| 00:00.03 | cr2 | i have written him, but did not get any reply. |
| 00:00.19 | cr2 | his homepage says : do docs for these chips. |
| 00:01.14 | lkcl | oops |
| 00:01.32 | *** join/#htc-linux am3 (n=root@59.93.241.246) |
| 00:03.23 | am3 | Is there any program to browse files through BlueTooth in Linux ? I have FC4, and can see the Dongle blinking but don't know how to see any files of the other machine.. |
| 00:03.36 | cr2 | lkcl: here - http://ovcam.org/ov511 |
| 00:04.18 | cr2 | am3: do you run bnep ? |
| 00:05.06 | am3 | I don't know what bnep is. |
| 00:05.23 | cr2 | ip over bluetooth linux driver. |
| 00:05.53 | cr2 | what do you have on the "other machine" ? |
| 00:06.13 | am3 | I will just try to find out.. |
| 00:07.40 | am3 | The other machine is usually a windows machine, Win 98, 2000, etc. The dongle had the IVT software which produces a file browser when Windows is on both sides, but not when one side is Linux |
| 00:10.02 | am3 | A locate bluetooth just now shows for instance /etc/bluetooth/rfcomm.conf and /etc/bluetooth/hcid.conf |
| 00:10.56 | am3 | Also /lib/modules/2.6.11-1.1369_FC4/kernel/net/bluetooth/bnep/bnep.ko |
| 00:11.52 | cr2 | i don't know bnep status on windows, maybe you need activesync-style link ? serial-on-usb ? |
| 00:13.40 | am3 | This is my dual boot laptop, with WinXP Home and FC4. When I boot WinXP, the Blue tooth works alright and I can browse files on the other machine, but in FC4 though the dongle is working there seems to be no program to browse filesalthough the |
| 00:16.06 | am3 | Two programs named Bluetooth Manager, and Bluetotoh File Sharing are listed in system tools |
| 00:16.07 | cr2 | am3: i have no clue about windows ;) |
| 00:16.52 | cr2 | BabelOued: i see only the gpiod2 in your log, which is OWM irq (battery monitor). |
| 00:17.46 | am3 | If we have Linux on both machines, then is there a program to browse files through Bluetooth, i.e., can it work like a wireless LAN ? |
| 00:18.41 | BabelOued | cr2: sorry i've made a correct test now http://linuxbox2.free.fr/blueangel/WIFI1.txt |
| 00:18.47 | bd2_ | am3, probably it's worth googling about "obex", "openobex" |
| 00:18.54 | BabelOued | cr2: with this script http://linuxbox2.free.fr/blueangel/testacx.txt |
| 00:19.14 | bd2_ | that is what windows probably use for the file browsing |
| 00:20.45 | psokolovsky | cr2, it seems that PXA reg definitions getting screwed in pxa2xx-i2sound. can you imagine such? |
| 00:21.09 | am3 | The IVT browser looks like a spreadsheet, though it does not have a "Folders" display.. |
| 00:24.05 | goxboxlive | night |
| 00:24.22 | *** join/#htc-linux rejon (n=rejon@69-12-152-62.dsl.dynamic.sonic.net) |
| 00:27.45 | cr2 | psokolovsky: strange. |
| 00:28.09 | cr2 | BabelOued: i'll look. |
| 00:28.21 | psokolovsky | cr2, more than. by remapping reg space using ioremap() all seems to work |
| 00:28.26 | psokolovsky | as expected |
| 00:30.19 | cr2 | but the pxa areas are remapped and then you only need to use the offsets ? |
| 00:30.26 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 00:31.42 | psokolovsky | cr2, all those SACR, etc stuff actually statically mapped, http://lxr.linux.no/source/include/asm-arm/arch-pxa/hardware.h?v=2.6.18#L42 |
| 00:31.57 | psokolovsky | cr2, maybe alsa stuff works in userspace? ;-D |
| 00:33.00 | cr2 | yes, this is the offset calculation. |
| 00:33.44 | cr2 | printk the virtual address of SACR in the driver. |
| 00:34.26 | psokolovsky | cr2, I can, but what that gives? |
| 00:35.03 | cr2 | if it is right, then nothing is messed up. |
| 00:36.15 | lkcl | psoko: in the header file, it says this: |
| 00:36.25 | lkcl | <PROTECTED> |
| 00:36.25 | lkcl | #ifndef _ASM_ARCH_SA1111 |
| 00:36.30 | lkcl | in pxa-regs.h |
| 00:36.37 | lkcl | what _that_ all about?? |
| 00:37.31 | cr2 | ? |
| 00:37.58 | lkcl | #ifndef _ASM_ARCH_SA1111 |
| 00:37.58 | lkcl | <PROTECTED> |
| 00:37.59 | lkcl | #define SACR0 __REG(0x40400000) /* Global Control Register */ |
| 00:38.11 | am3 | write |
| 00:38.35 | cr2 | it's a hack for pxa25x + sa1111 devices. |
| 00:38.41 | lkcl | ahh |
| 00:41.11 | psokolovsky | cr2, btw: |
| 00:41.11 | psokolovsky | # cat /proc/iomem |
| 00:41.12 | psokolovsky | 08000000-08ffffff : w100fb |
| 00:41.12 | psokolovsky | 0c000000-0c002000 : asic3.0 |
| 00:41.12 | psokolovsky | 0e000000-0e002000 : asic3.0 |
| 00:41.12 | psokolovsky | <PROTECTED> |
| 00:41.14 | psokolovsky | 15000000-15000020 : acx-mem.0 |
| 00:41.16 | psokolovsky | 20000000-2fffffff : PCMCIA socket 0 |
| 00:41.18 | psokolovsky | <PROTECTED> |
| 00:41.20 | psokolovsky | <PROTECTED> |
| 00:41.22 | psokolovsky | <PROTECTED> |
| 00:41.26 | psokolovsky | 40301680-403016a3 : pxa2xx-i2c.0 |
| 00:41.28 | psokolovsky | 40400000-40400083 : pxa2xx-i2s |
| 00:41.30 | psokolovsky | may there be some coherency issues due to this? |
| 00:43.01 | cr2 | no. |
| 00:44.37 | cr2 | BabelOued: i see only the gsm gpios in the log. |
| 00:45.35 | psokolovsky | cr2, cannot be? are you sure? |
| 00:46.15 | cr2 | psokolovsky: but the other pxa subsystems are mapped in the same way, and they work. |
| 00:46.26 | lkcl | cat /proc/iomem |
| 00:46.26 | lkcl | 0c000000-0c002000 : asic3.0 |
| 00:46.26 | lkcl | <PROTECTED> |
| 00:46.26 | lkcl | 10000000-10002000 : asic3.0 |
| 00:46.26 | lkcl | 40301680-403016a3 : pxa2xx-i2c.0 |
| 00:46.27 | lkcl | 40400000-40400083 : pxa2xx-i2s |
| 00:48.16 | BabelOued | cr2: i ve flight mode for this test. Is it correct script? |
| 00:48.29 | lkcl | ok - volume setting is now working, after i set GPIO A0 first and THEN did request_module('i2c-pxa') |
| 00:50.07 | cr2 | BabelOued: yes. |
| 00:50.17 | cr2 | lkcl: who lowers the A0 ? |
| 00:50.23 | *** part/#htc-linux rejon (n=rejon@69-12-152-62.dsl.dynamic.sonic.net) |
| 00:50.37 | lkcl | cr2: me, initially, in the asic3 initialisation htcsable.c |
| 00:50.58 | cr2 | i have some similar code in the universal, and the RTC access, but it seems unused. |
| 00:51.06 | cr2 | lkcl: then don't do it :) |
| 00:51.11 | lkcl | hehe |
| 00:51.38 | cr2 | lkcl: i mean, is there a legitimate ground for disabling i2c ? |
| 00:51.52 | lkcl | yes? save power? |
| 00:51.58 | cr2 | in such a way. i don't understand the purpose. |
| 00:52.28 | cr2 | no, it's more for multiplexing the i2c buses. |
