| 00:06.20 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 00:16.39 | *** join/#htc-linux thedicemaster2 (n=thedicem@24.132.89.51) |
| 00:20.51 | Orbital_sFear | tmzt: I got it, you can echo to /dev/kmsg and get messages to dmesg |
| 00:21.26 | Orbital_sFear | thanks for your help |
| 00:23.54 | *** part/#htc-linux Orbital_sFear (n=ldupin@bbb72-0-189-133.bendbroadband.com) |
| 00:29.38 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 00:51.22 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 01:00.49 | *** join/#htc-linux x29a_ (n=x29a@unaffiliated/x29a) |
| 01:14.36 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 01:15.14 | *** join/#htc-linux root2 (n=root@rgnb-5d87d95e.pool.mediaWays.net) |
| 01:18.30 | *** part/#htc-linux Renfield (n=Renfield@c-24-218-184-34.hsd1.nh.comcast.net) |
| 01:36.32 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 01:58.10 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 02:04.47 | luke-jr | http://wiki.xda-developers.com/index.php?pagename=RaphaelLinux suggests even Raph is nowhere near working? |
| 02:18.24 | *** join/#htc-linux swc|666 (n=infidel2@unaffiliated/swc666/x-4934821) |
| 02:23.54 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 02:45.44 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 03:00.59 | *** join/#htc-linux dcordes-kais (n=dcordes-@ip-77-25-120-76.web.vodafone.de) |
| 03:01.45 | dcordes-kais | luke-jr young jedi |
| 03:02.35 | dcordes-kais | dcordes-kais this information is probably misleading. |
| 03:03.35 | dcordes-kais | s/dcordes-kais/luke-jr/ |
| 03:03.46 | luke-jr | dcordes-kais: ? |
| 03:09.08 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 03:11.09 | *** join/#htc-linux ppman (n=mkern@pool-71-250-6-82.nwrknj.east.verizon.net) |
| 03:11.47 | ppman | hey, I'm interested in building some Angstrom-based stuff for kaiser, which machine do I use? |
| 03:13.52 | *** join/#htc-linux root2 (n=root@rgnb-5d87ce2a.pool.mediaWays.net) |
| 03:19.27 | *** join/#htc-linux cr2 (n=cr2@109.84.130.177) |
| 03:23.43 | *** join/#htc-linux Bry8Star (n=Bry8Star@71.137.228.95) |
| 03:32.54 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 03:41.43 | *** join/#htc-linux junalmeida (n=junalmei@201.19.126.132) |
| 03:46.42 | *** join/#htc-linux Bry8Star (n=Bry8Star@adsl-71-137-228-95.dsl.irvnca.pacbell.net) |
| 03:54.23 | *** join/#htc-linux dzo (n=dzo@mail.marginz.co.nz) |
| 04:16.53 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 04:27.44 | *** join/#htc-linux droid001 (n=g1@p4FDCF391.dip.t-dialin.net) |
| 04:39.49 | *** join/#htc-linux Bry8Star (n=Bry8Star@70.239.151.164) |
| 04:43.54 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 04:58.33 | *** join/#htc-linux townkat (n=town@86.127.6.111) |
| 05:01.39 | *** join/#htc-linux mrmoku|a` (n=mrmoku@ppp-88-217-24-48.dynamic.mnet-online.de) |
| 05:05.32 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 05:29.50 | *** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey) |
| 05:33.53 | *** join/#htc-linux dzo_ (n=dzo@mail.marginz.co.nz) |
| 07:36.03 | *** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey) |
| 08:07.47 | cr2 | morning |
| 08:16.37 | *** join/#htc-linux pleemans (n=toi@d54C2A96D.access.telenet.be) |
| 08:18.02 | cr2 | 1462 if (configMgr_start(drv->adapter.CoreHalCtx) != OK) |
| 08:18.03 | cr2 | 1463 { |
| 08:18.05 | cr2 | 1464 print_err("TIWLAN: Failed to start config manager\n"); |
| 08:22.37 | phh | cr2: yeah, just have to understand what this actually means. |
| 08:25.20 | phh | maybe it means it just can't talk to the card... or just to the userland... or anything else |
| 08:25.55 | cr2 | where is this function located ? |
| 08:26.50 | phh | you have to get the whole ti driver |
| 08:26.57 | cr2 | and does the compiler do wrong struct packing ? |
| 08:27.01 | cr2 | no |
| 08:27.16 | phh | mmm haven't though about compiler |
| 08:27.31 | cr2 | 96 CHECK_STRUCT_SIZE(beacon_probeRsp_t, SIZEOF_BEACON_FRM); |
| 08:27.44 | cr2 | 56 #define SIZEOF_BEACON_FRM 0x57 |
| 08:27.45 | cr2 | 57 #else |
| 08:27.47 | cr2 | 58 #define SIZEOF_BEACON_FRM 0x53 |
| 08:27.48 | cr2 | 59 #endif |
| 08:28.05 | cr2 | 71 print_info(KERN_INFO"\n.... ERROR in size of %s struct 0x%08x should be 0x%08x.....\n", \ |
| 08:28.07 | cr2 | 72 #_type, (int)sizeof(_type),(int)_size); \ |
| 08:32.53 | phh | do you know which compiler is used to build g1 ? |
| 08:33.32 | cr2 | what were your numbers ? let's check them directly |
| 08:34.07 | phh | http://pastebin.com/m24a88226 |
| 08:37.55 | cr2 | 55 #ifdef EXC_MODULE_INCLUDED |
| 08:38.05 | cr2 | what does it mean ? |
| 08:39.56 | cr2 | 49 typedef struct |
| 08:39.57 | cr2 | 50 { |
| 08:39.59 | cr2 | 51 int err_code; |
| 08:40.00 | cr2 | 52 char *desc; |
| 08:40.02 | cr2 | 53 } |
| 08:40.03 | cr2 | 54 configMgr_error_desc_t; |
| 08:40.15 | *** join/#htc-linux kiozen (n=kiozen@rgnb-5d87dc81.pool.mediaWays.net) |
| 08:40.42 | cr2 | which error number did you get ? |
| 08:45.13 | *** join/#htc-linux MethoS- (n=clemens@134.102.106.250) |
| 08:46.20 | phh | do you know where can i read it ? |
| 08:46.36 | cr2 | != OK |
| 08:46.53 | cr2 | is it printed then ? |
| 08:47.46 | phh | it doesn't seems to be printed anywhere |
| 08:50.49 | *** join/#htc-linux timebomb (n=tb@85.177.142.5) |
| 08:54.34 | phh | argh |
| 08:54.48 | phh | how do we print messages to kernel ? |
| 08:55.17 | phh | oops, i changed the wrong file |
| 08:56.52 | phh | the bad thing being that rmmod wlan crashes the kernel so i have to reboot at every crash ... |
| 08:58.04 | cr2 | <PROTECTED> |
| 08:58.07 | cr2 | hmm |
| 08:58.57 | cr2 | <PROTECTED> |
| 08:58.59 | *** join/#htc-linux sd00 (n=smac@82-39-194-231.cable.ubr02.jarr.blueyonder.co.uk) |
| 08:59.02 | phh | in wlan_loader there is something about it i'm wondering: |
| 08:59.04 | phh | <PROTECTED> |
| 08:59.04 | phh | <PROTECTED> |
| 08:59.04 | phh | <PROTECTED> |
| 08:59.04 | phh | <PROTECTED> |
| 08:59.05 | phh | <PROTECTED> |
| 08:59.07 | phh | <PROTECTED> |
| 08:59.17 | cr2 | what is the value of hSmeSm ? |
| 08:59.18 | phh | the comment seems to say that a cli has to be launched somwhere |
| 08:59.23 | phh | and maybe not just wlan_loader |
| 08:59.55 | cr2 | don't know what they mean 'cli' |
| 09:00.11 | cr2 | irq disabled ? |
| 09:00.24 | phh | cli useually means command line interface |
| 09:00.58 | phh | cr2: error code is 3 |
| 09:01.30 | phh | argh |
| 09:01.33 | phh | tiwlan.ini is wrong |
| 09:01.37 | cr2 | <PROTECTED> |
| 09:01.51 | phh | i'll try with G1's one then |
| 09:01.59 | phh | (the one i use is from the android git) |
| 09:02.11 | cr2 | where do you get the "wrong" one ? :) |
| 09:03.59 | phh | groumpf i have to restart at every single change |
| 09:04.19 | phh | i hate that driver. |
| 09:05.03 | phh | (yes, even just for a config change.) |
| 09:07.27 | phh | wth ?!? |
| 09:07.33 | phh | messages changed, error hasn't |
| 09:07.38 | phh | [ 44.462921] TIWLAN: 780.932921: Firmware running. |
| 09:07.38 | phh | [ 46.947479] TIWLAN: 783.417479: HAL_CTRL ,ERROR:Timeout waiting for the hardware to complete initialization |
| 09:07.38 | phh | [ 46.974273] TIWLAN: 783.444274: HAL_CTRL ,ERROR:Firmware download failed! |
| 09:07.38 | phh | [ 47.002227] TIWLAN: 783.472228: HAL_CTRL ,ERROR:whal_hwCtrl_Config: failed to initialize |
| 09:07.38 | phh | [ 47.042755] configMgr_start3 |
| 09:08.06 | phh | here's maybe the clock problem. |
| 09:08.32 | phh | [ 38.926788] clock-wince: pc_clk_set_rate: id=19 rate=11000000 |
| 09:08.32 | phh | [ 38.940643] clock-wince: set mdns: 19, 11000000; bitidx=7, offset=a4, ns=0 |
| 09:08.32 | phh | [ 38.962280] clock-wince: set_mdns_host_clock: 19, freq=7372800 |
| 09:08.36 | phh | but i guess you already know that. |
| 09:09.06 | cr2 | it takes the first frequency below the rate |
| 09:09.28 | cr2 | you can add the 12.28 line, and name it 11 |
| 09:09.34 | cr2 | then it will be picked |
| 09:10.21 | phh | ok |
| 09:10.38 | cr2 | the list should be sorted by frequency |
| 09:16.24 | phh | the line should be: |
| 09:16.27 | phh | <PROTECTED> |
| 09:16.29 | phh | right ? |
| 09:16.41 | phh | because i still get: |
| 09:16.44 | phh | [ 51.596282] clock-wince: set mdns: 19, 11000000; bitidx=7, offset=a4, ns=0 |
| 09:16.44 | phh | [ 51.619659] clock-wince: set_mdns_host_clock: 19, freq=7372800 |
| 09:18.45 | cr2 | have you softed the list ? |
| 09:18.51 | cr2 | s/sof/sor/ |
| 09:19.08 | cr2 | the 7372800 is before the SD block |
| 09:19.49 | phh | between MSM_CLOCK_REG(3686400*16, 6, 0x19, 0x0c, 0, 2, 4, 1), /* 3686400 */ and MSM_CLOCK_REG(19200000, 1, 0x0a, 0x05, 3, 3, 1, 1), /* 19.2MHz */ |
| 09:20.20 | cr2 | <PROTECTED> |
| 09:20.25 | cr2 | ... |
| 09:20.35 | cr2 | <PROTECTED> |
| 09:20.44 | cr2 | this is in head |
| 09:21.13 | cr2 | put it before the 12000000 |
| 09:22.26 | cr2 | <PROTECTED> |
| 09:22.27 | cr2 | <PROTECTED> |
| 09:22.36 | cr2 | i guess you can put 12 here |
| 09:22.44 | phh | ok |
| 09:22.47 | cr2 | 11 is only confusing |
| 09:22.58 | cr2 | or we can use 12.28 as in wince |
| 09:23.09 | phh | bt hu, where do you have this max_dtr ? |
| 09:23.28 | cr2 | arch/arm/mach-msm/board-htcraphael-mmc.c |
| 09:23.34 | phh | oh ok |
| 09:24.34 | cr2 | btw, does your stty support 4000000 speed ? |
| 09:24.59 | cr2 | then we may try to switch BT bps without .bts |
| 09:25.22 | cr2 | just came into my mind |
| 09:25.30 | phh | cr2: you already made me try |
| 09:25.44 | phh | hciattach -s 115200 ttyHS1 texas 4000000 flow |
| 09:25.55 | phh | (it's the command from G1 iirc) |
| 09:25.55 | cr2 | stty -F /dev/ttyHS1 speed 4000000 |
| 09:26.08 | cr2 | not hciattac |
| 09:26.37 | phh | you want to communicate with the BT chip by hand ? |
| 09:27.04 | cr2 | the easiest option. but wifi first :) |
| 09:27.18 | phh | yep |
| 09:28.47 | phh | [ 78.298706] clock-wince: set mdns: 19, 11000000; bitidx=7, offset=a4, ns=0 |
| 09:28.47 | phh | [ 78.322113] clock-wince: set_mdns_host_clock: 19, freq=7372800 |
| 09:28.50 | phh | i'm doing it wrong ... |
| 09:29.16 | cr2 | change .max_dtr = 11000000, to 12000000 |
| 09:29.24 | phh | i changed it to 13M .. |
| 09:29.34 | cr2 | hmm |
| 09:29.48 | phh | diamond does use this file too ? |
| 09:30.27 | cr2 | yes |
| 09:30.44 | cr2 | and we will merge the panel code, once it works |
| 09:30.50 | cr2 | for both |
| 09:31.16 | cr2 | obj-$(CONFIG_MACH_HTCDIAMOND) += |
| 09:31.29 | cr2 | board-htcraphael-mmc.o |
| 09:33.13 | phh | http://pastebin.com/m3928ee84 |
| 09:33.16 | phh | here is my current diff |
| 09:36.59 | cr2 | and it does not work for the clock ? |
| 09:38.01 | cr2 | <PROTECTED> |
| 09:38.03 | cr2 | <PROTECTED> |
| 09:38.12 | cr2 | <PROTECTED> |
| 09:38.13 | cr2 | <PROTECTED> |
| 09:38.23 | phh | it still tries to pick up 11M and get 7M ... |
| 09:38.33 | cr2 | hmm |
| 09:38.45 | cr2 | is 11 hardcoded somewhere else ? |
| 09:39.05 | phh | not in TI driver |
| 09:39.59 | phh | in other board mmc |
