00:16.32 | *** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring) |
00:45.05 | jonpry | d3tul3, what are all those commits to local? |
00:46.59 | d3tul3 | jonpry, stuff that i am currently testing |
00:47.29 | jonpry | looks interesting |
00:48.58 | d3tul3 | can be summarised as BT + nand (for wifi-nvram) + smd plus some other minor ones that are likely irrelevant |
00:49.10 | *** join/#htc-linux Bry8Star{EB (Bry8Star-@2607:f128:42:fb::2) |
00:50.11 | jonpry | working well? |
00:50.35 | d3tul3 | i am not sure about BT, i remember you saying some time ago seeing an interrupt storm with BT |
00:50.46 | jonpry | yes i never got it to suspend |
00:51.18 | jonpry | you don't happen to follow acl's nand tree? |
00:51.51 | d3tul3 | well this is the same patchset that's working (and suspending) in 3.0 and 3.2 but a couple of times now i get possibly unrelated storms of h2w .... strange thing is -> theinterrupt is not incrementing it's just that evdev won't shut up |
00:52.16 | d3tul3 | i check in on acl now and then, he's doing some fancy stuff with the battery driver |
00:52.44 | jonpry | any idea what patches we would need to support cdma nand booting? |
00:53.04 | d3tul3 | it might boot as is |
00:53.17 | jonpry | hrm |
00:53.36 | d3tul3 | i suspect the panel code needs to be revamped and then he's got some driver in mach-msm re: charging/battery in LK |
00:53.57 | jonpry | iirc there was some issue with clocks |
00:54.37 | jonpry | i wonder why he couldn't push that stuff into linux-on-qualcomm |
00:54.49 | d3tul3 | bootcheck.c |
00:55.14 | d3tul3 | oh yeah no idea about clocks...too bad gitorious is horrible about folder history |
00:55.21 | d3tul3 | or any kind of history for that matter |
00:57.33 | d3tul3 | oh i have a clone of his tree |
00:58.01 | jonpry | first nand patch: http://gitorious.org/~acl/linux-on-qualcomm-s-msm/linux-msm-rhod-nand/commit/fca314d45f8c949aa50eeeb5d5a26df948de8868 |
01:00.10 | jonpry | we seem to do this in msm_clock_a11_reset_imem(void) |
01:01.16 | d3tul3 | yes our clocks and his will unfortunately differ |
01:02.33 | jonpry | i don't really understand why |
01:02.50 | jonpry | wtf would an imem reset cause problems on nand |
01:04.13 | d3tul3 | rt if it survives haret how can it break nand |
01:06.18 | jonpry | not sure |
01:07.06 | jonpry | maybe MSM_IMEM_BASE is sec locked |
01:08.06 | jonpry | boot loaders live there. maybe wince is able to unlock it somehow |
01:09.13 | d3tul3 | perhaps we can pass a cmnd like argument "nand" and make a bunch of things contingent on it |
01:09.21 | d3tul3 | s/like/line/ |
01:09.44 | jonpry | i think the plan was to use custom atags |
01:10.11 | d3tul3 | hm his .27 kernel still passes some commandline |
01:12.03 | jonpry | it could also be that imem clocks are off when the reset happens |
01:14.07 | jonpry | add code to turn it on? |
01:15.42 | d3tul3 | you ready to flash your phone? |
01:17.52 | jonpry | no |
01:18.06 | jonpry | try to get someone else to try this stuff :p |
01:18.20 | jonpry | docs says IMEM_CONFIG requires CC_PBUS_CLK |
01:20.11 | d3tul3 | looks like he's also de-wince-fied rpcrouter |
01:22.44 | jonpry | strangely enough pbus clk appears to be i2c |
01:34.34 | jonpry | maybe this is enough if (!a11_clk_is_enabled(IMEM_CLK)) |
01:34.34 | jonpry | return; |
01:46.15 | *** join/#htc-linux ccube (ccube@nx.ccube.de) |
02:06.14 | *** join/#htc-linux d3tul3 (~detule@unaffiliated/d3tul3) |
02:17.17 | d3tul3 | jonpry, about to do some testing out of the chroot on this ppp speed up hack |
02:19.30 | jonpry | out of the chroot? |
02:19.58 | d3tul3 | i don't have to use those hackish speed test apps |
02:20.09 | d3tul3 | i can scp to my work machine and do some transfers |
02:21.52 | jonpry | have you confirmed breakage without the hacks? |
02:22.18 | d3tul3 | breakage? |
02:22.30 | jonpry | sod, crash etc |
02:23.08 | d3tul3 | no, nothing on this new smd....never really had much in the way of breakage on the old smd either, but I guess I am lucky to have good coverage unlike CptAJ |
02:24.27 | jonpry | eventually the unpatched smd will trash the slab allocator so much that only order 1 allocations are possible |
02:24.54 | jonpry | and after that a chunk of data will accumulate that is larger than 4k |
02:24.59 | jonpry | then bye bye |
02:25.35 | d3tul3 | nothing like that in multiple weeks now |
02:27.24 | jonpry | it required a lot of no reboot time to trash the slab |
02:27.28 | d3tul3 | there's also a MAX_TTY_BUF_SIZE in the new tty limitting the per-call allocation size not sure if that would make a difference |
02:27.50 | jonpry | smd_tty? |
02:27.59 | d3tul3 | yeah |
02:28.42 | jonpry | cotulla and letama messed with that on leo and it helps but you can't stop the linux tty layer from doing the damage |
