne.de) |
22:23.14 | *** join/#htc-linux apt (~apt@rikers.org) |
22:23.14 | *** topic/#htc-linux is Welcome to the HTC Linux project | Community portal & WiKi http://htc-linux.org | For IRC logs, HaRET & kernel mailing lists etc. see http://htc-linux.org/wiki/index.php?title=Contact | The htc-linux.org project is not affiliated with the HTC Corporation | This channel is for development purposes - Join #htc-linux-chat for offtopic |
22:23.16 | *** join/#htc-linux sailerboy (~sailerboy@sailerboy.net) |
22:25.24 | *** join/#htc-linux NetRipper (~netripper@tikkie.net) |
22:25.24 | *** join/#htc-linux snadge (~snadge@starbug.ugh.net.au) |
22:25.28 | *** join/#htc-linux lardman|home (~lardman@138-38-179-9.docking.bath.ac.uk) |
22:26.10 | *** join/#htc-linux arrrghhh (~arrrghhh@unaffiliated/arrrghhh) |
22:31.58 | *** join/#htc-linux infernix (nix@unaffiliated/infernix) |
22:35.05 | d3tul3 | jonpry, forgot to take out bootani |
22:36.32 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
22:38.09 | *** join/#htc-linux LjL (~ljl@unaffiliated/ljl) |
22:47.17 | d3tul3 | hey jonpry this look right to you: 12-11 17:42:35.832 E/libEGL ( 1455): load_driver(/system/lib/egl/libGLES_qcom.so): Cannot load library: get_lib_extents[759]: 1416 - /system/lib/egl/libGLES_qcom.so is not a valid ELF object |
22:56.10 | jonpry | no |
22:56.54 | jonpry | d3tul3, md5 d6cb69c27e99f8161c10def661e8cb91 |
22:58.36 | d3tul3 | 143167211e73d570c4b178caf7fe24cd |
22:58.55 | d3tul3 | perhaps i messed up the dl |
22:59.22 | jonpry | size 533456 |
23:00.18 | jonpry | i just downloaded it from github and got the same md5 |
23:01.49 | d3tul3 | i was messing up |
23:09.51 | d3tul3 | what did i do wrong http://pastebin.com/XhKSM19p |
23:11.39 | jonpry | you have a bootanimation? |
23:12.18 | jonpry | for whatever reason its not using the qcom renderer |
23:12.28 | jonpry | 12-11 18:09:11.857 I/SurfaceFlinger( 1455): renderer : Android PixelFlinger 1.4 |
23:12.31 | jonpry | is not right |
23:13.18 | jonpry | but it appears to have loaded it |
23:13.42 | jonpry | and this is a good sign 12-11 18:09:11.847 I/SurfaceFlinger( 1455): # of configs : 24 |
23:13.57 | jonpry | as pixelflinger only supports 8 configurations |
23:14.13 | d3tul3 | no bootanimation |
23:14.54 | d3tul3 | it kills adb at that point and all i have is a blank screen |
23:15.30 | jonpry | iirc this had something to do with libegl |
23:15.37 | jonpry | but its not in the diff |
23:20.17 | jonpry | d3tul3, EGLUtils.cpp |
23:21.45 | jonpry | not sure what yours looks like |
23:23.45 | d3tul3 | http://pastebin.com/QPuPmFLj |
23:25.38 | *** join/#htc-linux stroughtonsmith (~steven@86-43-208-204-dynamic.b-ras2.bbh.dublin.eircom.net) |
23:25.53 | *** join/#htc-linux stroughtonsmith (~steven@86-43-208-204-dynamic.b-ras2.bbh.dublin.eircom.net) |
23:26.58 | d3tul3 | i guess I am missing a few things |
23:31.12 | jonpry | ah yes that won't work |
23:33.44 | jonpry | http://pastebin.com/Y86s0s0W |
23:48.06 | jonpry | make sure that NO_RGBX is defined. that other codepath is the one that will actually allow selection of a hardware context |
23:49.21 | *** join/#htc-linux sykosoft (~sykosoft@67-61-60-19.cpe.cableone.net) |
23:49.32 | sykosoft | WisTilt2: Send out that rhod400 on Friday |
23:49.34 | sykosoft | sent |
23:49.36 | sykosoft | *sent |
23:49.37 | sykosoft | man |
23:50.02 | WisTilt2 | lol. got the email thanks. it will be put to good use over the next couple weeks. |
23:51.24 | WisTilt2 | d3tul3 kexec not working in 3.1 even though its enabled in kernel? |
23:51.32 | d3tul3 | WisTilt2, correct |
23:51.49 | WisTilt2 | k, thought it was me but after several tries i figured |
23:53.15 | WisTilt2 | between the fake vsync delete and moving fb where i normally run it i gained 1.2fps, not a lot but probably worthwhile |
23:55.10 | WisTilt2 | sykosoft did you send that 3 day or how? |
23:56.11 | sykosoft | checking (had my wife do it) |
23:57.04 | WisTilt2 | she did bill it to our fedex account i hope |
23:57.31 | d3tul3 | food time |