| 00:19.59 | BabelO | good night |
| 00:55.31 | *** join/#htc-linux noelcragg (n=noel@adsl-69-232-228-255.dsl.pltn13.pacbell.net) |
| 00:59.08 | noelcragg | apologies for the newbie question: is there a wiki page that contains pointers to the latest source / binaries for HTC Sable? |
| 01:16.47 | *** join/#htc-linux Marex (n=marex@gwfm10-3-250.802.cz) |
| 02:13.16 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 02:25.27 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
| 03:38.41 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfb10f.pool.einsundeins.de) |
| 03:50.10 | *** join/#htc-linux dcordes (n=snoopdog@unaffiliated/dcordes) |
| 04:01.13 | *** join/#htc-linux joschkop179 (n=joschi@ACAE2306.ipt.aol.com) |
| 04:59.41 | *** join/#htc-linux joschkopf79 (n=joschi@AC9E531D.ipt.aol.com) |
| 06:31.27 | *** join/#htc-linux CVirus (n=GoD@196.205.192.125) |
| 07:37.51 | *** join/#htc-linux goxboxlive (n=goxboxli@208.84-48-176.nextgentel.com) |
| 08:54.34 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 09:42.13 | *** join/#htc-linux dcordes (n=snoopdog@unaffiliated/dcordes) |
| 09:56.04 | *** join/#htc-linux pH5 (n=ph5@p5485DA1E.dip.t-dialin.net) |
| 09:59.26 | *** join/#htc-linux swc-666 (n=no@24.143.67.18) |
| 10:35.01 | *** part/#htc-linux swc-666 (n=no@24.143.67.18) |
| 11:09.50 | *** join/#htc-linux patoche (n=patoche@lns-bzn-44-82-249-192-208.adsl.proxad.net) |
| 11:16.36 | *** join/#htc-linux Marex (n=marex@gwfm10-3-250.802.cz) |
| 11:17.34 | *** join/#htc-linux kiozen (n=kiozen@rgnb-4db1d929.pool.einsundeins.de) |
| 11:52.45 | *** join/#htc-linux patoche (n=patoche@lns-bzn-44-82-249-192-208.adsl.proxad.net) |
| 12:04.20 | *** join/#htc-linux rob_w|laptop (n=rob@X0930.x.pppool.de) |
| 12:04.44 | Marex | pH5, cr2, hi |
| 12:05.00 | Marex | pH5, cr2, I managed to switch the camera on, so it appears at 0x30 on i2c |
| 12:05.31 | Marex | pH5, cr2, Now is the time to bind i2c-ov96xx (or ov9640? ) |
| 12:06.50 | pH5 | hi Marex, that would be my next step, yes. |
| 12:07.27 | Marex | well I compiled it as a module, set address to 0x30, but when I insmoded it, no function from that module was called |
| 12:07.32 | Marex | I did some stupid mistake :/ |
| 12:13.29 | pH5 | Marex: the current code is a bit ugly, the i2c_ov96xx_init etc. functions are called from the ov96xx driver in media/video |
| 12:13.47 | Marex | ah I see |
| 12:14.02 | Marex | I surely havent compiled that one, thanks |
| 12:33.43 | *** join/#htc-linux dcordes (n=snoopdog@unaffiliated/dcordes) |
| 12:41.58 | *** join/#htc-linux BabelO (n=BabelO@lun34-2-82-238-28-28.fbx.proxad.net) |
| 12:42.00 | *** join/#htc-linux BabelO_ (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 12:43.13 | BabelO | Hi |
| 12:45.10 | dcordes | hi BabelO |
| 12:50.11 | BabelO | pH5: i install scratchbox, not so easy, installer script contain an error |
| 12:52.50 | dcordes | BabelO, what's that? |
| 12:54.57 | pH5 | hi BabelO, I don't remember having to change anything in there to get it to work.. |
| 12:54.59 | pH5 | hi dcordes |
| 12:55.12 | dcordes | hi pH5 |
| 12:59.10 | BabelO | pH5: for me login executable was only set to root owner, and can not start it |
| 12:59.51 | BabelO | pH5: i try to compile my usb webcam driver to run on maemo ;) |
| 13:00.28 | BabelO | dcordes: it is the n800 emulator |
| 13:03.46 | pH5 | BabelO: very useful |
| 13:04.20 | pH5 | my usb gadgets are similarly useless on the n810. a usb sound card, a sd card reader, a bluetooth dongle... |
| 13:05.23 | *** join/#htc-linux kiozen (n=kiozen@rgnb-4db1dfe4.pool.einsundeins.de) |
| 13:06.57 | *** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-e848c737e2563018) |
| 13:07.22 | BabelO | pH5: yes same here, my webcam has a better image and need less light |
| 13:07.34 | BabelO | Hi kiozen |
| 13:07.51 | kiozen | hi BabelO et al |
| 13:08.16 | pH5 | BabelO: true, there is no way it could be worse :) |
| 13:08.26 | pH5 | hi kiozen |
| 13:08.37 | dcordes | hi kiozen |
| 13:08.43 | dcordes | BabelO, gspca? |
| 13:09.07 | BabelO | dcordes: yes, the only one |
| 13:13.34 | dcordes | BabelO, does n800 have usb2 or is only old usb required for webcam? |
| 13:14.00 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 13:14.20 | BabelO | dcordes: don t know, maybe pH5 know that |
| 13:16.33 | pH5 | dcordes: n8x0 is usb2 for both client/host |
| 13:28.57 | dcordes | neat |
| 13:29.17 | dcordes | I would like one of those n810 but for now my emedded device budget is burnout |
| 14:23.08 | *** join/#htc-linux TimRiker (n=timr@70.1.153.92) |
| 15:03.41 | *** join/#htc-linux bernt_ (n=bernt@dslb-084-061-247-156.pools.arcor-ip.net) |
| 15:05.19 | *** part/#htc-linux BabelO (n=BabelO@lun34-2-82-238-28-28.fbx.proxad.net) |
| 15:33.28 | cr2 | hi |
| 15:51.40 | *** join/#htc-linux CVirus (n=GoD@82.201.222.170) |
| 16:22.54 | *** join/#htc-linux ltxda0 (n=ltxda@unaffiliated/ltxda) |
| 16:29.51 | noelcragg | which source tree are you folks working from: xanadux.cvs.sourceforge.net or handhelds.org? |
| 16:30.24 | *** join/#htc-linux gers (n=grts@p54A2CC38.dip.t-dialin.net) |
| 16:31.44 | cr2 | noelcragg: hh.org |
| 16:31.49 | noelcragg | thx |
| 16:42.05 | *** join/#htc-linux ltxda00 (n=ltxda@c-98-196-125-205.hsd1.tx.comcast.net) |
| 16:53.08 | Marex | cr2, hi |
| 16:53.21 | Marex | cr2, Im now looking at the SoC audio on 6515d |
| 16:53.33 | Marex | looks like it's the same as in blueangel ... |