02:29.46 | d3tul3 | alright but this new smd_tty also has a contingency of avail < = 0 |
02:29.57 | d3tul3 | which i assume is the case when slab doesn't yield any buffer |
02:30.55 | jonpry | there are ways to avoid crashing the kernel but all operations requiring order > 0 will fail |
02:30.59 | jonpry | like wifi |
02:31.16 | jonpry | although on 39 etc we prealloc the buffers |
02:31.17 | *** join/#htc-linux rajkosto (~rajkosto@cable-94-189-233-89.dynamic.sbb.rs) |
02:31.25 | d3tul3 | right statically |
02:32.04 | d3tul3 | so you are saying this was an active issue before these ldisc patches went in in june/july? |
02:32.14 | *** join/#htc-linux Rajko (~rajkosto@wan.rajkonet.info) |
02:32.33 | jonpry | yes. just it wasn't noticable until we had uptimes more than a couple of days |
02:33.01 | jonpry | it was always noticable on leo. which is why they had to develop a way to hotpatch amss |
02:35.42 | jonpry | not sure if the dft has pulled this yet |
02:35.58 | d3tul3 | i mean how relevant is this to rmnet |
02:36.39 | jonpry | it isn't. but nobody likes rmnet. its slow. but they had to use it because pppd was trashing the slab and locking up |
02:37.42 | d3tul3 | 19.2KB/s, 12.1KB/s, 12.1KB/s, 16.5KB/s, 7.0KB/s five runs with the hack |
02:37.57 | d3tul3 | on a 2282KB file |
02:38.23 | jonpry | thats dirt slow |
02:38.57 | d3tul3 | my reception could be better i guess |
02:40.01 | jonpry | wistilt2 was getting 20mbit oer hsdpa |
02:41.40 | jonpry | thats almost like gprs speed |
03:05.31 | d3tul3 | without smd7 hack: 11.7KB/s, 14.6KB/s, 14.6KB/s, 21.9KB/s, 18.3KB/s |
03:06.33 | d3tul3 | i'll try the autobuild as well |
03:10.54 | jonpry | those speeds are so low it can't make a difference |
03:11.13 | jonpry | hyc was doing stuff with > 1 MB/s |
03:11.28 | jonpry | its also probably why you can't get smd induced crashes |
03:23.58 | d3tul3 | autobuild 3.3 : 18.0KB/s, 16.3KB/s, 21.9KB/s, 31.3KB/s, 22.2KB/s |
03:24.52 | jonpry | do you have 3g enabled? |
03:25.42 | d3tul3 | it's all on auto...and i see 3g symbol in taskbar so i assume it's working as it should |
03:28.42 | jonpry | need opensignal app |
03:29.35 | jonpry | what carrier are you using? |
03:29.38 | d3tul3 | sprint |
03:31.03 | jonpry | does sprint have hsdpa? |
03:31.19 | d3tul3 | don't know what the evdo equivalent of hsdpa is |
03:31.29 | *** join/#htc-linux ray|yar (~raymonddu@c-69-245-114-102.hsd1.mi.comcast.net) |
03:32.39 | d3tul3 | wonder how that app will play with our ril |
03:34.27 | jonpry | apparently well |
03:34.54 | jonpry | evdo i guess is at least 2.4mbit |
03:35.05 | jonpry | not sure why you get such horrible speed |
03:37.09 | d3tul3 | yeah it's pretty pathetic |
03:42.18 | jonpry | did you fix the h2w? |
03:43.18 | d3tul3 | my signal strength is no less than 84% |
03:43.48 | jonpry | your tower could be heavily used? |
03:44.06 | d3tul3 | h2w disappeared.....really mysterious getevent blows up and the event identifies as "h2w headset" but the h2w interrupts aren't counting |
03:44.48 | d3tul3 | so i am not quite sure how the events are getting to evdev |
03:45.51 | d3tul3 | finally backported getevent from ics so i can run 'getevent -l' to print out the actual report_codes (SYN, etc) just so i can see what the hell is being reported....but now i can't reporduce it |
03:46.10 | jonpry | oh well |
03:46.55 | jonpry | need to upgrade the whole shebang to ics |
03:47.12 | jonpry | or get a new phone |
03:48.49 | d3tul3 | i'll be completely lost on a native android device...mkbootimage is all greek to me |
03:49.57 | jonpry | fastboot and signatures |
03:50.27 | jonpry | i really want to crack some windows phone |
03:53.31 | d3tul3 | you should check out that haret + wp7 thread |
03:53.59 | d3tul3 | if i remember correctly there were several posts : people offering their spare wp7 phones to help the project |
03:56.13 | jonpry | guessing we need something along the lines of a kernel exploit |
03:56.19 | jonpry | like any jail break |
03:57.29 | d3tul3 | i guess MS wisened up and removed the publicly available setKmode api |
03:57.37 | d3tul3 | setK(ernel)mode |
03:58.47 | jonpry | yes and i believe there is also a chain of trust |
04:00.29 | jonpry | but any jail break should allow elevation |
04:01.40 | d3tul3 | technically they just need to write a driver that can then invoke kernel side apis and execute haret no? no idea how winmo security really works |
04:02.07 | jonpry | drivers are probably signed |
04:02.29 | jonpry | or else the whole security model is fubar |
04:06.16 | jonpry | also i wonder how they are getting this haret on there. don't all wp7 apps have to come from the market and be signed? |
