| 00:04.08 | *** join/#htc-linux myxor (n=myxor@pdbn-4d089c45.pool.mediaWays.net) |
| 00:04.19 | *** part/#htc-linux Jon_ (n=Jon@93-97-203-170.zone5.bethere.co.uk) |
| 00:08.33 | *** join/#htc-linux gh0ul (n=asds@host81-157-60-163.range81-157.btcentralplus.com) |
| 00:08.35 | *** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net) |
| 00:18.48 | *** join/#htc-linux nashpa (n=dliviu@dliviu.plus.com) |
| 00:21.01 | *** part/#htc-linux nashpa (n=dliviu@dliviu.plus.com) |
| 00:29.43 | *** join/#htc-linux ppman (n=mkern@pool-70-111-252-30.nwrk.east.verizon.net) |
| 00:29.53 | ppman | hiya |
| 00:30.16 | ppman | on kaiser, how do I tell the difference between winmo and the hardware calling I2C? |
| 00:36.19 | ppman | http://omploader.org/vMTZ3dw |
| 00:36.27 | ppman | look at all the I2C stuff, |
| 00:36.31 | ppman | about every 10 seconds |
| 00:38.17 | *** join/#htc-linux sdt555 (n=titus@70.114.152.91) |
| 00:40.41 | cr2 | ppman: you need 0 and c |
| 00:47.30 | cr2 | good night |
| 01:51.34 | ppman | cr2: what was that? |
| 01:51.52 | ppman | I got all the bytes related to I2c in that command, right? |
| 01:51.59 | ppman | oh wait |
| 01:52.03 | ppman | he disconnected :( |
| 02:11.22 | dream_kill | ppman u seen what i posted about battery? |
| 02:11.31 | ppman | nope |
| 02:11.33 | dream_kill | http://pastebin.com/d51a868fa |
| 02:12.05 | ppman | what is that? |
| 02:12.43 | ppman | what language is that? |
| 02:12.50 | dream_kill | battery stuff from diamond/raphael :P |
| 02:12.52 | dream_kill | c:P ? |
| 02:13.02 | ppman | it *almost* looks like c... |
| 02:13.07 | dream_kill | visual c :P |
| 02:13.11 | ppman | ewwww |
| 02:13.18 | ppman | but raph/diamond doesn't help me |
| 02:13.33 | dream_kill | should be similar in kaiser! |
| 02:13.43 | ppman | msm7x00A uses proc_comm/ read from memory |
| 02:13.57 | ppman | msm7x00, the arm9 doesn't read battery status |
| 02:14.59 | ppman | kaiser's battery driver reads the ds2746 directly in wince |
| 02:15.00 | dream_kill | Smem batterry id |
| 02:15.00 | dream_kill | yes? |
| 02:15.00 | ppman | there's no battery stuff in shared mem on kaiser |
| 02:21.16 | *** join/#htc-linux ALoGeNoff (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 02:23.01 | dream_kill | ppman did u check priv? |
| 02:23.06 | ppman | priv? |
| 02:23.33 | dream_kill | on irc |
| 02:23.39 | dream_kill | i sent u a query |
| 02:23.50 | ppman | priv is a user? |
| 02:24.29 | dream_kill | what do u use for irc conenction? |
| 02:24.41 | ppman | (I really should use IRC more often, but I don't - I don't know the lingo/anything else) |
| 02:24.44 | ppman | using irssi |
| 03:07.25 | *** join/#htc-linux ALoGeNo (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 03:25.07 | ppman | on kaiser, how do I know whether the USB or the AC is plugged in? |
| 03:29.35 | ppman | wait... do I need to know that? |
| 03:29.48 | ppman | the hardware handles fast vs. slow, correct? |
| 03:43.58 | tmzt | ppman: mmutrace to see what ce does |
| 03:46.30 | ppman | tmzt: I can only find the difference between chrging and not |
| 03:46.52 | ppman | I'm running android w/o a battery driver with the USB plugged in atm |
| 03:46.56 | tmzt | sorry I meant to paste that earlier |
| 03:47.07 | tmzt | what are you trying to do? |
| 03:47.09 | tmzt | with i2c |
| 03:47.19 | ppman | talk to the ds2746 |
| 03:47.23 | ppman | the battery chip |
| 03:48.47 | tmzt | yeah that's going to you about the battery not the charging I think |
| 03:49.32 | ppman | I see a bunch of traffic on 0xa9900000 and friends every ten seconds |
| 03:49.58 | ppman | that and a comment I saw in battdrvr.dll suggests the os is reading the chip directly. |
| 03:56.12 | ppman | schweet - I don't have to tell it which type of chrge it's doing. |
| 04:24.18 | *** join/#htc-linux ppman_ (n=mkern@pool-70-111-215-98.nwrk.east.verizon.net) |
| 04:39.24 | *** join/#htc-linux ALoGeNoff (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 05:00.56 | *** part/#htc-linux sdt555 (n=titus@70.114.152.91) |
| 05:06.17 | *** join/#htc-linux Dan__ (n=chatzill@cpe-74-69-91-35.rochester.res.rr.com) |
| 06:02.11 | *** join/#htc-linux tcccp (i=hey@ballbreaker.hey-ix.net) |
| 06:03.23 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 06:55.40 | *** join/#htc-linux goxboxlive (n=goxboxli@185.84-48-126.nextgentel.com) |
| 07:26.10 | ALoGeNo | morning |
| 07:26.37 | ALoGeNo | nice news for mp900c and 2.6.28 kernel :) |
| 07:26.57 | ALoGeNo | we have basic support for the MobilePro 900/c mini-laptop |
| 07:26.58 | tmzt | what news? |
| 07:27.11 | tmzt | video/keyboard? |
| 07:27.16 | ALoGeNo | http://kernelnewbies.org/Linux_2_6_28#head-44ac0f9cb64e8495d6cd5a3fc4dafc744fb93f5a |
| 07:27.18 | *** join/#htc-linux Zoolooc (n=lucian@nrbg-4dbfd174.pool.einsundeins.de) |
| 07:27.22 | ALoGeNo | i dunno :P |
| 07:28.40 | ALoGeNo | and hello tmzt :) |
| 07:28.53 | *** join/#htc-linux tcccp (i=hey@ballbreaker.hey-ix.net) |
| 07:29.57 | ALoGeNo | i go try now to compile the 2.6.28 with mp900default_config or so.. |
| 07:31.05 | ALoGeNo | hey you seen goboxlive in this days over here tmzt ? |
| 07:31.27 | ALoGeNo | the uni still stoped :/ |
| 07:32.19 | tmzt | ~seen goxboxlive |
| 07:32.44 | apt | goxboxlive is currently on #htc-linux (37m 4s) #handhelds (37m 4s) #openmoko (37m 4s) #opie (37m 4s), last said: 'hi cr2 . What device are you working on these days'. |
| 07:32.45 | goxboxlive | hi tmzt |
| 07:32.50 | tmzt | slow |
| 07:32.51 | ALoGeNo | heyyy |
| 07:32.52 | ALoGeNo | xD |
| 07:32.56 | ALoGeNo | is here ! |
| 07:32.56 | ALoGeNo | xD |
| 07:33.12 | tmzt | there's still an opie? |
| 07:33.36 | ALoGeNo | what do you mean? |
| 07:33.52 | goxboxlive | http://opie.handhelds.org - Opie 1.2.4 is out! |
| 07:34.04 | ALoGeNo | wow! |
| 07:34.07 | tmzt | what qtopia or qt/e is that? |
| 07:34.25 | ALoGeNo | but is a new uni kernel with that?? |
| 07:34.26 | tmzt | opie logo was the nicest thing to possibly see after switching from sharp rom |
| 07:34.29 | ALoGeNo | O_O |
| 07:35.13 | goxboxlive | afar qt-2.3.10 |
| 08:05.35 | *** join/#htc-linux Venny (i=Travis@h216-45-127-2.dynamic.platinum.ca) |
| 08:22.22 | *** join/#htc-linux Sti_0239 (n=Where_is@127.185-65-87.adsl-dyn.isp.belgacom.be) |
| 08:40.21 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 08:50.42 | *** join/#htc-linux kiozen (n=oeichler@BAH49d7.bah.pppool.de) |
| 08:55.31 | *** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net) |
| 09:13.07 | *** join/#htc-linux cr2 (n=cr2@ip-90-186-23-128.web.vodafone.de) |
| 09:21.42 | *** join/#htc-linux Jon__ (n=Jon@93-97-203-170.zone5.bethere.co.uk) |
| 09:31.58 | *** join/#htc-linux Jon___ (n=Jon@93-97-203-170.zone5.bethere.co.uk) |
| 09:57.33 | *** join/#htc-linux Moonlit (i=mooonlit@mario.dreamhost.com) |
| 10:00.09 | *** join/#htc-linux imfloflo (i=5bcfd11d@gateway/web/ajax/mibbit.com/x-0d229978fbfd257c) |
| 10:05.36 | *** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net) |
| 10:13.02 | cr2 | hi |
| 10:13.16 | cr2 | looking for some non-A people |
| 10:16.21 | *** join/#htc-linux Xime (n=xime@bankize.net) |
| 10:19.42 | *** join/#htc-linux metter (n=metter@137-171.62-81.cust.bluewin.ch) |
| 10:22.38 | Moonlit | non-A? |
| 10:22.42 | Moonlit | (hi, btw) |
| 10:23.11 | BabelO | msm7200 not ùsù7200A |
| 10:23.38 | BabelO | hi |
| 10:23.38 | BabelO | hi cr2 |
| 10:23.38 | Moonlit | oh, gotcha, I'm actually just idling to try and learn something |
| 10:23.49 | Moonlit | unfortunately I know nothing about any of this stuff but it does interest me, and you guys are awesome :P |
| 10:25.30 | cr2 | hi BabelO |
| 10:25.37 | cr2 | Moonlit: which cpu do you have ? |
| 10:26.51 | Moonlit | uh... you know, I couldn't tell you, what I can tell you is that I have a GSM Kaiser (kais130) |