| 16:53.57 | Marex | I see more and more differences between sable and beetles :E |
| 16:54.39 | noelcragg | Does anyone have info on which test points are JTAG pins on Sable? |
| 17:00.08 | BabelO_ | pH5: any idea why i have no network under scratchbox ? |
| 17:04.28 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
| 17:04.46 | *** join/#htc-linux BabelO (n=Fabrice@unaffiliated/babelo) |
| 17:20.43 | *** join/#htc-linux blingwadman (n=bling@CPE001839b40c4e-CM0011e6ed09da.cpe.net.cable.rogers.com) |
| 17:26.11 | gers | A webbrowser for qtopia? http://www.geekzone.co.nz/content.asp?contentid=7632 |
| 17:27.24 | *** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-a0e13288445bebfb) |
| 17:33.38 | goxboxlive | gers: this is great news, we relly need a web browser in Qtopia |
| 17:34.36 | cr2 | Marex: you need to check the clock source. BA is different from magician. and sable is different from beetles |
| 17:38.25 | Marex | cr2, I already checked that using haret |
| 17:38.50 | Marex | CKEN8 and CKEN23 - i2s and SSP |
| 17:39.38 | Marex | cr2, I mean ... blueangel driver might be possibly reused for beetles |
| 17:44.38 | *** join/#htc-linux LunohoD_ (n=alex@e180076110.adsl.alicedsl.de) |
| 18:00.56 | cr2 | Marex: yes, but does sable use the external SPI clock as BA ? |
| 18:01.04 | cr2 | s/sable/beetles/ |
| 18:01.44 | cr2 | since it's pxa27x, it's probably doing the same as magician. but i'm not sure. |
| 18:02.22 | Marex | hm, I dont see GPIO27 (SSP_EXT_CLK) changing |
| 18:02.30 | Marex | (using haret that is) |
| 18:02.47 | Marex | but 23,24 and 25 do change, so SSP is used for sure |
| 18:04.04 | Marex | cr2, btw I dont see that device on i2c at all :< |
| 18:04.05 | Marex | :/ |
| 18:06.58 | pH5 | Marex: is it using the alternate functions SSPSCLK, SSPSFRM and SSPTXD for 23,24,25? Does it set BCKD and ENB in the SACR0 register as magician? |
| 18:07.34 | Marex | nah, it doesnt use alt functions |
| 18:07.59 | Marex | but is that important to detect the device on i2c ? |
| 18:08.13 | Marex | I thought for I2C I needed just power and reset |
| 18:08.38 | pH5 | I don't remember, but with uda1380 there was something strange like some clock needed to drive the registers |
| 18:08.55 | pH5 | not sure if that was needed for detection though |
| 18:09.24 | Marex | pH5, yea, I already red about that strangeness |
| 18:09.27 | Marex | :-( |
| 18:11.55 | pH5 | hm, looking at the magician driver I only needed to power and reset in magician_init, the basic registers are run by the i2c clock |
| 18:12.49 | Marex | pH5, well I turned on SSP1 and SPI clock |
| 18:12.56 | Marex | s/SPI/I2C |
| 18:12.59 | Marex | s/SPI/I2C/ |
| 18:13.02 | Marex | :E |
| 18:13.09 | Marex | pH5, well I turned on SSP1 and I2S clock |
| 18:13.32 | Marex | i2c should be on since I was able to detect camera there |
| 18:13.47 | Marex | now I just need reset and power for that damned chip |
| 18:13.56 | Marex | pH5, how did you get the correct GPIOs ? |
| 18:14.27 | pH5 | Marex: watching gpios and playing sound |
| 18:14.33 | pH5 | (red button's clicks or whatever) |
| 18:15.07 | cr2 | Marex: check 'dump pxa27x' in gnuharet |
| 18:15.08 | Marex | pH5, hm, that only turns on and off GPIOA6 which is audio routing |
| 18:15.24 | Marex | cr2, ok, thank you |
| 18:15.33 | Marex | btw thanks for your patience ;-) |
| 18:15.39 | cr2 | Marex: that will dump and decode the most important registers. |
| 18:15.56 | pH5 | Marex: and also no pxa gpios switching? |
| 18:18.10 | Marex | pH5, just 23 24 25 |
| 18:18.21 | Marex | cr2, ok |
| 18:19.06 | pH5 | Marex: and they are all used as gpios, no alternate function? maybe you have your power/reset in there. |
| 18:19.48 | Marex | pH5, hm, three pins for that are a bit overkill, dont you think ? ;) |
| 18:20.09 | Marex | btw all are outputs with ALT 0 |
| 18:20.22 | pH5 | I agree. Though I wonder what else are they used for, as it's not SSP? |
| 18:21.49 | Marex | they can be SSP even without ALT |
| 18:22.13 | cr2 | http://wiki.xda-developers.com/index.php?pagename=BeetlesGPIO |
| 18:22.37 | cr2 | isn't it the same i2s IN, ssp out setup ? |
| 18:22.49 | cr2 | s/out/OUT/ |
| 18:24.34 | cr2 | Marex: just run 'dump pxa27x' while playing sound |
| 18:24.35 | pH5 | Marex: the ssp does drive those gpios even if they are set to af0?! or are they doing ssp in software? |
| 18:25.01 | pH5 | both would seem very strange. |
| 18:25.20 | Marex | pH5, Im not as advanced as you :) |
| 18:25.30 | pH5 | cr2: ah, you think they are set to af0 outputs to save power? |
| 18:25.38 | pH5 | (while not playing sound only) |
| 18:25.48 | pH5 | Marex: yet ;) |
| 18:26.26 | cr2 | pH5: they are switched to ALT only while sound outpput. |
| 18:26.29 | Marex | - CKEN : 0x01594ee9 = 00000001 01011001 01001110 11101001 (Page 0) |
| 18:26.29 | Marex | + CKEN : 0x01d94ee9 = 00000001 11011001 01001110 11101001 (Page 0) |
| 18:26.40 | Marex | SSP1 off -> on |
| 18:26.47 | Marex | - SSCR0_P1 : 0x4000001f = 01000000 00000000 00000000 00011111 (Page xxx) |
| 18:26.47 | Marex | + SSCR0_P1 : 0x4000009f = 01000000 00000000 00000000 10011111 (Page xxx) |
| 18:27.14 | pH5 | cr2: ok, I think they didn't do that on magician. might make sense though. or not, who knows.. |
| 18:27.16 | Marex | and some stuff with DMA, LCD, USB ... totally unrelated |
| 18:27.32 | Marex | the first was while not playing, the second while playing sound |