04:07.05 | d3tul3 | knowing MS they are probably storing all driver certs in registry or something |
04:07.56 | d3tul3 | i guess you need some sort of unlocked wp7 phone |
04:08.33 | jonpry | http://labs.mwrinfosecurity.com/files/Advisories/mwri_htc-htcutility-kernmem_2011-11-10.pdf |
04:09.09 | d3tul3 | yeah i've seen that this fellow in the thread GoodDayToDie wrote an exploit using that |
04:09.20 | d3tul3 | i mean in that advisory they pretty much give you the code |
04:09.43 | d3tul3 | not sure how many devices come with that .dll and expose that ioctl |
04:10.38 | d3tul3 | htcroot i think his project is called |
04:10.44 | jonpry | the dll is probably signed so you might be able to add it to devices that don't have the cert revoked yet |
04:13.46 | *** join/#htc-linux zmc (~zmc@lurian.net) |
04:15.15 | *** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star) |
04:16.22 | d3tul3 | bed time later |
05:46.32 | *** join/#htc-linux kiozen (~kiozen@p578a42db.dip0.t-ipconnect.de) |
06:00.11 | *** join/#htc-linux mes_ (~mes@sentry.lazo.ca) |
06:03.18 | *** join/#htc-linux netson (~netson-ub@202.159.118.21) |
06:37.26 | *** join/#htc-linux balans2 (~user@53543C6D.cm-6-5a.dynamic.ziggo.nl) |
06:57.40 | *** join/#htc-linux bitrot (~rajkosto@cable-94-189-233-89.dynamic.sbb.rs) |
06:59.25 | *** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info) |
07:29.12 | *** join/#htc-linux Rajko (~rajkosto@cable-94-189-233-89.dynamic.sbb.rs) |
07:30.15 | *** join/#htc-linux bitrot (~rajkosto@wan.rajkonet.info) |
08:00.47 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-107-19.dhcp.stcd.mn.charter.com) |
08:13.55 | *** join/#htc-linux marc1706 (~Marc@phpbb/modifications/marc1706) |
08:37.23 | *** join/#htc-linux |Jeroen| (~jeroen@d5153E8CC.access.telenet.be) |
09:06.57 | *** join/#htc-linux rob_w (~bob@unaffiliated/rob-w/x-1112029) |
10:02.39 | *** join/#htc-linux balans2 (~user@53546403.cm-6-5b.dynamic.ziggo.nl) |
10:07.20 | *** join/#htc-linux balans21 (~user@53546403.cm-6-5b.dynamic.ziggo.nl) |
10:08.28 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
10:21.17 | *** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net) |
10:23.35 | *** join/#htc-linux Ondalf (~ondalf@cable-roi-fe98dc00-121.dhcp.inet.fi) |
10:27.14 | *** join/#htc-linux Mavy-bnc (mavfree@91.196.169.2) |
10:34.59 | *** join/#htc-linux zeusk (77eb360e@gateway/web/freenode/ip.119.235.54.14) |
10:50.33 | *** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info) |
11:03.08 | *** join/#htc-linux balans21 (~user@53546403.cm-6-5b.dynamic.ziggo.nl) |
11:51.38 | *** join/#htc-linux balans2 (~user@53543BAE.cm-6-5a.dynamic.ziggo.nl) |
12:13.10 | *** join/#htc-linux zeusk (77eb360e@gateway/web/freenode/ip.119.235.54.14) |
12:23.09 | *** join/#htc-linux Rajko (~rajkosto@wan.rajkonet.info) |
12:26.37 | *** join/#htc-linux ahigerd (~ahigerd@173.218.142.90) |
12:26.37 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
13:17.01 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-107-19.dhcp.stcd.mn.charter.com) |
13:39.05 | *** join/#htc-linux ahigerd (~ahigerd@173.218.142.90) |
13:39.05 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
13:41.50 | *** join/#htc-linux LordDeath (~LordDeath@cable-81-173-165-254.netcologne.de) |
13:48.11 | *** join/#htc-linux simpson (77eb360e@gateway/web/freenode/ip.119.235.54.14) |
13:48.26 | simpson | fakker |
13:48.37 | simpson | privet |
13:59.24 | *** join/#htc-linux balans2 (~user@53543BAE.cm-6-5a.dynamic.ziggo.nl) |
14:01.22 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-86-76.dynamic.mnet-online.de) |
14:02.41 | *** join/#htc-linux balans21 (~user@53543BAE.cm-6-5a.dynamic.ziggo.nl) |
14:19.42 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
14:28.52 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
15:43.05 | *** join/#htc-linux ccxCZ (~ccxCZ@new.webprojekty.cz) |
16:00.32 | *** join/#htc-linux DuperMan (~Duper@93-173-58-42.bb.netvision.net.il) |
16:06.41 | *** join/#htc-linux Alex[sp3dev] (~alexander@178.76.204.13) |
16:21.04 | *** join/#htc-linux ali1234 (~ajbuxton@s15821883.onlinehome-server.info) |
16:21.20 | *** part/#htc-linux ali1234 (~ajbuxton@s15821883.onlinehome-server.info) |
16:23.04 | *** join/#htc-linux balans21 (~user@53546403.cm-6-5b.dynamic.ziggo.nl) |
17:52.36 | *** join/#htc-linux helicopter88 (~helicopte@host169-113-dynamic.26-79-r.retail.telecomitalia.it) |
18:04.32 | *** join/#htc-linux Entvex (~entvex@83.91.89.186) |
18:04.35 | Entvex | Hello |
18:04.45 | Entvex | do any one here have a HTC X one ? |
18:14.58 | CptAJ | Entvex, maybe not but you might as well lay out your question. Worst case we can point you in the general direction of the answer |
18:19.17 | Entvex | okay CptAJ |
18:19.35 | Entvex | i am looking for the right phone to buy |