| 10:27.15 | cr2 | ok, it's a non-A cpu :) |
| 10:27.35 | Moonlit | takes notes |
| 10:27.46 | cr2 | do you know to to telnet to haret ? |
| 10:28.20 | Moonlit | I'd probably have a rough idea, but sadly my USB socket's borked and I haven't fixed it yet |
| 10:28.34 | cr2 | wifi ? |
| 10:28.38 | Moonlit | doh |
| 10:29.03 | Moonlit | was thinking of connecting via a serial terminal emulator |
| 10:29.19 | Moonlit | you'll have to give me a while to catch up here :P |
| 10:29.26 | Moonlit | but yeah |
| 10:29.28 | Moonlit | I can do that |
| 10:30.24 | *** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 10:31.01 | cr2 | kais does not have a serial port. |
| 10:31.30 | Moonlit | nah, I'm new to HTC phones, I'm used to phones having serial via usb cables |
| 10:31.44 | Moonlit | so it's taking a little adjustment :P |
| 10:34.57 | cr2 | there is serial over usb here too |
| 10:35.14 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 10:35.38 | Moonlit | I can't imagine needing it all that much though |
| 10:35.54 | Moonlit | considering these phones have connectivity out the wazoo |
| 10:36.56 | Moonlit | hrm, so this is what telnetting to haret's like |
| 10:37.15 | Moonlit | guess I wouldn't have expected anything different, I just didn't realise you could telnet to it |
| 10:42.56 | Moonlit | I'd admit that I'm sitting here turning the LCD on and off via telnet, but I'd feel like a 5 year old in here lol |
| 10:45.34 | cr2 | type |
| 10:45.54 | cr2 | pd 0xa8600000 0x100 |
| 10:46.26 | cr2 | like at the bottom here : |
| 10:46.30 | cr2 | http://wiki.xda-developers.com/index.php?pagename=MSM_CLK |
| 10:46.30 | Moonlit | hang on, resetting, got the LCD stuck off >_> |
| 10:47.44 | Moonlit | SDC2 Md? |
| 10:49.00 | Moonlit | ok, that gave me some garbage that I can only imagine means something more to you than it does to me |
| 10:50.04 | Moonlit | you want the result pastebinned? |
| 10:54.59 | cr2 | yes |
| 10:57.53 | Moonlit | http://slexy.org/view/s21UAWjva5 |
| 11:00.54 | cr2 | thanks, it looks good. |
| 11:01.26 | Moonlit | well, I wouldn't put it up on my wall, but I'll take your word for that ;) |
| 11:02.12 | Moonlit | but uh... I gather that's a dump of something specific, what's that address range for? |
| 11:02.20 | *** join/#htc-linux Skitzo (n=DCLXVI@eth582.vic.adsl.internode.on.net) |
| 11:02.46 | cr2 | MSM_CLK |
| 11:02.59 | cr2 | the internal clock setup. |
| 11:03.17 | Moonlit | oh, heh, I guess that makes sense when you look at it |
| 11:03.22 | cr2 | it is documented here http://wiki.xda-developers.com/index.php?pagename=MSM_CLK |
| 11:03.23 | Moonlit | what with the CLK in the name and all |
| 11:04.17 | Moonlit | yeah, I did look at that page but it didn't make a huge amount of sense to me, though I'd gathered it had something to do with timing, given that it's SDC2 Md (where M is multiplier and d is PWM duarion) |
| 11:04.31 | Moonlit | when I say I'm new to this, I mean I'm *really* new to this |
| 11:05.19 | Moonlit | my experience with Linux is minimal, my coding experience is pure Visual Basic (and the predecessors thereof) and I only just got this Kaiser |
| 11:05.26 | Moonlit | so all in all, I'm pretty clueless lol |
| 11:06.23 | Moonlit | but hey, if there are moments where I can learn and/or contribute then I'm pretty happy to hang around |
| 11:11.49 | cr2 | ok, then it will take too long to explain :) |
| 11:12.25 | Moonlit | yeah, no biggy, I just pick up things as I go along, it'll probably make sense soon enough :) |
| 11:25.15 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |
| 11:25.58 | *** join/#htc-linux BabelO (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 11:32.27 | *** join/#htc-linux imfloflo (i=5bcfd11d@gateway/web/ajax/mibbit.com/x-15d7521ba7bc0446) |
| 11:37.06 | *** join/#htc-linux imfloflo (i=5bcfd11d@gateway/web/ajax/mibbit.com/x-e2bfa3fd6929b0bd) |
| 11:38.50 | *** join/#htc-linux Untouchab1e (n=Untoucha@cm-84.209.89.86.getinternet.no) |
| 11:38.54 | Untouchab1e | Good afternoon! :) |
| 11:38.58 | Moonlit | af'noon |
| 11:39.03 | Untouchab1e | how's it going? |
| 11:39.08 | Moonlit | it's good, you? |
| 11:39.16 | Untouchab1e | not too bad :) |
| 11:39.23 | Moonlit | good good :) |
| 11:39.54 | Untouchab1e | Up to anything special? |
| 11:39.59 | Untouchab1e | or just hanging out? :) |
| 11:40.07 | Moonlit | nah, not really, just hanging out :) |
| 11:40.39 | Moonlit | just kinda watching this channel, thought maybe I'd learn something lol |
| 11:40.49 | Untouchab1e | :) |
| 11:40.57 | Untouchab1e | its a good place to learn, thats for sure ^^ |
| 11:41.32 | Moonlit | yeah, well I kinda mentioned earlier that I'm totally new to pretty much all this stuff, since I can't really code and my Linux experience is limited |
| 11:41.54 | Moonlit | but I've already figured out Git and how to compile a kernel, which is good progress I think :) |
| 11:42.01 | Untouchab1e | Absolutely ;) |
| 11:43.36 | Moonlit | I have to say, embedded Linux is much easier to deal with as a user than "desktop" Linux |
| 11:44.05 | Untouchab1e | heh, I see what you mean |
| 11:44.16 | Moonlit | I mean I guess it has to be in order to be usable, but it's definitely a positive thing |
| 11:45.09 | Untouchab1e | yeah |
| 11:45.57 | Untouchab1e | Im still more comfortable in a Windows environment. Mostly because I'm a Windows developer (.NET), but I still love Linux |
| 11:46.16 | Moonlit | yeah, I'm a Windows guy, I've done some VB.net stuff |
| 11:46.36 | Moonlit | I do like Linux and I want to be able to use it, it just isn't there for me yet (on the desktop at least) |
| 11:47.24 | Untouchab1e | heh.. to be honest, most of my problems with Linux has almost always been Compiz's fault |
| 11:48.10 | Moonlit | I just get the little niggly stuff, stuff not always working how I want or occasionally malfunctioning for no reason |
| 11:48.43 | Moonlit | and tbh another reason I prefer the idea of embedded Linux is the UIs in a lot of the desktop Linux world leave a lot to be desired |
| 11:49.10 | cr2 | Moonlit: do you have a bt headset ? |
| 11:49.18 | Moonlit | nope, not yet anyway |
| 11:49.29 | Untouchab1e | I have one.. but I dont have it on me atm |
| 11:49.38 | Moonlit | a lot of use we are, eh? lol |
| 11:49.42 | Untouchab1e | haha |
| 11:49.51 | Moonlit | "not yet" and "not with me" :P |
| 11:50.19 | cr2 | ok |
| 11:52.55 | Untouchab1e | haha |
| 11:53.10 | Untouchab1e | My point being, I will have access to mine on sunday |
| 11:54.47 | Untouchab1e | I need some brainstorming help here btw.. |
| 11:54.56 | Untouchab1e | Im working on my HTC Android Config tool.. |
| 11:55.25 | *** join/#htc-linux BabelO (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 11:55.34 | Moonlit | nice idea |
| 11:55.42 | Untouchab1e | yeah, the first version is already released.. but |
| 11:56.11 | Untouchab1e | and someone over at XDA suggested implementing a check for newest Android builds for the Raph and Diam (which I host).. Now, I am stuck with the dilemma of either implementing a direct download feature. kinda like (download the latest version now). or just a simple check that reports which version is the newest one |
| 11:57.13 | Untouchab1e | on one hand, its handy to just being able to download the latest build from the app, but on the other hand, by saying "Hey, its a new build out, wanna go and get it?" will send them to the download page on my site where they can also read up on the changelog and usage instructions |
| 11:59.07 | Moonlit | you could give both options |
| 11:59.18 | Moonlit | a "download now" and a "read more" |
| 11:59.36 | Untouchab1e | Hmm, yea |
| 11:59.38 | Untouchab1e | yeah* |
| 12:04.39 | Untouchab1e | For some reason, I didnt think of that ^^ |