| 18:27.49 | cr2 | Marex: LCD is just the pxafb, you'd mask these gpios anyway. |
| 18:28.26 | cr2 | Marex: you should check the SSP1 clock register setup. |
| 18:28.28 | Marex | cr2, sure, I said those changes are unrelated |
| 18:28.44 | cr2 | i guess it's the same setup as magician. |
| 18:28.50 | Marex | yea, just checking PXA manual |
| 18:28.52 | cr2 | i.e. pxa27x clock. |
| 18:29.27 | pH5 | Marex: perhaps GPIOA6 is the codec power and you just don't see the reset gpio.. |
| 18:29.43 | pH5 | cr2: doesn't gnuharet still use the busy loop with register readout for gpio watching? |
| 18:29.52 | Marex | pH5, when I plug in the headphone reduction (without headphones) it changes to GPIOA3 |
| 18:29.59 | pH5 | in that case it can miss a short reset gpio toggle. |
| 18:30.02 | Marex | so those are both audio routing GPIOs |
| 18:30.13 | pH5 | Marex: oh, ok, I see. |
| 18:30.31 | pH5 | they are the mic and speaker amp power gpios |
| 18:30.56 | Marex | which are not needed now as well ;) |
| 18:31.10 | Marex | pH5, really ... now it's the most important to even turn that IC on |
| 18:31.31 | pH5 | Marex: do you have anything else on that board which could have gpios on it? |
| 18:31.35 | pH5 | a pasic2/3? |
| 18:31.41 | Marex | ASIC3 |
| 18:32.02 | Marex | and also there is some CPLD, but I think that's used for camera only |
| 18:32.16 | Marex | (I get loads of EGPIO stuff in gnuharet when camera is on) |
| 18:32.29 | pH5 | Marex: argh, nope, please watch this one again |
| 18:32.43 | pH5 | You just remind me that magician has its sound gpios on the cpld, too. |
| 18:33.05 | Marex | GOD DAMN :E |
| 18:33.16 | Marex | doesnt magician have xilinx CPLD ? |
| 18:33.23 | Marex | I have lattice iirc |
| 18:33.43 | pH5 | Marex: yes |
| 18:34.04 | *** join/#htc-linux joschkopf79 (n=joschi@f051107150.adsl.alicedsl.de) |
| 18:38.19 | Marex | hm, it only turns on SSE (synchronous serial transfers) in SSCR0 |
| 18:39.30 | Marex | and turns on SSP1 clock |
| 18:40.13 | pH5 | what about SACR? |
| 18:40.27 | pH5 | (and never mind about the clocks until you have found the power gpio ;)) |
| 18:40.50 | Marex | is it possible, that it's still on ? |
| 18:41.12 | pH5 | Marex: it is only powered while playing or recording sound |
| 18:41.22 | Marex | ok |
| 18:41.48 | Marex | <PROTECTED> |
| 18:44.06 | Marex | so BCKD is set to output |
| 18:46.48 | pH5 | that looks all like the magician setup |
| 18:47.52 | pH5 | SSCR0[ACS] selects the audio clock to drive the uda1380 interpolator via ssp, decimator is driven via the i2s clock. |
| 18:48.21 | pH5 | I hope you can find power/reset gpios on the cpld and then just copy the magician.c setup. |
| 18:48.44 | Marex | pH5, any idea how to trace that CPLD> |
| 18:48.45 | Marex | ? |
| 18:49.20 | Marex | ah, google |
| 18:51.05 | pH5 | http://wiki.xda-developers.com/index.php?pagename=MagicianCPLD has a small recipe |
| 18:52.04 | pH5 | though I think that probably is not current anymore. |
| 18:53.40 | Marex | I just got to that page too, thank you |
| 18:55.23 | Marex | hm, I havent managed to connect wince PDA with haret over usb yet |
| 18:55.31 | Marex | I will just stick with a recipe |
| 19:04.23 | BabelO | pH5: i m not able to get maemo kernel under scratchbox , does it works for you ? |
| 19:07.12 | pH5 | yes. compiling, patching and reflashing the kernel was one of the first things I tried :) |
| 19:07.19 | pH5 | how does it fail for you? |
| 19:07.44 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfb10f.pool.einsundeins.de) |
| 19:10.34 | BabelO | pH5: ok i found why, it seem the maemo-4.x tutorial miss something, i create a new target with and add debian-etch, now it works ;) |
| 19:10.46 | BabelO | pH5: it is really nice to have all these tools |
| 19:12.30 | pH5 | BabelO: I just used the default CHINOOK_ARMEL target |
| 19:14.32 | BabelO | pH5: ok, i ve created my own target like in tutorial |
| 19:23.11 | cr2 | Marex: do you know where is the CPLD located ? somehow i doubt that it's used. |
| 19:23.37 | Marex | it's used for camera |
| 19:23.50 | cr2 | Marex: sable has it for wifi/bt coexistance. |
| 19:23.54 | cr2 | camera ? |
| 19:24.08 | Marex | well I get loads of EGPIOxx stuff |
| 19:24.12 | Marex | when using camera |
| 19:24.17 | cr2 | the CPLD as gpio extender does not make sense. because you already have asic3. |
| 19:24.36 | Marex | I know that myself too :S |
| 19:24.50 | cr2 | well, it's because gnuharet is "tuned" for the universal. |
| 19:24.50 | Marex | but as pH5 said, magician uses it as gpio extender |
| 19:25.03 | cr2 | magician is a much more simple device. |
| 19:25.08 | cr2 | it does not have asic3 |
| 19:25.54 | cr2 | n560 has a CPLD too. as a gpio extender. |
| 19:26.12 | pH5 | cr2: even hx4700 has a cpld (used for wifi power for example) |
| 19:27.02 | cr2 | pH5: beetles does not have wifi. |
| 19:27.06 | Marex | 00572fab: debug 80083c14: 01c300b0(strh) 000001e3 a9200088 |
| 19:27.13 | Marex | what's this in wi trace > |
| 19:27.14 | Marex | ? |
| 19:27.34 | Marex | there are loads of it |
| 19:27.48 | pH5 | I'm just saying it is possible that they have gpios on the cpld used to power the codec. |
| 19:27.57 | pH5 | If they are not on pxa and not on asic3, where else to look? |
| 19:27.57 | cr2 | Marex: addr2mod 01c300b0 |
| 19:28.09 | pH5 | bbl, food |
| 19:28.18 | cr2 | pH5: it's only 1 reg there. |