| 09:40.10 | phh | but they aren't supposed to be builds |
| 09:40.13 | phh | -s |
| 09:41.17 | phh | oops |
| 09:41.18 | phh | my bad |
| 09:41.35 | phh | i pushed the kernel in the wrong place |
| 09:41.42 | cr2 | hehe |
| 09:42.08 | *** join/#htc-linux kvaster (n=kvaster@live.bn.by) |
| 09:45.31 | phh | [ 58.628540] clock-wince: set mdns: 19, 13000000; bitidx=7, offset=a4, ns=0 |
| 09:45.31 | phh | [ 58.643402] clock-wince: set_mdns_host_clock: 19, freq=12288000 |
| 09:45.43 | phh | [ 63.586456] TIWLAN: 3072.919402: Working on a 1251 PG 1.2 board. |
| 09:45.43 | phh | [ 63.615173] TIWLAN: 3072.948119: Starting to process NVS... |
| 09:45.43 | phh | [ 63.660766] TIWLAN: 3072.993682: NVS version 6.3.7 found. |
| 09:45.43 | phh | [ 63.671661] TIWLAN: 3073.004607: Radio type is 0x1C. |
| 09:45.43 | phh | [ 63.694793] TIWLAN: 3073.027739: Finished processing NVS. |
| 09:45.45 | phh | [ 63.718261] configMgr_start3 |
| 09:45.47 | phh | [ 63.727905] TIWLAN: TIWLAN: Failed to start config manager |
| 09:45.52 | phh | grrr |
| 09:45.56 | phh | no "firmware running" this time |
| 09:47.24 | cr2 | Working on a 1251 PG 1.2 board. |
| 09:47.34 | cr2 | that means the sdio works |
| 09:47.49 | cr2 | why is the firmware not loaded ? |
| 09:48.30 | phh | maybe some timeout mmm |
| 09:48.50 | phh | let's try building kaiser's driver |
| 09:51.47 | townkat | android booting hangs at ieee80211_crypt:registred algorithm 'NULL', is it a known problem ? |
| 10:03.11 | phh | kaiser's driver seems to do the same :( |
| 10:03.18 | phh | err no |
| 10:03.44 | phh | i just get NOK /o\ |
| 10:04.54 | *** join/#htc-linux Squarc (n=Squarc@82-217-32-29.cable.quicknet.nl) |
| 10:34.36 | *** join/#htc-linux GNUtoo (n=gnutoo@host134-152-dynamic.51-79-r.retail.telecomitalia.it) |
| 10:37.14 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 10:38.34 | phh | cr2: anyway i think we'd better try with wl12xx, since it is most likely to be goes mainline |
| 10:39.02 | cr2 | ok |
| 10:39.15 | phh | but it seems to need some platform code, and i don't know anything of this |
| 10:44.15 | *** join/#htc-linux GlemSom (n=glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk) |
| 10:46.49 | *** join/#htc-linux Zinbolic (n=zinbolic@84.238.80.215) |
| 10:51.00 | phh | yes, TI driver do some thing to wake up wifi chip, and then wl12xx is able to detect... but conflicts and friends said kernel should crash. |
| 10:52.13 | phh | haha just an sdio_enable_func |
| 10:52.39 | phh | mm maybe not |
| 11:12.39 | DJWillis | phh: the platform code for WL12xx is trivial and WL12xx is already well on the way to mainline (if it is not already there). It has not taken me very long to get it running on several devices BUT there is a snag, you have to read the calibration file (NVS) from the EEPROM and that is not really working, I am trying to sort that. |
| 11:14.18 | phh | yeaha \o/ |
| 11:14.18 | phh | driver loaded |
| 11:14.23 | phh | hum nothing in iwlist scan. |
| 11:14.36 | phh | DJWillis: calibration is no big deal for DIAM/RAPH afaik |
| 11:14.45 | phh | it's always at the same place |
| 11:14.59 | DJWillis | phh: true ;-) |
| 11:15.25 | phh | [ 398.706939] wl1251: WARNING WLAN not ready |
| 11:15.25 | phh | [ 398.724304] firmware: requesting wl1251-nvs.bin |
| 11:15.25 | phh | [ 399.371276] wl1251: 181 tx blocks at 0x3b788, 35 rx blocks at 0x3a780 |
| 11:15.25 | phh | [ 399.396362] wl1251: firmware booted (Rev 6.0.4.40) |
| 11:15.31 | phh | but nothing works then ... |
| 11:15.43 | DJWillis | phh: It would be nice to get EEPROM based calibration reading into the mainline deiver. |
| 11:15.58 | phh | DJWillis: i don't think it is possible |
| 11:16.04 | phh | i think every board has to do it |
| 11:16.10 | DJWillis | phh: you DO realise that you hace just loaded a calibration file from the file system. |
| 11:16.30 | phh | uh ? |
| 11:16.47 | DJWillis | phh: no, the WL1251 (and WL1271) modules come with factory calibration setups burnt onto the EEPROM on the module. |
| 11:17.12 | phh | on the *wifi* module ? |
| 11:17.16 | phh | why is there on nand then ? |
| 11:17.35 | phh | and i thaught theses ones were eeprom less |
| 11:18.38 | DJWillis | phh: hmmm, ok, TI 1251 modules 'can' have the EEPROM with the calibration data, I assumed that the RALPH/DIAM etc. had the full module not running in EEPROM'less mode. |
| 11:19.18 | phh | well i'm not totally sure they are eeprom less, but as both calibration and firmware are in the NAND, i assume it is eepromless |
| 11:19.25 | DJWillis | phh: I'll confess I thought the modules had the EEPROM's and they where not being read out. Can't say I have checked in detail my WL1251 work is on other devices. |
| 11:20.14 | DJWillis | phh: have you compaired a few NVS dumps from devices to check they are not all the same and some hang over from development? Just checking ;-) |
| 11:21.01 | phh | err, i'm just thinking i did it the wrong way |
| 11:21.18 | phh | my device had three calibration data, and two lasts are identical |
| 11:21.28 | phh | i assumed the right one was one of those but mmm |
| 11:21.39 | phh | and no i haven't checked against other devices since i've got only one |
| 11:22.06 | DJWillis | phh: I know the Google G1 has an EEPROM and that one of the calibration dumps floating around is taken from that. That is why I would be supprised if the RALPH/DIAM are EEPROM'less. Last time I had WiFi on the RAPH was with the WiLink4 driver however. |
| 11:22.54 | phh | any idea on how to check ? |
| 11:24.25 | DJWillis | phh: with the WL1251 driver, that is the problem I am trying to work around ;-), you can try the TI WiLink driver with EEPROM reading (i.e. do not give it an NVS just the firmware and see if you can do a scan ;-)). The code is inside the Wilink4 state machine but it's evil ;-) |
| 11:24.52 | phh | DJWillis: wilink is the one in android ? |
| 11:25.33 | DJWillis | phh: close enough, yep, Android is based on TI's WiLink4 driver so I tend to ground all those 'driver forks' as WiLink4 |
| 11:25.42 | phh | ok |
| 11:26.08 | phh | this one doesn't work, but no idea why and i can't understand it's code, so i decided togive wl12xx a try |
| 11:26.35 | phh | but if you think it's mistake i'll go back on TI's driver |
| 11:26.55 | DJWillis | phh: join the club, that was the WHOLE reason for the WL1251 driver, the TI code makes your eyes bleed ;-) |
| 11:27.02 | phh | :) |
| 11:27.46 | phh | DJWillis: is there something to do to switch radio with wl12xx ? |
| 11:28.11 | DJWillis | phh: no, WL1251 is slower unless you move the SDIO IRQ to a GPIO IRQ but it's a lot easier to understand, supports 'modern' Linux and IS the way to go but it does not support reading the calibration EEPROM at the moment. |
| 11:29.08 | DJWillis | phh: the fact you have loaded a calibration NVS from the filesystem and IW scan does nothing implies you have duff calibration ;-) |
| 11:29.23 | cr2 | DJWillis: why reading EEPROM from nand is such a bg deal ? |
| 11:29.24 | phh | ok |
| 11:29.30 | DJWillis | phh: switch radio? |
| 11:29.57 | DJWillis | cr2: well if you have pre-calibrated modules ;-), it's not HTC specific. |
| 11:30.07 | cr2 | DJWillis: isn't the radio id coming from eeprom ? |
| 11:30.29 | cr2 | ok |
| 11:30.48 | cr2 | DJWillis: in unix everything is a file :) |
| 11:30.57 | DJWillis | cr2: no, I am talking about reading the EEPROM into the ARM on the WL1251, not reading the calibration NVS from NAND/SD. The device firmware will always have to come from NAND/SD/etc. etc. |
| 11:31.01 | cr2 | imho the eeprom should be a file in the filesystem. |
| 11:31.06 | phh | 0[ 1416.144775] wl1251: ERROR sdio write failed (-84) |
| 11:31.06 | phh | [ 1416.161010] mmc0: Data CRC error |
| 11:31.06 | phh | [ 1416.168762] msmsdcc_data_err: opcode 0x00000035 |
| 11:31.07 | phh | [ 1416.176452] msmsdcc_data_err: blksz 4, blocks 1 |
| 11:31.09 | phh | and looping |
| 11:31.11 | phh | hu. |
| 11:31.33 | cr2 | phh: too fast ? dma ? |
| 11:32.15 | DJWillis | cr2: /proc/calibration under WiLink4 (some varirents) but you have to set it up. So as WL1251 does not read the calibration part of the EERPOM (and when we try and read it we get back 0x0 :-o) that does not expose it. |
| 11:32.21 | cr2 | DJWillis: the loader i see has 3 parameters: firmware, eeprom and tiwlan.ini |
| 11:32.29 | phh | haha |
| 11:32.43 | phh | cr2: debugfs/htcraphaelmmc_dbg/wifi_pwr has to be to 1 i guess ? :D |
| 11:33.08 | cr2 | DJWillis: what's the purpose of /proc/calibration ? i prefer plain file |
| 11:33.10 | DJWillis | cr2: I am well aware of that, your talking about Google ti_loader and not wlan_cu I assume? |
| 11:33.18 | cr2 | yes |
| 11:33.30 | cr2 | i find this way more logical |
| 11:33.37 | DJWillis | cr2: I did not put it there, that comes from the WiLink4 driver. |
| 11:33.56 | cr2 | google does it in a braindead way too |
| 11:34.05 | cr2 | by creating an ATAG |
| 11:34.31 | cr2 | to play some stupid games with the nand layout |
| 11:35.01 | DJWillis | cr2: ok, tell me this, if you assume you have, say 4000 devices, all with a unique calibration from the factory on the WL1251's EEPROM do you suggest flashing a unique image to every device with a rip of that calibration to a file ;-) |
| 11:36.23 | cr2 | DJWillis: why not mount /dev/mtd/eeprom_cal ? |
| 11:37.03 | cr2 | it's position for the HTC nands is known and fixed. |
| 11:37.17 | cr2 | the same in the g1 nand |
| 11:37.25 | phh | cr2: trout_wifi_power does (dis|en)able the radio of the wifi chip ? |
| 11:37.38 | DJWillis | cr2: the 'non braindead' way to do it is read the device specific parts (i.e. NVS) from the EEPROM on driver load and just embed the firmware into the image that is common to all devices. I am at a loss as to how you plan to mount a device specific EEPROM as a damm MTD device. |
| 11:37.47 | cr2 | but ones create ATAG instead, and others the HTC_READ_EEPROM_IOCTL |
| 11:38.12 | DJWillis | phh: it cuts/enables the power to the MMC interface the chip is on. |
| 11:38.21 | phh | hu ? |
| 11:38.28 | cr2 | DJWillis: we talk about 0x800 bytes, that are stored on the device. |
| 11:38.30 | phh | how can it detect the card when it is set to 0 /o\ |
| 11:39.05 | cr2 | DJWillis: and how to access these 0x800 bytes |
| 11:39.33 | cr2 | i think that a plain file, or an mtd partition is the most user-friendly way to do it |
| 11:39.50 | cr2 | and not creating ATAGs or reading from /proc/ |
| 11:41.01 | *** join/#htc-linux StarLite (n=nnscript@s55916ca6.adsl.wanadoo.nl) |
| 11:42.06 | phh | dev-friendly i'd say :p |
| 11:42.48 | phh | cr2: so for CRC errors, i should go back to 7MHz ? |
| 11:43.30 | *** join/#htc-linux x-fak (n=WinGuru@reactos/tester/x-fak) |
| 11:43.50 | DJWillis | cr2: I sort of agree but I am trying to say 'why' when the driver should be doing this all itself. By all means dump the calibration data to a file/device etc. but there is NO need for the driver to depend on that at all. |