| 12:08.01 | Moonlit | sometimes it's the simple things :) |
| 12:11.16 | Untouchab1e | yeah, very true |
| 12:11.16 | cr2 | a dumb question here |
| 12:11.20 | Untouchab1e | shoot |
| 12:11.45 | cr2 | are there any miniUSB stick available ? |
| 12:11.55 | Moonlit | not that I'm aware of |
| 12:12.01 | cr2 | me too |
| 12:12.28 | Moonlit | you could use an adapter I guess if a device was capable of being a USB host |
| 12:12.33 | Untouchab1e | hmm, you mean like miniUSB sticks that would be able to act as a memory card for mobile devices or something? |
| 12:12.44 | cr2 | yes |
| 12:12.54 | cr2 | i have not seen them too |
| 12:12.57 | Moonlit | it'd be a little inconvenient though |
| 12:13.11 | cr2 | why ? |
| 12:13.31 | Moonlit | well, compared to a microSD it'd be awkward |
| 12:13.32 | Untouchab1e | why not use a regular microSD card? |
| 12:13.34 | Untouchab1e | hehe |
| 12:13.52 | Moonlit | having things hanging off your device is never particularly fun |
| 12:13.59 | cr2 | i have a microSD-USB-B adapter |
| 12:15.02 | cr2 | it seems that raph100 supports miniUSB stick, but the details should be sorted out |
| 12:15.29 | cr2 | hanging ? it'll be just a mini instead of a normal USB |
| 12:16.22 | Moonlit | I dunno if we're thinking the same thing, I'm imagining a USB stick with a miniUSB plug on it instead of a normal USB plug |
| 12:16.30 | cr2 | i must have a miniUSB-A cable somewhere |
| 12:16.32 | Moonlit | which you'd plug into the USB port on a phone or whatever |
| 12:16.51 | cr2 | Moonlit: exactly. |
| 12:16.59 | cr2 | but i've never seen it |
| 12:17.18 | Moonlit | well I can see how it could be useful but I've never seen or heard of them existing |
| 12:17.32 | cr2 | probably because there are no miniUSB hosts ? |
| 12:17.38 | Moonlit | would make sense |
| 12:17.57 | Moonlit | I think microSD would make them kinda redundant though |
| 12:18.20 | cr2 | diamond does not have microSD |
| 12:18.26 | Moonlit | fair point |
| 12:18.36 | cr2 | but may have usb otg |
| 12:18.58 | cr2 | at least it seems that raph100 has it |
| 12:19.05 | Moonlit | hrm, I see where you're going now, I hadn't considered the lack of a microSD slot |
| 12:22.44 | cr2 | http://forum.xda-developers.com/showthread.php?t=411905 |
| 12:23.58 | Moonlit | that's an idea I had a while back, jagnet mentions a device that allows you to access USB sticks via bluetooth |
| 12:24.08 | Moonlit | that could be another option for phones without microSD |
| 12:24.44 | Moonlit | the host thing sounds interesting though |
| 12:31.36 | cr2 | http://www.willcom-fan.com/wzero3/entries/mobachiki/000380/03.jpg |
| 12:33.52 | cr2 | but i think we don't have an msm-ohci driver ? |
| 12:39.52 | cr2 | 18bit lcd |
| 12:40.53 | swetland | ehci |
| 12:44.57 | Untouchab1e | Ok, updated the app now :) 0.4.. added a "latest build" checker |
| 13:00.39 | cr2 | swetland: ehci driver is in the tree ? |
| 13:01.15 | cr2 | this h2w port seems to be misused in all possible ways. |
| 13:01.37 | cr2 | audio,video,rs232... |
| 13:02.24 | cr2 | at least we know the gpio alt setting for ehci now. |
| 13:10.40 | *** join/#htc-linux ALoGeNoff (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 13:19.45 | *** join/#htc-linux exco (n=exco@e181118017.adsl.alicedsl.de) |
| 13:41.34 | dcordes | aye |
| 13:42.39 | exco | do you have a parrot on your shoulder? |
| 13:47.13 | cr2 | dcordes: i have cleaned up the MSM_CLK |
| 13:47.13 | Moonlit | yarr? |
| 13:47.25 | cr2 | do you have a bt headset ? |
| 13:48.07 | cr2 | polaris_cpld page does not exist ? |
| 13:52.06 | exco | cr2: no |
| 13:52.39 | exco | but if you tell me how to get the affected data - I can create it |
| 13:53.00 | *** join/#htc-linux imfloflo (i=5bcfd11d@gateway/web/ajax/mibbit.com/x-fbd362c453364822) |
| 13:54.07 | exco | You can access my Polaris through haret if you want |
| 13:57.05 | cr2 | exco: copy the kaiser_cpl page, but remove all the descriptions |
| 13:57.19 | cr2 | exco: do you have a bt headset ? |
| 13:58.01 | exco | cr2: unfortunately just sold mine |
| 13:58.19 | Moonlit | lol, you'd think someone would have a headset, right? |
| 13:58.37 | exco | any reason for "_" instead of "space" in wiki pagenames? |
| 13:59.53 | exco | I have the E100 headset (don't think that helps) |
| 14:03.39 | cr2 | exco: %20 is ugly |
| 14:04.11 | cr2 | my bt headset is too far away |
| 14:04.33 | cr2 | because i never use it ;) |
| 14:04.58 | exco | there are 2 pages for titan cpld http://wiki.xda-developers.com/index.php?pagename=Titan_CPLD http://wiki.xda-developers.com/index.php?pagename=TitanCPLD (also the second one says vogue cpld) |
| 14:05.04 | exco | *although |
| 14:14.21 | cr2 | hehe |
| 14:14.25 | cr2 | 401 /* 0x32 normal, 0x50 for Toshiba display */ |
| 14:14.26 | cr2 | 402 mddi_writel(0x0050, DRIVE_LO); |
| 14:14.53 | cr2 | raph100 and blac100 have 0xd0, raph800 has 0xa0 |
| 14:16.29 | cr2 | 409 /* disable periodic rev encap */ |
| 14:16.29 | cr2 | 410 mddi_writel(MDDI_CMD_PERIODIC_REV_ENCAP, CMD); |
| 14:16.29 | cr2 | not on wince |
| 14:16.48 | *** join/#htc-linux pH5 (n=ph5@e178237090.adsl.alicedsl.de) |
| 14:17.27 | cr2 | 420 /* Recommendation from PAD hw team */ |
| 14:17.27 | cr2 | 421 mddi_writel(0xa850f, PAD_CTL); |
| 14:17.34 | cr2 | only on raph800 |
| 14:17.46 | cr2 | raph100 and blac100 have 0xa8505 |
| 14:18.05 | exco | cr2: http://wiki.xda-developers.com/index.php?pagename=Polaris_CPLD |
| 14:23.18 | tmzt | cr2: got some data from lama for i780, the framebuffer is 320x320x16 and is the pxafb address, writing to it in haret works |
| 14:24.13 | tmzt | haret/haretconsole don't support pxa3xx, I was able to trace the matrix keyboard registers and got some activity there on keypress, but nothing on i2c (optical mouse maybe) |
| 14:25.02 | cr2 | tmzt: there are no pxa3xx docs ? only the linux kernel source ? |
| 14:25.25 | tmzt | I used kernel source, marvell might have them somewhere, I didn't look |
| 14:25.26 | cr2 | tmzt: where is the fb located ? |
| 14:26.15 | cr2 | exco_: does the current code use cpld on polaris ? |
| 14:26.37 | tmzt | 00:21 <tmzt> HaRET(1)# print "%08x" VRAM |
| 14:26.38 | tmzt | 00:21 <tmzt> bc8c0000 |
| 14:27.10 | cr2 | tmzt: i'm not familiar with the pxa3xx memory map |
| 14:27.17 | cr2 | is it sram ? |
| 14:27.32 | tmzt | I didn't try the sd card gpio because the gpios var was empty, I addlisted the linux headers values once but forgot to the last time |
| 14:27.42 | cr2 | the sdram is at 0xc afair, like on the sa1100 ? |
| 14:27.50 | tmzt | I'm not sure how haretconsole detects the device |
| 14:28.05 | tmzt | from the haret session |
| 14:28.18 | cr2 | through the list |
| 14:28.41 | cr2 | do you have the earlyharetlog ? |
| 14:28.48 | tmzt | 01:27 <tmzt> 8b200000 | 44000000 | 1MB section | CB AP=1 |
| 14:28.49 | exco_ | cr2: you'd need to ask Rogro82 about that (or dcordes) - because his changes aren't committed yet |
| 14:28.49 | tmzt | 01:27 <tmzt> ab200000 | 44000000 | 1MB section | AP=1 |
| 14:28.49 | tmzt | 01:28 <tmzt> HaRET(14)# pd 0x44000000 0x14 |
| 14:28.49 | tmzt | 01:28 <tmzt> 44000000 | 03b008f9 17173d3f 0301053f 04f0000c | ....?=..?....... |
| 14:28.52 | tmzt | 01:28 <tmzt> 44000010 | 8400fe92 | .... |
| 14:28.54 | tmzt | 01:28 <tmzt> that should be it |
| 14:28.57 | tmzt | no |
| 14:28.58 | cr2 | exco_: ok |
| 14:29.07 | tmzt | ask lama for that I only have some haretconsole logs |
| 14:30.49 | cr2 | ok |
| 14:30.50 | tmzt | 00:27 <tmzt> ======== Connecting to haret ======== |
| 14:30.50 | tmzt | 00:27 <tmzt> Minimal virtual address: 00010000, maximal virtual address: 7FFFFFFF |
| 14:30.54 | tmzt | 00:27 <tmzt> Detected machine Default/generic (Plat='PocketPC' OEM='SGH-i780') |
| 14:30.57 | tmzt | 00:27 <tmzt> CPU is Intel ARM arch 5TE revision 2 product 9 stepping 1 running in system mode |