| 19:29.26 | Marex | cr2, gnuharet doesnt support that :/ |
| 19:29.55 | Marex | I see |
| 19:31.27 | cr2 | Marex: use gnuharet only for 'dump asic3', 'dump asic3gpio' and 'dump pxa27x' |
| 19:31.43 | Marex | ok, I see |
| 19:31.46 | cr2 | all these 3 functions are not implemented in mainline. |
| 19:32.08 | cr2 | but are extremely helpful on pxa2xx/asic3 devices. |
| 19:33.20 | cr2 | i'll look at the sable source, and check the wiki. |
| 19:34.43 | Marex | thanks |
| 19:36.01 | cr2 | can you add the cam gpios to wiki ? |
| 19:37.07 | Marex | sure |
| 19:37.15 | Marex | I have some more too |
| 19:37.21 | Marex | check the LEDS patch |
| 19:37.53 | Marex | I will add them soon |
| 19:40.31 | cr2 | ok. |
| 19:44.45 | Marex | something changed with set traceignore in haret 0.5.1? |
| 19:45.54 | Marex | I get loads of these 000000: 000d7868: mem IRQS(0) aed00000=00000800 (00000800) |
| 19:45.54 | Marex | 000000: 000d7ace: debug 80083c14: 01c300b0(strh) 000001e3 a9200088 |
| 19:46.32 | cr2 | you use haretconsole ? |
| 19:47.33 | Marex | yes, over telnet |
| 19:47.52 | Marex | strh is probably haret detecting itself |
| 19:50.04 | cr2 | you should use the 'console' script, not the telnet directly. if you want that these messages were filtered out. |
| 19:50.18 | cr2 | but then you need to adapt the asic3 tracing script. |
| 19:52.08 | Marex | isnt there a way to filter them ? |
| 19:52.15 | Marex | using traceignore ? |
| 19:53.06 | cr2 | the interrupts should be filtered out by pxa27x.py or something like that. |
| 19:53.21 | cr2 | ibit ** |
| 19:53.47 | Marex | ah, now I know what you mean |
| 19:58.22 | cr2 | regs_asic3.py |
| 19:58.31 | cr2 | and regs_pxa.py |
| 19:59.28 | cr2 | look at regs_misc.py |
| 20:00.37 | cr2 | let's fix them for beetles/sable |
| 20:02.44 | Marex | haretconsole doesnt look way too stable |
| 20:04.04 | cr2 | ? |
| 20:04.22 | Marex | it dies all the time on some exceptions |
| 20:05.21 | cr2 | cvs.handhelds.org seems to be very unhappy. |
| 20:05.32 | Marex | why ? |
| 20:06.13 | cr2 | first it said 'no free space at /tmp/cvs-*' , and now i can't connect at all. |
| 20:06.56 | Marex | :S |
| 20:09.00 | cr2 | can't create temporary directory /tmp/cvs-serv20052 |
| 20:09.00 | cr2 | No space left on device |
| 20:10.33 | Marex | poor you :S |
| 20:12.19 | cr2 | not me. the cvs.hh.org |
| 20:13.50 | BabelO | pH5: he he gspca works, but i need hub, so it is no more a mobile device ;) |
| 20:15.23 | cr2 | BabelO: athena rules :) |
| 20:15.41 | BabelO | cr2: does athena drive engouh usb current ? |
| 20:15.43 | cr2 | Marex: do you have pwm1 on ? |
| 20:15.50 | cr2 | BabelO: yes. |
| 20:15.57 | BabelO | cr2: cool |
| 20:16.05 | cr2 | BabelO: i connect the 1GB stick and boot qtopia off it. |
| 20:16.13 | gers | goxboxlive: i have another news: poky 3.1 has experimental support for universal. http://pokylinux.org/releases/pinky-3.1/ and http://pokylinux.org/getit/ |
| 20:16.54 | BabelO | cr2: yes for usb stick or card reader, it works, but webcan ormy bluetooth dongle , it is not enough |
| 20:17.26 | cr2 | BabelO: then i use a hub. |
| 20:18.08 | cr2 | BabelO: n560 has an external power directly on the connector, but i'm not sure about the max_current kernel setting... |
| 20:18.11 | Marex | cr2, iirc not |
| 20:18.43 | BabelO | cr2: does resume works on n560 now ? |
| 20:19.16 | Marex | cr2, - PWM1 : off |
| 20:19.28 | cr2 | BabelO: LCD does not resume properly. |
| 20:19.39 | cr2 | Marex: ok. the same here. |
| 20:23.59 | Marex | cr2, any idea how to dump that CPLD ? |
| 20:24.40 | cr2 | Marex: you need to know where it is mapped first. |
| 20:25.21 | Marex | I guess it'd be so one of 6 PXA MMIOs |
| 20:25.29 | Marex | that's not too much possibilities |
| 20:25.33 | Marex | 0x0 is flash |
| 20:26.18 | cr2 | Marex: there are 2 suspicious areas |
| 20:26.33 | cr2 | http://wiki.xda-developers.com/index.php?pagename=BeetlesMemoryMap |
| 20:26.33 | Marex | 0x0c000000 and ? |
| 20:26.36 | cr2 | 0x04 and 0x08 |
| 20:27.03 | cr2 | where is asic3sdio ? |
| 20:27.14 | cr2 | the wiki is very incomplete ;) |
| 20:27.50 | cr2 | hehe. no memory map for sable |
| 20:27.52 | Marex | not more than my knowledge |
| 20:28.09 | cr2 | check 0x04 and 0x08 |
| 20:28.20 | cr2 | if you can trace something there. |
| 20:28.30 | cr2 | or just 'pd' first |
| 20:29.35 | cr2 | ok, added sdio at 0x0e |
| 20:30.00 | cr2 | 0x14* and 0x15 are probably the debug card. |
| 20:30.01 | Marex | HaRET(3)# pd 0xa4000000 4 |
| 20:30.01 | Marex | a4000000 | 00000000 | .... |
| 20:30.03 | Marex | nothing |
| 20:30.06 | Marex | Im setting trace |
| 20:30.10 | cr2 | at least 0x100 |
| 20:30.16 | cr2 | 4 is too little. |
| 20:30.26 | cr2 | and the EGPIO may be write-only |
| 20:32.21 | Marex | all zeros |
| 20:33.08 | Marex | HaRET(17)# pd 0x0c000000 0x100 |
| 20:33.09 | Marex | 0c000000 | ffffffff ffffffff ffffffff ffffffff | ................ |
| 20:33.46 | Marex | both 0x08 and 0x00 are zeros |
| 20:34.23 | cr2 | Marex: 0x04000000 and 0x08000000 |
| 20:34.47 | cr2 | 0x0c000000 doesn't show up at all. |
| 20:34.57 | Marex | 000.958(0002011) IRQS GEDR0: GPIO13(77)=1 |
| 20:35.01 | Marex | while playing |
| 20:35.27 | cr2 | 0x08000000 is acx100 wifi on sable. |
| 20:35.31 | Marex | I need to get rid of these |
| 20:35.32 | Marex | 000.988(0007831) IRQS ICIP: USBc(11)=1 |
| 20:35.41 | Marex | cr2, I see |
| 20:35.43 | cr2 | ibit irqs 11 |