| 11:44.19 | phh | cr2: i'm just thinking, the 12.288Mhz you gave me is supposed to work ? since it's the clock it tries to use |
| 11:44.30 | cr2 | DJWillis: ok |
| 11:44.35 | phh | ( MSM_CLOCK_REG(12288000, 1, 0x5, 0x2, 3, 2, 4, 1), /* 12.288MHz */) |
| 11:44.43 | cr2 | phh: i've dumped these settings from my raph |
| 11:44.55 | cr2 | phh: but you may try to run at 144400 |
| 11:45.08 | *** join/#htc-linux leaigor (n=laigor@188.134.16.241) |
| 11:45.09 | cr2 | why not. |
| 11:45.13 | DJWillis | cr2: it's just that adding any FS dependancies tha do no need to be there seems 'brain-dead' to me ;-) |
| 11:45.25 | phh | cr2: by setting max_dtr ? |
| 11:45.41 | cr2 | DJWillis: /lib/firmware/* is used by many drivers |
| 11:45.51 | cr2 | phh: yes |
| 11:46.01 | cr2 | phh: select the slowest speed |
| 11:46.14 | cr2 | phh: it's used for SD init already |
| 11:46.22 | phh | ok |
| 11:46.35 | DJWillis | cr2: of course it is (and WL1251 uses it) for the device agnostic bits (such as the firmware), the calibration is unique to each and every WL1251 module. |
| 11:47.32 | cr2 | DJWillis: ok, i only care about a working driver in the end :) |
| 11:48.01 | DJWillis | cr2: I am not that aware of Android specific crapage as I spent more time with the origional WiLink4 driver. Yep, funny that, my aim is to get a good working driver. |
| 11:48.36 | cr2 | DJWillis: the g1 tiacx driver has only 1 advantage for us. it works. and the source looks really horrible. |
| 11:49.22 | DJWillis | cr2: on the 1251 there are 2 parts, the device firmware (that runs on the WL1251's ARM etc.) and the device calibration (unique to that module). Yep, I used that driver 1st on the Pandora, it works but, man it is SO nasty even TI disown it ;-). |
| 11:49.44 | DJWillis | cr2: and Nokia took one look and wrote there own (what became WL12xx) ;-) |
| 11:51.08 | *** join/#htc-linux pr0p (n=asdf@87.70.54.117) |
| 11:51.33 | DJWillis | cr2: WiLink4 is also going to become harder and harder to keep working with newer kernels (IMHO) with WEXT being refactored down and more and more work going into the MAC80211 space etc. etc. |
| 11:52.04 | phh | cr2: same thing ... |
| 11:52.12 | cr2 | nokia always reinvents the wheel |
| 11:52.23 | cr2 | phh: idon't see an easy waytodisable dma |
| 11:52.25 | phh | and it's freq is actually 144000 :/ |
| 11:52.56 | cr2 | 144kHz |
| 11:54.05 | phh | btw, disabling wifi reboots the system (either with trout_wifi_pwr or TI driver) |
| 11:54.35 | phh | but i guess that the first thing to do is to get it working :) |
| 11:59.06 | phh | [ 205.490753] mmc0: Command timeout |
| 11:59.12 | phh | maybe 144000 is too low now ? |
| 11:59.58 | GNUtoo | you're talking about trout...did someone got wifi+alsa? |
| 12:00.19 | phh | sorry it's not trout, it's just the function that's named so |
| 12:00.26 | GNUtoo | ah ok |
| 12:01.13 | DJWillis | cr2: that maybe true but in this case I can see the reason Nokia did it ;-) |
| 12:16.13 | phh | grrr |
| 12:16.17 | phh | it loaded successfully once |
| 12:16.20 | phh | and now it won't even load |
| 12:16.27 | *** join/#htc-linux thedicemaster (n=thedicem@j89051.upc-j.chello.nl) |
| 12:38.19 | cr2 | phh: i'm not sure if the driver power/reset logic is the same as in wince |
| 12:53.57 | *** join/#htc-linux W8TVI (n=nnme@71.238.235.30) |
| 13:09.05 | phh | grr, now i know why when driver loaded iwlist scan gave nothing, my wifi router was down... but i can't get this driver up again ... |
| 13:09.27 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 13:17.11 | *** join/#htc-linux ppman (n=mkern@pool-70-111-208-218.nwrk.east.verizon.net) |
| 13:17.54 | ppman | anyone on here that maight be able to help me with some angstrom development for these phones? |
| 13:22.47 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 13:23.42 | phh | cr2: i think it's some timeout thing for the CRC errors |
| 13:23.53 | phh | if i load the module without firmware nor calibration |
| 13:23.59 | phh | then reload it with firmware without calibration |
| 13:24.05 | phh | and then with firmware and calibration |
| 13:24.21 | phh | then i get the "[ 383.775543] wl1251: firmware booted (Rev 6.0.4.40)" |
| 13:25.16 | phh | YUEEEEEEEAAAAAAAHHHHHHHHAAAAAAaaa |
| 13:25.17 | cr2 | phh: timeouts in which program ? |
| 13:25.18 | phh | iwlist scan works !!! |
| 13:25.22 | phh | cr2: wifi driver |
| 13:25.22 | cr2 | :D |
| 13:25.52 | phh | groumpf no wpa_supplicant built |
| 13:26.13 | cr2 | you should describe it |
| 13:27.04 | phh | for the record i used wince firmware not android one (don't know if that changes anything) |
| 13:27.14 | phh | and wifi_pwr doesn't set power on MMC, but on the radio |
| 13:27.43 | phh | (the chip respond if wifi_pwr isn't setted, but no wifi respond, if wifi_pwr is setted iwlist scan just works) |
| 13:30.35 | *** join/#htc-linux rayman18 (i=opera@85.49.147.79) |
| 13:33.32 | cr2 | which gpio is that ? |
| 13:33.46 | cr2 | 1 gpio is shared between bt and wifi |
| 13:38.46 | phh | http://htc-linux.org/wiki/index.php?title=WL1251 here it is |
| 13:41.02 | cr2 | great |
| 13:41.23 | ppman | just curious, how much work is being put into rhodium? |
| 13:41.40 | ppman | because that's an awesome phone - if it was running linux, it's be perfect... |
| 13:43.34 | cr2 | nothing |
| 13:43.45 | cr2 | ppman: do you have rhodium ? |
| 13:43.56 | ppman | no, but I might soon |
| 13:44.20 | phh | i successfully connected to an open wifi network \o/ |
| 13:44.41 | ppman | phh: which device? |
| 13:44.46 | phh | ppman: diam100 |
| 13:45.08 | ppman | cool |
| 13:48.51 | cr2 | ppman: http://www.htc-linux.org/wiki/index.php?title=Topaz |
| 13:49.19 | cr2 | compare that to http://www.htc-linux.org/wiki/index.php?title=Raphael |
| 13:49.47 | ppman | topaz is diamond? |
| 13:50.09 | ppman | wow.. |
| 13:50.28 | phh | diamond2 i think |
| 13:50.31 | ppman | then there's us - all we're missing now is camera and bt and audio at the same time... |
| 13:51.13 | cr2 | ppman: ? |
| 13:51.13 | phh | cr2: btw, what's the status of the camera on raph/diam ? |
| 13:51.21 | cr2 | phh: do you have 2 cams ? |
| 13:51.23 | ppman | cr2: kaiser |
| 13:51.30 | phh | cr2: yes |
| 13:51.45 | cr2 | ppman: bt and audio work |
| 13:51.53 | ppman | at the same time? |
| 13:51.55 | cr2 | phh: the gpios are in wiki |
| 13:52.24 | cr2 | phh: but the clcoks need to make an rpc call |
| 13:52.31 | phh | ok |
| 13:52.36 | cr2 | i don't know if it belongs to clock-wince.c |
| 13:53.17 | cr2 | phh: but you may try to add the CAMERA pmem area, and the structs from sapphire |
| 13:53.30 | cr2 | phh: but do one simple thing first. |
| 13:53.41 | cr2 | dump the gpios 'watch gpios 0' |
| 13:54.05 | cr2 | when the cams are off, cam3mpix is on and camvga is on |
| 13:54.33 | phh | well not today, i have some other work to do |
| 13:54.39 | phh | i'll ask you back next weekend i guess |
| 13:54.45 | cr2 | then you will know which one is mux, and its state |
| 13:54.51 | cr2 | ok |
| 13:55.02 | cr2 | for me it's not a priority |
| 13:55.18 | phh | there is not many things to do left you know ? :p |
| 13:55.24 | phh | only 4MHz bt and GPS |
| 13:55.29 | phh | oh panels first |
| 13:55.48 | cr2 | tvout |
| 13:55.52 | cr2 | and fm radio |
| 13:56.17 | cr2 | but for tvout the codeaurora kernel is better |
| 13:57.05 | cr2 | ppman: why shouldn't bt work at the same time as audio ? |
| 13:57.28 | cr2 | ppman: don't know about bt pcm audio |
| 13:57.33 | ppman | the bt audio conflicts with the regular audio? not sure... |
| 13:57.49 | ppman | the name of one of the zImages got me |
| 13:57.51 | phh | ppman: it should |
| 13:58.05 | ppman | it was like... zImage-working-bt-broken-audio or something |
| 13:58.31 | cr2 | does bt really use uart1dm on kaiser ? |
| 13:58.41 | cr2 | or is it an uart1 hack ? |
| 13:58.58 | cr2 | i.e. can you switch to 4Mbit |
| 13:59.04 | ppman | I haven't delved too deep into kernel work in a long time... |
| 13:59.28 | cr2 | it's not really kernel |
| 13:59.34 | ppman | back when I was developing, I wrote the first terrible battery and modified keyboard drivers... |
| 13:59.39 | cr2 | do you load the bt firmware ? |
| 14:00.03 | phh | cr2: oh we need a good battery driver too |
| 14:00.11 | ppman | phh: what chip? |
| 14:00.25 | ppman | cr2: I don't know anything about bt on here.. |
| 14:00.39 | cr2 | ok |
| 14:00.58 | phh | ppman: don't know |
| 14:01.04 | cr2 | phh: the calculation is not good ? |
| 14:01.16 | phh | cr2: yes |
| 14:01.27 | cr2 | ok |
| 14:01.35 | ppman | can't help you there, I just write terrible battery drivers :P |
| 14:01.41 | cr2 | it's some hardcore fp asm |
| 14:02.07 | phh | cr2: if you have a disassembler you can look at HTC's wince tool |
| 14:02.28 | cr2 | wince tool ? |
| 14:02.31 | phh | there is a debug tool for battery |
| 14:02.49 | cr2 | how is it called ? |
| 14:02.55 | phh | see http://forum.xda-developers.com/showthread.php?t=416334 |
| 14:03.39 | cr2 | phh: what is your wifi fw md5sum ? |
| 14:03.52 | phh | read the wiki :p |
| 14:04.02 | phh | ebf5c2036d37bc56b4d41ddcbda4311e /lib/firmware/wl1251-fw.bin |
| 14:04.12 | phh | file was called Fw1251r1c.bin |
| 14:05.06 | cr2 | http://www.htc-linux.org/wiki/index.php?title=WL1251 |
| 14:05.18 | cr2 | added mine |
| 14:05.19 | cr2 | ebf5c2036d37bc56b4d41ddcbda4311e Fw1251r1c.bin, used on raph100 |
| 14:05.27 | cr2 | ok, it's the same |
| 14:06.55 | phh | i made a little script to get wifi "automatically", let's see if it works. |
| 14:07.39 | cr2 | "The firmware is a 192-byte file" |
| 14:07.44 | cr2 | kilobyte |
| 14:07.51 | cr2 | and only about it |
| 14:08.14 | cr2 | 185388 |
| 14:09.10 | phh | cr2: if you want to try the driver, there is some little (dirty) patch against compat-wireless i had to set |
| 14:10.08 | cr2 | hehe |
| 14:10.35 | phh | http://pastebin.com/m2ce8eb55 |
| 14:10.43 | cr2 | can we merge the wl12xx into kernel tree |
| 14:10.45 | phh | it tries to link against usb host functions |
| 14:11.02 | phh | cr2: if you know how to, you can |
| 14:11.22 | phh | but i think you'll have to update to 2.6.30 or something like that |
| 14:11.33 | cr2 | .29 now |
| 14:12.07 | cr2 | the tools seem to assume usb |
| 14:12.23 | cr2 | anchor->urb_list |
| 14:12.46 | cr2 | irrelevant for us |
| 14:12.59 | phh | yes i know |
| 14:13.16 | cr2 | g1 fw is bigger |
| 14:13.39 | phh | strange |
| 14:16.13 | cr2 | bs=8192 count=1 skip=4048 |
| 14:16.21 | cr2 | which offset is it ? |
| 14:16.34 | cr2 | bs=1 skip=64 |
| 14:17.02 | phh | it's 4048*8192 |
| 14:17.24 | cr2 | 33161216+64 |
| 14:17.27 | phh | yep |
| 14:17.38 | phh | this doesn't seems right to me. |
| 14:17.47 | cr2 | 33161280 |
| 14:18.20 | cr2 | 0x02500000 , and +0x800 |
| 14:18.39 | phh | 38 797 312 + 2048 |
| 14:18.53 | phh | =38 799 360 |
| 14:18.59 | phh | yep, the wiki seems wrong. |