18:19.48 | Entvex | but is the Samsung Galaxy S II good ? |
18:20.22 | Entvex | i wanna buy a Samsung Galaxy S II or HTC S or HTC X what do you think ? and what phone have best custom rom support :) ? |
18:20.27 | CptAJ | jonpry, did some tests. panel fails to unblank with autobacklight turned off on winmo. When I turn it on (on winmo) I can unblank it by covering and uncovering it with my hand. The light/proximity sensor is definitely playing a role here. |
18:20.31 | Alex[sp3dev] | nokia n9 ftw |
18:23.09 | jonpry | CptAJ, i'm guessing that the backlight control stuff is all different on your phone. does changing brightness have any effect in android? |
18:23.34 | CptAJ | changing brightness from within android you mean? |
18:23.45 | Entvex | do any one have the Samsung Galaxy S II ? |
18:24.15 | Alex[sp3dev] | i would suggest there are thousands of people having galaxy s2 |
18:24.21 | Alex[sp3dev] | including me unfortunately |
18:24.23 | jonpry | CptAJ, yes |
18:24.44 | Entvex | Alex[sp3dev] are you happy with it ? |
18:25.01 | CptAJ | jonpry, none at all |
18:25.23 | Alex[sp3dev] | Entvex: it is fine as is, fast and has a good camera. but I think I should've bought nokia n9 because I passionately hate android and java and absolutely love qt4 |
18:25.39 | CptAJ | s2 has better resolution, processor and screen size than n9 |
18:25.50 | Entvex | :) but symbian is kinda dead right ? |
18:25.56 | Alex[sp3dev] | n9 is not symbian |
18:26.00 | ahigerd | The N9 is Meego |
18:26.02 | Alex[sp3dev] | it's meego (linux + x11 + qt4) |
18:26.05 | jonpry | CptAJ, probably need to mmutrace the i2c traffic while turning on and off autobacklight and changing brightness |
18:26.10 | ahigerd | And I'm a big fan of Qt4 myself |
18:26.15 | Entvex | ah sorry :D okay |
18:26.21 | CptAJ | which is also dead Entvex XD |
18:26.30 | ahigerd | Alex[sp3dev]: If you're a Qt4 fan you might have heard of me before. :P |
18:26.33 | Entvex | do Meego dead ? |
18:26.37 | Entvex | oh okay :) |
18:26.37 | ahigerd | Not technically |
18:26.39 | ahigerd | But close enough |
18:26.40 | Entvex | *s* too slow |
18:26.43 | ahigerd | Nobody's shipping it |
18:26.56 | CptAJ | the n9 is asking for trouble. alex likes trouble cause he's russian I think |
18:27.23 | Entvex | but what phone is the best ? i mean is s2 still have new roms |
18:27.28 | Entvex | or is it on thr way out ? |
18:27.29 | Alex[sp3dev] | ahigerd: hum.. sorry, i've not played with qt much, mostly with gtk. i just love how qt4 is documented and its build system |
18:27.37 | Entvex | i'm so new to all this :) |
18:28.26 | Alex[sp3dev] | Entvex: s2 is a fine phone but both nexus s and galaxy nexus have a larger community and bleeding edge roms from google |
18:28.55 | Entvex | i'll tjek the price on galaxy nexus |
18:29.04 | ahigerd | Alex[sp3dev]: Funny. I agree with you but plenty of people HATE Qt4's build system. XD |
18:29.26 | Alex[sp3dev] | ahigerd: how can you possibly hate anything that works everywhere and even supports c++0x |
18:29.40 | CptAJ | I'll have to agree about the documentation. Beautiful job there from qt |
18:30.14 | ahigerd | Eh, I can understand the animosity towards qmake. It's great for simple projects but it doesn't handle dependency tracking or autoconfiguration very well. |
18:30.22 | Alex[sp3dev] | looks like my laptop now lasts longer than my galaxy s2 |
18:30.37 | Alex[sp3dev] | ahigerd: one good thing about qmake is that it is not autohell |
18:30.41 | ahigerd | But I think that's something that could be corrected with a qmake 3.x instead of switching to something as ugly as cmake |
18:31.16 | Alex[sp3dev] | does any of you guys have a laptop with switchable graphics? |
18:31.20 | jonpry | i do |
18:31.42 | Alex[sp3dev] | jonpry: did you notice it consumes +10W power after a resume? if so, could you test one patch? |
18:32.06 | jonpry | i only use the intel graphics |
18:32.17 | jonpry | gave up long ago |
18:32.31 | Entvex | Galaxy Nexus also seem cheap :D! |
18:32.45 | Alex[sp3dev] | jonpry: i also use intel but turns out radeon powers on itself after a resume and just drains power. and doing on/off via switcheroo causes it to hang for a minute. so i made sort of a hack |
18:33.03 | Alex[sp3dev] | still waiting from David Airlie to review it, but Alan Cox said it sucks ;) |
18:33.06 | jonpry | i have nv graphics |
18:33.20 | Alex[sp3dev] | i see |
18:33.34 | Alex[sp3dev] | intel is really nice. at least webgl and compositing doesn't crash x |
18:33.38 | Entvex | is it better to wait for the Samsung Galaxy S 3? |
18:34.15 | jonpry | it actually runs catia properly. my last laptop was a sandybridge with non switchable ati and it had all kinds of rendering errors |