| 00:52.49 | cr2 | i thought that they set A0=0 and access the RTC |
| 00:52.58 | cr2 | but it's not the case on the universal. |
| 00:53.46 | cr2 | you can check it in linux. |
| 00:53.58 | cr2 | i2cdetect with A0=1 and A0=0 |
| 00:55.16 | lkcl | gimme 1sec... |
| 00:55.47 | lkcl | from what i remember when i did that last, it gives 0x12 in the 0x12 slot, UU in the 0x51 slot and XX everywhere else |
| 00:56.27 | cr2 | independent of the A0 ? |
| 00:57.46 | lkcl | oh - no. when A0=1 and A1=1 |
| 00:57.57 | lkcl | sorry. |
| 01:04.10 | cr2 | BabelOued: how do i init pppd to connect to BA over usb ? |
| 01:08.17 | BabelOued | i use synce with WM2003 |
| 01:08.54 | BabelOued | synce-serial-config /dev/ttyUSB0;vdccm &;synce-serial-start |
| 01:09.20 | cr2 | hmm. what is in you /etc/ppp/peers/synce-device ? |
| 01:09.36 | cr2 | i don't want to install full synce |
| 01:10.02 | psokolovsky | cr2, last news ;-E. it seems all that SACR crap is read-before-write %) |
| 01:10.09 | BabelOued | /dev/ttyUSB0 115200 |
| 01:10.13 | BabelOued | connect '/usr//bin/synce-serial-chat' |
| 01:10.17 | BabelOued | nodefaultroute |
| 01:10.21 | BabelOued | noauth |
| 01:10.25 | BabelOued | local |
| 01:10.29 | BabelOued | 192.168.131.102:192.168.131.201 |
| 01:10.29 | BabelOued | ms-dns 192.168.131.102 |
| 01:10.30 | BabelOued | crtscts |
| 01:10.30 | BabelOued | linkname synce-device |
| 01:10.54 | cr2 | psokolovsky: ok. |
| 01:11.15 | cr2 | BabelOued: what is in /usr//bin/synce-serial-chat ? |
| 01:11.15 | psokolovsky | cr2, no! not ok! tell me how that could be! ;-) |
| 01:11.51 | BabelOued | cr2: it's a binary |
| 01:12.12 | cr2 | psokolovsky: what do you mean ? |
| 01:12.31 | psokolovsky | cr2, did hear about such cases or hit something like that? |
| 01:12.34 | cr2 | BabelOued: probably i don't need this weird ms stuff. |
| 01:12.46 | cr2 | psokolovsky: no. |
| 01:13.15 | cr2 | i can look what wince does. |
| 01:13.19 | psokolovsky | that's it - ;-( |
| 01:13.37 | cr2 | but then it must be documented... |
| 01:14.21 | psokolovsky | ok, let me prepare patch and we'll see |
| 01:14.35 | psokolovsky | maybe I'm just too tired and circle around... |
| 01:19.24 | BabelOued | cr2: if you wan http://linuxbox2.free.fr/blueangel/synce-serial-chat |
| 01:19.30 | BabelOued | it's from SUSE 10.1 |
| 01:19.45 | cr2 | psokolovsky: a bad patch that makes it work is better than nothing. |
| 01:20.06 | cr2 | BabelOued: i gave up and will install synce. i have amd64 notebook. |
| 01:20.10 | psokolovsky | cr2, well, good patch is even better ;-) |
| 01:21.20 | cr2 | i hate suse ;) it installs tons of 32bit shit for no reason ;) |
| 01:21.53 | cr2 | time for a switch to debian. |
| 01:23.33 | BabelOued | cr2: suse support standard compile and i can compile mythtv with it :) |
| 01:29.18 | cr2 | BabelOued: i' |
| 01:29.33 | cr2 | m also in the flight mode, but gpio39 changes . |
| 01:33.03 | BabelOued | you have WM3 ? or WM5 ? |
| 01:33.19 | cr2 | wince2003, otherwise synce will not work. |
| 01:34.07 | cr2 | 84 is active from time to time, maybe the backup battery status. need to check with wi, |
| 01:35.14 | cr2 | A4 is reset |
| 01:37.09 | BabelOued | cr2: i use this for WM5, but it work only one time, the first day :) |
| 01:37.10 | BabelOued | http://www.synce.org/index.php/Building_SynCE_with_Windows_Mobile_2005_support_from_Subversion |
| 01:37.52 | cr2 | A3,A5,B13 are power, A4 -reset. |
| 01:38.25 | cr2 | A4=0, A4=1, A4=0 |
| 01:38.25 | BabelOued | A4 reset, ok, in this order power on ? |
| 01:39.43 | cr2 | yes. |
| 01:40.16 | BabelOued | do you have 17 or 75 as irq ? |
| 01:41.06 | cr2 | i don't think 17 is irq. it's something else. |
| 01:42.21 | *** join/#htc-linux Ro_En_PC (n=roen@p54A67A3C.dip.t-dialin.net) |
| 01:43.03 | BabelOued | how many time reset ? mdelay(5); ? |
| 01:45.37 | cr2 | use the same values as on universal. it's the same chip after all. |
| 01:45.45 | BabelOued | ok |
| 01:48.28 | BabelOued | on htcuniversal you keep RESET to 1 ? or it is inverted ? |
| 01:50.17 | BabelOued | how can we find WLAN_BASE ? |
| 01:50.40 | cr2 | with 'wi', but i don't see anything ;) |
| 01:51.06 | cr2 | you can try the h4000 way and set the pcmcia0 as the addr. |
| 01:55.59 | psokolovsky | so, anyone? |
| 01:56.04 | psokolovsky | Kevin2, still online? |
| 01:58.32 | BabelOued | cr2: ok file modified |
| 02:09.28 | cr2 | BabelOued: which file ? |
| 02:09.56 | BabelOued | my htcblueangel_acx.c |
| 02:10.36 | cr2 | ok. |
| 02:10.45 | cr2 | does it work ? :) |
| 02:11.07 | Kevin2 | psokolovsky: I'm here. |
| 02:11.45 | BabelOued | long time to compile :) not an amd64 |
| 02:12.43 | psokolovsky | Kevin2, I'm, committing a magic patch (will make you laught loudly). It solves for me issues with sound working out of the book. |
| 02:13.00 | psokolovsky | Kevin2, could you try it soon or now? |
| 02:13.15 | psokolovsky | (need 5 more mins to write good commit msg) |
| 02:13.30 | lkcl | ooo, psoko - you got it? |
| 02:14.03 | Kevin2 | psokolovsky: Why don't you just email me the patch? |
| 02:14.42 | psokolovsky | Kevin2, wasn't sure you were around, and I want get -hh1 after all, maybe even today ;-) |
| 02:14.48 | psokolovsky | lkcl, it seems %) |
| 02:14.54 | lkcl | good grief. |
| 02:15.09 | lkcl | then i might have a chance to check only one problem rather than two. hooray. |
| 02:15.30 | lkcl | Rule Of Reverse-Engineering Number One: only have one thing to test. |
| 02:16.44 | Kevin2 | lkcl: What's rule 2? Have a long list of alternate things to test because the first one _never_ works? |
| 02:16.57 | lkcl | well.... yep! |
| 02:18.03 | cr2 | Kevin2: why i can trace the SSP (touchscreen) on pxa270 and can't on pxa263 ??? |
| 02:18.34 | Kevin2 | Does tracing work at all on pxa263? |
| 02:18.56 | cr2 | yes. |
| 02:19.11 | cr2 | i2c, owm asic, ffuart |
| 02:19.12 | Kevin2 | Maybe they're using DMA on that machine? |
| 02:19.21 | cr2 | hmm. maybe. |
| 02:19.32 | Kevin2 | Maybe they've mapped in the PIO space at a different virtual address? |
| 02:19.34 | cr2 | i can't trace wifi (should be on pcmcia) |
| 02:19.48 | cr2 | unlikely |
| 02:20.02 | Kevin2 | No one's been able to trace wifi. I don't know the reason - maybe they're using dma, alternate address, etc. |
| 02:20.21 | cr2 | on universal it is traceable |