| 14:19.40 | cr2 | 38797312 |
| 14:19.44 | cr2 | unlikely |
| 14:19.45 | phh | <PROTECTED> |
| 14:19.48 | phh | <PROTECTED> |
| 14:19.55 | cr2 | your value is inside SPL |
| 14:19.55 | phh | here is the command i used to get the calibration(s) |
| 14:20.36 | cr2 | err |
| 14:20.58 | cr2 | how does your partition table look like ? |
| 14:21.09 | *** join/#htc-linux Shihan (n=pjr@124-170-54-129.dyn.iinet.net.au) |
| 14:21.44 | phh | hu ? there is a partition table ? |
| 14:22.01 | cr2 | mtd |
| 14:22.07 | phh | anyway i checked and it is the values as stated in http://www.htc-linux.org/wiki/index.php?title=RaphaelNAND |
| 14:22.09 | *** join/#htc-linux rayman18 (i=opera@79.pool85-49-147.dynamic.orange.es) |
| 14:23.03 | cr2 | ash, you mean the WL1251 is wrong |
| 14:23.10 | phh | yes sorry |
| 14:23.23 | cr2 | you don't have nand |
| 14:23.33 | cr2 | well. not in the raph100 sense |
| 14:23.37 | phh | ? |
| 14:23.58 | cr2 | you talk about the builtin MMC |
| 14:24.05 | phh | no |
| 14:24.06 | phh | the NAND. |
| 14:24.16 | cr2 | hmm |
| 14:24.24 | cr2 | and you can read the first 36MB ? |
| 14:24.49 | cr2 | <PROTECTED> |
| 14:24.57 | phh | no they are all 0xff in my dump |
| 14:25.14 | cr2 | ok, it's nand mpu. |
| 14:25.30 | cr2 | you can flip 1 bit, and it goes away :) |
| 14:25.34 | cr2 | but be careful. |
| 14:26.40 | cr2 | 38799360 and 38797312 are values in raph wiki |
| 14:26.53 | phh | in the RaphaelNAND page yep |
| 14:26.57 | phh | (if my calculator is right :) |
| 14:27.10 | cr2 | 38799424 and 38797376 |
| 14:27.13 | cr2 | are your values |
| 14:27.33 | cr2 | the WL1251 wiki speaks of DIAM500 |
| 14:27.43 | phh | argh |
| 14:27.48 | cr2 | i'm a bit confused now. |
| 14:27.57 | phh | diam500 might be not the same as raph/diam 100 ... |
| 14:28.24 | cr2 | but why do yo uhave offset to raph wiki ? |
| 14:28.41 | cr2 | 64bytes |
| 14:29.08 | phh | err you're right |
| 14:29.17 | phh | don't know |
| 14:29.38 | phh | maybe a header header ? |
| 14:30.19 | cr2 | the wiki speaks of 4048*8192+64 |
| 14:30.48 | cr2 | we have 0x2500000 |
| 14:31.01 | phh | isn't the "header header" thing possible ? |
| 14:31.05 | phh | or maybe CRC |
| 14:31.09 | phh | or such things |
| 14:31.12 | cr2 | The calibration file starts with the header 0x02 0x11 0x56 0x06 0x1C 0x06. |
| 14:31.24 | cr2 | 0x800 wifi eeprom. 0x12 0x51 0xee 0x00 |
| 14:31.27 | phh | (mine does starts with 0x02 0x11 ... |
| 14:31.28 | phh | ) |
| 14:31.39 | cr2 | and the -64 offset ? |
| 14:32.00 | cr2 | 1251 ee is obvoiusly an id |
| 14:32.10 | phh | +64 to RaphaelNAND and +0 to WL1251 |
| 14:33.29 | cr2 | 4736*8192+64 on raph |
| 14:33.42 | cr2 | 4048*8192+64 on diam500 ? |
| 14:34.02 | phh | it seems so |
| 14:34.06 | phh | but maybe the wiki poster failed somehow |
| 14:34.19 | cr2 | where did he loose 736-48 ? |
| 14:35.13 | cr2 | it's about 5.5MB |
| 14:35.36 | phh | where does the RaphaelNAND page comes from ? |
| 14:35.52 | phh | it's just the partition table ? |
| 14:36.16 | cr2 | th first 36MB should be protected |
| 14:36.56 | phh | oh the diam500 address should be protected you mean ? |
| 14:37.13 | phh | yup that's weird. |
| 14:38.20 | cr2 | yes |
| 14:38.44 | cr2 | must be some read failure on his side |
| 14:39.25 | cr2 | do the first 64bytes contain something after the sig ? |
| 14:39.31 | cr2 | i'll check the dll |
| 14:39.44 | phh | cr2: do you know how to embed code (ie shell script) in the wiki ? |
| 14:41.01 | cr2 | <verbatim> |
| 14:41.54 | phh | doesn't work |
| 14:41.57 | cr2 | we need to recheck the powerup/reset sequence in the driver |
| 14:42.13 | cr2 | <verbatim> FOO </verbatim> ? |
| 14:42.21 | phh | yep |
| 14:42.34 | phh | it'll just show <verbatim> FOO </verbatim |
| 14:43.28 | cr2 | leave the trailing space |
| 14:43.46 | cr2 | like in 'dd' lines |
| 14:43.51 | cr2 | and md5sums |
| 14:43.52 | Shihan | howdy guys, i've got a possibly daft question.. the kernel that the hero boots, is the source used to boot it generally available? (i.e. the one htc themselves build off) |
| 14:44.32 | phh | cr2: the dd lines use <pre>, thanks haven't thaught of reading theses :) |
| 14:45.21 | cr2 | the length is at +0xc offset to the header |
| 14:45.53 | cr2 | version is 45,49,4a |
| 14:46.32 | phh | cr2: i added a shell script to enable wifi with working timings |
| 14:46.51 | phh | but it needs 40s to start. |
| 14:47.36 | cr2 | 0x02 0x11 0x56 0x06 0x1C 0x06. |
| 14:47.52 | cr2 | 41 42 43 44 45 46 |
| 14:48.49 | cr2 | why 3 wlans ? |
| 14:49.11 | phh | udev loves messing this up |
| 14:49.27 | phh | at first load it's wlan0 then wlan1 and then wlan2 or something like that |
| 14:50.02 | *** join/#htc-linux sdt555 (n=titus@147.145.40.44) |
| 14:50.31 | cr2 | ok |
| 14:50.34 | *** part/#htc-linux sdt555 (n=titus@147.145.40.44) |
| 15:04.39 | cr2 | the +0x800 copy is a reserve copy |
| 15:04.43 | cr2 | raid1 |
| 15:07.38 | *** join/#htc-linux dcordes-kais (n=dcordes-@ip-77-24-93-43.web.vodafone.de) |
| 15:08.55 | cr2 | phh: spl starts wifi @144kHZ |
| 15:12.36 | *** part/#htc-linux sd00 (n=smac@82-39-194-231.cable.ubr02.jarr.blueyonder.co.uk) |
| 15:13.09 | phh | well i don't even know what spl actually is |
| 15:13.58 | cr2 | 0xee1251, 0x0, 0x0 |
| 15:14.06 | cr2 | wince bootloader |
| 15:14.14 | cr2 | can test wifi too |
| 15:14.32 | cr2 | which speed do you use now ? |
| 15:14.45 | cr2 | 0x7c0 |
| 15:14.46 | phh | 7M |
| 15:15.00 | cr2 | 0xee1251, 0x0, 0x0, 0x7c0 |
| 15:15.10 | cr2 | 0x7c0 is length |
| 15:15.18 | phh | [ 98.134063] clock-wince: pc_clk_set_rate: id=19 rate=11000000 |
| 15:15.18 | phh | [ 98.157379] clock-wince: set mdns: 19, 11000000; bitidx=7, offset=a4, ns=0 |
| 15:15.18 | phh | [ 98.182464] clock-wince: set_mdns_host_clock: 19, freq=7372800 |
| 15:15.18 | cr2 | +x40=0x800 |
| 15:15.22 | phh | so it's the current kernel |
| 15:15.32 | phh | current git i mean |
| 15:15.45 | phh | (in the wifi point of view i mean) |
| 15:15.49 | cr2 | will it work with 144kHz ? |
| 15:16.13 | cr2 | also need to check the powerup sequence and timing |
| 15:16.34 | phh | i think i got timeout when i ran it @ 144kHz |
| 15:16.40 | phh | but maybe i did it wrong |
| 15:16.45 | cr2 | it is in board-htcraphael-mmc.c ? |
| 15:16.55 | phh | yes |
| 15:16.58 | phh | the max_dtr value |
| 15:17.14 | cr2 | all sdio stuff runs at 144kHz in the wince bootloader. |
| 15:17.43 | phh | there is actually part of the process done @ 144kHz |
| 15:17.43 | UnknownArtists | hi, i hve an htc arte110 and want to run linux on it |
| 15:17.57 | cr2 | ok |
| 15:18.07 | UnknownArtists | so first i have cloned that git repo by: git clone git://linuxtogo.org/home/groups/mobile-linux/kernel.git |
| 15:18.33 | UnknownArtists | i alos have buuilt a toolchain: crossdev -t armv5te-softfloat-linux-gnueabi |
| 15:18.52 | UnknownArtists | but if i try to run: make HTCARTEMIS_DEFAULTCONFIG |
| 15:19.09 | UnknownArtists | i got the following error: make: *** No rule to make target `HTCARTEMIS_DEFAULTCONFIG'. Stop. |
| 15:19.25 | phh | no capslock ... |
| 15:19.29 | UnknownArtists | whats the problem? have i cloned the wrong git-repo? |
| 15:19.55 | phh | try lowcase. |
| 15:19.59 | cr2 | make htcartemis_defconfig ? |
| 15:20.40 | cr2 | phh: int trout_wifi_power(int on) looks a bit different from wince. |
| 15:21.12 | UnknownArtists | ok first same error and if i write defconfig instead of defaultconfig is got: Can't find default configuration "arch/x86/configs/htcartemis_defconfig"! |
| 15:21.53 | cr2 | UnknownArtists: are you sure that it's the right git for omap ? |
| 15:21.56 | Shihan | ARCH=arm perhaps? |
| 15:22.05 | cr2 | ali1234: |
| 15:22.09 | UnknownArtists | i also couldn't find any occurance of "arte" in the Makefile |
| 15:22.14 | cr2 | hm, he is not here |
| 15:22.49 | UnknownArtists | cr2, some time ago miknix told me in this channel to use that git |
| 15:23.25 | cr2 | UnknownArtists: maybe, i don't know. |
| 15:23.25 | UnknownArtists | where could i find out if it is the right one? |
| 15:23.41 | UnknownArtists | or where to find alternatives |
| 15:23.41 | cr2 | ask miknix or ali1234 |
| 15:23.47 | cr2 | linwizard |
| 15:23.51 | cr2 | #linwizard |
| 15:24.06 | UnknownArtists | ok thanks |
| 15:24.58 | townkat | cr2, is kernel still broken for blac / |
| 15:24.59 | townkat | ? |
| 15:25.16 | phh | townkat: it will until someone fix it |
| 15:25.44 | townkat | what is to be done then ? |
| 15:25.58 | phh | set the correct driver name |
| 15:26.03 | cr2 | phh: what is the call sequence for trout_wifi_power / int trout_wifi_reset(int on) ? |
| 15:26.03 | phh | and i think there is some panel code to write |
| 15:26.20 | phh | cr2: what do you mean with "call sequence" ? |
| 15:26.25 | cr2 | townkat: as a hack you may add the toshiba name, but it's wrong |
| 15:26.34 | phh | if you want the order used to start the wifi on linux, read my script shell |
| 15:26.42 | phh | the driver doesn't do anything by himself |
| 15:26.44 | cr2 | phh: you've posted some tiacx log, where these functions were clalled |
| 15:28.11 | phh | http://pastebin.com/m24a88226 that's what you want ? |
| 15:28.51 | townkat | where should i add toshiba name cr2 ? |
| 15:29.19 | cr2 | phh: yes |
| 15:29.53 | cr2 | <PROTECTED> |
| 15:30.08 | cr2 | change the name to TC358720XBG |
| 15:30.13 | cr2 | but it's a hack |
| 15:31.08 | phh | cr2: i do it the other way (first CD then power) and it works so don't think it is releavant |
| 15:31.16 | cr2 | because the client name for blac100 is S1D13774 |
| 15:31.38 | cr2 | phh: i see what wince does |
| 15:31.57 | cr2 | townkat: the right fix is different |
| 15:32.33 | phh | cr2: but you'll have to check what pwr(0); does, it seems kernel code does it wrong (as said before, it crashes the kernel |
| 15:32.34 | phh | ) |
| 15:32.52 | cr2 | phh: looking at it right now |
| 15:33.43 | cr2 | townkat: err. add S1D13774 as name |
| 15:34.16 | cr2 | it is the right fix, although incomplete, because the proper epson init is not yet written. |
| 15:34.53 | phh | cr2: without panel code, what happens is that panel never suspend ? |
| 15:35.02 | cr2 | yes |
| 15:35.17 | phh | ok |
| 15:35.29 | cr2 | phh: you have hitachi or sharp panel ? |
| 15:35.52 | phh | the first one you coded i guess |
| 15:36.14 | phh | (since it worked on the first kernel update) |
| 15:36.28 | phh | but don't know which one is it |
| 15:36.29 | cr2 | there was no init then |
| 15:36.54 | cr2 | print the TC358720XBG gpio value |
| 15:37.09 | cr2 | look at the raph-panel.c code for type detection |
| 15:37.44 | cr2 | townkat: does it boot now ? |
| 15:38.45 | townkat | it may take a while |
| 15:38.52 | cr2 | ok |
| 15:38.53 | townkat | i am learning linux too now |
| 15:38.54 | townkat | :) |
| 15:39.07 | cr2 | trout_wifi_power: 1 |
| 15:39.16 | cr2 | trout_wifi_reset: 0 |
| 15:39.25 | cr2 | trout_wifi_set_carddetect: 1 |