| 14:31.17 | cr2 | that's enough |
| 14:31.41 | tmzt | how? |
| 14:31.56 | cr2 | but you need to modify the mach-pxa.cpp |
| 14:32.23 | cr2 | enough to assign a pxa3xx cpu to it |
| 14:33.03 | cr2 | this virtual address range message is really useless |
| 14:33.57 | tmzt | so the sram is important because it needs noncontig patches? |
| 14:34.15 | cr2 | not really |
| 14:34.22 | cr2 | just it's faster than sdram |
| 14:34.38 | cr2 | pxafb will be happy using sdram for fb |
| 14:35.06 | tmzt | ok, a virtual write to that framebuffer drew on the screen and we have the timings (lccr's) posted above (the pd) |
| 14:35.19 | cr2 | tmzt: maybe if we will setup pmdh as in wince, the hw refresh will work again ;) |
| 14:35.28 | cr2 | linux != android os. |
| 14:35.33 | tmzt | ah |
| 14:35.46 | dcordes | exco_: no the parrot.. he left me |
| 14:35.58 | tmzt | have you been able to dump pmdh regs? |
| 14:36.02 | tmzt | dcordes !!! |
| 14:36.38 | dcordes | me? |
| 14:36.49 | cr2 | tmzt: if the subsystem addresses are the same as on pxa27x, you can try to run my old px27x parser |
| 14:37.16 | cr2 | aka haret-w.exe |
| 14:37.28 | dcordes | cr2: got no bt headset |
| 14:37.33 | tmzt | I have a dumper somewhere |
| 14:37.39 | dcordes | but I can lend one |
| 14:38.40 | cr2 | dcordes: i'm interested in the MSM_CLK area 0x100 dump with BT headset playing sound, or phone talking |
| 14:38.55 | tmzt | hsuart? |
| 14:39.27 | cr2 | tmzt: that's the question |
| 14:39.43 | tmzt | not i2s with the bt chip? |
| 14:39.45 | cr2 | tmzt: there is some other bt audio related gpio alt config. |
| 14:39.53 | cr2 | afaik no. |
| 14:39.56 | dcordes | I thought kaiser bt is no uart |
| 14:40.23 | cr2 | bt audio should use some other pcm port |
| 14:40.31 | cr2 | not uart. |
| 14:40.43 | tmzt | ok, but if it's pcm why does it need hs-uart? |
| 14:40.59 | cr2 | control line |
| 14:41.08 | cr2 | to send bts commands |
| 14:41.36 | cr2 | i've lost my raph headset somewhere, can't test fm too ;) |
| 14:41.37 | tmzt | I see this as the problem android had, they want to implement a2dp with audioflinger |
| 14:41.57 | tmzt | so this clk would be for the uart2dm also, even if not in high speed mode? |
| 14:42.03 | cr2 | no |
| 14:42.12 | cr2 | the uart2dm clocks are known. |
| 14:42.17 | *** join/#htc-linux ALoGeNo (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 14:42.48 | tmzt | you figure out the clk grp? |
| 14:43.19 | dcordes | cr2: when I got the bt and audio is active on it, what command must I run? |
| 14:43.20 | cr2 | yes, but it is also doing some scary stuff at +0x2xx |
| 14:43.35 | cr2 | dcordes: pd 0xa8600000 0x100 |
| 14:43.39 | dcordes | ok will do |
| 14:44.03 | *** join/#htc-linux StarLite (n=nnscript@s55916ca6.adsl.wanadoo.nl) |
| 14:44.07 | cr2 | tmzt: check the MSM_CLK wiki page |
| 14:44.58 | Moonlit | you must get sick of typing that lol |
| 14:46.12 | tmzt | do you know where to dump the h3wd area? |
| 14:48.01 | cr2 | Moonlit: it's a bit cryptic page, i know :) |
| 14:48.29 | cr2 | tmzt: i'm not concerned about the hw3d now. |
| 14:49.44 | cr2 | tmzt: it's a separate project, that includes disassembling g1 binaries |
| 14:50.26 | cr2 | tmzt: i want to get the working and suspending/resuming framebuffer device. and autorefreshing too ;) |
| 14:50.28 | tmzt | yeah, I'm just wondering about the register space and grp clk |
| 14:51.30 | *** join/#htc-linux imfloflo (i=5bcfd11d@gateway/web/ajax/mibbit.com/x-7c25bed5b989d0fb) |
| 14:51.38 | cr2 | tmzt: mdp ? |
| 14:52.03 | tmzt | no, whatever the linux hw3d exposes |
| 14:52.32 | cr2 | #define MDP_TV_OUT_CTL ( 0xc0000) |
| 14:52.50 | cr2 | need a tvout cable for it though ;) |
| 14:52.54 | cr2 | ok |
| 14:53.02 | tmzt | shouldn't that be on a seperate chip? |
| 14:53.14 | cr2 | msm has tvout |
| 14:53.28 | imfloflo | hey guys do you try to search on google anything and click on the first link |
| 14:53.31 | cr2 | don't now where is the encoder |
| 14:53.37 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 14:53.47 | tmzt | it doesn't use the epson/toshiba for that? I thought the pdf pretty much confirmed it |
| 14:54.06 | tmzt | the two page one from epson I mean |
| 14:54.10 | cr2 | hm. g1 does not have tvout |
| 14:54.23 | *** join/#htc-linux exco (n=exco@e181118017.adsl.alicedsl.de) |
| 14:54.26 | cr2 | yeah, a full datasheet will be nice |
| 14:54.35 | cr2 | epson also has PiP |
| 14:56.04 | tmzt | not sure it matters but I enabled the pxa overlay on my q kernel and it caused some corruption. this could apply because they are working on simplyfing support for overlays and other multiple framebuffer devices if we ever wanted picture-in-picture or something else to work |
| 14:56.32 | cr2 | it's hardware pip |
| 14:59.32 | tmzt | #define HW3D_REGS_LEN 0x100000 |
| 14:59.56 | tmzt | yeah, that's hardware too but they decide to use multiple framebuffer memory devices in this version of the driver |
| 15:00.17 | tmzt | it even has yuv support on one of the overlays apparently |
| 15:01.14 | tmzt | hah, static struct clk *grp_clk/*imem_clk are both in mach-msm/hw3d.c |
| 15:02.05 | exco | cr2: I didn't notice - Rogro's changes have been committed 16h ago. |
| 15:02.15 | cr2 | raph manual says that the lcd is 666, but wince supports only 565 |
| 15:02.27 | tmzt | 18-bit? |
| 15:03.11 | cr2 | yes |
| 15:03.35 | cr2 | but you probably need to send some spi commands to configure it. |
| 15:03.54 | cr2 | but not on raph100 it seems |
| 15:05.04 | tmzt | it has pmem_pdata for bot pmem_gpu0 and pmem_gpu1 |
| 15:05.28 | tmzt | s/bot/both/ |
| 15:05.51 | cr2 | what is gpu0 ? |
| 15:05.52 | tmzt | 0x00000000 0x00700000 what?? |
| 15:06.04 | *** join/#htc-linux foobarman (n=foobarma@81.193.82.81) |
| 15:06.12 | cr2 | raph manual says 32MB sram |
| 15:06.24 | cr2 | that's tricky |
| 15:06.25 | tmzt | gpu1: MSM_RAM_CONSOLE_BASE-MSM_PMEM_GPU1_SIZE |
| 15:06.38 | *** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix) |
| 15:06.41 | tmzt | how can gpu0 base be at 0x0? |
| 15:06.44 | cr2 | i need to find a portable way to detect the sram size. |
| 15:06.53 | miknix | what the heck |
| 15:06.55 | cr2 | why not |
| 15:07.04 | tmzt | http://android.git.kernel.org/?p=kernel%2Fmsm.git&a=search&h=android-msm-2.6.27&st=grep&s=MSM_PMEM |
| 15:07.07 | miknix | "This site may harm your computer." |
| 15:07.11 | tmzt | not sure how to do that |
| 15:07.12 | cr2 | spl is preserved only on wince |
| 15:07.24 | tmzt | miknix: lama said something similar |
| 15:07.26 | miknix | is it only me or Google searches are filtering every single result? |
| 15:07.55 | cr2 | miknix: works for me |
| 15:08.07 | miknix | strange |
| 15:08.12 | tmzt | lama: hi |
| 15:08.25 | cr2 | miknix: but i've seen such things on other computer |
| 15:08.35 | cr2 | and thought, wtf ? |
| 15:09.26 | cr2 | miknix: you have an msm device now ? |
| 15:09.39 | miknix | not really, why? |
| 15:09.43 | *** join/#htc-linux patoche (n=patoche@lns-bzn-58-82-251-208-84.adsl.proxad.net) |
| 15:09.52 | miknix | funny: http://pastebin.com/d2b871015 |
| 15:10.05 | cr2 | you search fr msm-related things :) |
| 15:10.15 | lama | well |
| 15:10.19 | lama | whole google is borked |
| 15:10.26 | lama | no matter what you search |
| 15:10.29 | *** join/#htc-linux MethoS- (n=lem@host-091-097-242-061.ewe-ip-backbone.de) |
| 15:10.33 | lama | even their own sites are badware :P |
| 15:10.34 | exco | http://www.google.com/interstitial?url=http://www.microsoft.com/windows/windows-7/beta-download.aspx |
| 15:10.34 | exco | Warning - visiting this web site may harm your computer! - *harrharr* |
| 15:10.41 | cr2 | lol |
| 15:10.48 | Moonlit | hehe, was just talking about that in another channel |