18:35.01 | CptAJ | Entvex, waiting is a tricky game. The window where waiting is the right choice is very narrow because there will always be a better rig about to come out |
18:35.10 | Alex[sp3dev] | jonpry: if you have time, could you give it a try to see if it doesn't break video on resume? i just need this stuff for my laptop but need more reviews http://comments.gmane.org/gmane.linux.kernel/1289242 |
18:36.09 | Entvex | Spot on CptAJ! if you were going to buy a phone today which one would you buy ? |
18:36.32 | CptAJ | depends on the budget really |
18:36.55 | CptAJ | I don't know much about the high end because I can't afford it XD |
18:37.10 | Alex[sp3dev] | on the other hand, we could just match all radeon 6xxx via DMI for this hack. i just want airlied to reply because he wrote the switcheroo in the first place |
18:37.31 | jonpry | what does switcheroo do? |
18:37.49 | Alex[sp3dev] | jonpry: switches between discrete/integrated graphics. |
18:37.55 | jonpry | how? |
18:38.09 | jonpry | my integrated graphics has no connections to any display connectors |
18:38.17 | Alex[sp3dev] | well, it has callbacks from drivers. acpi, i think |
18:38.54 | jonpry | the framebuffer has to be accessed over pcie and copied to the intel's memory |
18:39.06 | Alex[sp3dev] | radeon turns on after a suspend, but switcheroo thinks it's off. if you try to turn it ON, radeon firmware hangs. so i think it is a sane idea to forcibly turn off all the cards that aren't active |
18:39.54 | Alex[sp3dev] | jonpry: are you sure it has no hardware mux? need to look up |
18:39.55 | CptAJ | Entvex, of all the ones mentioned the s2 and the htc x are the top dogs |
18:39.57 | jonpry | so you don't have to use a headless x server with ati? |
18:40.05 | jonpry | yeah i am sure |
18:40.18 | CptAJ | Entvex, pick the one with more posts on the dev forum XD |
18:40.21 | Entvex | CptAJ s2 a lot cheaper then HTC x |
18:40.26 | CptAJ | s2 then |
18:40.37 | CptAJ | they're pretty similar |
18:40.38 | Alex[sp3dev] | jonpry: how does one tell? is there a datasheet? |
18:40.40 | Entvex | and i bet it have a better battary! |
18:40.49 | Alex[sp3dev] | jonpry: do you mean, this *patch* will break ati? |
18:41.01 | Entvex | beacuse with 4 cores! the HTC X LOVES to eat power! |
18:41.07 | CptAJ | check that there are good custom roms and no showstoppers (locked bootloaders or whatever) |
18:41.23 | Entvex | do the s2 have a GPU ? |
18:41.29 | jonpry | Alex[sp3dev], i guess it won't break anything that doesn't have switcheroo |
18:41.30 | CptAJ | probably |
18:41.31 | Entvex | i like to play a few games aswell :) |
18:41.43 | jonpry | Alex[sp3dev], https://wiki.archlinux.org/index.php/Bumblebee#Bumblebee:_Optimus_for_Linux first sentence |
18:41.44 | Entvex | (gamedev) |
18:41.53 | Entvex | if you have played that! |
18:42.01 | Entvex | i love that game xD |
18:42.20 | Alex[sp3dev] | jonpry: regarding ati, i have not even tried to get it running. intel has better pm and better drivers. and my opengl experiments are not advanced yet to make intel not enough |
18:42.29 | Entvex | jonpry and Alex[sp3dev] i love Arch linux! |
18:43.03 | jonpry | Alex[sp3dev], yes the ati is some dodgy stuff |
18:43.08 | Entvex | CptAJ did you have the s2 ? |
18:47.25 | Alex[sp3dev] | with intel, power drain is quite ok. no wifi + min bl => 7W, good bl => 9W, eclipse + device connected => 12W. compiling linux should be around 20W. my thinkpad has radeon and it never goes below 21W |
18:53.48 | CptAJ | Entvex, nope. I dont have one |
18:54.46 | CptAJ | ARM Mali-400 GPU I think thats what it has for a gpu |
18:54.55 | Entvex | ahh okay |
18:55.07 | CptAJ | check out pdadb.net when you need specs for a phone |
18:55.30 | Entvex | i would like to do androide apps :D when i get one! the androide emulator in the SDK Runs like shit :( |
18:57.09 | Alex[sp3dev] | galaxy s2 is a good phone overall. although wvga for 4.3 inch seems like a low resolution. but in cm9 everything except tv-out works |
18:57.10 | CptAJ | jonpry, what should I trace for the I2C writes? the a9900000 phys addy thats on msm_iomap.h? |
18:58.14 | Alex[sp3dev] | 0xb2300000 iirc. need to check |
18:58.31 | Alex[sp3dev] | yes |
19:00.54 | Entvex | Alex[sp3dev] do you have a s2 ? |
19:01.35 | Alex[sp3dev] | Entvex: yes as i have already stated |
19:01.54 | CptAJ | Alex[sp3dev], is that virtual or physical_ |
19:02.01 | CptAJ | ? |
19:02.27 | Entvex | sorry for not paying attaction i'm talking about this in a few channels to find good advise :) Alex[sp3dev] |
19:02.44 | Alex[sp3dev] | CptAJ: just add that to mmutrace |
19:03.12 | Alex[sp3dev] | CptAJ: and then dump mmu 2 0xb2300000 iirc. or i may be mistaken. it was long ago but you can always google the logs |