| 02:20.24 | Kevin2 | Also, you can't trace anything that is running in "abort mode". |
| 02:20.40 | Kevin2 | cr2: Oh, I didn't know that - it can't be traced on Apache. |
| 02:20.55 | cr2 | this is strange. |
| 02:21.46 | Kevin2 | cr2: Which machine are you looking at? |
| 02:23.03 | lkcl | ARSE. B10 i think is reset. B9 is headphone. |
| 02:23.26 | cr2 | Kevin2: blueangel. |
| 02:24.09 | bd2 | psokolovsky, you solved sound issue?! |
| 02:25.29 | psokolovsky | bd2, yes. but you curse me once you see the patch %) |
| 02:25.42 | cr2 | BabelOued: set 0x2c000000 as the wifi base. |
| 02:25.54 | BabelOued | ok |
| 02:26.27 | bd2 | psokolovsky, is it that ugly? :-) |
| 02:28.13 | cr2 | BabelOued: but 75 is not an irq ? |
| 02:28.14 | psokolovsky | it's actually relatively clean and it is one-liner |
| 02:28.22 | psokolovsky | committed |
| 02:28.35 | psokolovsky | And the patch was: |
| 02:28.35 | psokolovsky | <PROTECTED> |
| 02:28.35 | psokolovsky | + printk("SACR0 == 0x%x, SADIV == 0x%x\n", SACR0, SADIV); |
| 02:28.36 | psokolovsky | <PROTECTED> |
| 02:28.41 | psokolovsky | LOL |
| 02:28.46 | lkcl | o dear |
| 02:28.56 | psokolovsky | Kevin2, bd2: please test it if you can! |
| 02:30.45 | psokolovsky | lkcl, plese do |
| 02:33.23 | bd2 | psokolovsky, hmmm.. printk? :-) you're kidding :-) |
| 02:33.34 | Kevin2 | psokolovsky: Hrrm.. Seems like a mis-compile issue. But I disassembled the object code earlier and found nothing wrong. |
| 02:33.39 | psokolovsky | bd2, yes, consider it magic spell ;-I |
| 02:34.05 | psokolovsky | Kevin2, well, maybe. I actually didn't even think about that. |
| 02:34.24 | Kevin2 | psokolovsky: Have you tried putting an mdelay(1) in there instead? This could be a timing issue - the clock might not yet be on by the time the register write occurs. |
| 02:34.28 | lkcl | it may be a timing issue |
| 02:34.28 | psokolovsky | mere "int tmp = SACR0" doesn't work, even though it's volatile |
| 02:34.29 | lkcl | yep. |
| 02:34.37 | bd2 | psokolovsky, did you tried to insert delay? |
| 02:34.40 | lkcl | not long enough a delay |
| 02:34.52 | psokolovsky | Kevin2, mdelay is too much, if old code worked w/o delay at all |
| 02:34.57 | psokolovsky | I tried udelay(100) |
| 02:35.02 | psokolovsky | had them few around |
| 02:35.30 | psokolovsky | well, just try that patch ;-) |
| 02:35.42 | psokolovsky | obviously, it can't be left in that state ;-I |
| 02:36.22 | bd2 | "It turns out that registers exhibit read-before-write behavior." -- it could be possible, btw |
| 02:36.34 | lkcl | fucking hell. |
| 02:36.35 | lkcl | sound. |
| 02:36.47 | psokolovsky | bd2, yep, possible. |
| 02:36.55 | lkcl | i don't believe it. |
| 02:36.59 | bd2 | lol ;-) |
| 02:37.09 | lkcl | dang me. |
| 02:37.37 | lkcl | well it looks like i found the right gpio for that reset, then, too. |
| 02:38.12 | lkcl | dang. |
| 02:38.14 | cr2 | Kevin2: the SSP2 on pxa2[56]x is at a different address ;) now it works. |
| 02:38.25 | lkcl | hey, cr2 - i got sound, dude! |
| 02:38.33 | cr2 | :) |
| 02:38.41 | cr2 | i have uda1380 ;) |
| 02:38.43 | bd2 | psokolovsky, but.. if "int tmp = SACR0" doesn't work.. then it's not read-b-write issue, no? |
| 02:39.12 | lkcl | ohhh maaaaaan, the mic works too. |
| 02:39.17 | psokolovsky | bd2, I really wanted to check if gcc still optmizes that away ;-\ |
| 02:39.27 | lkcl | ah hahahaahahahaaa |
| 02:39.56 | lkcl | it didn't help that the GPIO B10 was set to an input instead of an output |
| 02:39.57 | psokolovsky | lkcl, is that related to gpio or that patch? |
| 02:40.02 | lkcl | both, dude. |
| 02:40.09 | psokolovsky | ok, thanks |
| 02:40.18 | psokolovsky | +1, who's next? |
| 02:40.59 | bd2 | psokolovsky, heh.. if you're still hacking it now, please check .o file with objdump -D |
| 02:41.00 | cr2 | psokolovsky: it's great that it works now, but i still don't understand it. |
| 02:41.25 | psokolovsky | cr2, me too. that's why I asked before ;-) |
| 02:41.38 | bd2 | I've done "make clean" recently, thus it will take time to try it and examine .o file myself |
| 02:41.42 | psokolovsky | My bet that leenouks is screwy at times with memory |
| 02:41.56 | cr2 | which compiler are you using ? |
| 02:42.00 | lkcl | also i'm checking the asic3 GPIO pin A0 to 'on' and _then_ loading pxa-i2c |
| 02:42.06 | psokolovsky | bd2, that's cool that you have clean tree. I need to do that myself too |
| 02:42.12 | psokolovsky | cr2, gcc 3.4.1 |
| 02:42.17 | cr2 | ok. me too. |
| 02:43.38 | Kevin2 | psokolovsky: It works on apache. |
| 02:43.48 | psokolovsky | Kevin2, ok |
| 02:43.58 | Kevin2 | $ arm-linux-gcc --version |
| 02:43.58 | Kevin2 | arm-linux-gcc (GCC) 4.0.2 |
| 02:43.59 | psokolovsky | cr2, so you tried it too? |
| 02:44.17 | bd2 | okay.. thus, we got compilers zoo... :-) |
| 02:45.06 | psokolovsky | yep ;-) |
| 02:45.27 | cr2 | psokolovsky: i'm on blueangel now. there is no uda1380 sound here ;) |
| 02:45.28 | lkcl | arm-linux-gnu-gcc (GCC) 4.1.2 20061028 (prerelease) (Debian 4.1.1-19) |
| 02:45.29 | lkcl | haha |
| 02:46.02 | lkcl | sound, sound, i have sound |
| 02:46.05 | psokolovsky | cr2, ak4641 you mean. patch is for it. btw, not yet tested for uda regressions. |
| 02:46.12 | psokolovsky | ;-) |
| 02:46.25 | cr2 | psokolovsky: blueangel has uda1380. |
| 02:47.14 | psokolovsky | ok |
| 02:47.58 | lkcl | as does himalaya. |
| 02:48.10 | cr2 | and magician |
| 02:48.18 | cr2 | and weird SSP+I2S setup. |
| 02:48.46 | psokolovsky | yep, indeed, usa180 & ak4641 cover 80% pdas about which we care ;-) |
| 02:52.30 | cr2 | uuuu. 010ea806: debug 090d4a88: 11d200b0(ldrh) 00000000 a920091c |
| 02:52.31 | psokolovsky | lkcl, well, I'm glad you're happy! hope you'll get pretty fully working sable soon! ;-) |
| 02:52.39 | cr2 | sdio is traceable. |
| 02:52.57 | lkcl | all the major stuff is done-ish. |
| 02:53.15 | cr2 | suspend, lcd abd backlight. |
| 02:53.26 | lkcl | four things left on the list: lcd, backlight, core and keyboard |
| 02:53.30 | lkcl | you forgot keyboard, cr2 :) |
| 02:53.34 | cr2 | battery monitor and charging. |
| 02:53.37 | lkcl | oh yes. |
| 02:53.41 | cr2 | keyboard is there. |
| 02:53.43 | lkcl | six. |
| 02:53.44 | lkcl | it is? |
| 02:53.49 | cr2 | pxa27x_keyboard. |
| 02:53.52 | lkcl | ahhh |
| 02:53.57 | cr2 | look at apache. |