| 15:40.37 | cr2 | int trout_wifi_reset(int on) is more or less empty |
| 15:41.04 | phh | and as far as i've seen it doesn't change a thing |
| 15:46.00 | cr2 | 66,67=0; msleep x64; vreg c=1, vreg d=1,66,67=1, msleep x14 |
| 15:47.59 | cr2 | reset does nothing |
| 15:48.06 | cr2 | power does |
| 15:50.05 | cr2 | ok, msleep 0x64 maybe added before the vreg on |
| 15:50.16 | cr2 | otherwise looks ok |
| 15:50.30 | cr2 | phh: the eeprom header is well docuented |
| 15:51.51 | phh | the 64 "missed" bytes you mean ? |
| 15:52.11 | cr2 | sig, 0, 0 , len, crc, die0, die1, die2, die3, countryid |
| 15:52.12 | cr2 | yes |
| 15:53.33 | cr2 | you may check with your data |
| 16:06.41 | *** join/#htc-linux UnknownArtists (n=unknown@84.19.198.69) |
| 16:07.38 | Shihan | theres nothing to stop you using a second partition on an SD card as an ext2 root partition is there? |
| 16:12.00 | ali1234 | back |
| 16:13.56 | cr2 | phh: cant find my nand dump. |
| 16:16.32 | townkat | cr2, i did that, same error |
| 16:16.38 | townkat | android booting hangs at ieee80211_crypt:registred algorithm 'NULL', is it a known problem ? |
| 16:17.26 | phh | no error above this message ? |
| 16:17.42 | townkat | it may be, i only noticed this becaus is the last line |
| 16:17.46 | townkat | what should i look for ? |
| 16:19.31 | cr2 | NULL is not a problem |
| 16:19.42 | cr2 | townkat: but the lcd works |
| 16:21.00 | townkat | it was "working" the same before changing client name |
| 16:21.09 | cr2 | ok |
| 16:21.31 | phh | townkat: do a "screenshot" maybe |
| 16:22.00 | townkat | k |
| 16:22.03 | townkat | i try |
| 16:22.56 | phh | i'd bet on broken initramfs, which gives really tough to find messages |
| 16:25.24 | townkat | i am using the diamond android, with blackstone zimage and only mtype changed for blackstone, is it ok ? |
| 16:30.50 | *** join/#htc-linux BHSPitMonkey (n=stephen@unaffiliated/bhspitmonkey) |
| 16:38.47 | *** join/#htc-linux sd00 (n=smac@82-39-194-231.cable.ubr02.jarr.blueyonder.co.uk) |
| 16:48.15 | *** join/#htc-linux balans (n=Administ@53543B53.cable.casema.nl) |
| 16:53.52 | phh | townkat: maybe some drivers doesn't work... give a screenshot we'll se |
| 16:53.52 | phh | e |
| 16:58.00 | *** join/#htc-linux pr0p (n=asdg@87.70.54.117) |
| 17:02.27 | Shihan | am i in the wrong channel or am i asking stupid questions? |
| 17:03.56 | rayman18 | Shinhan: what are you asking? |
| 17:03.57 | phh | first one i think, afaik, most people here have non-android phones |
| 17:04.39 | Shihan | well, mine is non-android, kaiser |
| 17:04.57 | phh | you're asking about hero |
| 17:05.44 | Shihan | yes, mostly out of curiosity and i've not found anyone who knows an answer to that... there is a link in the htc-linux wiki page, but its dead |
| 17:09.02 | Shihan | personally though im more interested in how i can mess with a kaiser, hence the sd card question |
| 17:16.19 | Shihan | hmmm, well i guess no one knows, or no one knows what ext2/3 are? |
| 17:16.30 | phh | linux "basic" filesystems |
| 17:16.37 | phh | for desktop |
| 17:18.42 | *** join/#htc-linux timebomb (n=tb@dslb-092-076-094-128.pools.arcor-ip.net) |
| 17:20.59 | townkat | is this clear enough ? http://img213.imageshack.us/img213/5082/cute0115.jpg |
| 17:21.27 | townkat | the screen also seem corrupted a bit |
| 17:21.33 | townkat | phh |
| 17:22.11 | phh | enough to be read yep |
| 17:22.42 | phh | can't find anything wrong :/ |
| 17:22.54 | phh | good luck tracing that |
| 17:23.18 | townkat | thnx |
| 17:23.32 | townkat | and thank you for looking at it |
| 17:23.33 | townkat | ;) |
| 17:27.09 | Shihan | hmmm, so where would a kaiser user go who interested in messing with linux on his phone? (most places seem to lead here) but i feel like im in the wrong place |
| 17:27.49 | phh | well yes it is here, but not on such times :p |
| 17:27.57 | phh | it's more like 2am ~ |
| 17:28.06 | phh | GMT |
| 17:28.25 | Shihan | ahhh, fair enough:) good to know im not taking crazy pills |
| 17:29.01 | *** join/#htc-linux balans (n=Administ@53543B53.cable.casema.nl) [NETSPLIT VICTIM] |
| 17:29.01 | *** join/#htc-linux W8TVI (n=nnme@71.238.235.30) |
| 17:29.03 | phh | (and i hate that since i'm GMT+2 and they begin talking when i have to go to bed.) |
| 17:29.45 | Shihan | im gmt+10.. 3:30am here... :) |
| 17:30.17 | townkat | i am gmt+2 too |
| 17:30.22 | phh | 3:30 am saturday sunday or monday ? |
| 17:30.31 | phh | monday i'd guess |
| 17:30.47 | Shihan | monday, and i have an early start which i shall regret later |
| 17:40.09 | *** join/#htc-linux W8TVI (n=nnme@71.238.235.30) |
| 17:43.49 | *** part/#htc-linux sd00 (n=smac@82-39-194-231.cable.ubr02.jarr.blueyonder.co.uk) |
| 17:47.39 | *** join/#htc-linux zwon (n=sven@ip-78-94-250-13.unitymediagroup.de) |
| 17:49.50 | *** join/#htc-linux W8TVI (n=nnme@71.238.235.30) |
| 17:51.14 | *** join/#htc-linux balans (n=barry@53543B53.cable.casema.nl) |
| 17:54.11 | *** join/#htc-linux W8TVI (n=nnme@71.238.235.30) |
| 18:01.37 | luke-jr | um |
| 18:01.45 | luke-jr | is it possible to blow WinCE away at all? |
| 18:04.15 | zwon | hi everybody |
| 18:04.33 | *** join/#htc-linux W8TVI (n=nnme@71.238.235.30) |
| 18:06.24 | zwon | cr2 told me i have to add "board_htcraphael.disable_wifi=1" to the kernel command line...stupid question but where is this command line? |
| 18:06.26 | *** join/#htc-linux GNUtoo (n=gnutoo@host134-152-dynamic.51-79-r.retail.telecomitalia.it) |
| 18:07.05 | townkat | in startup.txt |
| 18:07.44 | zwon | hm should i add this command to the set cmdline tag? |
| 18:07.46 | *** join/#htc-linux balans1 (n=Administ@53543B53.cable.casema.nl) |
| 18:08.17 | townkat | u should have something like this |
| 18:08.18 | townkat | set cmdline "msmts_calib=115.105.863.911 msmvkeyb_toggle=off mddi.width=320 mddi.height=480 lcd.density=240 ppp.apn=movistar.es ppp.username=MOVISTAR ppp.password=MOVISTAR no_console_suspend" |
| 18:08.28 | townkat | add that between the " " |
| 18:08.53 | zwon | ok thank you |
| 18:09.01 | townkat | np |
| 18:09.12 | townkat | what device do u have ? |
| 18:14.27 | *** join/#htc-linux leobaillard (n=leobaill@63.48.73-86.rev.gaoland.net) |
| 18:14.41 | leobaillard | hi all ! |
| 18:17.43 | leobaillard | I'm currently trying to get a working version of your wonderful project on my DIAM100 : it boots and all, I get to the locking screen but when I try to press the menu button, the screen trembles and goes to black and the phone crashes... |
| 18:17.49 | leobaillard | do you have any ideas ? |
| 18:20.06 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 18:25.48 | phh | leobaillard: where do you get your kernel from ? |
| 18:26.23 | leobaillard | well, let me remember... |
| 18:26.47 | phh | or the date should be enough |
| 18:27.09 | phh | if it's > 20 it's most probably known to be broken |
| 18:27.09 | leobaillard | the last one I think |
| 18:27.15 | leobaillard | around 09/22 |
| 18:27.24 | phh | ok, known to be broken then |
| 18:27.28 | leobaillard | okay |
| 18:27.39 | leobaillard | is there one that I can try ? |
| 18:27.39 | phh | get it from http://cs-alb-pc3.massey.ac.nz/vogue/files/?C=M;O=D |
| 18:28.40 | leobaillard | thx, I'm gonna try that, but I have another problem : the ppp.stuff settings for the data connection ar not taken into account by haret |
| 18:28.51 | phh | it's not a problem with haret :p |
| 18:28.57 | leobaillard | it throws me errors at booting |
| 18:29.04 | leobaillard | default.txt ? |
| 18:29.25 | phh | it says so when there is the tux logo or before ? |
| 18:29.47 | phh | if there is the tux logo, then it's normal |
| 18:29.50 | leobaillard | before, when I launch Linux, haret shows me errors |
| 18:29.57 | phh | ok |
| 18:30.02 | phh | then you misplaced quotes i guess |
| 18:30.31 | leobaillard | quotes are needed ? |
| 18:30.36 | phh | yes |
| 18:30.41 | phh | you should have something like: |
| 18:30.45 | leobaillard | http://pastebin.com/d5006ea3d |
| 18:31.02 | phh | set cmdline "msmts_calib=115.105.863.911 msmvkeyb_toggle=off mddi.width=320 mddi.height=480 lcd.density=240 ppp.apn=movistar.es ppp.username=MOVISTAR ppp.password=MOVISTAR no_console_suspend" |
| 18:31.20 | phh | (on only one line) |
| 18:32.12 | leobaillard | oh okay, I thought it was classic variables |
| 18:32.24 | leobaillard | the rest of my .txt seems correct to you ? |
| 18:32.29 | *** join/#htc-linux stickboy (n=anonymou@cpe-74-68-145-150.nyc.res.rr.com) |
| 18:32.43 | phh | ho you putted these var there... yep that's wrong :p |
| 18:32.57 | phh | don't know about other stuff |
| 18:33.09 | phh | seems fine |
| 18:33.19 | phh | but just copy and modify existing default.txt |
| 18:33.25 | leobaillard | okay |
| 18:37.16 | phh | if you need more informations please note that i'm french too and that i may accept pvs :) |
| 18:38.15 | leobaillard | t'es français ? |
| 18:38.16 | leobaillard | erf |
| 18:38.22 | leobaillard | làje me sens con :þ |
| 18:38.48 | leobaillard | c'est bon àsavoir en tout cas :) |
| 18:39.00 | *** join/#htc-linux ppman (n=mkern@pool-70-111-211-39.nwrk.east.verizon.net) |
| 18:39.15 | phh | leobaillard: french's only for private messages |
| 18:39.21 | ppman | how does the framebuffer work? |
| 18:39.24 | ppman | kaiser, specifically |
| 18:40.54 | *** join/#htc-linux balans (n=barry@53543B53.cable.casema.nl) |
| 18:41.10 | ppman | like, I can make a device node for it, right? |
| 18:41.31 | phh | ppman: yes it's normal framebuffer |
| 18:41.37 | phh | on most systems it's called /dev/msmfb |
| 18:41.39 | phh | (don't know why) |
| 18:42.26 | phh | if your system hasn't got it, check /proc/devices to know which major fb is, then /proc/fb to check the minor and do mknod /dev/fb c major minor |
| 18:42.28 | *** join/#htc-linux MethoS- (n=clemens@134.102.106.250) |
| 18:42.55 | rayman18 | wiffi now works on andriod? |
| 18:43.07 | phh | rayman18: it always worked |
| 18:43.13 | ppman | rayman18: for which device? |
| 18:43.17 | phh | but i think you're thinking about a particular device :p |
| 18:43.33 | rayman18 | oh, sorry, im talkin about diam100, of course |
| 18:43.34 | rayman18 | :) |
| 18:43.46 | phh | rayman18: not on android yet |
| 18:43.49 | phh | but it does on linux |
| 18:44.09 | rayman18 | ok, it means soon on adroid i guest.. |
| 18:44.11 | rayman18 | :) |
| 18:44.19 | phh | s/guest/guess/ but yes |
| 18:44.25 | phh | but it means a lot of troubles |
| 18:44.39 | phh | as far as i've seen it means a total rebuild of the rootfs |
| 18:44.59 | rayman18 | thanks for the info phh |
| 18:45.28 | phh | if noones does that until next weekend i will, so you'll have wifi in android next week i think :) |
| 18:45.33 | phh | (and bluetooth) |
| 18:45.51 | rayman18 | really? :) |
| 18:46.00 | rayman18 | thanks for the work |
| 18:46.21 | rayman18 | if i could help..... |
| 18:46.21 | phh | all the thanks must go to cr2 and dcordes (and maybe other) |