19:03.15 | CptAJ | ah, thats the same as the a99 one (just checked my memory map =P) |
19:03.15 | Entvex | are you all talking about programming on phones :) ? |
19:03.41 | Entvex | i would like to try that |
19:03.47 | Alex[sp3dev] | Entvex: people here are mostly porting android to old windows mobile phones (htc touch pro, xperia x1) |
19:03.49 | Entvex | i mean make my own rom |
19:04.04 | Entvex | ahh okay :) |
19:04.06 | jonpry | CptAJ, yeah have to use virt |
19:04.28 | CptAJ | or p2v(0xphys) >_> |
19:04.34 | Entvex | have any of you got androide 4.0 working ? |
19:04.36 | helicopter88 | Entvex: I'd join #xda-devs |
19:04.51 | Entvex | thanks :) |
19:04.52 | jonpry | yeah its just too slow |
19:06.04 | Entvex | do you code in assmbly then ? |
19:06.17 | Entvex | to get androide working on a windows phone ? |
19:07.33 | Alex[sp3dev] | mostly c. last time i used assembly was to boot the custom bootloader. now everything more or less works, including booting from nand but it took like 20 developers and 3 years |
19:08.16 | Entvex | i have only coded a little bit in assmbly |
19:09.11 | jonpry | reversing is done in assembly |
19:09.45 | CptAJ | reversers are like the rockstars of programming |
19:10.07 | Entvex | is that beacuse it's so hard ? |
19:10.12 | CptAJ | pretty much XD |
19:10.32 | ahigerd | Hard, obscure, and gets perceived as glamorous :P |
19:10.52 | CptAJ | assembly is annoying enough to write. reverse engineering basically is decompiling programs and reading the assembly to figure out what makes them tick |
19:10.53 | Entvex | well i can decompile c# and java programs |
19:11.03 | Alex[sp3dev] | lol |
19:11.04 | Entvex | i only write in c# and java |
19:11.24 | Entvex | and some c++ |
19:12.06 | Entvex | http://eu3.memecdn.com/lady-gaga_c_297512.jpg |
19:12.17 | Entvex | i looled! |
19:12.43 | CptAJ | I knew all along |
19:13.10 | *** join/#htc-linux mastermerlin (~Adium@p4FEE5303.dip.t-dialin.net) |
19:14.36 | Entvex | xD! |
19:14.48 | jonpry | nah binhacking is rockstar stuff |
19:15.17 | CptAJ | I figure thats more like jazz |
19:15.45 | CptAJ | way cooler but only musicians will pay attention |
19:15.58 | jonpry | could be |
19:16.15 | jonpry | until you want camera or 3d to work |
19:16.58 | jonpry | guess your 3d is already good? |
19:17.14 | Entvex | i have never tryed to reversing a "real" app |
19:17.32 | Entvex | only some stupid small onces just to try it |
19:22.09 | CptAJ | jonpry, http://paste2.org/p/2003913 traced. Each event is labelled there |
19:22.54 | CptAJ | Entvex, I guess thats how it starts. I'm just learning assembly now so you got a good head start ;) |
19:23.11 | jonpry | that seems pretty much like the stuff we have |
19:23.33 | Entvex | well yeah but i am using it at uni amt |
19:23.35 | Entvex | atm :) |
19:28.20 | CptAJ | jonpry, what could be the problem then? |
19:29.01 | jonpry | no i2c errors in dmesg? |
19:33.43 | CptAJ | http://paste2.org/p/2003923 lot of timeouts on vsync |
19:33.50 | jonpry | http://pastebin.com/7D6wU6Qz |
19:33.52 | CptAJ | let me get a log of the wakeup sequence |
19:34.19 | jonpry | try this. the initialization is strange to me. so you might have to toggle autobacklight in android for it to actually end up on or off |
19:34.34 | jonpry | once autobl is really off. then brightness should work |
19:49.28 | Entvex | have any of you seen this http://bluestacks.com/ |
19:50.18 | CptAJ | alright, first test run. It's unblanking on its own. Backlight settings now respond properly. however, it eventually crashes like so http://paste2.org/p/2003942 |
19:50.44 | CptAJ | dex error and modem reset. then it stops responding to key events |
19:54.19 | jonpry | probably should remove that vreg stuff from the -panel.c |
19:55.16 | CptAJ | oh yeah, I should mention that there's still no panel during bootup |
19:56.08 | jonpry | i don't really understand that. maybe try forcing it to blank and unblank in bootup somewhere |
19:57.25 | CptAJ | alright, the vreg stuff. what should I change? |
19:58.49 | jonpry | get rid of that vreg_disable/enable |
19:59.13 | jonpry | and all the calls to msm_proc_comm_wince |
20:00.17 | jonpry | i'm not sure this is the right course of action but its worth a shot i guess |
20:02.59 | *** join/#htc-linux ahigerd (~ahigerd@173.218.142.90) |
20:03.06 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
20:07.34 | jonpry | gtg |
20:36.36 | *** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info) |
20:37.20 | detule | jonpry, possibly his backlight doesn't support this 0-10 backlight, looks like we should be writing 0-255 instead of 0-9? |
20:50.08 | detule | or maybe not |
20:58.10 | CptAJ | well, it does seem to function with those values |
20:58.52 | detule | your backlight works? |