| 02:54.02 | lkcl | okie! |
| 02:54.26 | cr2 | battery monitor needs a common code with universal and hx4700. |
| 02:54.38 | cr2 | asic3_owm, based on ds1wm |
| 02:55.05 | lkcl | oh, and camera. |
| 02:56.16 | cr2 | camera is the same on magician and universal. |
| 02:56.53 | cr2 | now i need to convince the fcking wince2003 to accept the sdio gps driver. |
| 02:56.58 | lkcl | :) |
| 03:00.02 | lkcl | my god, it's only been a month. |
| 03:01.21 | lkcl | ok. |
| 03:01.31 | lkcl | a sensible person, at 3am, would go to sleep. |
| 03:01.45 | cr2 | at 4am |
| 03:01.49 | cr2 | good night. |
| 03:01.51 | BabelOued | yes 4am |
| 03:02.00 | psokolovsky | good night! |
| 03:02.16 | psokolovsky | don't even tell how much it here ;-) |
| 03:02.27 | psokolovsky | tagging will go to tomorrow ;-) |
| 03:04.48 | lkcl | night cr2 |
| 03:16.55 | BabelOued | good night all |
| 03:24.21 | *** join/#htc-linux skodde (n=skodde@85-18-136-66.fastres.net) |
| 03:26.04 | bd2 | heh... he finally killed by i2s.. bastards! |
| 08:00.46 | *** join/#htc-linux dullard (n=jim@adsl-static-1-30.uklinux.net) |
| 09:02.29 | *** join/#htc-linux BHSPitMonkey (n=stephen@adsl-67-64-112-116.dsl.rcsntx.swbell.net) |
| 09:02.32 | BHSPitMonkey | hi all |
| 09:02.33 | *** join/#htc-linux RoEn_PC (n=roen@p54A6517C.dip.t-dialin.net) |
| 09:03.04 | BHSPitMonkey | anybody home |
| 09:09.30 | *** join/#htc-linux rob_w (n=bob@p85.212.147.176.tisdip.tiscali.de) |
| 09:31.42 | *** join/#htc-linux pH5 (n=ph5@p5485CA82.dip.t-dialin.net) |
| 09:46.33 | *** join/#htc-linux booba (n=booba@AAnnecy-256-1-79-119.w90-10.abo.wanadoo.fr) |
| 10:07.27 | *** join/#htc-linux buz (n=buz@84-73-95-139.dclient.hispeed.ch) |
| 10:56.46 | *** join/#htc-linux alex-begin (n=345rt@84.232.212.62) |
| 11:41.10 | *** join/#htc-linux asylumed (n=insanity@196.211.28.91) |
| 14:08.39 | lkcl | no, nobody home BHSPitMonkey :) |
| 14:09.11 | *** part/#htc-linux booba (n=booba@AAnnecy-256-1-79-119.w90-10.abo.wanadoo.fr) |
| 14:10.13 | cr2 | lkcl: good morning |
| 14:10.21 | lkcl | ouch. mornin cr2 |
| 14:10.43 | cr2 | what about resume/suspend ? |
| 14:10.58 | lkcl | ha ha |
| 14:11.11 | lkcl | let me get some breakfast first |
| 14:11.16 | cr2 | look at h4000 and set the address from universal |
| 14:11.19 | cr2 | ok. |
| 14:11.23 | lkcl | but yes. ok |
| 14:13.31 | lkcl | eek some assembly instructions direct into .... ??? |
| 14:13.42 | cr2 | into RAM. |
| 14:14.19 | lkcl | how odd. |
| 14:14.35 | cr2 | if you don't change the IPL in G3 it will jump to 0xa0040000 |
| 14:14.43 | lkcl | ahh. |
| 14:14.57 | lkcl | what about the asic3_set_clock_cdex |
| 14:15.11 | lkcl | PWER PFER PRER etc? |
| 14:15.15 | cr2 | where ? |
| 14:15.27 | cr2 | P* is at PXA. |
| 14:15.37 | cr2 | dump pxa27xpower |
| 14:15.38 | lkcl | htcuniversal_core.c |
| 14:15.40 | lkcl | ok. |
| 14:15.57 | cr2 | aah. ok. |
| 14:16.18 | cr2 | then "dump asic3gpio" too. |
| 14:16.22 | lkcl | <PROTECTED> |
| 14:16.23 | lkcl | <PROTECTED> |
| 14:16.23 | lkcl | <PROTECTED> |
| 14:16.23 | lkcl | <PROTECTED> |
| 14:16.23 | lkcl | <PROTECTED> |
| 14:16.24 | lkcl | <PROTECTED> |
| 14:16.26 | lkcl | <PROTECTED> |
| 14:16.28 | lkcl | <PROTECTED> |
| 14:16.30 | lkcl | <PROTECTED> |
| 14:16.32 | lkcl | <PROTECTED> |
| 14:16.34 | lkcl | <PROTECTED> |
| 14:16.36 | lkcl | <PROTECTED> |
| 14:16.40 | lkcl | <PROTECTED> |
| 14:16.42 | lkcl | from: http://wiki.xda-developers.com/index.php?pagename=Ipaq6915_Research |
| 14:17.05 | cr2 | you see the bits. |
| 14:17.10 | lkcl | eek - so i need all those, ahh ok. |
| 14:17.17 | cr2 | not all. |
| 14:17.41 | cr2 | clock_cdex is seen on "dump asic3" |
| 14:18.30 | lkcl | ack. |
| 14:19.09 | lkcl | <PROTECTED> |
| 14:19.16 | lkcl | no breakfast yet :) |
| 14:23.18 | *** join/#htc-linux cr2 (n=konversa@crpl22.physik.uni-wuppertal.de) |
| 14:24.46 | lkcl | what about these? |
| 14:24.48 | lkcl | pxa27x memctl |
| 14:24.48 | lkcl | <PROTECTED> |
| 14:24.48 | lkcl | <PROTECTED> |
| 14:24.48 | lkcl | <PROTECTED> |
| 14:24.48 | lkcl | <PROTECTED> |
| 14:24.49 | lkcl | <PROTECTED> |
| 14:24.51 | lkcl | <PROTECTED> |
| 14:25.23 | cr2 | they are set by IPL |
| 14:25.54 | cr2 | unless you will reflash it, it's not your business. |
| 14:26.17 | lkcl | hehe |
| 14:30.08 | lkcl | hmmm... h4000 uses pxa_pm_ll_ops |
| 14:30.18 | lkcl | htcuniversal uses a 'core device' concept |
| 14:30.47 | cr2 | core device is needed only because of egpio |
| 14:31.01 | cr2 | you don't have it, so do pxa_pm_ll_ops |
| 14:31.52 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 14:32.04 | cr2 | even on universal it is core device is questionable |
| 14:32.05 | lkcl | ok. |
| 14:32.08 | lkcl | errrr |
| 14:32.13 | BabelOued | hi |
| 14:32.27 | cr2 | BabelOued: hi |
| 14:33.02 | cr2 | configuring wifi on wince2003 is a hell. how can i find out my ip ? |
| 14:33.17 | BabelOued | cr2: i've tried with adress you give me... but always same error |
| 14:33.21 | lkcl | ok am just copying htcuniveral_pm.c |
| 14:33.59 | cr2 | BabelOued: ok, then we need to find this memory by other means. |
| 14:34.17 | lkcl | i start to get it no |
| 14:34.18 | lkcl | w |
| 14:34.21 | BabelOued | cr2: i use a little software to get IP adress or connect to my linux box and use netstat |
| 14:34.59 | cr2 | i'm connected, but i can't access internet with exploiter for some reason. how can i debug it? |
| 14:36.01 | *** join/#htc-linux skodde (n=skodde@85-18-136-66.fastres.net) |
| 14:37.15 | lkcl | btw how am i gonna wake this thing up? :) |
| 14:38.23 | cr2 | powerr button, PWER enables the irq. |
| 14:38.58 | lkcl | ok - suspend worked.... |
| 14:39.20 | lkcl | or maybe it crashed, cos the phone buttons are on |
| 14:39.20 | cr2 | fscking wince. the wep key is autogenerated, thank you very much. |
| 14:39.25 | lkcl | duh |
| 14:39.41 | lkcl | two lights were still on but nobody home... |
| 14:40.21 | cr2 | lkcl: sorry. probably you need the bl and lcd drivers too. |
| 14:40.26 | lkcl | oh ok :) |
| 14:40.28 | cr2 | bl is very simple. |
| 14:40.44 | lkcl | you're just trying to trick me into finishing this device, aren't you? |
| 14:40.51 | cr2 | lcd should be also doable, since you know the gpios. |
| 14:41.10 | cr2 | yes, we should finish all devices till christmas :) |