| 20:35.52 | cr2 | ibit irqs 77 |
| 20:35.58 | cr2 | ibit irqs 11,77 |
| 20:36.05 | Marex | thank you |
| 20:36.55 | cr2 | i don't see the lattice cpld used on sable. |
| 20:37.05 | cr2 | all acx100 gpios are pxa+asic3 |
| 20:37.36 | Marex | why do they have it there then? |
| 20:37.46 | Marex | btw sable has differnt chip |
| 20:38.15 | cr2 | don't know. |
| 20:41.15 | Marex | yea, I have UDA1380 and sable has AK14.. something |
| 20:41.56 | cr2 | yes. |
| 20:43.01 | cr2 | hmm. |
| 20:43.50 | cr2 | pH5: are you back from eating ? |
| 20:44.31 | pH5 | yup |
| 20:44.48 | cr2 | i get a flood of |
| 20:44.51 | cr2 | 059999: mem aee00004=f4df8902 (08000000) |
| 20:44.51 | cr2 | 059999: mem aee00008=f0f7df3b (000010c4) |
| 20:44.55 | cr2 | is it DMA ? |
| 20:45.08 | Marex | DMA25 is active while playing |
| 20:45.14 | cr2 | aee = 0x400 |
| 20:45.47 | Marex | you mean 0x04 ? |
| 20:46.13 | cr2 | Marex: what do you see at 0x04000000 0x100 ? |
| 20:46.20 | Marex | nothing |
| 20:46.22 | Marex | all zeros |
| 20:46.26 | pH5 | cr2: phys 0x40000004 ? |
| 20:46.26 | cr2 | ok. |
| 20:46.30 | pH5 | that's dma, yes. |
| 20:46.32 | cr2 | pH5: yes. |
| 20:46.37 | cr2 | pH5: can i mask it ? |
| 20:47.08 | cr2 | pH5: it seems i need to get some more python knowledge :) |
| 20:48.25 | pH5 | cr2: magician has mapped 0xaee to 0x40e (gpio), maybe you copied the gpio setup from a machine that has the same? |
| 20:49.40 | cr2 | pH5: aah. sorry. |
| 20:49.58 | pH5 | cr2: aee somehow ended up in the IRQS or GPIOS list when it shouldn't, I guess. |
| 20:50.15 | cr2 | pH5: i've did |
| 20:50.18 | cr2 | Regs_Sable = regs_pxa.Regs_pxa27x.copy() |
| 20:50.18 | cr2 | #Regs_Universal.update(getEGPIOdefs(0x0a000000, 1)) |
| 20:50.18 | cr2 | Regs_Sable.update(regs_asic3.getASIC3Defs(base=0x10000000, sd_base=0x0c000000 |
| 20:50.18 | cr2 | <PROTECTED> |
| 20:50.19 | cr2 | memalias.RegsList['Sable'] = Regs_Sable |
| 20:50.44 | cr2 | hmm, but it seems that they are not used. |
| 20:51.14 | pH5 | cr2: what does p2v(0x40e00004) return? |
| 20:51.22 | cr2 | pH5: these are the LCD gpio lines i guess. |
| 20:51.55 | cr2 | pH5: the problem is that the python stuff does not decode 0x40e0000* as gpios. |
| 20:52.04 | cr2 | for some reason... |
| 20:53.21 | cr2 | 4 and 8 are GPLR0/1 |
| 20:54.28 | pH5 | GPLR1/2 |
| 20:55.01 | pH5 | (http://www.xkcd.com/386/) |
| 20:55.04 | cr2 | yes. |
| 20:55.11 | cr2 | show gpios looks ok. |
| 20:55.14 | cr2 | HaRET(1)# show gpios |
| 20:55.15 | cr2 | <PROTECTED> |
| 20:55.15 | cr2 | <PROTECTED> |
| 20:55.15 | cr2 | <PROTECTED> |
| 20:55.38 | cr2 | hm. but no asic3 gpios. |
| 20:56.55 | pH5 | cr2: I think you have to copy what Kevin2 did for Apache and Hermes in machlist.txt |
| 20:57.11 | pH5 | add the asic3 registers manually. the python stuff is just for decoding |
| 20:58.24 | cr2 | pH5: ok, i'll try. |
| 21:01.02 | cr2 | pH5: it still does not explain why the pxa gpios are not decoded... |
| 21:04.04 | cr2 | ok, checking with the latest haret. |
| 21:04.39 | cr2 | great, now it works. |
| 21:04.59 | cr2 | it seems i had a too old haret version on the miniSD card. |
| 21:05.13 | pH5 | ah, good :) |
| 21:05.35 | cr2 | a lot of 001.998(0012162) IRQS GEDR0: GPIO9(73)=1 |
| 21:05.45 | cr2 | this is the asic3 gpio mux |
| 21:07.04 | cr2 | the machlist.txt compiled per-machine stuff seems to be illogical. |
| 21:07.52 | cr2 | can i hook some per-machine exec after the console connect, depending on the mach id ? |
| 21:11.35 | Marex | 0c000800 | 02003c12 149a3400 00020100 00000b00 | .<...4.......... |
| 21:11.35 | Marex | 0c000810 | 00000900 00000b00 00000900 00800780 | ................ |
| 21:11.35 | Marex | 0c000820 | 837f031d 02000000 000040ee 00002000 | .........@... .. |
| 21:11.35 | Marex | 0c000830 | 00000000 00001000 00000000 00000000 | ................ |
| 21:11.35 | Marex | 0c000840 | 002a003f 00000000 00000000 00000000 | ?.*............. |
| 21:11.39 | Marex | 0c000850 | 00000000 00000000 00000000 00000000 | ................ |
| 21:12.43 | cr2 | Marex: is it (mmu)traceable ? |
| 21:13.13 | cr2 | pH5: hmm. so there is no real asic3 support for machines.txt ? |
| 21:14.15 | Marex | set trace 0x0c000000 |
| 21:14.17 | Marex | wi 5 |
| 21:14.19 | Marex | right > |
| 21:14.20 | Marex | ? |
| 21:14.34 | cr2 | trace size ? |
| 21:15.03 | cr2 | addlist mmutrace p2v(0x0c000000) 0x100000 |
| 21:15.15 | cr2 | watch mmutrace 5 |
| 21:15.26 | pH5 | cr2: looks like it. Kevin2 would know. |
| 21:15.43 | Marex | line 22: Expected <watch list var> |
| 21:15.57 | cr2 | pH5: didn't see him for a very long time. |
| 21:16.44 | cr2 | pH5: the asic3 gpios are ok, but how do i trace 4 asic3 irq regs, indirectly through the level changes ? |
| 21:18.12 | pH5 | cr2: ouh, I don't know how that is possible |
| 21:18.56 | *** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 21:19.32 | cr2 | ok. |
| 21:19.36 | cr2 | testing the gpios. |
| 21:26.40 | cr2 | hm. ibit gpios 58-76 does not work |
| 21:26.52 | cr2 | how did i mask the gpio range.... |
| 21:29.12 | cr2 | ah, help vars |
| 21:34.24 | Marex | <PROTECTED> |
| 21:34.24 | Marex | <PROTECTED> |
| 21:34.24 | Marex | <PROTECTED> |
| 21:34.31 | Marex | <PROTECTED> |
| 21:34.32 | Marex | <PROTECTED> |
| 21:34.32 | Marex | <PROTECTED> |
| 21:34.38 | Marex | BINGO! |
| 21:37.05 | cr2 | ibit gpios 56..76 |
| 21:37.18 | cr2 | to mask out the px27x LCD |