20:59.08 | CptAJ | yeah |
20:59.27 | CptAJ | its just the unblanking routine that's been giving trouble |
20:59.42 | detule | does it dim and undim when you use the slider in android? |
21:00.04 | CptAJ | yeah, jonpry fixed that today =) |
21:00.40 | detule | is it in the autobuild? |
21:01.00 | detule | oh |
21:01.10 | detule | i see it was the autobacklight setting that was broken |
21:01.35 | CptAJ | I dont think so, jonpry pastebinned it |
21:02.18 | *** join/#htc-linux mes_ (~mes@S0106000ce55186df.cc.shawcable.net) |
21:04.28 | detule | xool |
21:04.33 | detule | s/xool/cool/ |
21:04.45 | detule | what are you using for panel code? |
21:08.01 | detule | did you get rid of the vreg-s and dex calls in middi_power_client? |
21:08.39 | CptAJ | http://pastebin.com/q70GwjUp this without the vregs and dex calls right now |
21:09.39 | detule | doubt that would compile |
21:09.45 | CptAJ | it did |
21:09.48 | CptAJ | and its working |
21:10.20 | detule | what's on line 176 i don't understand "dex.da ta" |
21:11.13 | CptAJ | must be a pastebin error. look at the raw paste data |
21:11.48 | detule | this panel looks different than what's in the autobuild you must be working off a newer itteration hard to follow but great that it's working for you |
21:12.52 | CptAJ | its still a bit funky but I guess we're getting there |
21:14.12 | detule | oh i see it's that version of *panel but you've taken out the dex/proc_comm calls.... |
21:14.31 | detule | a little slow here |
21:15.46 | detule | CptAJ, you've shared your good news with the few other imagio users in the world? |
21:16.35 | CptAJ | I wanted to pack it up as an early christmas present for them in several forums where I know they check sporadically |
21:16.51 | CptAJ | but the suspend issues are really getting in the way =( |
21:17.42 | CptAJ | I set myself the milestone of having it at least relatively stable with the temporary key configuration (no microp keys) and data |
21:17.55 | CptAJ | so there's some point to using it |
21:18.41 | CptAJ | I dont see a point in releasing if its just gonna work tethered to the PC with suspend off |
21:20.37 | CptAJ | by the way, I got acoustic kinda working the other day |
21:21.03 | CptAJ | sounds are coming through the earpiece |
21:21.10 | CptAJ | no loudspeaker though |
21:21.38 | CptAJ | I've abandoned that line of tweaking until suspend is finally licked |
21:22.33 | detule | you have all the csv's in place? |
21:22.55 | detule | that sounds like a routing issue |
21:23.01 | CptAJ | csv's? |
21:23.12 | CptAJ | (also, no incall sound for some reason) |
21:23.14 | detule | from winmo? |
21:23.32 | detule | need to copy all *.csv files from your windows directory to the root of your sdcard |
21:23.39 | CptAJ | oh my |
21:23.59 | CptAJ | I'll get on that, I guess. Sounds simple enough |
21:24.01 | detule | also can you copy paste your board file with acoustic |
21:24.08 | detule | curious about your endpoints |
21:25.16 | CptAJ | http://paste2.org/p/2004024 boardfile |
21:26.17 | detule | see if it works with the csv files |
21:26.31 | detule | should fix at least one of your problems :) |
21:26.34 | *** join/#htc-linux Alex[sp3dev] (~alexander@178.76.204.13) |
21:29.06 | detule | in htcwhitestone_init |
21:29.30 | detule | you should do all the acousic_* snafu before patform_add_devices |
21:30.30 | detule | that's pretty important |
21:36.40 | *** join/#htc-linux ahigerd (~ahigerd@173.218.111.126) |
21:36.56 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
21:38.04 | detule | i guess not quite as important as i thought |
21:48.24 | *** join/#htc-linux ahigerd (~ahigerd@173.218.111.126) |
21:48.24 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
21:53.21 | CptAJ | alright, moved the acoustic_ stuff and placed the csv files in the root of the sd card |
21:53.24 | CptAJ | trying it now |
21:58.48 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
21:59.27 | *** join/#htc-linux ahigerd (~ahigerd@173.218.111.126) |
21:59.27 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
22:00.21 | *** join/#htc-linux furtardo (~mks@nat/yahoo/x-qxazwexvijkzjfmu) |
22:10.31 | CptAJ | mic is working =) |
22:10.43 | CptAJ | at least we have calls now |
22:11.39 | CptAJ | unblank is being stupid though. It wont unblank while untethered. I think it's just taking too long and the suspend timeout kicks in before it unblanks or something. It seems to take a variable amount of time to unblank when plugged in |
22:11.52 | CptAJ | at leastit isn't crashing anymore though |
22:12.01 | CptAJ | almost there |
22:20.16 | *** join/#htc-linux raymonddull (~raymonddu@c-69-245-114-102.hsd1.mi.comcast.net) |
22:29.15 | *** join/#htc-linux ahigerd (~ahigerd@173.218.111.126) |