| 14:41.23 | lkcl | hehe. ok _really_ i have to get breakfast. |
| 14:41.23 | cr2 | as a present to htc. |
| 14:43.49 | cr2 | now it works. wtf wince says it is connected to AP when using its own autogenerated key ? |
| 14:44.38 | BabelOued | cr2: so that's why i don't use WEP o mine :) |
| 14:45.07 | cr2 | BebelOued: i'm forced to use it, it was open before. |
| 14:48.14 | BabelOued | cr2: i m trying to use pcmcia1 memory |
| 14:51.40 | *** join/#htc-linux pwr (n=pwr@86.121.147.203) |
| 14:53.13 | cr2 | BabelOued: the gpio17 stream is really heavy. |
| 14:55.42 | BabelOued | it trigger on wifi traffic ? |
| 14:55.59 | cr2 | i think it has something to do with the LCD. |
| 14:56.36 | cr2 | you can try to switch off wifi, login over usb and watch if it still the same. |
| 14:56.56 | cr2 | then you can powerdown the LCD and the backlight. |
| 14:57.43 | cr2 | hmm. i need the per-machine startup.txt |
| 14:58.05 | cr2 | i have the "universal" gpios masked. |
| 15:09.19 | BabelOued | cr2: i use a task manager app an i can see Dll attached to device battdrvr.dll is to BAT1 |
| 15:13.34 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 15:15.36 | cr2 | BabelOued: i can trace the OWM asic. we need a common driver with magician and alpine for it. pH5 has it more or less done. |
| 15:17.15 | BabelOued | we have code into cvs ? |
| 15:20.48 | cr2 | http://en.pastebin.ca/255994 |
| 15:21.39 | cr2 | the .bus_shift = 1 |
| 15:21.59 | cr2 | and some other things are not correct for blueangel. |
| 15:22.18 | cr2 | but it's a good start |
| 15:22.18 | lkcl | ok wheeeeeeeee |
| 15:22.51 | lkcl | cr2 - any suggestions on code to start from for the lcd? |
| 15:23.05 | BabelOued | cr2: thanks |
| 15:23.19 | lkcl | h4000? |
| 15:23.22 | cr2 | lkcl: h4000 ? |
| 15:23.26 | lkcl | yehhh |
| 15:23.27 | lkcl | okie |
| 15:23.37 | cr2 | it's the most simple one. |
| 15:25.29 | lkcl | k. |
| 15:25.38 | lkcl | right. |
| 15:25.58 | lkcl | what's those haret power lcd commands again? |
| 15:26.22 | lkcl | wakey wakey wince |
| 15:26.30 | lkcl | ping ping come oooonnn... |
| 15:26.37 | lkcl | zzzzz |
| 15:27.06 | cr2 | pwrlcd 1; pwrlcd 4 |
| 15:27.25 | cr2 | BabelOued: it seems you are right, and the gpio17 is the wifi irq ? |
| 15:27.50 | lkcl | ack. |
| 15:27.51 | cr2 | i also get some bursts of gpio85, but it's not an irq. |
| 15:29.03 | lkcl | GPIOD12 S 0 ? 2952430 |
| 15:29.03 | lkcl | GPIOD14 S 0 ? 2952549 |
| 15:29.03 | lkcl | EGPIO0 S 0 594ea9 584ea8 2952563 |
| 15:29.03 | lkcl | EGPIO16 S 0 594ea9 584ea8 2952576 |
| 15:29.03 | lkcl | CKEN0 PWM0 off |
| 15:29.04 | lkcl | CKEN16 LCD off |
| 15:29.06 | lkcl | GPIOA15 S 0 ? 2952619 |
| 15:29.10 | lkcl | GPIOC9 S 0 ? 2952632 |
| 15:29.12 | lkcl | GPIOC10 S 0 ? 2952646 |
| 15:29.14 | lkcl | is off |
| 15:29.59 | BabelOued | cr2: ok and pcmcia1 memory have strange effect on LCD |
| 15:30.40 | cr2 | BabelOued: i'm not really sure about the irq. pcmcia1 is unlikely. |
| 15:30.56 | *** join/#htc-linux psokolovsky (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com) |
| 15:31.06 | cr2 | lkcl: what does wiki say ? |
| 15:31.18 | lkcl | wiki sayyyy aaaanything :) |
| 15:31.37 | cr2 | lkcl: keyboard backlight and lcd backlight gpios ? |
| 15:31.48 | lkcl | finding things |
| 15:31.50 | lkcl | allo psoko |
| 15:32.21 | cr2 | lkcl: it's all in the sableasic3 wiki page. |
| 15:32.24 | lkcl | oh ok. |
| 15:32.28 | lkcl | i just needed the timings |
| 15:32.30 | lkcl | confirmed |
| 15:32.54 | cr2 | d12 and d14 are backlight |
| 15:33.34 | *** join/#htc-linux RoEn_PC (n=roen@p54A6517C.dip.t-dialin.net) |
| 15:34.21 | Kevin2 | psokolovsky / cr2 : can one of you commit my apache patch to hh.org cvs? (I sent it to hh-kernel.) |
| 15:34.34 | psokolovsky | Hi! |
| 15:34.48 | psokolovsky | Kevin2, they probably go after -hh1 unless you insist |
| 15:35.23 | lkcl | SableASIC3 only lists 3 power lines, and i see 5 going up/down and i see CKEN 16 bouncing, too |
| 15:35.55 | Kevin2 | psokolovsky: No problem. Thanks. |
| 15:35.56 | lkcl | oh D12 is backlight ha ha |
| 15:36.01 | psokolovsky | ok |
| 15:36.17 | lkcl | and D14 is keyphone backlight |
| 15:38.17 | lkcl | do i _have_ to have a get_power function? |
| 15:39.10 | cr2 | lkcl: look how i have solved it for universal, but you can check the 3 gpios too. universal is PITA when it comes to LCD gpios ;) |
| 15:39.16 | lkcl | o dear |
| 15:39.29 | cr2 | it has 7 of them. |
| 15:41.12 | lkcl | waah! |
| 15:41.52 | cr2 | psokolovsky: we need some kind of wiki for asic3 API. |
| 15:42.28 | psokolovsky | cr2, if possible, let's host in in hh.org wiki |
| 15:42.54 | cr2 | yes, ok. |
| 15:43.41 | cr2 | i'd like to get rid of exporting the foobar_asic3 for the child devices. |
| 15:44.36 | cr2 | and i think that exporting the asic3*clock_cdex is a bad idea too. |
| 15:45.30 | bd2 | hx4700 using asic3*clock_cdex a lot. What's an alternative? |
| 15:46.03 | bd2 | or righter, why it's a bad idea in first place? :-) |
| 15:46.15 | cr2 | asic3_add_clock_cdex() and asic3_del_clock_cdex() |
| 15:46.30 | cr2 | because many devices use the same clock, and there is not refcounting. |
| 15:46.56 | cr2 | that leads to ugly comments in the hx4700* code. |
| 15:47.03 | bd2 | yup |
| 15:47.19 | bd2 | (some made by me in hx4700_core.c) |
| 15:47.30 | cr2 | :) |
| 15:47.50 | bd2 | okay, your hint seems cool indeed |
| 15:49.34 | Kevin2 | psokolovsky: Does hx4700 have a MAX1587? Did you try tracing the PI2C interface? (Apache has direct control of the 1587 and it alters core voltage when changing cpu speeds.) |
| 15:49.55 | lkcl | ok how am i gonna test this power lcd thing on linux, then? |
| 15:50.01 | psokolovsky | Kevin2, it has some CPU power regulator. bd2 worked on that. |
| 15:50.42 | bd2 | Kevin2, MAX1586B. I've posted patch to k-d almost year ago |
| 15:51.19 | bd2 | hx4700_voltage first hit in google |
| 15:52.09 | cr2 | i have seen the dpm driver in the motorola a780 kernel. maybe there are some good bits there too. |
| 15:52.10 | bd2 | ugh.. no, that is old and ugly patch. anyway, newer one somewhere there too. |
| 15:52.57 | bd2 | I'm thinking of doing it mach unspecific.. or indeed look into dpm |
| 15:55.24 | lkcl | ok let's see what damage that does :) |
| 15:56.59 | Kevin2 | bd2: Any reason your patch wasn't checked into hh.org cvs? |