| 15:11.10 | cr2 | Or you can continue to http://www.microsoft.com/windows/windows-7/beta-download.aspx at your own risk. For detailed information about the problems we found, visit Google's Safe Browsing diagnostic page for this site. |
| 15:11.19 | tmzt | so framebuffer has same address as GPU1? |
| 15:12.04 | cr2 | tmzt: on g1 ? |
| 15:12.05 | lama | tmzt: are you going to compile test kernel for i780? |
| 15:12.05 | tmzt | anyway this is about grp clk, if you have that working I guess we don't need this |
| 15:12.06 | miknix | cr2, heh! Just checking out Google was not tagging "linux" only searches |
| 15:12.12 | tmzt | in android kernel (for g1) |
| 15:12.17 | lama | I don't have tools, and... skills :P |
| 15:12.43 | cr2 | tmzt: NetRipper uses a different memory layout. |
| 15:13.07 | tmzt | would the trout one work? |
| 15:13.32 | cr2 | need to ask NetRipper |
| 15:13.44 | tmzt | I'm hoping to see if someone can try a JesusFreke image extracted from yaffs2, for testing maejrep's wifi support |
| 15:13.51 | cr2 | but you must be sure that you have 64MB smem first |
| 15:14.05 | tmzt | ah, so it won't work on raphael? |
| 15:14.35 | cr2 | tmzt: the raph800 spl does something to detect 32 vs 64 |
| 15:15.58 | tmzt | I assume maps of pmem devices use the physical addresses so that will break on raph with NetRipper map |
| 15:16.20 | tmzt | do you know what clk_imem is? |
| 15:16.31 | cr2 | no |
| 15:16.44 | cr2 | i get exception reading from 0x02000000 |
| 15:16.59 | tmzt | it's disabled when gpu is in android kernel |
| 15:17.00 | cr2 | but it may be MPU, who knows |
| 15:17.08 | cr2 | in wince |
| 15:17.20 | lama | http://www.smh.com.au/news/technology/biztech/relief-for-those-caught-by-street-views-dunny-deal/2009/01/29/1232818599129.html |
| 15:18.05 | cr2 | because i can't read the amss area too |
| 15:18.18 | cr2 | it was possible on titan, btw ;) |
| 15:18.55 | tmzt | what are you trying to read? |
| 15:21.00 | cr2 | tmzt: ram |
| 15:21.09 | cr2 | lama: lol. image of a man in Spain who flipped his middle finger at a passing Google car. |
| 15:21.14 | tmzt | I mean why that address |
| 15:21.27 | miknix | Internet is fixed now |
| 15:23.26 | exco | miknix: was it broken? |
| 15:23.37 | Moonlit | yeah, google went boom |
| 15:23.39 | tmzt | cr2: INT_GRAPHICS? |
| 15:23.53 | miknix | Google IS the Internets :P |
| 15:23.55 | exco | so google == internet? |
| 15:23.59 | cr2 | http://images.smh.com.au/ftsmh/ffximage/2008/11/18/gmaps_bird_spain_gallery__554x400.jpg |
| 15:24.00 | tmzt | http://android.git.kernel.org/?p=kernel/msm.git;a=blob;f=arch/arm/mach-msm/hw3d.c;h=b013b57077af9c60f1728f0f65e0494be6c5ff7 |
| 15:25.39 | *** join/#htc-linux Mullins (n=bw@89.204.239.174) |
| 15:27.08 | cr2 | tmzt: it_graphics is known |
| 15:27.39 | exco | gotta try Win 7 ... bbl (hopefully) |
| 15:28.21 | tmzt | missed one, pdata in the h3wd.c driver itself |
| 15:28.53 | tmzt | h3wd 0xA0000000 0x1000000 .cached = 0 (something in the memory controller) |
| 15:28.54 | cr2 | 179 static struct android_pmem_platform_data pmem_data = { |
| 15:28.55 | cr2 | 180 .name = "hw3d", |
| 15:28.57 | cr2 | 181 .start = 0xA0000000, |
| 15:28.58 | cr2 | 182 .size = 0x100000, |
| 15:28.59 | tmzt | yes |
| 15:29.07 | cr2 | is it virtual of physical address ? |
| 15:29.13 | *** part/#htc-linux exco (n=exco@e181118017.adsl.alicedsl.de) |
| 15:29.14 | cr2 | s/of/or/ |
| 15:29.26 | tmzt | this is all phys that I know of (pmem) |
| 15:29.35 | tmzt | pmem is physical version of /dev/mem |
| 15:30.00 | *** join/#htc-linux pH5 (n=ph5@e178237090.adsl.alicedsl.de) |
| 15:31.32 | cr2 | 5c000000 | a0000000 | Extended (4K) | AP=3 |
| 15:31.34 | cr2 | <PROTECTED> |
| 15:31.35 | cr2 | 91b00000 | a0000000 | 1MB section | AP=1 |
| 15:31.37 | cr2 | b1b00000 | a0000000 | 1MB section | AP=1 |
| 15:31.38 | cr2 | H |
| 15:33.01 | tmzt | only 1mb? |
| 15:33.05 | cr2 | ok, it's not in the raphaelmemorymap |
| 15:34.19 | cr2 | no, it's included |
| 15:34.23 | cr2 | b1b00000 a0000000 1 |
| 15:38.07 | *** join/#htc-linux ALoGeNoff (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 15:49.04 | tmzt | cr2: does the cpld driver support the i2c mode? |
| 15:49.29 | tmzt | (this is for wizard actually but I'm wondering if there is a generic cpld driver) |
| 15:50.26 | cr2 | htc-egpio ? no. |
| 15:50.26 | miknix | tmzt, not really. each cpld behave differently |
| 15:50.35 | cr2 | miknix has written one |
| 15:50.37 | cr2 | ok |
| 15:51.04 | miknix | it works with htc herald and htc wizard so far |
| 15:51.08 | *** join/#htc-linux stefan_schmidt (n=stefan@p5B034270.dip.t-dialin.net) |
| 15:51.16 | cr2 | tmzt: the generic gpio+irq cpld driver is htc-egpio |
| 15:51.37 | cr2 | miknix: artemis is the only omap850 exception |
| 15:51.48 | cr2 | which does not use i2c-cpld |
| 15:53.13 | *** join/#htc-linux ALoGeNo (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 15:53.52 | miknix | cr2, so, there are other omap850 devices using it? |
| 15:54.38 | cr2 | i'm looking at the MSM_CLK list and at the clock-7x01a.h |
| 15:54.45 | cr2 | miknix: the only one i know |
| 15:54.54 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 15:55.26 | tmzt | maybe excalibur, not sure if we confirmed if it does/does not use it |
| 15:55.28 | cr2 | usb is after uart ... |
| 15:55.48 | cr2 | tmzt spl will tell you |
| 15:56.19 | tmzt | I don't have it, penguinz was working on it some time ago |
| 15:56.32 | cr2 | hmm, +0xe0 is usb clock then |
| 15:56.55 | tmzt | why do you say usb is after uart? |
| 15:57.06 | tmzt | and how does this tell you the memory order |
| 15:57.09 | cr2 | in the list |
| 15:57.31 | cr2 | they don't match 100%, but the order is very similar |
| 15:58.04 | cr2 | goolge is probably not evil, they have taken the spec and didn't mix it :) |
| 15:58.33 | cr2 | the g1 vreg order 100% fits the dex masks |
| 15:59.27 | miknix | the google thing was slashdoted |
| 16:00.25 | *** join/#htc-linux ppman (n=mkern@pool-70-111-235-140.nwrk.east.verizon.net) |
| 16:03.16 | cr2 | if my therory is right, then "Stereo DAC clock " is between the pmdh and SD |
| 16:04.05 | cr2 | hm. there is an unknown clock there in between |
| 16:04.57 | cr2 | the i2c is at the wrong place, though |
| 16:05.52 | tmzt | ah, I see |
| 16:06.02 | tmzt | what's missing then? |
| 16:06.14 | tmzt | I assume the grp and imem are together |
| 16:06.18 | cr2 | emdh is cam ? |
| 16:06.28 | tmzt | probably, it's mddi isn't it? |
| 16:06.36 | cr2 | external mddi host |
| 16:06.49 | tmzt | yeah, I mean cam is mddi isn' it? |
| 16:06.56 | cr2 | should be the camera, accordig to the qcom patents ;) |
| 16:07.15 | cr2 | i don't know, but there are mddi cams |
| 16:07.26 | tmzt | hah, you have two grp_clk's? |
| 16:07.31 | cr2 | but then the cam is a host, and msm is the client |
| 16:07.38 | cr2 | 1 |
| 16:07.44 | tmzt | in wiki |
| 16:07.51 | cr2 | but the 0x80 register is related to grp_clk |
| 16:07.57 | tmzt | +0x80 and +0x84 |
| 16:08.23 | cr2 | i have evin more sdc_clk |
| 16:08.33 | cr2 | s/evin/even/ |
| 16:08.49 | tmzt | I know, but I assume those are sdc1,2,3,4 and Md/Ns? |
| 16:09.01 | tmzt | or, pclk and clk |
| 16:09.02 | tmzt | sorry |
| 16:09.28 | cr2 | clock-7x01a.c |
| 16:09.51 | tmzt | I'm saying the hw3d.c suggests one of those is imem if you grp-related means they change together |
| 16:09.57 | cr2 | yes, we need to think about clk_api |
| 16:10.14 | cr2 | how to implement it in a non-evil way |
| 16:10.42 | *** join/#htc-linux j0b0 (n=jobo@5ED40048.cable.ziggo.nl) |
| 16:10.58 | cr2 | icdec_rx and tx are after i2 ? |
| 16:11.16 | cr2 | looks like that in the list |
| 16:11.17 | tmzt | icdec? |
| 16:11.23 | cr2 | icodec |
| 16:11.49 | tmzt | which is? Stereo DAC? |
| 16:11.54 | cr2 | no |
| 16:11.59 | cr2 | sdac is later |