| 18:46.26 | phh | i had almost nothing to do |
| 18:47.24 | rayman18 | but i havent enought....knowledge |
| 18:47.29 | rayman18 | to be helpfull |
| 18:47.37 | rayman18 | only can test.....and test... |
| 18:47.49 | rayman18 | but i love this proyect |
| 18:47.51 | rayman18 | :) |
| 18:48.25 | rayman18 | thanks again for the info |
| 18:48.29 | ppman | 29, 0, c is my fb.. kthx |
| 18:48.30 | phh | i might need some testers next weekend :p |
| 18:48.48 | rayman18 | heheh |
| 18:48.53 | rayman18 | if you need....tell me |
| 18:48.55 | rayman18 | :) |
| 18:48.56 | *** join/#htc-linux balans (n=barry@53543B53.cable.casema.nl) |
| 18:51.39 | phh | well i'll need someone to check some code won't brick my diamond ;) |
| 18:52.03 | rayman18 | mmmm |
| 18:52.11 | rayman18 | that kind of test.... |
| 18:52.19 | rayman18 | :p |
| 18:52.45 | phh | well i'm kidding, as far as i know there is no reason to brick a diamond with that |
| 18:53.06 | rayman18 | i know :D |
| 18:53.32 | rayman18 | and hope so |
| 18:53.33 | rayman18 | too |
| 18:53.37 | rayman18 | <PROTECTED> |
| 18:54.57 | rayman18 | but that could be the perfect excuse to get a new android terminal like Hero..... |
| 18:55.02 | rayman18 | hehehe |
| 18:55.35 | ppman | android sucks... |
| 18:55.38 | ppman | google killed it. |
| 18:56.56 | phh | yeah |
| 18:57.08 | phh | they could have done a GUI of linux |
| 18:57.14 | phh | but no, they wanted there own kernel. |
| 18:57.44 | ppman | my sarcasm detector hates you. |
| 18:57.53 | rayman18 | but......WM sucks more...... |
| 18:57.54 | phh | :p |
| 18:58.19 | ppman | android is great, but not so much if they won't let us use stuff like market |
| 18:59.28 | phh | well you just have to copy it from an official one |
| 18:59.33 | phh | but that kinda suck :/ |
| 18:59.39 | ppman | which is illegal |
| 19:00.05 | phh | i know :( |
| 19:02.41 | phh | rayman18: that's weird you're the only one who have reacted to my announce |
| 19:06.35 | rayman18 | everybody sleeping...... |
| 19:06.41 | rayman18 | <PROTECTED> |
| 19:07.09 | *** join/#htc-linux balans1 (n=barry@212-123-149-239.ip.telfort.nl) |
| 19:07.22 | rayman18 | it's my turn soon... |
| 19:07.47 | rayman18 | now 9:07 PM |
| 19:07.49 | ppman | everyone lives in really weird timezones.. |
| 19:08.39 | phh | :) |
| 19:09.16 | cr2 | 7,b22 |
| 19:09.16 | rayman18 | <PROTECTED> |
| 19:09.27 | cr2 | dumb me. it's vibra |
| 19:14.36 | *** join/#htc-linux rafyvitto (n=rafyvitt@24.54.253.34) |
| 19:17.25 | rafyvitto | hey c2 how's the bluetooth radio work going? |
| 19:17.59 | phh | rafyvitto: just works :p |
| 19:18.08 | phh | (ok, only low rate) |
| 19:18.42 | phh | like 100kbps |
| 19:20.47 | cr2 | #define SUPPORT_WRONG_ECC_CONFIG 1 |
| 19:21.01 | cr2 | phh: how can we check if we need this ? |
| 19:21.10 | phh | cr2: no idea |
| 19:21.15 | phh | i tried to look at what it does |
| 19:21.18 | phh | but i haven't found |
| 19:21.37 | cr2 | #if SUPPORT_WRONG_ECC_CONFIG |
| 19:21.38 | cr2 | <PROTECTED> |
| 19:21.40 | cr2 | <PROTECTED> |
| 19:21.41 | cr2 | #endif |
| 19:22.09 | cr2 | <PROTECTED> |
| 19:22.26 | cr2 | it's part of the dma buffer |
| 19:22.57 | cr2 | maybe by looking how wince does the dma transfer |
| 19:23.19 | cr2 | if you are brave, you can disable it :) |
| 19:23.25 | cr2 | and try to read from nand |
| 19:23.33 | phh | and try to erase it ? :D |
| 19:23.53 | phh | dma transfer won't erase with reading i think but hum |
| 19:23.55 | phh | seems dangerous :p |
| 19:24.11 | cr2 | if #define SUPPORT_WRONG_ECC_CONFIG 1 |
| 19:24.24 | cr2 | is wrong, then it "was" dangerous :) |
| 19:24.52 | cr2 | and such small mines are all over the g1 code. |
| 19:25.18 | cr2 | wtf prevented them to write a small comment in the source ;) |
| 19:26.16 | phh | qualcomm i'd say. |
| 19:26.29 | cr2 | hehe |
| 19:27.11 | cr2 | NAND_EBI2_ECC_BUF_CFG |
| 19:29.02 | cr2 | <PROTECTED> |
| 19:29.04 | cr2 | <PROTECTED> |
| 19:29.25 | cr2 | do you have this value from dmesg ? |
| 19:29.54 | phh | my 2MB dmesg isn't enough ... |
| 19:30.19 | cr2 | initramfs supports /var/log/messages* |
| 19:30.29 | phh | (and mtd isn't built in this kernel) |
| 19:30.51 | cr2 | yes, it was considered dnagerous. |
| 19:31.01 | cr2 | because of this issue, and no partitioning |
| 19:31.14 | phh | don't worry i changed a lot of thing in the kernel configuration :p |
| 19:31.19 | *** join/#htc-linux balans (n=barry@53543B53.cable.casema.nl) |
| 19:31.22 | phh | oh |
| 19:31.28 | phh | that one does have mtd actually. |
| 19:31.41 | cr2 | :) |
| 19:31.50 | phh | cr2: /var/log/messages doesn't have boot logs ... |
| 19:32.03 | cr2 | <PROTECTED> |
| 19:32.13 | phh | arf |
| 19:32.28 | phh | my dmesg is still not large enough :D |
| 19:32.30 | phh | i'm missing 4s log |
| 19:32.40 | *** join/#htc-linux luke-jr (n=luke-jr@2002:62b3:1d4c:0:20e:a6ff:fec4:4e5d) |
| 19:32.45 | phh | but the one you want is on the second line (you lucky !) |
| 19:32.49 | phh | Sep 27 16:36:39 htcraphael user.info kernel: [ 4.253326] <6>NAND_EBI2_ECC_BUF_CFG: 1ff |
| 19:32.54 | cr2 | phh: ok |
| 19:33.10 | cr2 | phh: can you telnet to haret ? |
| 19:33.24 | phh | not today |
| 19:33.29 | cr2 | ok |
| 19:34.05 | cr2 | i'm interested if we can mmutrace the nand |
| 19:34.15 | cr2 | or it will hang like BT |
| 19:35.29 | cr2 | in msm_nand it's all dma |
| 19:35.36 | cr2 | so, unlikely |
| 19:35.46 | phh | so i got lucky to be able to dump my whole nand ? |
| 19:36.15 | cr2 | if you really want to do it, you can do it. |
| 19:37.06 | phh | i just did a dd if=/dev/mtdblock0ro of=/media/sdcard bs=8192 |
| 19:37.24 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 19:38.29 | cr2 | with "wrong_ecc_config" ? |
| 19:38.48 | phh | can't remember |
| 19:38.56 | cr2 | i had some errors in the dump |
| 19:39.02 | cr2 | compared to the wince dump |
| 19:39.12 | phh | ah, i haven't checked |
| 19:39.21 | cr2 | that's why i'd like to verify this issue |
| 19:39.26 | phh | wifi works, so i don't care. |
| 19:39.36 | cr2 | lol |
| 19:39.51 | phh | yeah i'm selfish i know :p |
| 19:40.14 | townkat | cr2, did u see my screenshot ? |
| 19:40.30 | cr2 | #define NAND_EBI2_ECC_BUF_CFG NAND_REG(0x00F0) |
| 19:40.37 | cr2 | townkat: yes. it looks ok |
| 19:40.56 | cr2 | townkat: can you boot with initramfs ? |
| 19:41.07 | townkat | how to do that ? |
| 19:41.54 | cr2 | i normally build linload.exe |
| 19:43.33 | phh | linload ... ? |
| 19:43.58 | phh | just copy the angstrom initramfs into initrd.gz in the same place as zImage |
| 19:43.58 | *** join/#htc-linux zwon_ (n=sven@ip-78-94-250-13.unitymediagroup.de) |
| 19:44.18 | zwon_ | hi again |
| 19:44.26 | zwon_ | http://pastebin.com/m23290d25 |
| 19:44.45 | zwon_ | could somebody help me? |
| 19:44.53 | phh | what's the problem ? |
| 19:45.06 | phh | kernel works at least |
| 19:45.11 | townkat | if i give you full access to my android dir can u do that? i am not sure i know what to do |
| 19:45.32 | phh | zwon_: root=/dev/ram0 ?!? |
| 19:45.35 | phh | that seems *really* weird |
| 19:45.57 | zwon_ | what should itype else |
| 19:46.01 | phh | ram0 is only for initrd (not initramfs), and it's initramfs according to your log |
| 19:46.14 | phh | just drop the root= thing |
| 19:46.31 | cr2 | #if SUPPORT_WRONG_ECC_CONFIG |
| 19:46.32 | cr2 | <PROTECTED> |
| 19:47.34 | cr2 | <PROTECTED> |
| 19:47.35 | cr2 | <PROTECTED> |
| 19:47.43 | cr2 | <PROTECTED> |
| 19:47.45 | cr2 | <PROTECTED> |
| 19:47.46 | cr2 | <PROTECTED> |
| 19:48.47 | cr2 | but you have 1ff |
| 19:48.50 | cr2 | Sep 27 16:36:39 htcraphael user.info kernel: [ 4.253326] <6>NAND_EBI2_ECC_BUF_CFG: 1ff |
| 19:49.06 | cr2 | phh: i suggest to disable :) |
| 19:49.15 | phh | ok |
| 19:49.41 | phh | and how safe on a range from 0 to 10 is it ? :D |
| 19:49.46 | phh | -10 ? :D |
| 19:49.58 | cr2 | phh: where ? |
| 19:50.11 | phh | how safe it is to disable it i mean |
| 19:50.16 | cr2 | <PROTECTED> |
| 19:50.18 | cr2 | <PROTECTED> |
| 19:50.19 | cr2 | #if SUPPORT_WRONG_ECC_CONFIG |
| 19:50.21 | cr2 | <PROTECTED> |
| 19:50.22 | cr2 | <PROTECTED> |
| 19:50.23 | cr2 | #endif |
| 19:50.54 | cr2 | it's just an extra feature |
| 19:50.54 | phh | ok i'm crazy, i'm trying. |
| 19:50.59 | cr2 | ok :) |
| 19:51.35 | cr2 | your saved value is 0x1ff, and the "working" value is 0x203 then |
| 19:51.46 | cr2 | otherwise you stay with 0x1ff |
| 19:51.55 | phh | if it stops, it's maybe for a good reason you know. |
| 19:52.01 | cr2 | you may see ecc errors at read |
| 19:52.16 | cr2 | but it's unlikely |
| 19:52.26 | phh | zwon_: remove the root=/dev/ram0 thing in your cmdline |
| 19:52.33 | cr2 | unless you write into nand, nothing happens. imho |
| 19:52.42 | phh | cr2: i wrote only read only driver |
| 19:52.49 | phh | built* |
| 19:52.51 | cr2 | ok |
| 19:52.57 | phh | i'm not THAT crazy ! :p |
| 19:53.04 | cr2 | LOL |
| 19:53.23 | cr2 | it's a pity that druidu is not here |
| 19:53.52 | phh | crossing fingers. |
| 19:55.04 | phh | *grrrr* |
| 19:55.22 | phh | i still get that stupid [ 66.660369] msm_nand_read_oob: unsupported ops->len, 512 thing |
| 19:55.31 | cr2 | it's normal |
| 19:55.47 | phh | i DO bs=8192 ! |
| 19:55.54 | phh | and strace confirms dd does 8192 ! |
| 19:56.20 | phh | and actually that was the problem even before WRONG_ECC_CONFIG |
| 19:56.37 | phh | when i successfully dumped my nand, it was on android's initramfs |
| 19:56.39 | cr2 | (ops->len % mtd->writesize) != 0 |
| 19:56.52 | phh | wait |
| 19:56.53 | phh | *write* size ? |
| 19:57.12 | phh | ... |
| 19:57.15 | phh | 131072 works. |
| 19:57.29 | cr2 | 0x20000 ? |
| 19:57.38 | phh | ah no, 131072 doesn't |
| 19:57.57 | phh | wtf |
| 19:58.05 | phh | it's discountinuous |
| 19:58.24 | cr2 | the physical block size is 0x800 |
| 19:58.29 | cr2 | +0x40 |
| 19:58.36 | phh | +0x40 ? |
| 19:58.57 | cr2 | ecc and things |
| 19:59.01 | phh | just thinking ... i might have wrong build option... |
| 19:59.45 | phh | groumpf can't remember which option i used |
| 19:59.52 | phh | was it direct char or block device ? |
| 20:00.08 | phh | let's try again with char. |
| 20:00.54 | cr2 | 0x10 extra per 512byte |
| 20:01.11 | phh | let's reenable wrong_ecc_config but using char device |
| 20:01.43 | tmzt | wifi works? |
| 20:01.50 | phh | tmzt: yes :) |
| 20:02.28 | cr2 | B11 |
| 20:05.23 | phh | cr2: that was it |
| 20:05.31 | zwon | phh: http://pastebin.com/m1557916f |
| 20:05.31 | phh | mtdchar do works even with WRONG_ECC_CONFIG |
| 20:05.34 | phh | mtd block doesn't |
| 20:05.40 | phh | either with or without WRONG_ECC_CONFIG |
| 20:06.05 | phh | zwon: root=/dev/ram0 is still here ... |
| 20:06.21 | cr2 | zwon: mmc0 is wifi ? |
| 20:06.28 | zwon | hm i deleted it^^ |
| 20:07.00 | phh | cr2: i think the block layer is stupid and/or need some settings to set block size to something else than 512bytes |