| 15:58.22 | bd2 | Kevin2, I think I'm not insisted on it it that time plus it's hx4700-specific. better I'll convert it to pxa27x_pmic.c and machs will pass only PMIC address to it |
| 15:58.29 | lkcl | cr2: hmmm.... i'm going to need to know what the power button is, aren't i |
| 15:59.15 | lkcl | nuts - need to add the devices. |
| 15:59.37 | cr2 | not really. it's handled by the hardware. |
| 16:00.02 | cr2 | it's gpio0 if you are curious. on all devices i know. |
| 16:00.13 | bd2 | Kevin2, here it is http://www.handhelds.org/pipermail/hx4700-port/19/1953.html |
| 16:00.49 | bd2 | and this irreproducible fixup http://www.handhelds.org/pipermail/hx4700-port/19/1954.html |
| 16:01.50 | lkcl | ok. |
| 16:04.40 | lkcl | forgot to add core and lcd to main init duh |
| 16:05.09 | lkcl | which would explain why resume didn't work ha ha |
| 16:06.42 | lkcl | btw cr2: any idea why CKEN PWM goes off on lcd power down as _well_ as CKEN LCD? |
| 16:06.54 | lkcl | should i be switching that one up/down as well? |
| 16:07.07 | cr2 | lkcl: can you write the ovcamchip maintainer about ov6650 and 9650 ? maybe i've landed in the spam foler ? the drivers are available and only need to be integrated. |
| 16:07.16 | lkcl | sure. |
| 16:07.43 | cr2 | because you don't want to have working lcd backlight whwn the lcd is off. |
| 16:08.01 | cr2 | learned that hard way on the universal :) |
| 16:08.07 | lkcl | oops. |
| 16:08.08 | lkcl | ok. |
| 16:08.29 | lkcl | ha ha |
| 16:08.36 | lkcl | ok well that suspend experiment failed, too. |
| 16:09.25 | BabelOued | cr2: you have memory adress for ds1 ? |
| 16:10.19 | cr2 | BabelOued: http://wiki.xda-developers.com/index.php?pagename=BlueangelVirtualMemoryMapping |
| 16:10.29 | cr2 | 0x11000000 0xa9400000 0x01 BT/wifi led, etc ? |
| 16:10.42 | cr2 | etc ? is the owm :) |
| 16:11.47 | cr2 | the relevant registers are +0 +2 +4 +6 +8. |
| 16:12.05 | BabelOued | cr2: ah ok it's the same as for led drivre virt=ioremap(0x11000000, 4096); |
| 16:12.28 | cr2 | yes, the led control registers are a bit higher up. |
| 16:14.01 | lkcl | hm... ok - now what? |
| 16:14.12 | lkcl | how am i gonna debug this suspend stuff? |
| 16:14.55 | cr2 | bbl |
| 16:15.47 | lkcl | k dude |
| 16:20.55 | BabelOued | cr2: eh eh w1_asic3.c is full coment line (no code) |
| 16:23.23 | Kevin2 | bd2: Thanks. I'll take a look at it. |
| 16:44.30 | BabelOued | cr2: on BA we have an offset of 0x18 for GPIO register on AIC2 ? |
| 16:59.16 | *** join/#htc-linux RoEn_PC (n=roen@p54A67A38.dip.t-dialin.net) |
| 17:00.25 | *** join/#htc-linux pleemans (n=peter@d51A5E421.access.telenet.be) |
| 17:00.38 | lkcl | ok i needed a button registered. there's only one, but... |
| 17:01.03 | lkcl | it's _my_ button, all miiiine |
| 17:01.30 | *** join/#htc-linux JTRipper (n=jtripper@p54B1E2C5.dip.t-dialin.net) |
| 17:02.06 | lkcl | oops. |
| 17:14.42 | *** join/#htc-linux rob_w (n=bob@p85.212.147.176.tisdip.tiscali.de) |
| 17:22.50 | Speedy2 | Hi lkcl |
| 17:28.17 | BabelOued | cr2: sorry stupid question GPIO register are for LED on BA |
| 17:34.03 | lkcl | allo Speedy2 |
| 17:34.17 | lkcl | we got the audio working and acx! |
| 17:34.41 | lkcl | i'm on to lcd, backlight and power management now |
| 17:34.46 | lkcl | brb gotta go get bins in |
| 17:39.22 | rob_w | *zipzap |
| 17:39.57 | Speedy2 | What was the problem with the audio? |
| 17:39.57 | Speedy2 | And congrats! |
| 17:42.33 | rob_w | i cant mount my cd drives anymore ?!? |
| 17:43.47 | rob_w | but i need to get that powerpc docu ... grrrr |
| 17:48.02 | *** join/#htc-linux rob_w (n=bob@p85.212.147.176.tisdip.tiscali.de) |
| 18:05.57 | *** join/#htc-linux TripleDES (n=sergio@222.red-62-57-170.user.auna.net) |
| 18:09.44 | *** join/#htc-linux rmoravcik (n=rmoravci@icm4-orange.orange.sk) |
| 18:10.15 | *** join/#htc-linux alex-begin (n=345rt@84.232.212.62) |
| 18:16.02 | cr2 | rmoravcik: hi |
| 18:22.53 | rmoravcik | cr2: hi |
| 18:24.40 | cr2 | rmoravcik: did you dump all the s324* regs by hand ? |
| 18:25.49 | cr2 | there are many more samsung-based phones coming (eten500,hermes,etc.) which need the same data dumped in wince. |
| 18:31.14 | pierrox | hi |
| 18:31.36 | pierrox | how things are going on htc phones ? |
| 18:33.26 | cr2 | pierrox: very good. it is more or less clear how to do an UMTS video call now. |
| 18:34.01 | pierrox | cr2: :-) that is so easy |
| 18:34.59 | cr2 | the samsung cpu docs i have seen are not so detailed like pxa2xx. is it true ? |
| 18:35.27 | pierrox | cr2: yes it is true, they are not good |
| 18:35.55 | pierrox | cr2: and even their example source code are awfull |
| 18:36.05 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 18:36.09 | cr2 | but the rx3715 seems to be rather advanced to me. |
| 18:36.42 | pierrox | thanks to the great work on asic3 |
| 18:38.17 | psokolovsky | cr2, rx3715 is in mainline, as you know. bretty basic there, though |
| 18:38.48 | cr2 | i still have one puzzling question about asic3. what are the alt functions of A13 and A14. the rest is more or less clear. |
| 18:39.24 | cr2 | psokolovsky: but that means some good knowledge is there. unlike omap850 & co. |
| 18:39.25 | psokolovsky | cr2, you know anything about asic3 alt functions at all? 8-o |
| 18:39.35 | *** join/#htc-linux rob_w (n=bob@p85.212.135.90.tisdip.tiscali.de) |
| 18:39.43 | psokolovsky | cr2, Ben Dooks did it. He has knowledge ;-) |
| 18:39.53 | cr2 | :D |
| 18:39.57 | cr2 | http://wiki.xda-developers.com/index.php?pagename=ASIC3 |
| 18:40.07 | cr2 | i think i'll copy this page to hh.org |
| 18:40.26 | cr2 | all known registers and ALT functions and clocks. |
| 18:41.23 | psokolovsky | cr2, there's single bit for alt function, right? |
| 18:41.42 | rmoravcik | cr2: i have somewhere dump of s3c2440 regs from haret |
| 18:41.45 | cr2 | yes. |
| 18:43.06 | cr2 | rmoravcik: i did the same for pxa27x, but later gave up, and written the 'dump pxa27x*' commands for haret. because when you have 5-6 devices, doing manual dumps are painful. |
| 18:45.01 | pierrox | cr2: an interesting feature would be to add a generalised description of the registers : which bit mean what |
| 18:45.18 | cr2 | you can't ask somebody who has a device to do raw register dumps, but with haret it's feasible. |