| 16:12.18 | tmzt | then what? do you know? |
| 16:12.35 | cr2 | 6c-78 may be icodec_rx and tx |
| 16:12.40 | cr2 | are they used somewhere ? |
| 16:13.22 | cr2 | sdac is between mddi and sd |
| 16:13.43 | cr2 | maybe 98-9c |
| 16:13.45 | tmzt | no |
| 16:13.52 | tmzt | not used in git-grep of android |
| 16:14.53 | cr2 | ok |
| 16:14.56 | cr2 | sdac ? |
| 16:15.13 | cr2 | the regs in the table are those referenced by oemsbl |
| 16:15.19 | tmzt | no, odd |
| 16:15.31 | cr2 | ok |
| 16:15.47 | cr2 | they may use rpc et al. |
| 16:15.47 | tmzt | they are in clock-7x01a.c also but nowhere else |
| 16:15.55 | tmzt | not clock api? |
| 16:16.07 | cr2 | g1 code has many hacks |
| 16:16.37 | tmzt | it enables codecs by name, that could activate different clocks on the arm9 side? |
| 16:17.15 | cr2 | yes |
| 16:17.39 | cr2 | wince also implements the proc_comm clk api. but does not use it ;) |
| 16:17.51 | cr2 | so there are 2 clock groups |
| 16:18.01 | cr2 | one is accessng the regs directly |
| 16:18.06 | cr2 | they are all documented |
| 16:18.24 | cr2 | and there are some sec_clk calls |
| 16:18.29 | tmzt | what address? |
| 16:18.31 | cr2 | that go into nowhere |
| 16:18.39 | cr2 | at least i can't track them |
| 16:18.50 | cr2 | address ? |
| 16:19.16 | tmzt | sorry, I was thinking we could search for a writel to it but it would be macro |
| 16:20.33 | cr2 | there are 3 0xe0 groups: |
| 16:21.05 | cr2 | 10,20; 400,800; 1000,2000 |
| 16:21.16 | cr2 | if they are for usb |
| 16:21.33 | cr2 | then clocks-7x01a.h has 3 usb clocks |
| 16:22.05 | cr2 | hs clk, hs pclk and otg |
| 16:22.18 | cr2 | makes sense |
| 16:22.29 | cr2 | but needs to be traced somehow. |
| 16:23.10 | tmzt | over bluetooth/wifi? |
| 16:24.19 | tmzt | http://android.git.kernel.org/?p=kernel/msm.git;a=blob;f=arch/arm/mach-msm/clock-7x01a.c#l47 |
| 16:24.34 | tmzt | USB_OTG_CLK 38 /* Full speed USB clock */ |
| 16:24.35 | cr2 | the idx 0,a,c are one group |
| 16:24.47 | tmzt | assuming then otg host only supports 1.1 speeds |
| 16:25.03 | tmzt | (although otg is really 2.0) |
| 16:25.07 | cr2 | swetland said ehci, when i've said ohci :) |
| 16:25.35 | cr2 | so it is 2.0 |
| 16:26.08 | cr2 | 0 enables bit 0x200, a is mddi and c is grp |
| 16:26.21 | *** join/#htc-linux |patoche| (n=patoche@lns-bzn-58-82-251-208-84.adsl.proxad.net) |
| 16:26.23 | cr2 | so this bit 0x200 is mddi/grp related |
| 16:27.10 | cr2 | then 1,5,3,6,7,b |
| 16:27.16 | cr2 | b is some generic |
| 16:27.43 | *** join/#htc-linux Andrew3 (n=Andy@71-211-46-165.clsp.qwest.net) |
| 16:27.44 | cr2 | 6,7 sdc1 and 2 |
| 16:27.56 | *** part/#htc-linux Andrew3 (n=Andy@71-211-46-165.clsp.qwest.net) |
| 16:28.01 | cr2 | 3 is uart2dm |
| 16:28.15 | cr2 | sorry |
| 16:28.20 | cr2 | 5 is uart2dm |
| 16:28.35 | *** join/#htc-linux AndyCR (n=Andy@71-211-46-165.clsp.qwest.net) |
| 16:28.41 | tmzt | 20:15 < swetland> yeah, the hsusb controller is otg capable, has a ehci host controller |
| 16:28.46 | cr2 | 1 and 3 are usb something |
| 16:29.02 | cr2 | tmzt: but there is no driver for it. |
| 16:29.16 | tmzt | for what, the otg part? |
| 16:29.18 | cr2 | but we can find the ehci location |
| 16:29.29 | cr2 | for ehci host |
| 16:29.43 | tmzt | I think the biggest differences are ulpi registers |
| 16:29.59 | tmzt | like that was the block that had to be changed between non-A and A |
| 16:30.54 | tmzt | #define USB_HS_PCLK 37 /* High speed USB pbus clock *? |
| 16:30.59 | tmzt | what is pbus? |
| 16:31.25 | tmzt | and what is pbus_clk (searched for pbus in clock-7x) |
| 16:31.29 | *** join/#htc-linux _patoche_ (n=patoche@lns-bzn-58-82-251-208-84.adsl.proxad.net) |
| 16:31.39 | cr2 | remote_apis_clnt.dll where is it ? |
| 16:31.54 | cr2 | peripheral bus |
| 16:31.56 | tmzt | what? |
| 16:32.08 | cr2 | pbus=peripheral bus |
| 16:32.12 | tmzt | I've only ever seen rapiclnt |
| 16:32.25 | tmzt | maybe in the .NET dlls |
| 16:32.29 | cr2 | no |
| 16:32.34 | cr2 | in the phone |
| 16:32.41 | cr2 | i don't see such dll |
| 16:32.50 | tmzt | yeah, there is .net in wm |
| 16:33.10 | *** join/#htc-linux dream_kill (n=nospam@92.56.48.66) |
| 16:33.11 | cr2 | but clkregim.dll says some clocks are exported by remote_apis_clnt.dll |
| 16:33.32 | tmzt | what is clkregim.dll? |
| 16:34.24 | tmzt | are all the pclks's related to pbus_clk? |
| 16:34.32 | cr2 | no idea |
| 16:34.39 | cr2 | maybe |
| 16:35.12 | tmzt | IMEM Internal graphics memory clock |
| 16:35.55 | cr2 | found remore_apis_ |
| 16:36.00 | cr2 | it's a xip dll |
| 16:36.41 | cr2 | and it's some rpc stuff |
| 16:36.55 | cr2 | not surprising for the xip |
| 16:38.09 | cr2 | time for the rpc wiki page ;) |
| 16:39.56 | *** join/#htc-linux _patoch3 (n=patoche@lns-bzn-58-82-251-208-84.adsl.proxad.net) |
| 16:40.00 | cr2 | hehe. vbatt_read |
| 16:40.02 | tmzt | VFE_CLK 40 /* Camera / Video Front End clock */ |
| 16:40.28 | tmzt | VFE_MDC_CLK 41 /* VFE MDDI client clock */ |
| 16:40.56 | tmzt | Internal CODEX RX clock |
| 16:41.04 | tmzt | Internal CODEX TX clock |
| 16:41.16 | tmzt | this is onboard audio adc/dac? |
| 16:42.22 | cr2 | probably |
| 16:42.40 | tmzt | not used then if external codec is used |
| 16:43.03 | tmzt | PCM_CLK 15 |
| 16:50.15 | *** join/#htc-linux |patoche| (n=patoche@lns-bzn-58-82-251-208-84.adsl.proxad.net) |
| 16:55.37 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87ed28.pool.einsundeins.de) |
| 17:00.03 | *** join/#htc-linux ALoGeNoff (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 17:01.32 | *** join/#htc-linux balsa1 (n=kll@87.72.13.87) |
| 17:03.19 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 17:04.21 | *** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net) |
| 17:07.43 | *** join/#htc-linux Mullins (n=bw@89.204.239.174) |
| 17:17.54 | *** join/#htc-linux timebomb (n=tb@e176127158.adsl.alicedsl.de) |
| 17:35.06 | *** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey) |
| 17:45.49 | *** join/#htc-linux ALoGeNo (n=alogeno@137.pool85-60-2.dynamic.orange.es) |
| 17:46.55 | *** join/#htc-linux ykut_johny (n=ykut_joh@op.niser.org.my) |
| 17:54.44 | *** join/#htc-linux shoragan_ (n=shoragan@sicherheitsschwankung.de) |
| 17:55.54 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) [NETSPLIT VICTIM] |
| 17:56.16 | *** join/#htc-linux konsta (n=asds@host81-157-60-163.range81-157.btcentralplus.com) |
| 17:56.33 | *** join/#htc-linux _Nono (n=arnaud@rps4056.ovh.net) [NETSPLIT VICTIM] |
| 17:57.33 | *** join/#htc-linux goxboxlive (n=goxboxli@185.84-48-126.nextgentel.com) |
| 18:00.00 | *** join/#htc-linux Mullins (n=bw@89.204.239.174) [NETSPLIT VICTIM] |
| 18:05.06 | *** join/#htc-linux stefan_schmidt (n=stefan@p5B0376FD.dip.t-dialin.net) |
| 18:08.02 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) [NETSPLIT VICTIM] |
| 18:09.15 | *** join/#htc-linux LemHxC (n=nimp@cnv94-1-82-232-227-160.fbx.proxad.net) |
| 18:13.22 | LemHxC | is away: vadrouille |
| 18:21.04 | *** join/#htc-linux nashpa (n=dliviu@dliviu.plus.com) |
| 18:44.36 | *** join/#htc-linux Marajin_ (n=marajin@87-194-102-189.bethere.co.uk) |
| 18:58.41 | *** join/#htc-linux patoche (n=patoche@lns-bzn-54-82-251-114-125.adsl.proxad.net) |
| 19:15.33 | *** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes) |
| 19:16.12 | *** join/#htc-linux Jon__ (n=Jon@94-194-181-77.zone8.bethere.co.uk) |
| 19:23.11 | *** join/#htc-linux jack_ (n=ASUS@f048251105.adsl.alicedsl.de) |
| 19:23.16 | jack_ | hi |
| 19:25.37 | *** join/#htc-linux AstainHellbring (n=arto@unaffiliated/astainhellbring) |
| 19:26.50 | jack_ | where are the htc-msm-2.6.27 autobuilds? |
| 19:27.54 | jack_ | I need the kernel |
| 19:31.28 | jack_ | I have new 7201a board |
| 19:43.10 | cr2 | jack_: there is no such thing as 7201a board |
| 19:53.57 | *** join/#htc-linux kill_me2 (n=km2@99-7-110-97.lightspeed.lsanca.sbcglobal.net) |