| 20:07.04 | cr2 | [ 3.438415] mmc0: Qualcomm MSM SDCC at 0x00000000a0500000 irq 26,102 dma 8 |
| 20:07.13 | cr2 | no, it's sdc2 |
| 20:07.16 | *** join/#htc-linux Nanto (n=Vegita@54031C33.catv.pool.telekom.hu) |
| 20:07.26 | phh | while char is direct |
| 20:07.32 | phh | so dd gives the right block size |
| 20:07.44 | cr2 | phh: something is incomplete at the mtd layer setup |
| 20:08.22 | phh | cr2: well if mtd char works, you won't need anything else for a long time (until using linux as NAND os i'd say.) |
| 20:09.04 | cr2 | phh: ok. it's a noble goal :) |
| 20:09.12 | cr2 | zwon: mmc0: MMC clock 144000 -> 32000000 Hz, PCLK 66000000 Hz |
| 20:09.38 | cr2 | zwon: can you reduce the speed ? |
| 20:09.54 | *** join/#htc-linux c4software (n=Adium@roo49-2-88-161-139-221.fbx.proxad.net) |
| 20:09.56 | zwon | to 25MHz? |
| 20:10.01 | cr2 | 24 |
| 20:10.12 | phh | mmc is 32Mhz ?!? |
| 20:10.15 | cr2 | also wit h the commandline parameter |
| 20:10.19 | phh | isn't that *really* high ? |
| 20:10.23 | cr2 | phh: SD |
| 20:10.36 | *** join/#htc-linux swc|666 (n=infidel2@unaffiliated/swc666/x-4934821) |
| 20:10.36 | cr2 | phh: mmc is 20MHz |
| 20:10.38 | phh | ah not the same thing ? |
| 20:10.42 | cr2 | no |
| 20:10.48 | phh | i still haven't understand the difference -_- |
| 20:10.50 | c4software | hi |
| 20:11.01 | cr2 | it was political at the beginning |
| 20:11.07 | townkat | phh, there is no initramfs file in initrd.gz |
| 20:11.13 | cr2 | so the sd driver in linux is called mmc |
| 20:11.15 | phh | townkat: ?!? |
| 20:11.20 | phh | townkat: initrd.gz *is* an initramfs |
| 20:11.30 | *** part/#htc-linux rayman18 (i=opera@79.pool85-49-147.dynamic.orange.es) |
| 20:11.45 | tmzt | no-console-suspend?what does that do? |
| 20:12.13 | phh | tmzt: on default linux configuration, console "suspend" by himself on a timeout |
| 20:12.15 | cr2 | zwon: "msmsdcc_fmax=24000000 |
| 20:12.19 | phh | meaning it blanks the screen |
| 20:12.22 | townkat | phh, i i only need to extract initrd.gz ? |
| 20:12.23 | phh | which might me annoying |
| 20:13.04 | c4software | i have left this week end whats new on current kernel ? (wifi and BT enable for everyone ?) |
| 20:13.44 | phh | c4software: not for everyone |
| 20:13.54 | phh | well don't know if it's for everyone |
| 20:14.00 | phh | only diam100 tested for wifi |
| 20:14.05 | phh | and BT only on diam100 & raph100 |
| 20:14.09 | c4software | nice |
| 20:14.16 | c4software | i will pm you phh |
| 20:15.01 | townkat | \o/ ? |
| 20:15.43 | rafyvitto | i cant enable my bluetooth it says starting and then goes back to turn on< does it torn on on angstrong? |
| 20:16.19 | cr2 | rafyvitto: echo 1 > /sys/class/rfkill/rfkill0/state |
| 20:16.29 | cr2 | rafyvitto: but it's the same on g1 afair |
| 20:16.40 | phh | cr2: it won't be enough |
| 20:16.46 | phh | the G1 command line won't work on diam100 |
| 20:16.53 | phh | also the default kernel doesn't have UART HS |
| 20:17.01 | phh | (if we are speaking about raph/diam) |
| 20:17.15 | cr2 | i think it has |
| 20:17.21 | cr2 | i've added it long ago |
| 20:17.43 | phh | cr2: not default configuration |
| 20:17.49 | phh | on diamond at least |
| 20:17.51 | phh | maybe raphael |
| 20:18.15 | *** join/#htc-linux W8TVI (n=nnnme@71.238.235.30) |
| 20:18.27 | cr2 | CONFIG_SERIAL_MSM_HS=y |
| 20:18.53 | cr2 | ok, not on diam |
| 20:19.10 | phh | who is "in charge" of diamond ? |
| 20:19.17 | phh | because there are many lacks against raphael |
| 20:19.27 | cr2 | you :) |
| 20:19.38 | townkat | :) |
| 20:19.54 | phh | i don't know a thing about MSM /o\ |
| 20:20.11 | cr2 | you only need to know a thing about git |
| 20:20.18 | phh | true |
| 20:20.34 | cr2 | everything else you can discuss here |
| 20:21.24 | townkat | i extracted the initrd.gz and changed startup.txt accordingly, it hang at the same line |
| 20:21.35 | phh | cr2: there is noone else who usually does this ? |
| 20:21.50 | cr2 | what ? |
| 20:22.11 | phh | who commited *diamond* ? |
| 20:22.25 | cr2 | many people |
| 20:22.27 | c4software | hmmm a doc exist for power management for the lcd panel |
| 20:22.42 | cr2 | but they seem to be busy now |
| 20:23.34 | phh | ok |
| 20:23.40 | *** join/#htc-linux balans1 (n=barry@212-123-149-239.ip.telfort.nl) |
| 20:24.11 | cr2 | and the devices are so similar, that it does not really matter |
| 20:24.26 | cr2 | raph just has most hardware |
| 20:24.30 | cr2 | features |
| 20:24.57 | phh | well i bought the cheapest useful smartphone i found |
| 20:25.17 | phh | 150euro for such a thing, it's enough to me :) |
| 20:25.48 | townkat | i tried with the old initrd.cpio.gz i have from the old working android we have for blac, still hang at the same error, this should mean its not the initramfs isnt it ? |
| 20:25.48 | phh | (i bought it second hand two months ago.) |
| 20:25.59 | phh | townkat: guess so yes |
| 20:26.05 | phh | maybe framebuffer is corrupted |
| 20:26.09 | phh | don't know how to see that |
| 20:26.20 | *** join/#htc-linux rafyvitto (n=rafyvitt@24.54.253.34) |
| 20:26.35 | cr2 | townkat: you boot with wifi enabled ? |
| 20:34.04 | townkat | yup |
| 20:34.08 | townkat | should i disable it ? |
| 20:36.41 | cr2 | townkat: yes |
| 20:36.55 | townkat | i disabled it now |
| 20:36.56 | townkat | same |
| 20:37.31 | townkat | what about bluetooth? |
| 20:37.44 | townkat | board_htcraphael.disable_wifi=1 does not disable bt isnt it ? |
| 20:39.20 | zwon | http://pastebin.com/m5d7f9b93 |
| 20:39.33 | zwon | still same mmc errors |
| 20:44.04 | townkat | zwon, how do you get that data out of console ? |
| 20:44.11 | townkat | is there some log file or... ? |
| 20:46.46 | zwon | in haretconsole type PWF config.txt 0xe000d 0x20000 |
| 20:47.15 | zwon | then you get a dmesg file to your haret dir |
| 20:47.28 | townkat | ahh, i see |
| 20:48.35 | townkat | thnx |
| 20:48.44 | townkat | u have a transcend card ? |
| 20:50.53 | *** join/#htc-linux balans (n=barry@212-123-149-239.ip.telfort.nl) |
| 20:52.10 | zwon | right xD |
| 20:52.18 | zwon | how do you know |
| 20:52.20 | zwon | ? |
| 20:54.39 | townkat | i seen u have a blacstone |
| 20:54.45 | townkat | i have one too |
| 20:55.05 | zwon | ah ok |
| 20:55.08 | townkat | but i dont have a transcend card |
| 20:55.19 | zwon | you have a transcend card too? |
| 20:55.26 | zwon | ah ok^^ |
| 20:55.34 | townkat | mine boots more, but still not complete |
| 20:55.42 | *** join/#htc-linux pr0p (n=asf@87.70.54.117) |
| 20:56.01 | townkat | but it boots the old android package in the wiki completly |
| 20:56.27 | *** join/#htc-linux timebomb (n=tb@e177142005.adsl.alicedsl.de) |
| 20:56.53 | zwon | hm doesnt work with mine too |
| 20:57.05 | zwon | xD |
| 20:58.06 | zwon | i think i need a new card...but if i buy one android will start as NAND |
| 20:58.18 | *** join/#htc-linux pr0p (n=asf@87.70.54.117) |
| 20:58.42 | townkat | did u try to boot from internal flash ? |
| 20:59.09 | zwon | hm no i dont found a tutorial |
| 20:59.27 | townkat | just put the files in the internal flash i think |
| 20:59.31 | townkat | tmp |
| 20:59.47 | zwon | hm ok i'll try |
| 21:00.03 | tmzt | Glem: what machines are enabled in your kernels? |
| 21:01.04 | townkat | what mkes haret to boot directly without presenting the window? |
| 21:02.50 | *** join/#htc-linux ipauldev (n=ipauldev@phassinger.dsl.visi.com) |
| 21:03.18 | tmzt | cr2: cli is clear interrupts |
| 21:04.00 | tmzt | well not in that case maybe |
| 21:04.07 | ipauldev | G1 Screen Replacement DIY - http://digg.com/d315ixu |
| 21:04.13 | *** part/#htc-linux ipauldev (n=ipauldev@phassinger.dsl.visi.com) |
| 21:04.34 | *** join/#htc-linux ipauldev (n=ipauldev@phassinger.dsl.visi.com) |
| 21:05.53 | *** part/#htc-linux ipauldev (n=ipauldev@phassinger.dsl.visi.com) |
| 21:06.34 | tmzt | townkat: startup.txt or linboot instead of haret |
| 21:06.41 | tmzt | ask r2 about the econd one |
| 21:06.51 | tmzt | s/ec/sec/ |
| 21:07.32 | townkat | abt the first one |
| 21:07.36 | townkat | what in startup.txt ? |
| 21:08.34 | *** join/#htc-linux dream_kill (n=nospam@89.131.127.37) |
| 21:09.02 | townkat | s/startup/star |
| 21:09.11 | townkat | s/startup/star/ |
| 21:09.48 | townkat | what in startup.txt ? |
| 21:09.51 | townkat | s/startup/star/ |
| 21:09.59 | townkat | :D |
| 21:13.22 | tmzt | same as default.txt |
| 21:13.44 | zwon | tmzt: it wont work on internal flash |
| 21:13.45 | *** join/#htc-linux rashire2 (n=ed1112wa@pool-98-114-209-170.phlapa.fios.verizon.net) |
| 21:13.51 | tmzt | but if you put boot in startup.txt it starts linux without showing anything |
| 21:14.02 | townkat | ahh |
| 21:14.06 | townkat | so boot is the key |
| 21:14.07 | townkat | ;) |
| 21:14.10 | townkat | thank you |
| 21:14.11 | tmzt | zwon: the loop files can't be on nand |
| 21:14.12 | townkat | :D |
| 21:14.35 | tmzt | zwon: they can only be internal on diamond |
| 21:14.48 | tmzt | as it has embedded sd |
| 21:15.04 | zwon | which are the loop files? |
| 21:15.09 | zwon | e2img ...? |
| 21:15.19 | cr2 | tmzt: why not on nand ? |
| 21:15.51 | tmzt | cr2: it ha to loop mount things and linux can't access nand yet I think |
| 21:16.01 | tmzt | cr2: how do I get wifi working? |
| 21:16.21 | tmzt | yes, e2img |
| 21:16.21 | cr2 | tmzt: wiki |
| 21:16.22 | tmzt | in git? |
| 21:17.32 | cr2 | http://www.htc-linux.org/wiki/index.php?title=WL1251 |
| 21:17.39 | tmzt | I think I'm going to work on getting an shr image to work so I can have current X |
| 21:17.55 | cr2 | the eeprom data for diam500 is strange |
| 21:18.29 | cr2 | tmzt: why did you patch the fbmem.c file ? |
| 21:18.40 | cr2 | how does it work on other phones |
| 21:19.31 | phh | cr2: i'm currently trying to get wifi work for c4software's diam100 btw |
| 21:19.31 | zwon | hm i got the loop files on both |
| 21:20.24 | *** join/#htc-linux leaigor (n=laigor@188.134.16.241) |
| 21:22.19 | tmzt | can we export panel id in htchw or somewhere |
| 21:22.26 | tmzt | class mddi-client maybe |
| 21:22.44 | tmzt | cr2: I did? |
| 21:23.02 | tmzt | msm-fb? |
| 21:23.28 | cr2 | - if (i) return i; |
| 21:23.29 | cr2 | + //if (i) return i; |
| 21:23.31 | cr2 | + if (i) return 0; |
| 21:23.35 | cr2 | drivers/video/fbmem.c |
| 21:23.49 | cr2 | i = fb_set_var(info, &var); |
| 21:23.54 | tmzt | ah |
| 21:24.01 | cr2 | what's wrong with this call ? |
| 21:24.14 | tmzt | that should be remove pixclock check instead |
| 21:24.34 | tmzt | because X sends 0 for pixclock |
| 21:24.58 | tmzt | that's an upstream patch but not sure if it goes in that file |
| 21:25.02 | cr2 | what should be returned for pixclock ? |
| 21:25.11 | tmzt | returned? |
| 21:25.16 | tmzt | this is the set |
| 21:25.32 | cr2 | ok, why not set it ? |
| 21:25.38 | tmzt | it should succeed reguardless or what pixclock is |
| 21:25.51 | tmzt | becaue it can't be 0 |
| 21:26.05 | cr2 | can't it be solved in fb_set_var ? |
| 21:26.12 | tmzt | yes |
| 21:26.16 | tmzt | it should be |