| 21:37.35 | cr2 | asic3 gpio names are still not decoded though. |
| 21:38.48 | Marex | cr2, how can I trace GPIO changes ? |
| 21:39.53 | Marex | sorry |
| 21:41.20 | cr2 | watch gpios NN |
| 21:41.31 | cr2 | but you should mask the pwm and LCD gpios. |
| 21:41.39 | cr2 | ibit gpios 56..76 |
| 21:41.42 | cr2 | for the LCD |
| 21:41.46 | cr2 | and |
| 21:41.50 | cr2 | ibit gpios 16 |
| 21:41.53 | cr2 | for pwm0 |
| 21:42.00 | cr2 | backlight |
| 21:43.16 | Marex | yup, I see |
| 21:45.18 | Marex | ibit gpios 56..76 doesnt mask them :/ |
| 21:45.40 | cr2 | which haret version do you have ? |
| 21:46.02 | cr2 | haha. wifi is called 'phone' on n560 |
| 21:46.12 | Marex | XD |
| 21:46.34 | Marex | ah. I used wgpio |
| 21:46.40 | Marex | instead of watch gpios |
| 21:46.43 | Marex | sorry |
| 21:59.37 | Marex | cr2, is there something like watch asic3 ? |
| 22:00.42 | cr2 | no, it should be included in the gpio list |
| 22:00.59 | cr2 | i'm trying it for sable and n560 now |
| 22:01.10 | cr2 | somehow the names are not decoded. |
| 22:01.16 | cr2 | do you know python ? |
| 22:01.54 | Marex | not at all :S |
| 22:02.19 | *** join/#htc-linux patoche (n=patoche@lns-bzn-24-82-64-164-28.adsl.proxad.net) |
| 22:03.42 | *** join/#htc-linux kiozen (n=kiozen@rgnb-4db1dfe4.pool.einsundeins.de) |
| 22:03.59 | kiozen | cr2: need python? |
| 22:07.06 | cr2 | kiozen: haretconsole does not want to decode the CPLD reg names |
| 22:07.28 | kiozen | any pointers into code |
| 22:07.31 | kiozen | ? |
| 22:07.46 | cr2 | kiozen: hmm. looking |
| 22:08.14 | kiozen | cr2: where can I get your version |
| 22:08.56 | cr2 | i jave it only locally. i'll pastebin in 2min. |
| 22:09.41 | kiozen | ok |
| 22:10.21 | cr2 | kiozen: http://handhelds.org/cgi-bin/cvsweb.cgi/haret/haretconsole/regs_misc.py?rev=1.5&content-type=text/x-cvsweb-markup |
| 22:10.27 | cr2 | this is the original version |
| 22:10.53 | cr2 | look at Kaiser |
| 22:11.02 | cr2 | i've replaced it with Loox5xx |
| 22:11.11 | cr2 | but it does not decode the names. |
| 22:11.49 | cr2 | kiozen: btw, i've compiled qlandkarte 0.7.2 today, before i've used 0.6.2 |
| 22:12.38 | cr2 | kiozen: the StringTbl10 decoding is broken in 0.7.2, it was working in 0.6.2 |
| 22:13.13 | cr2 | i seem to hit a lot of bugs once i touch any software :) |
| 22:13.24 | kiozen | StringTbl10: ok have to check, there where some "fixes" for OSM :/ |
| 22:13.53 | kiozen | Regs_Kaiser = regs_msm.Regs_msm7500.copy() |
| 22:14.07 | kiozen | the intelligence is done in regs_msm et al |
| 22:14.58 | cr2 | kiozen: is have regs_pxa |
| 22:15.18 | cr2 | the problem is with EGPIO decoding. |
| 22:15.29 | cr2 | def getEGPIOdefs(base, count, start=0): |
| 22:15.38 | cr2 | in regs_misc |
| 22:15.58 | cr2 | it was working before, so i'm surprised. |
| 22:16.43 | kiozen | any exception or just nothing? |
| 22:16.55 | cr2 | just nothing |
| 22:17.22 | cr2 | i see the raw addresses not being parsed. |
| 22:17.37 | kiozen | best to place some print commands in getEGPIOdefs |
| 22:18.20 | cr2 | kiozen: forget it. was some weird glitch. |
| 22:18.27 | cr2 | works again now ;) |
| 22:18.29 | kiozen | ok :) |
| 22:18.47 | kiozen | just keep hitting it and it will owrk |
| 22:19.00 | cr2 | yeah. |
| 22:19.02 | kiozen | s/owrk/work/ |
| 22:19.44 | cr2 | btw, GT is excellent :) some minor enhancements here and there are needed, but it's already very useful. |
| 22:20.00 | cr2 | i'm missing 1 thing though. |
| 22:20.05 | kiozen | :) |
| 22:20.47 | cr2 | is it possible to drag-n-drop the wpts from one GT window to another as gcps ? |
| 22:21.12 | kiozen | 2 instances of GT? |
| 22:21.30 | cr2 | that's the only feature i'm missing compared to qgis. |
| 22:21.38 | cr2 | well. maybe. |
| 22:22.01 | kiozen | not quite sure what you want |
| 22:22.16 | cr2 | i open one GT and load the wpts. |
| 22:22.28 | cr2 | they show up as nice 'balloons' |
| 22:22.44 | cr2 | i open the second window, and begin calibrating. |
| 22:23.12 | kiozen | 2nd window? |
| 22:23.22 | cr2 | so i don't want to enter the gcp coordinates manually, but just pull then from the first window. |
| 22:23.50 | kiozen | wpts == wpts or wpts = ref.points? |
| 22:23.53 | cr2 | yes, 2 GT programs running simultaneously. |
| 22:24.22 | cr2 | wpt coords "dropped" into gcp Coord. field. |
| 22:24.50 | kiozen | ok so you have wpts in one GT and want to drop them as gcp in another GT |
| 22:24.56 | cr2 | ye.s |
| 22:25.18 | kiozen | ok, never have done drag-n-drop. time to start |
| 22:25.20 | cr2 | you can't have them both in one GT instance. |
| 22:25.33 | cr2 | :) |
| 22:25.55 | cr2 | maybe just 2 tabs. |
| 22:25.57 | kiozen | well you could place wpts on the map switch to edit map and drag-n-drop them from the wpt toolview |
| 22:26.00 | cr2 | but DND is more fun. |
| 22:26.10 | cr2 | yes. |
| 22:26.22 | kiozen | let's see the nect days |
| 22:26.29 | cr2 | ok. |
| 22:26.45 | cr2 | imho that's a killer feature. |
| 22:26.51 | kiozen | what's wrong with CGarminStrTbl10 |
| 22:26.52 | cr2 | ozi can't do it |
| 22:27.09 | kiozen | who cares about ozi ;) |
| 22:27.15 | cr2 | and i've hacked it into qgis some time ago, because a lot of people wanted it. |
| 22:27.39 | cr2 | ozi is used by many people. it's like wince for raster gps maps :) |
| 22:27.53 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
| 22:27.59 | kiozen | yes it's quite confi if you reference raster via osm/Garmin maps |