| 19:56.37 | *** join/#htc-linux MethoS-- (n=lem@dyndsl-085-016-165-155.ewe-ip-backbone.de) |
| 20:00.06 | *** join/#htc-linux patoche (n=patoche@lns-bzn-54-82-251-114-125.adsl.proxad.net) |
| 20:02.34 | *** part/#htc-linux balsa1 (n=kll@87.72.13.87) |
| 20:10.57 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 20:31.56 | *** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 20:33.49 | *** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net) |
| 20:41.53 | *** join/#htc-linux Jon__ (n=Jon@94-194-181-77.zone8.bethere.co.uk) |
| 20:55.53 | *** join/#htc-linux chab7 (n=kvirc@212.92.4.114) |
| 21:07.10 | *** join/#htc-linux Pure4Real (n=pure4rea@89-97-140-219.ip17.fastwebnet.it) |
| 21:09.13 | cr2 | it's quiet today |
| 21:09.42 | Pure4Real | so i notice |
| 21:10.34 | Pure4Real | was just reading up a bit on irc logs |
| 21:10.44 | Pure4Real | who's working on the wifi driver atm? |
| 21:11.36 | cr2 | maejrep compiled and run it, but the driver needs some fixes |
| 21:12.36 | Pure4Real | and that's based on the acx100 driver? |
| 21:12.51 | *** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde) |
| 21:17.49 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 21:19.33 | cr2 | no |
| 21:19.50 | cr2 | it's the tiacx driver written by ti+android people |
| 21:20.20 | Pure4Real | ah |
| 21:23.48 | *** part/#htc-linux Jon__ (n=Jon@94-194-181-77.zone8.bethere.co.uk) |
| 21:26.35 | *** join/#htc-linux reformatt (n=chatzill@97-121-30-66.bois.qwest.net) |
| 21:28.59 | *** join/#htc-linux MethoS- (n=lem@host-091-097-241-018.ewe-ip-backbone.de) |
| 21:48.23 | *** join/#htc-linux Venny (i=Travis@h216-45-127-2.dynamic.platinum.ca) |
| 22:14.55 | nashpa | cr2, where can the tiacx driver be found? |
| 22:17.20 | cr2 | it's better to wait for maejrep |
| 22:17.51 | maejrep | the code is in android's git |
| 22:18.11 | maejrep | but does take some tweaks to be usable |
| 22:18.36 | Pure4Real | do you have a description of the tweaks somewhere? |
| 22:18.47 | maejrep | not yet |
| 22:25.43 | *** join/#htc-linux Jon__ (n=Jon@94-194-181-77.zone8.bethere.co.uk) |
| 22:28.28 | *** join/#htc-linux vmpu (n=pui@89-33-138-94.u-nite.ro) |
| 22:31.50 | cr2 | maejrep: i've detected most clocks. the rest is accessed over rpc (if at all) |
| 22:32.05 | maejrep | cool |
| 22:32.18 | maejrep | yeah rpc is another beast we haven't started on yet :p |
| 22:32.49 | nashpa | what device are you guys working on? |
| 22:33.03 | cr2 | i've also documented all mddi init differences between g1,raph100,raph800 and blac100 |
| 22:33.28 | cr2 | maejrep: imvho g1 disables the autorefresh |
| 22:33.39 | *** join/#htc-linux Jon__ (n=Jon@94-194-181-77.zone8.bethere.co.uk) |
| 22:34.12 | cr2 | nashpa: mainly raph?00 |
| 22:34.29 | maejrep | ah nice |
| 22:34.38 | nashpa | 7201A? or 7200A? |
| 22:34.39 | maejrep | autorefresh of what? |
| 22:34.45 | maejrep | 7x01a |
| 22:34.55 | maejrep | raph100 is 7201a, raph800 is 7501a |
| 22:34.58 | cr2 | maejrep: mddi |
| 22:35.32 | maejrep | cr2: which means what? :x it doesn't refresh the screen? :x or handles it all in software |
| 22:35.34 | nashpa | ah, the difference is only in the companion chips, one is GSM, other is CDMA |
| 22:35.45 | *** join/#htc-linux timebomb (n=tb@e176099053.adsl.alicedsl.de) |
| 22:35.54 | cr2 | maejrep: i think it handles it in software |
| 22:36.03 | maejrep | nashpa: right, but i think raph500 is actually 7500a (?) |
| 22:36.14 | maejrep | well that's lame |
| 22:36.24 | cr2 | 7501a _is_ 7500a :) |
| 22:36.29 | maejrep | orl |
| 22:36.31 | maejrep | +y |
| 22:37.09 | cr2 | maejrep: it may save some mA. and we don't know the special 3D requirements |
| 22:37.27 | maejrep | true |
| 22:37.32 | cr2 | maejrep: i've also added the HW3D area to memory map. |
| 22:37.58 | cr2 | and the per-device vreg table |
| 22:38.24 | cr2 | and the usb host alt gpios |
| 22:38.38 | cr2 | but there is no ehci-msm driver afaik |
| 22:39.16 | maejrep | hmm, wifi vreg level is set to 0 ? |
| 22:39.17 | cr2 | and the raph manual says 32MB stacked ram |
| 22:39.37 | cr2 | 0 means don't touch |
| 22:39.39 | cr2 | in enable |
| 22:39.46 | cr2 | idx,1,0 |
| 22:39.48 | maejrep | so just leave it as is? |
| 22:39.51 | cr2 | yes |
| 22:39.56 | maejrep | hmm |
| 22:40.22 | maejrep | what happens if it wasn't initialized properly by windows yet? |
| 22:40.22 | cr2 | enable without changing the level |
| 22:40.33 | cr2 | by oemsbl |
| 22:40.39 | maejrep | ah |
| 22:40.48 | cr2 | g1 does not init a _lot_ of things too |
| 22:40.59 | maejrep | i see |
| 22:41.18 | cr2 | and i've documented all your alt gpios |
| 22:41.26 | cr2 | your = raph800 |
| 22:41.59 | maejrep | what is "a/m" in the gpio page? |
| 22:42.55 | cr2 | application/modem = arm11/arm9 owner |
| 22:43.00 | maejrep | ah |
| 22:43.22 | cr2 | but this column is unfinished |
| 22:43.37 | cr2 | and some gpios may change the owner and runtime :) |
| 22:43.41 | maejrep | you saw my gpio dump at the bottom? |
| 22:43.49 | cr2 | oemsbl has some variations too. |
| 22:43.56 | cr2 | where ? |
| 22:44.06 | maejrep | link at the bottom of that page: http://wiki.xda-developers.com/index.php?pagename=Raph800GPIODump |
| 22:45.07 | cr2 | yes. but it's more interesting for haret |
| 22:46.00 | maejrep | any idea why I don't see it enabling/disabling vregs when turning wifi on and off? :o |
| 22:46.19 | cr2 | 25 0x19 0 9 O soft reset=0 |
| 22:46.30 | cr2 | GPIO25: Owner=Arm9, Dir=O, Level=H |
| 22:47.02 | cr2 | this is the gpio that changes the owner btw. |
| 22:47.06 | maejrep | i've tried gpio_set_value(25,0) and it doesn't work :x |
| 22:47.18 | cr2 | you need to change the owner first |
| 22:47.25 | maejrep | oh? |
| 22:47.32 | maejrep | i guess that makes sense |
| 22:47.44 | cr2 | in your table arm9 is the owner |
| 22:48.08 | cr2 | btw you can try to modify the arm9 gpio directly :) |
| 22:48.15 | cr2 | but then it may crash |
| 22:48.54 | cr2 | i doubt that the MPU can have control on per-register basis |
| 22:48.57 | maejrep | so you change that at a9100504 ? |
| 22:49.14 | cr2 | since we have access to the arm9 gpio area, we can control it completely. |
| 22:49.36 | cr2 | change the owner bit for this gpio |
| 22:49.49 | maejrep | k i'll try that |
| 22:49.50 | cr2 | at least it's how wince does it |
| 22:50.04 | maejrep | will also work on putting patches together for the small tweaks i've done |
| 22:50.14 | cr2 | writing directly to the arm9 address seems risky |
| 22:50.38 | cr2 | ok |
| 22:51.07 | cr2 | do you have write access to linuxtogo git ? |
| 22:51.54 | maejrep | no |
| 22:51.59 | cr2 | hmm |
| 22:52.05 | maejrep | i have an account, but haven't asked to be added to the project yet |
| 22:52.07 | cr2 | NetRipper: are you alive ? |
| 22:54.22 | cr2 | maejrep: do you have a bt headset ? |
| 22:54.49 | maejrep | somewhere :p |
| 22:55.35 | cr2 | i'm interested in the MSM_CLK dump, while the bt audio is active |
| 22:56.11 | maejrep | as in, while on a call? or just while paired? |
| 22:56.16 | maejrep | or while listening to music |
| 22:57.28 | maejrep | cr2: owner bit is.. 0=arm9,1=arm11? or the other way |
| 22:57.41 | maejrep | i guess I could dump the regs to find out :p |
| 22:58.10 | cr2 | i forgot it :) |
| 22:58.40 | cr2 | call or listening musi |
| 22:58.50 | cr2 | i think call is better |
| 22:59.14 | cr2 | because for music they can use sco over 4mbit serial |
| 22:59.56 | maejrep | 0=arm9 |
| 23:00.17 | maejrep | HaRET(1)# pdump 0xa9100504 0x4 |
| 23:00.17 | maejrep | a9100504 | 07fffcbe | .... |
| 23:00.36 | maejrep | at least, if my gpio dump is close to accurate (I assume windows doesn't change everything ;) |
| 23:00.54 | maejrep | all the arm9 bits are 0 in that value ^ |