| 18:45.42 | cr2 | pierrox: i've done it for all interesting asic3 and pxa2xx registers. |
| 18:45.53 | rmoravcik | pierrox: somethings like this http://morgan.fbi.cz/pub/rx3715/gpio_info/s3c2440_gpio.txt |
| 18:46.28 | cr2 | rmoravcik: but it's device-dependent ? |
| 18:47.16 | pierrox | i mean that we could add a description : register address+value+mask+shift -> description of the fiels |
| 18:47.19 | pierrox | field |
| 18:47.29 | cr2 | i have http://wiki.xda-developers.com/index.php?pagename=UniversalGPIO and http://wiki.xda-developers.com/index.php?pagename=UniversalASIC3 |
| 18:47.42 | rmoravcik | cr2: all discription is in s3c2440 datasheet |
| 18:48.52 | cr2 | rmoravcik: if i've had a samsung phone i've done it. but i only have 3 pxa2xx phones :) |
| 18:49.45 | pierrox | tell me where the right haret repository is and i will do it |
| 18:51.22 | cr2 | pierrox: the only repository supporting samsung cpu is the mainline haret. on hh.org |
| 18:51.46 | pierrox | cr2: that's not the only one |
| 18:52.21 | cr2 | if you want to create the patches that can be reused by everybody, you should use it. |
| 18:52.23 | pierrox | cr2: i have seen one with a gpio watch function that is note on hh.org |
| 18:52.52 | rmoravcik | pierrox: there is a fork of haret for s3c2442 cpu |
| 18:53.04 | cr2 | ok, but then rip out this code and submit it as a patch to Kevin2. |
| 18:53.27 | cr2 | i don't see any reason for creating new forks. |
| 18:53.39 | pierrox | neither do i, it is a real pain |
| 18:53.57 | pierrox | things are hard enough |
| 18:54.44 | cr2 | i've done haret-gnu because the mainline was completely stuck for years, did not support gcc and did not support any really useful hardware monitoring functions. |
| 18:55.00 | rmoravcik | pierrox: http://linux.sp.ru/ |
| 18:55.21 | cr2 | but now it's not that important, because the mainline haret in actively maintained, and makes good progress. |
| 18:55.30 | pierrox | cr2: haret did need an heavy lifting, that is true |
| 18:55.44 | pierrox | rmoravcik: yes, this is the one with s3c2442 ? |
| 18:56.24 | rmoravcik | pierrox: yes |
| 18:57.43 | pierrox | by the way have someone already played with s3c24xx dma ? |
| 18:58.57 | rmoravcik | pierrox: fluff (ben dooks) |
| 18:59.22 | pierrox | rmoravcik: yes, he wrote the thing :-) |
| 18:59.48 | rmoravcik | pierrox: yes, he wrote it |
| 18:59.53 | cr2 | guys, when the openmoko will come in january (?) i think this will be a non-issue. |
| 19:00.17 | cr2 | but the haret dumper will become the main wince->linux porting tool. |
| 19:00.17 | pierrox | cr2: a non-issue ? |
| 19:00.33 | lkcl | allo darlins |
| 19:00.35 | *** join/#htc-linux rmoravcik (n=rmoravci@icm4-orange.orange.sk) |
| 19:00.39 | cr2 | pierrox: you'll get complete linux source with a bunch of drivers. |
| 19:00.57 | pierrox | cr2: i wait this moment with no patience |
| 19:01.01 | lkcl | oh btw pierrox: i got an htc tytn which has the same processor |
| 19:01.25 | pierrox | lkcl: s3c2440 too ? |
| 19:01.46 | lkcl | rmoravcik: no, absolutely zero reason for a fork of haret. that's a stupid windows developer mentality. |
| 19:01.59 | cr2 | tytn==hermes |
| 19:02.04 | lkcl | i believe so, pierrox. |
| 19:02.09 | pierrox | ok |
| 19:02.21 | lkcl | because stupid windows developers don't understand subversion or cvs. |
| 19:02.22 | rmoravcik | lkcl: i only find it, it not my fork |
| 19:02.26 | lkcl | good man :) |
| 19:02.37 | cr2 | http://wiki.xda-developers.com/index.php?pagename=HermesMemoryMap |
| 19:02.45 | lkcl | well done - cos i'm going to need it, to merge into haret-gnu! |
| 19:03.04 | cr2 | http://wiki.xda-developers.com/index.php?pagename=Hermes_HardwareOverview |
| 19:03.11 | lkcl | cr2: and because haret required msvc-arm |
| 19:03.24 | lkcl | ok. sable suspend. |
| 19:03.36 | cr2 | lkcl: you can remember how i could not compile haret on amd64 ;-) |
| 19:03.42 | lkcl | ngggh :) |
| 19:19.06 | lkcl | argh. |
| 19:20.05 | lkcl | how am i going to debug this suspend stuff? |
| 19:20.10 | lkcl | i haven't a clue! |
| 19:20.35 | pierrox | lkcl: don't know, i have the same problem |
| 19:21.29 | cr2 | lkcl: did you commit all your patches to CVS ? |
| 19:22.25 | lkcl | cr2: most of them, yes |
| 19:22.40 | lkcl | np pierrox |
| 19:23.09 | lkcl | just creating lcd as a module, so i can load/unload it |
| 19:23.17 | lkcl | at least that will test it |
| 19:24.38 | cr2 | pierrox: what is the WM chip on eten? is it the sound codec ? |
| 19:26.19 | *** join/#htc-linux rmoravcik (n=rmoravci@icm2-orange.orange.sk) |
| 19:26.45 | lkcl | oh, oops. |
| 19:27.07 | lkcl | errr the lcd errrmm... oops! |
| 19:27.18 | *** join/#htc-linux rmoravcik (n=rmoravci@icm2-orange.orange.sk) |
| 19:27.22 | lkcl | that would explain why things aren't working :) |
| 19:29.48 | cr2 | pierrox: i don't think they use SPI for sdio. |
| 19:32.49 | lkcl | hm, i wonder why the green and blue led lights have gone on |
| 19:33.02 | *** join/#htc-linux rejon (n=rejon@adsl-71-146-8-178.dsl.pltn13.sbcglobal.net) |
| 19:34.33 | pierrox | cr2: i don't know what is the WM chip, i will have to re-open the device |
| 19:35.08 | pierrox | cr2: and you are probably right about spi : i wasn't able to read/write data on the bus |
| 19:35.25 | cr2 | pierrox: it must be sound chip, i don't see any other. |
| 19:35.39 | pierrox | cr2: and under wince the gpio seem to be used for something else |
| 19:35.52 | cr2 | is bluetooth multiplexed with the gsm ??? |
| 19:36.06 | cr2 | what about gprs ? |
| 19:36.07 | pierrox | cr2: according to the registry : yes |
| 19:36.29 | pierrox | cr2: but i don't know how it is possible, and why |
| 19:36.53 | cr2 | is the BT headset supported ? |
| 19:36.56 | pierrox | cr2: you mean gps instead of gprs ? |
| 19:36.57 | *** join/#htc-linux rob__w (n=bob@p85.212.191.167.tisdip.tiscali.de) |
| 19:37.20 | cr2 | no. gprs. it must also have some datapath. |
| 19:37.21 | pierrox | cr2: yes (not sure) |
| 19:37.52 | cr2 | so if br+gsm are on the port 1, where is gps ? |
| 19:38.02 | cr2 | s/br+/bt+/ |
| 19:38.07 | pierrox | cr2: gps is on the third uart |
| 19:38.13 | cr2 | ok. |
| 19:38.27 | cr2 | is the usb host activated ? |
| 19:38.27 | pierrox | cr2: i can read bytes if the gps is on before booting linux |
| 19:38.51 | cr2 | ok, that's probably a powerup/reset gpio issue. |