| 22:28.08 | BabelO | cr2: http://pda-artemis.narod.ru/files/Artemis-Inside.pdf :) |
| 22:28.47 | cr2 | BabelO: since i'm at haret CPLD hacking now, i'll add the artemis CPLD decoder now. |
| 22:29.33 | *** join/#htc-linux joschkopf79 (n=joschi@e179051116.adsl.alicedsl.de) |
| 22:30.56 | cr2 | BabelO: i can only see the usual CPLD on these pics. |
| 22:31.33 | BabelO | cr2: do you understand what is written ? http://pda-artemis.narod.ru/ |
| 22:32.10 | BabelO | i use google translate but ... |
| 22:32.19 | cr2 | kiozen: Tbl10 -> if i take the TW basemap, i only see the binary mess in the labels instead of the chinese glyphs in 0.7.2 in 0.6.2 it was correct. |
| 22:32.47 | kiozen | ok, I'll check teh next days |
| 22:32.53 | cr2 | BabelO: lol. they obviously used it vice versa :) |
| 22:33.28 | goxboxlive | <BabelO> goxboxlive: magic guy at #git give me the correct command git rm `git ls-files -d` "Sorry, need to have it in the log" |
| 22:33.45 | BabelO | goxboxlive: ok :) |
| 22:33.52 | goxboxlive | :-) |
| 22:36.40 | cr2 | BabelO: it's a lot of text. i can only guess that the HTC artemis firmware is buggy as hell |
| 22:37.40 | BabelO | yes it is, i can confirm, i ll try since many hours now to get bluetooth pan working :(, wm6 is speedy but ... |
| 22:37.43 | *** join/#htc-linux ltxda0 (n=ltxda@unaffiliated/ltxda) |
| 22:40.15 | cr2 | BabelO: did you put 0x0, 0x1, 0x2 into ArtemisCPLD wiki ? |
| 22:40.46 | cr2 | yes, in the Version 4 |
| 22:41.12 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
| 22:42.33 | cr2 | it seems all these regs are 8bit, so we may use 0xff00 mask |
| 22:43.04 | Marex | cr2, is it possible to switch asic3 regs from running linux ? |
| 22:43.12 | Marex | over sysfs or so ... ? |
| 22:43.25 | BabelO | Marex: yes, with x tools :) |
| 22:43.43 | BabelO | Marex: or devmem |
| 22:46.02 | cr2 | Marex: i think only /dev/mem (devmem2 or something like that). or x tools :) |
| 22:46.13 | cr2 | hehe. |
| 22:46.19 | Marex | hm ok, thanks |
| 22:46.23 | cr2 | BabelO: you was faster :) |
| 22:46.40 | Marex | btw I was unable to watch asic gpios with haret |
| 22:46.52 | Marex | the command watch gpios x just doesnt show asic gpios |
| 22:47.41 | cr2 | Marex: you need |
| 22:47.43 | cr2 | <PROTECTED> |
| 22:47.43 | cr2 | <PROTECTED> |
| 22:47.44 | cr2 | <PROTECTED> |
| 22:47.44 | cr2 | <PROTECTED> |
| 22:48.16 | cr2 | i've added that for sable, but the python doesn't parse it as i expect. |
| 22:48.18 | Marex | I see ... thanks |
| 22:48.26 | cr2 | but you'll see them in raw dump. |
| 22:52.30 | *** join/#htc-linux dcordes (n=snoopdog@unaffiliated/dcordes) |
| 23:00.03 | Marex | cr2, yea, I do see them |
| 23:01.06 | cr2 | Marex: good. i don't have write access to this CVS. |
| 23:01.18 | cr2 | Marex: so i can't update it. |
| 23:01.40 | Marex | can you mail me the updated version ? |
| 23:01.45 | cr2 | maybe just update the haret in xanadux CVS. |
| 23:02.28 | Marex | btw I checked and I dont see any changes other than GPIO 23 24 25 27 29 31 113 (thats normal) and GPIOA6 (loudspeaker select) |
| 23:03.43 | cr2 | ok. |
| 23:03.59 | cr2 | sable has a bit different memory map |
| 23:04.08 | cr2 | can you make 'dump mmu' ? |
| 23:07.04 | Marex | I did, how do I save it ? |
| 23:07.56 | Marex | I did "> somefile", but I cant find somefile :D |
| 23:08.19 | Marex | hang on |
| 23:10.31 | Marex | cr2, see PM |
| 23:13.07 | *** join/#htc-linux joschkopf79 (n=joschi@e179057225.adsl.alicedsl.de) |
| 23:13.36 | cr2 | Marex: OK. |
| 23:14.28 | Marex | cr2, well I have added both ASIC3 and GPIO regs on that device and yet I still cant see power and reset for the soundcard |
| 23:14.33 | Marex | it has to be on that CPLD |
| 23:15.11 | cr2 | Marex: can't download. |
| 23:15.45 | Marex | click on "dump.dump" |
| 23:15.46 | cr2 | well, then you need to check ida. |
| 23:16.32 | cr2 | ok :) |
| 23:16.44 | cr2 | i really doubt that the cpld is used. |
| 23:16.52 | cr2 | i'll look myself :) |
| 23:17.47 | cr2 | Marex: wtf is 0x02 ? |
| 23:18.31 | cr2 | hm. and 0x01. |
| 23:18.32 | Marex | where? how should I know that, you are the expert |
| 23:18.39 | cr2 | how big is your ROM ? |
| 23:18.51 | Marex | 64megs |
| 23:18.56 | Marex | same as sables |
| 23:19.04 | cr2 | your mmap has a lof of stuff mapped at these addresses. |
| 23:19.07 | cr2 | pxa272 ? |
| 23:19.12 | Marex | yes |
| 23:19.16 | Marex | I see |
| 23:19.27 | cr2 | sorry, it's been over a year i've hacked on beetles/sable ;) |
| 23:19.31 | cr2 | ok. |
| 23:19.56 | Marex | you really arent in position to apologize ;) |
| 23:20.29 | Marex | 02000040 | 43454345 826cebf0 9000013c 00100000 | ECEC..l.<....... |
| 23:20.32 | Marex | that's ROM |
| 23:20.41 | cr2 | yes, it's ROM |
| 23:20.48 | cr2 | from 0x00 to 0x04 |
| 23:20.51 | cr2 | 64MB. |
| 23:21.06 | Marex | right |
| 23:21.20 | cr2 | nice. then you may flash linux . eventually. |
| 23:21.32 | cr2 | but sable had mdoc g4 too ? |
| 23:21.55 | Marex | beetles doesnt have doc g4 |
| 23:21.58 | dcordes | hi all |
| 23:22.08 | dcordes | cr2, Marex which machine are you h4xing on? |
| 23:22.18 | Marex | htcbeetles |
| 23:22.26 | cr2 | Marex: sable spec says 128MB ROM |
| 23:22.39 | cr2 | dcordes: and n560 at the same time ;) |
| 23:23.14 | cr2 | Marex: pxa270 says the spec. so it's only mdoc g4 on sable |
| 23:23.27 | cr2 | http://www.pdadb.net/index.php?m=specs&id=151&view=1 |