| 23:03.07 | maejrep | hmm, msm_gpio_configure() doesn't have any kind of lock |
| 23:03.55 | cr2 | wince disables interrupts there |
| 23:04.06 | cr2 | which makes sense |
| 23:05.26 | maejrep | right |
| 23:05.31 | maejrep | but msm_gpio doesn't :x |
| 23:05.38 | maejrep | unless its handled further u |
| 23:05.40 | maejrep | up* |
| 23:05.51 | maejrep | which could make sense.. gpio support is handled at a higher level |
| 23:05.55 | maejrep | <PROTECTED> |
| 23:05.55 | maejrep | <PROTECTED> |
| 23:06.12 | maejrep | (i left the 25 there just for you :D ) |
| 23:06.44 | maejrep | so, if you pass it either GPIOF_OWNER_ARM9 or GPIOF_OWNER_ARM11, it'll either ORR or BIC the respective bit |
| 23:09.15 | NetRipper | cr2, i am, but im a bit busy |
| 23:09.17 | NetRipper | whats up |
| 23:10.15 | NetRipper | maejrep, if you got patches to commit, just pm me the link :) to get access to git yourself, you can go to the mobile linux project on projects.linuxtogo.org and "apply" somewhere on the right-side of the main 'linux mobile' project page |
| 23:10.31 | NetRipper | pH5 or Kevin2 will then receive an e-mail to allow you access |
| 23:10.41 | maejrep | ah |
| 23:10.48 | maejrep | I'm working on creating diffs right now |
| 23:11.02 | maejrep | not in any hurry to get access myself ;p |
| 23:11.05 | NetRipper | ;) |
| 23:11.27 | NetRipper | only thing i want to keep nice in the git is support for multiple devices.. i.e. proper #ifdefs and machine_is_* stuff |
| 23:11.30 | maejrep | and i figure it's better that patches go through one person, to review code, etc |
| 23:11.41 | maejrep | nods |
| 23:12.30 | NetRipper | well im not the most experienced guy.. i also like the idea if people check up on my commits when i put something on linuxtogo ;) |
| 23:12.51 | cr2 | NetRipper: ok |
| 23:13.44 | NetRipper | cr2, btw, i havent been using the trout memory layout yet.. i first wanted to sort the ts/vkeyb/msmfb out.. im hitting performance issues |
| 23:13.53 | cr2 | maejrep: i'll put the mddi init into wiki. the code there needs some changes |
| 23:14.05 | NetRipper | i have added a 'callback' mechanism in the msmfb to allow vkeyb and ts calibration to draw on the framebuffer just before the dma starts |
| 23:14.21 | cr2 | NetRipper: the manual says raph has 32MB. and i can't read from smem for some reason. |
| 23:14.36 | NetRipper | but its causing a major impact on graphics, too much to be true ;) |
| 23:14.51 | NetRipper | cr2, read from within wince? |
| 23:14.59 | cr2 | yes, in haret |
| 23:15.13 | NetRipper | is it protected memory? locked by oemsbl? |
| 23:15.14 | cr2 | it may be protected on several levels |
| 23:15.40 | cr2 | maybe locked, maybe has the evil page table bits set |
| 23:15.46 | NetRipper | cr2, you could ask cmonex in #xda-devs, he knows pretty much about what they locked.. and if it's unlockable... |
| 23:15.52 | cr2 | locked = locked by MPU |
| 23:16.12 | NetRipper | there's also an sdcard based unlock |
| 23:16.33 | cr2 | the evil bits can be changed, but the MPU will probably need an oemsbl reflash. if it 's possible ;) |
| 23:16.35 | NetRipper | don't we disable the MPU in haret when linux boots? |
| 23:16.51 | NetRipper | or is MPU != MMU? |
| 23:16.57 | cr2 | no. |
| 23:17.07 | cr2 | MemoryProtectionUnit |
| 23:17.28 | NetRipper | ok |
| 23:17.33 | cr2 | it's some DMA/MA restriction |
| 23:17.49 | cr2 | NetRipper: everything that SPL does, linux can do too. |
| 23:18.06 | *** join/#htc-linux Dvyjones (n=dvyjones@unaffiliated/dvyjones) |
| 23:18.06 | cr2 | because the SPL has the same level of permissions as linux. |
| 23:18.27 | Dvyjones | Hi everybody! |
| 23:18.42 | Dvyjones | will start porting Linux to HTC Pharos again :P |
| 23:18.57 | NetRipper | cr2, ok, but SPL has nothing to do with smem, so it could be normal that we cant access it |
| 23:19.04 | cr2 | NetRipper: we need to fix more urgent problems first, before reflashng oemsbl :) |
| 23:19.12 | NetRipper | yes |
| 23:19.17 | NetRipper | of course ;) |
| 23:19.45 | NetRipper | but i wont reflash oemsbl on my device... if it would be possible at all.. |
| 23:19.54 | cr2 | i'll add detecting the smem size to my wiki todo list |
| 23:20.02 | NetRipper | we'd need jtag and the gold card in case the flash fails |
| 23:20.13 | Kevin2 | evin |
| 23:20.29 | Kevin2 | Heh, wrong window. :-)] |
| 23:20.32 | NetRipper | ;) |
| 23:20.33 | cr2 | no gold card will help you on oemsbl failure |
| 23:21.00 | NetRipper | cr2, gold card is needed to do jtag, isnt it? and with jtag it should be possible to recover from oemsbl error |
| 23:21.11 | *** join/#htc-linux ale_____ (n=alessand@host248-205-dynamic.19-79-r.retail.telecomitalia.it) |
| 23:21.18 | NetRipper | it's a low-level msm7k trick afaik |
| 23:22.39 | cr2 | ok |
| 23:23.10 | cr2 | it's beyond my knowlegde level :) |
| 23:24.26 | NetRipper | hehe same here i guess ;) |
| 23:26.03 | maejrep | cr2: to be sure I'm reading this right.. LDR R2, =0xBC0FC120 <-- means R2 = the value at address bc0fc120? not the number bc0fc120 |
| 23:26.30 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 23:27.51 | cr2 | maejrep: number |
| 23:27.53 | maejrep | actually, no I guess that's the other way around... cause then it uses [R2] in ldr and str calls |
| 23:28.27 | cr2 | maejrep: do you think that proc_comm_dex is a good idea ? |
| 23:28.32 | maejrep | so what's the difference between LDR R2, =0xDEADBEEF; and MOV R2, #0xDEADBEEF; ? |
| 23:28.45 | maejrep | you mean instead of _wince ? |
| 23:28.59 | cr2 | not instead, but an additional api |
| 23:29.15 | maejrep | why would it need to be additional? |
| 23:29.27 | cr2 | mov r2, #deadbeef does not work as a single machine op. afair |
| 23:29.31 | maejrep | shouldn't proc_comm cover everything that data exchanger does? |
| 23:29.55 | cr2 | it's easier to compare with wince drivers |
| 23:30.08 | cr2 | while you use the g1 api for proc_comm |
| 23:30.22 | cr2 | which looks a bit artificial to me :) |
| 23:30.40 | cr2 | with things like 0x18a instead of 0x8a |
| 23:30.51 | maejrep | I don't pass 0x18a |
| 23:31.02 | maejrep | the driver adds the 0x100 when data exists |
| 23:31.40 | maejrep | which is how wince does it :x |
| 23:31.58 | maejrep | actually i'm reading through it again now, to make sure I fix any remaining bugs in the _wince implementation |
| 23:32.11 | maejrep | if you want, I can rename it to msm_data_exchanger in the process ;) |
| 23:34.46 | cr2 | nothing against proc_comm, but it's easier to track the wince api. |
| 23:36.34 | maejrep | i don't care either way |
| 23:37.10 | maejrep | if it's easier to name things DEX and track the wince dex function, then that makes sense |
| 23:42.51 | *** join/#htc-linux ginge (n=kvirc@78-105-175-81.zone3.bethere.co.uk) |
| 23:44.05 | maejrep | cr2: what's the point of this? |
| 23:44.06 | maejrep | STRB R3, [R4] |
| 23:44.06 | maejrep | LDR R3, [R4] |
| 23:44.25 | maejrep | is it just to truncate R3 to a 0xff mask? |
| 23:46.15 | cr2 | strange code, but yes |
| 23:48.00 | maejrep | so, it looks like the wince dex code actually does send 0x100 with the command, to tell the driver when there's data to be read... is that better than what we're doing, which is check if data1 != 0? |
| 23:48.16 | maejrep | s/to be read/to be sent/ |
| 23:49.04 | cr2 | wince dex puts 0x1 byte before the cmd in the buffer |
| 23:49.20 | cr2 | if there are external parameters |
| 23:51.03 | *** join/#htc-linux Kuma (n=John@e179103181.adsl.alicedsl.de) |
| 23:54.22 | maejrep | so why not just take care of adding that in the driver, instead of doing it in the function call? |
| 23:55.12 | maejrep | ie, why pass dex( 0x18a, data1, 0), vs dex(0x8b, 0, 0), when you could just call dex( 0x8a, data1, 0 ), and let dex() add the 0x100 to the command, when data1 != 0 |