| 21:26.27 | tmzt | look at 2.6.31 to see how it was fixedx |
| 21:26.31 | cr2 | where is fb_set_var for msmfb ? |
| 21:26.36 | tmzt | I did this because it worked |
| 21:26.40 | tmzt | look i opsn |
| 21:26.41 | cr2 | hehe |
| 21:26.42 | tmzt | ops |
| 21:26.56 | tmzt | then I found out why it was broken |
| 21:26.59 | cr2 | we have .27, or .29 |
| 21:27.18 | tmzt | if strace dumped the struct it would have been obviou |
| 21:27.22 | tmzt | upstream |
| 21:27.31 | tmzt | there's a patch committed |
| 21:27.39 | cr2 | well, what should we patch in .27 git ? |
| 21:28.03 | tmzt | I believe they removed the pixclock check |
| 21:28.14 | tmzt | on linus tree |
| 21:28.22 | cr2 | ok |
| 21:28.37 | cr2 | what was the problem with msm_ts ? |
| 21:28.55 | tmzt | it returned cooked coordinates |
| 21:29.03 | tmzt | and X/tslib wants raw |
| 21:29.10 | cr2 | i think we had msm_ts2 ? |
| 21:29.28 | townkat | hmm, i managed to ouput a txt from haret and it seem to go further than it show on screen |
| 21:29.36 | townkat | http://pastebin.com/mefa001d |
| 21:29.56 | townkat | the screen output stops at [ 44.696868] ieee80211_crypt: registered algorithm 'NULL' |
| 21:30.00 | tmzt | new X has it's own calibration which intend to use |
| 21:30.11 | tmzt | maybe didn't are one |
| 21:30.14 | cr2 | the current msm_ts contains such android things, that there is no chance it will be used anywhere |
| 21:30.15 | tmzt | see |
| 21:30.31 | cr2 | then we can purge all this |
| 21:30.56 | tmzt | with the vkeyb changed to a hook its ok |
| 21:31.02 | cr2 | btw, what about msm_ts from .29 tree ? |
| 21:31.21 | tmzt | haven't looked |
| 21:32.02 | cr2 | i think current msm_ts.c needs to be moved into msm_ts_android.c |
| 21:32.34 | *** join/#htc-linux rafyvitto (n=rafyvitt@24.54.253.34) |
| 21:33.33 | tmzt | I want tup use darkstars method of using a userspace program for calibration but be able to configure kernel with the results |
| 21:33.45 | tmzt | but X doesn't need it |
| 21:33.56 | tmzt | right |
| 21:34.12 | cr2 | it's not the kernel issue anyway |
| 21:34.18 | tmzt | or config upotion to disable that |
| 21:34.29 | tmzt | as well as vkeyb |
| 21:34.41 | tmzt | true |
| 21:34.45 | cr2 | if android needs it, leave it as an android-dependent hack, but dont spoil the kernel source |
| 21:35.09 | tmzt | but until android user space can do calibration |
| 21:35.14 | tmzt | they need it |
| 21:35.27 | tmzt | but with darkatae |
| 21:35.28 | cr2 | ok |
| 21:35.51 | tmzt | darkstar method it could be done before starting android |
| 21:35.56 | cr2 | does adsp/sound depend on pmem ? |
| 21:36.11 | tmzt | not for playsc |
| 21:36.22 | tmzt | playwav I think |
| 21:36.35 | tmzt | but audiod |
| 21:36.45 | tmzt | but audioflinger does |
| 21:37.05 | tmzt | it should be alas dma |
| 21:37.17 | tmzt | alsa |
| 21:37.45 | cr2 | yes, but if i will disable pmem and other android stuff, what will i loose ? |
| 21:38.17 | cr2 | playwav is good enough as /dev/dsp |
| 21:38.40 | cr2 | for now |
| 21:39.04 | tmzt | not sure |
| 21:39.12 | cr2 | because making the mixer and equalizer properly work with alsa is a challenge. |
| 21:39.31 | tmzt | yes |
| 21:40.01 | tmzt | that will probably emaib I htchw |
| 21:40.26 | tmzt | unless why use request firmware or something |
| 21:41.14 | cr2 | yes, i mean only right now |
| 21:41.35 | cr2 | when you do cat file.wav > /dev/dsp, it's more or less the same as playwav |
| 21:42.38 | cr2 | we can even write the custom /dev/dsp driver for it, to hide the whole ioctl business |
| 21:43.04 | cr2 | - fb_info->var.yres_virtual = msmfb->yres * 2; |
| 21:43.05 | cr2 | + fb_info->var.yres_virtual = msmfb->yres; |
| 21:43.18 | cr2 | can this be done in some better way ? |
| 21:43.26 | phh | cr2: wifi works for c4software too :) |
| 21:43.39 | cr2 | phh: congratulations :) |
| 21:43.51 | c4software | Yep, but we need to launch the load.sh script twice |
| 21:43.55 | cr2 | phh: but we need to find a less hacky way :) |
| 21:44.09 | phh | cr2: yeah |
| 21:44.17 | tmzt | cr2: oss is not worth it |
| 21:44.21 | phh | i'll make a build with it still |
| 21:44.42 | cr2 | tmzt: yeah |
| 21:45.55 | tmzt | I've split drivers/video/msm into msm-single with its own Kconfig option |
| 21:49.17 | townkat | anyone pls take a look at my dmesg ? |
| 21:50.01 | cr2 | tmzt: can we malloc the framebuffer ram too ? |
| 21:50.25 | cr2 | instead of |
| 21:50.28 | cr2 | <PROTECTED> |
| 21:50.30 | cr2 | <PROTECTED> |
| 21:50.55 | cr2 | it can be put into SMI though |
| 21:51.32 | tmzt | didn't we discuss that? |
| 21:51.47 | tmzt | if it could allocate enough ram |
| 21:51.48 | cr2 | i'm only a bit afraid about non EBI1 dma |
| 21:52.05 | cr2 | how does it work for pxafb ? |
| 21:52.12 | cr2 | on universal |
| 21:52.20 | tmzt | it's in devices.c |
| 21:52.22 | cr2 | also a vga 16bit screen |
| 21:52.30 | cr2 | ok, here it's 18bit |
| 21:53.04 | cr2 | yes, a good question if the SMI is faster |
| 21:53.48 | tmzt | does g1 use smi or ebi? |
| 21:53.52 | *** join/#htc-linux pr0p (n=ag@87.70.54.117) |
| 21:54.20 | tmzt | so there's no new code for wifi? that debugfs entry is already committed? |
| 21:54.25 | cr2 | find ./kernel -name Kconfig | xargs grep PMEM |
| 21:54.27 | cr2 | ./kernel/drivers/misc/Kconfig:config ANDROID_PMEM |
| 21:54.47 | cr2 | who has said that ADSP depends on it ? |
| 21:54.52 | tmzt | ANDROID PMEM is the allocation of the pmem devices |
| 21:55.02 | tmzt | that's for userspace |
| 21:55.12 | tmzt | PMEM itself is needed I think |
| 21:55.14 | cr2 | == not needed for X |
| 21:55.24 | tmzt | not sure how this affects non-android |
| 21:55.33 | tmzt | X only uses msmfb |
| 21:55.43 | cr2 | isn't pmem something arm9 related ? |
| 21:56.03 | cr2 | i think it's some qualcomm innovation, originally |
| 21:56.11 | tmzt | yeah |
| 21:56.19 | tmzt | like rpc is used to allocate it |
| 21:56.29 | cr2 | tmzt: does your wifi fw match the wiki ? |
| 21:56.33 | cr2 | yes |
| 21:56.43 | tmzt | and their drvers use physical addresses |
| 21:56.52 | tmzt | I think so |
| 21:59.41 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 21:59.47 | cr2 | can you dump the eeprom ? |
| 22:00.18 | cr2 | phh: what are the differences between your eeprom and c4software ? |
| 22:00.41 | phh | c4software: give me your calibration to phhusson@free.fr |
| 22:00.54 | c4software | no problem |
| 22:01.04 | *** join/#htc-linux Captnoord (n=Captnoor@dc5147a47b.adsl.wanadoo.nl) |
| 22:01.15 | c4software | just a minute |
| 22:02.03 | cr2 | c4software: with the 0x40 header is even better |
| 22:02.47 | c4software | mail send |
| 22:03.09 | phh | c4software: calibration.2 then :D |
| 22:03.42 | cr2 | i wish we can teach the mtd layer about our blocksize |
| 22:05.44 | cr2 | [ 10.459887] msm_nand_read_oob: unsupported ops->len, 512 |
| 22:05.47 | cr2 | [ 10.474108] end_request: I/O error, dev mtdblock3, sector 0 |
| 22:05.47 | cr2 | [ 10.487597] Buffer I/O error on device mtdblock3, logical block 0 |
| 22:06.04 | phh | cr2: use mtd char |
| 22:06.46 | cr2 | i have a partition table |
| 22:06.55 | cr2 | need to remeber where is what |
| 22:07.36 | tmzt | I think you can |
| 22:07.53 | tmzt | ust that yaffs2 sets up the ops itself |
| 22:08.39 | tmzt | so you have to take the code from there |
| 22:09.07 | cr2 | tmzt: it's google-git-only ? |
| 22:10.17 | *** join/#htc-linux thedicemaster (n=thedicem@j89051.upc-j.chello.nl) |
| 22:10.17 | tmzt | what is? |
| 22:10.27 | cr2 | yaffs2 |
| 22:10.34 | tmzt | from that driver I mean |
| 22:10.45 | tmzt | no we should have it |
| 22:11.07 | tmzt | I mean that only that fs works |
| 22:11.39 | tmzt | len is 2k? |
| 22:16.10 | Captnoord | re |
| 22:16.11 | Captnoord | yo all |
| 22:16.26 | cr2 | 0x800 |
| 22:16.27 | cr2 | yes |
| 22:16.47 | Captnoord | how's wifi? |
| 22:16.47 | Captnoord | :P |
| 22:16.56 | phh | on diam/raph ? good and you ? :p |
| 22:17.15 | Captnoord | just got back from my girl so i'm good.... |
| 22:17.16 | Captnoord | :P |
| 22:17.18 | Captnoord | you? |
| 22:17.40 | phh | i'm too :p |
| 22:18.23 | cr2 | phh: i've dumped a good chunk of nand with my wrong partition table from May :) |
| 22:18.46 | cr2 | it didn't account for the 36MB offset |
| 22:19.42 | phh | hu ? |
| 22:19.46 | tmzt | apps partition table? |
| 22:20.12 | *** join/#htc-linux jummama (n=AndroidI@m420536d0.tmodns.net) |
| 22:20.20 | phh | cr2: if you want some prebuilt kernel/wifi driver, i've got on http://husson.hd.free.fr/zImage-27-09-2009 / http://husson.hd.free.fr/d2.tar |
| 22:20.35 | phh | i don't know how it will do on raphael |
| 22:20.50 | phh | but i think keyboard isn't build :/ |
| 22:22.23 | tmzt | phh: raph 500 support? |
| 22:22.37 | tmzt | wait |
| 22:22.37 | phh | tmzt: don't think so |
| 22:22.40 | tmzt | that breaks thkbga |
| 22:22.53 | phh | thkgba .. ? |
| 22:23.05 | tmzt | that's ok |
| 22:23.09 | tmzt | things |
| 22:23.10 | phh | tmzt: it has been built with raphael CDMA |
| 22:23.34 | phh | but i don't think wifi will work as easily |
| 22:23.49 | phh | but first you have to get your calibration data :p |
| 22:31.41 | c4software | its time for me to go :) |
| 22:31.47 | c4software | gnight everyone |
| 22:33.53 | cr2 | err. |
| 22:33.57 | cr2 | wince dmesg in nand |
| 22:34.30 | cr2 | x260000 offset |
| 22:35.43 | cr2 | x2e0000 end |
| 22:35.43 | jummama | hey, just wondering if there's been any work toward an X server on dream other than Xfbdev patches |
| 22:36.25 | cr2 | jummama: 2d acceleration for msm7201A ? |
| 22:36.55 | jummama | yeah |
| 22:37.39 | cr2 | afaik no |
| 22:37.49 | jummama | 3d would be nice, but i'll take what i can get |
| 22:38.46 | jummama | thanks |
| 22:38.48 | cr2 | ask tmzt |
| 22:39.11 | phh | how does android do ? |
| 22:39.17 | phh | maybe there is some EGL ? |
| 22:39.25 | phh | (in which case you could use Xegl) |
| 22:40.47 | jummama | android has opengles, maybe it'd be possible to get an xserver running inside android with the chrooted debian trick |
| 22:41.25 | phh | or just copy the egl libs to your own system |
| 22:41.32 | phh | i don't think they heavily rely on android |
| 22:42.58 | *** join/#htc-linux sd00 (n=smac@82-39-194-231.cable.ubr02.jarr.blueyonder.co.uk) |
| 22:43.21 | jummama | hmmm... thanks for the idea! |
| 22:53.16 | *** join/#htc-linux |Jason8|| (i=JayAte@server1.tiltshellz.org) |
| 22:54.51 | *** join/#htc-linux Orbital_sFear (n=ldupin@bbb72-0-189-133.bendbroadband.com) |
| 22:56.24 | *** part/#htc-linux sd00 (n=smac@82-39-194-231.cable.ubr02.jarr.blueyonder.co.uk) |
| 23:02.52 | *** join/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146) |
| 23:02.59 | Untouchab1e | Good evening |
| 23:11.53 | cr2 | good night |
| 23:14.37 | *** join/#htc-linux luke-jr (n=luke-jr@2002:62b3:1d4c:0:20e:a6ff:fec4:4e5d) |
| 23:43.31 | *** part/#htc-linux Orbital_sFear (n=ldupin@bbb72-0-189-133.bendbroadband.com) |