| 23:24.00 | Marex | the device says 272 :S |
| 23:24.44 | Marex | it's stupid though, 272 has onboard flash |
| 23:24.59 | Marex | though this device has some integrated chips around the CPU |
| 23:25.16 | Kevin2 | cr2: I'm still around. |
| 23:25.34 | Kevin2 | Marex: If you see an exception from haretconsole - please email it to haret@handhelds.org. |
| 23:26.01 | Marex | Kevin2, ok, it might have been my fault |
| 23:26.05 | cr2 | Marex: beetles is pxa272, you must be a happy guy. |
| 23:27.09 | Marex | those ICs are rams ... |
| 23:27.14 | Marex | yea, that makes it clear |
| 23:27.16 | cr2 | Kevin2: i've patched machtypes.txt and regs_misc.py to support more machines. but i can't get the asic3 to decode the bits. and i don't understand to to deal with asic3 irqs |
| 23:27.40 | cr2 | Kevin2: the haret CVS says 'disk full', so i can't make cvs diff |
| 23:28.40 | Kevin2 | Hrmm. cvs diff works for me. |
| 23:28.53 | cr2 | Marex: beetles does not have g4 |
| 23:29.06 | cr2 | Kevin2: trying again. |
| 23:29.14 | Marex | cr2, I know it doesnt |
| 23:29.18 | Kevin2 | You can use the git repo at : http://git.linuxtogo.org/ -- it should be identical. |
| 23:29.31 | cr2 | Kevin2: ok, it works now. |
| 23:29.43 | cr2 | i'll send the patch to haret@handhelds.org |
| 23:30.19 | Kevin2 | So, what are you trying to do with asic3 and asic3 irqs? |
| 23:30.42 | cr2 | i've added asic3 gpio read addresses to GPIOS |
| 23:31.09 | cr2 | and added the asic3 'handler' to regs_misc.py |
| 23:31.24 | cr2 | so i see the raw regs, but not the decoded bits. |
| 23:31.57 | Kevin2 | Hrmm. What are the lines you see? |
| 23:33.01 | cr2 | need to reconnect. |
| 23:33.43 | Kevin2 | Okay. If you pastebin the patch (in addition to emailing it), I'll try to take a look through it. |
| 23:34.04 | cr2 | 000.000 GPIOS a9200018=c087 |
| 23:34.04 | cr2 | 000.000 GPIOS a9200098=9662 |
| 23:34.11 | cr2 | ok, sending. |
| 23:35.22 | cr2 | Kevin2: http://pastebin.ca/944241 |
| 23:36.26 | cr2 | Kevin2: if it'll be sorted out, i'll add some other asic3-based machines. |
| 23:40.15 | Kevin2 | cr2: What machine doesn't show decoded asic3? |
| 23:40.21 | cr2 | Sable |
| 23:40.53 | cr2 | Marex can test it on beetles too. |
| 23:41.09 | cr2 | the CPLD decoder works on Loox N560. |
| 23:41.34 | Marex | cr2, can I try that CPLD decoder ? |
| 23:42.05 | cr2 | Marex: i'm talking about the asic3 decoder here. |
| 23:42.32 | Kevin2 | cr2: I think I'm going to need a cut-and-paste of the raw line you end up seeing. |
| 23:42.34 | Marex | I was talking about "<cr2> the CPLD decoder works on Loox N560." |
| 23:42.51 | cr2 | you need to prove that you see this CPLD first. and to find out it's base address and number of registers |
| 23:43.02 | cr2 | Kevin2: ok. |
| 23:43.17 | dcordes | how can I fetch this directory recrusively? http://www.algepop.net/users/alge/angstrom-ba/kernel-patches/ |
| 23:44.52 | cr2 | Kevin2: http://pastebin.ca/944256 |
| 23:46.06 | Marex | CPLD on kaiser is at 0x98000000 ... wtf :E |
| 23:46.47 | cr2 | Marex: kaiser != beetles |
| 23:47.16 | Kevin2 | cr2: Are you sure you should have "shift=1" ? |
| 23:47.17 | Marex | cr2, I know, but I thought there were just those six MMAPable places |
| 23:48.17 | cr2 | Kevin2: yes. 0x18 is shift 1 |
| 23:48.48 | cr2 | Kevin2: http://wiki.xda-developers.com/index.php?pagename=ASIC3 |
| 23:49.29 | cr2 | Marex: msm7xxx is a completely different beast |
| 23:49.42 | Marex | sorry, I see |
| 23:49.47 | Marex | Im tired |
| 23:50.11 | cr2 | Marex: we support not only the pxa. omap850 has the cpld in a completely different place. |
| 23:50.29 | Marex | cr2, I hacked on palmz71/omap310 :) |
| 23:50.39 | Marex | I know the mapping is different |
| 23:50.46 | cr2 | ok. |
| 23:51.08 | Kevin2 | cr2: Can you try it with "shift=0"? |
| 23:51.15 | cr2 | Kevin2: ok. |
| 23:51.59 | Marex | http://www.handhelds.org/moin/moin.cgi/EtenG500Hardware has the same CPLD |
| 23:52.13 | Marex | but really "CPLD : Lattice ispMACH 4064ZC (for what purpose ?)" |
| 23:52.54 | cr2 | Kevin2: it's just plain wrong. |
| 23:52.57 | cr2 | 000.000 GPIOS a9200198=5208 |
| 23:52.57 | cr2 | 001.035 GPIOS GPIO_B_SleepMask(576)=87c3 |
| 23:52.57 | cr2 | 001.097 GPIOS GPIO_B_SleepMask(576)=87c2 |
| 23:53.49 | cr2 | Marex: CPLD is a custom chip. it may be non user-programmable at all. |
| 23:54.02 | Kevin2 | Oh - what should the register name of a9200018 be? |
| 23:54.05 | cr2 | do you see any CPLD driver for G500 ? |
| 23:54.24 | cr2 | Kevin2: GPIOA_Status +0x0030 +0x018 Read here |
| 23:54.42 | cr2 | it's all in http://wiki.xda-developers.com/index.php?pagename=ASIC3 |
| 23:55.11 | cr2 | ASIC3_GPIO_OFFSET = { |
| 23:55.17 | cr2 | <PROTECTED> |
| 23:55.30 | cr2 | for shift=1 it should be 0x18 |
| 23:55.57 | Marex | cr2, I know that CPLD is preprogrammed too :p |
| 23:56.28 | Marex | cr2, nah, they probably dont know what is that CPLD for there |
| 23:56.39 | Marex | I guess it's used for radio in G500 |
| 23:56.41 | cr2 | Marex: actually you can reprogram it on some htc phones. at runtime :) but you probably don't want to do it. |
| 23:57.10 | Marex | cr2, if you can, then it's not CPLD , iirc CPLD was programmable only once |
| 23:57.41 | cr2 | Marex: ok, it's called CPLD in the HTC binaries :) |
| 23:58.43 | cr2 | http://en.wikipedia.org/wiki/Complex_Programmable_Logic_Device |
| 23:59.55 | Marex | I was wrong then |