22:29.16 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
22:55.46 | *** join/#htc-linux furtardo_ (~mks@nat/yahoo/session) |
22:57.51 | *** join/#htc-linux raymonddull (~raymonddu@c-69-245-114-102.hsd1.mi.comcast.net) |
23:00.37 | *** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd) |
23:00.37 | *** join/#htc-linux Alex[sp3dev] (~alexander@178.76.204.13) |
23:00.37 | *** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173-18-84-218.client.mchsi.com) |
23:00.37 | *** join/#htc-linux ChanServ (ChanServ@services.) |
23:00.37 | *** join/#htc-linux nitroboost (j@nitrology.com) |
23:00.37 | *** join/#htc-linux EdLin (~EdLin@midway.packetsense.net) |
23:00.37 | *** join/#htc-linux fakker (fakker@cpc13-hitc6-2-0-cust129.9-2.cable.virginmedia.com) |
23:00.37 | *** join/#htc-linux detule (~detule@unaffiliated/d3tul3) |
23:00.38 | *** join/#htc-linux phh (~quassel@137.194.15.151) |
23:00.38 | *** mode/#htc-linux [+o ChanServ] by wolfe.freenode.net |
23:01.08 | *** join/#htc-linux furtardo_ (~mks@nat/yahoo/x-ofbbtcsqwalfvrni) |
23:01.08 | *** join/#htc-linux polyrhythmic (~polyrhyth@c-71-197-239-27.hsd1.wa.comcast.net) |
23:01.32 | *** join/#htc-linux Mavy-bnc (mavfree@91.196.169.2) |
23:01.32 | *** join/#htc-linux ElFinLazz (~elfinlazz@182.215.84.22) |
23:01.32 | *** join/#htc-linux tmzt (~tmzt@adsl-76-253-137-36.dsl.akrnoh.sbcglobal.net) |
23:01.32 | *** join/#htc-linux FlawlesStyle (~LOL@unaffiliated/flawlesstyle) |
23:02.06 | *** join/#htc-linux bukington (~bukington@fac34-2-82-228-151-145.fbx.proxad.net) |
23:02.06 | *** join/#htc-linux sado1 (~sado1@meiserver.de) |
23:02.06 | *** join/#htc-linux CIA-100 (~CIA@cia.atheme.org) |
23:02.06 | *** join/#htc-linux Entropy512 (~quassel@cpe-69-205-145-233.stny.res.rr.com) |
23:05.55 | jonpry | no moar crashing? |
23:06.38 | CptAJ | nope |
23:06.48 | CptAJ | so thats a win right there |
23:06.52 | jonpry | so suspend in general is fixed? |
23:06.56 | CptAJ | no |
23:07.00 | jonpry | hrm |
23:07.04 | CptAJ | it still wont unblank properly |
23:07.18 | jonpry | like it takes a long time? |
23:07.24 | CptAJ | it seems to unblank whenever it feels like, yeah |
23:07.36 | CptAJ | when its unplugged, it never unblanks |
23:07.58 | jonpry | plugged in it doesn't actually suspend |
23:08.01 | CptAJ | unless I hit the power button again, in which case it quickly unblanks and blanks into suspend |
23:09.07 | CptAJ | it seems like its just taking to long to call novtec unblank |
23:09.10 | CptAJ | *too |
23:09.26 | jonpry | is this no unblank thing new? |
23:09.38 | CptAJ | no, its the same issue we've had from the beginning |
23:09.45 | CptAJ | it doesn't call on that function in time |
23:10.35 | CptAJ | in 27 I could trigger it by unlocking. Yesterday SOMETIMES it triggered by covering up the phone. So I'm guessing that a change in backlight sometimes triggers it |
23:11.33 | jonpry | er. i meant is the no blank thing new |
23:11.44 | CptAJ | it does blank |
23:11.48 | CptAJ | all the time |
23:11.51 | CptAJ | it rocks at blanking |
23:12.17 | CptAJ | if there was a blanking competition then... ok you get the picture XD |
23:12.46 | jonpry | lol |
23:14.13 | jonpry | maybe try to remove some of those sleeps in the mddi power client |
23:14.35 | jonpry | also those remote_writes in panel_unblank might be good |
23:17.06 | CptAJ | I'll try enabling the remote writes |
23:18.29 | CptAJ | http://paste2.org/p/2004136 check out this log. I suspended, unplugged the cable and tried to wake it up. |
23:18.57 | CptAJ | then after it didn't, I plugged back in to get the log. but as I was doing that, it woke up. (so yeah, it took a long time) |
23:19.50 | jonpry | we call that the slow wake. or failed wake |
23:20.15 | jonpry | another guy was responsible for fixing it on touch pro2 |
23:20.43 | jonpry | from what i understand it is a matter of timing. and that waiting in mddi_power_client is very dangerous |
23:21.23 | jonpry | once you have removed all the wasted time possible there are some other parameters that can be tweaked |
23:25.21 | *** join/#htc-linux FlawlesStyle` (~LOL@173.255.136.122) |
23:27.03 | *** join/#htc-linux tmzt_ (~tmzt@adsl-76-253-137-36.dsl.akrnoh.sbcglobal.net) |
23:33.12 | *** join/#htc-linux ElFinLazz (~elfinlazz@182.215.84.22) |
23:34.28 | *** join/#htc-linux wk (~wk@141.0.136.22) |
23:41.33 | *** join/#htc-linux Mavy-bnc (mavfree@91.196.169.2) |
23:43.46 | CptAJ | I'll tweak that stuff in a bit. Making some dinner |
23:45.39 | *** join/#htc-linux furtardo (~mks@nat/yahoo/x-nebipwtnmhvfguka) |
23:46.11 | *** join/#htc-linux Bry8Star{EB_ (Bry8Star-@2607:f128:42:fb::2) |
23:51.15 | *** join/#htc-linux friehmaen (freeman@fm.xers.de) |
23:55.38 | *** join/#htc-linux toer (~tore@179.81-166-86.customer.lyse.net) |