| 19:39.05 | pierrox | cr2: usb host is on, but there is no device |
| 19:39.26 | pierrox | or is it a special way to see if there are usb devices ? |
| 19:39.37 | cr2 | ok. may be a gps command issue. |
| 19:39.48 | lkcl | hm, i'm getting weird lcd module load/unload behaviour... |
| 19:39.50 | cr2 | at@usbpower=1 on universal. |
| 19:40.16 | cr2 | pierrox: do you know the gsm commands ? |
| 19:40.29 | cr2 | lkcl: use /sys for testing instead |
| 19:40.34 | rob__w | latest gtk2.10.6 oe build has a amazingly speed increase |
| 19:40.42 | pierrox | cr2: no, i just know stantard AT |
| 19:40.52 | pierrox | standard |
| 19:40.57 | cr2 | pierrox: it there any way to trace it ? |
| 19:41.13 | lkcl | okie oh yeh |
| 19:41.14 | cr2 | we have 2: 'wi' and registry key. |
| 19:41.31 | cr2 | you can't do 'wi'. |
| 19:41.57 | cr2 | the registry key log strips binary mux data. |
| 19:42.20 | cr2 | if available. on universal AT command channel is cleartext. |
| 19:42.23 | pierrox | cr2: currently i didn't have a close look at gsm, if focus on the sd card |
| 19:42.32 | cr2 | ok. |
| 19:42.37 | cr2 | do you have rilgsm.dll ? |
| 19:43.21 | lkcl | got it! |
| 19:43.25 | lkcl | ok - lcd works. |
| 19:43.34 | pierrox | cr2: yes |
| 19:43.35 | cr2 | rob__w: i have finally found how to monitor the main battery on blueangel. |
| 19:44.00 | cr2 | pierrox: run 'strings' and 'strings -el'. you'll learn a lot of interesting things. |
| 19:44.21 | cr2 | you can also post the 'ls \windows' output on wiki. |
| 19:44.41 | cr2 | lkcl: backlight ? |
| 19:45.10 | pierrox | cr2: AT+CPIN? : seems interesting :-)) |
| 19:47.43 | pierrox | cr2: but won't be of interested as long as i can't get this stupid SD controller working |
| 19:47.49 | pierrox | grrr |
| 19:48.30 | psokolovsky | cr2, I just committed asylumed's keys patch, if you work on BA |
| 19:48.55 | asylumed | :) |
| 19:49.12 | asylumed | psokolovsky:thx |
| 19:49.35 | psokolovsky | asylumed, Hi! see my mail reply too |
| 19:50.12 | cr2 | asylumed: a new battery driver is needed, based on ds1wm. i've corrected the blueangelasic3 wiki page |
| 19:50.29 | cr2 | for owm irq. |
| 19:50.48 | rob__w | how .. so its one wire one |
| 19:50.49 | cr2 | still can't find the wifi base address. |
| 19:51.26 | cr2 | rob__w: yes. the tsc2200 spi driver monitors the backup battery. like on hx4700. |
| 19:51.36 | rob__w | nice ... |
| 19:51.43 | asylumed | cr2: i see... |
| 19:52.32 | cr2 | i looked at the battery.dll and can trace the register accesses at 0x11000000 with 'wi'. |
| 19:53.01 | cr2 | the same owm CPLD is used on magician and alpine. |
| 19:53.11 | rob__w | cr2, so what values does it report in general , percentages or what ? |
| 19:54.17 | cr2 | rob__w: no idea. we need to compare the raw sampled data with 'powermon' under similar conditions. |
| 19:54.59 | cr2 | discharge the battery completely, and then charge in wince with logging, and charge in linux with logging. |
| 19:55.36 | cr2 | i think it will be easier than to disassemble the drivers with lot of multiplications and divisions ;-) |
| 19:57.31 | lkcl | cr2: yep - backlight power. |
| 19:58.45 | lkcl | ok backlight driver |
| 20:00.06 | cr2 | asylumed: i can try to repair the dpram driver. don't know what to do with the uda1380, this thing is really complex. |
| 20:00.42 | asylumed | cr2: audio is rather a big lump... |
| 20:01.08 | asylumed | cr2: the dpram driver would be awesome :) |
| 20:03.56 | cr2 | asylumed: the uda1380 should be completely rewritten ;) |
| 20:04.36 | asylumed | asylumed: the blueangel_audio interface? or the whole uda1380 driver? |
| 20:04.48 | asylumed | ahem... cr2: |
| 20:04.55 | lkcl | is there suspend/resume for backlight? |
| 20:04.56 | psokolovsky | Kevin2, committed your patches too |
| 20:05.47 | psokolovsky | lkcl, yes, and it works perfectly with corgi-bl. just write set_intensity() right ;-) |
| 20:05.55 | lkcl | oh ok! |
| 20:06.56 | Speedy2 | lkcl: Are you discovering what GPIOs map to pins on parts by manually flipping them on/off, etc? |
| 20:07.24 | lkcl | Speedy2: yep :) but that only was less risky when we already had 80% of the pins mapped out already |
| 20:07.43 | lkcl | it's how i found the acx GPIO 111 and its missing ASIC3 GPIO pin. |
| 20:07.43 | cr2 | psokolovsky: what will happen with the sa1111.c driver ? will it be moved to drivers/soc ? |
| 20:08.37 | cr2 | lkcl: you are lazy to change the string in haret :) |
| 20:08.47 | lkcl | _what_ dude? |
| 20:08.54 | lkcl | what have i not done now?? |
| 20:09.06 | BHSPitMonkey | morning |
| 20:09.06 | cr2 | i'll add this string as a parameter to the pwrctl. |
| 20:09.18 | psokolovsky | cr2, it's from mainline IIRC. so, once we have our stuff in mainline, we should lobby them to move it there ;-) |
| 20:09.57 | cr2 | psokolovsky: then i'll put sa1101.c into drivers/soc. i'm the only user afaik. |
| 20:10.01 | psokolovsky | cr2, for now, let it lie there, I guess. and yes, I know that SA code structured less clean then PXA (at least as I can see it) |
| 20:10.16 | psokolovsky | cr2, yes, new stuff please put there ;-) |
| 20:10.20 | cr2 | ok. |
| 20:11.03 | lkcl | where's the damn CONFIG_BACKLIGHT??? |
| 20:12.32 | lkcl | oops found it |
| 20:23.42 | rob__w | cr2, which wiki did you update ? |
| 20:25.30 | asylumed | hehe, was just wondering that myself :) |
| 20:35.18 | cr2 | http://wiki.xda-developers.com/index.php?pagename=BlueangelASIC3 |
| 20:41.12 | cr2 | asylumed: it's also useful to cleanup the LCD gpio mess. |
| 20:42.24 | BabelOued | cr2: asylumed patch are working fine :) and suspend resume too |
| 20:42.51 | cr2 | on 0x5 ? |
| 20:42.56 | BabelOued | yes |
| 20:43.12 | cr2 | <PROTECTED> |
| 20:43.35 | cr2 | is it for 0x6 ? does somebody here have it ? |
| 20:43.42 | BabelOued | i 've reported some lines from board id 6 |
| 20:44.09 | cr2 | ok, but you can't test. |
| 20:44.12 | asylumed | i have a boardid=6 |
| 20:44.18 | cr2 | cracked, but working ? |
| 20:44.19 | cr2 | ok. |
| 20:44.30 | lkcl | yes working. mostly. touchscreen is _really_ fscked. |
| 20:45.05 | cr2 | asylumed: and suspend/resume does not work for you ? |
| 20:45.45 | asylumed | cr2: working fine now :) |
| 20:45.57 | asylumed | cr2: just not the lcd anymore... |
| 20:46.11 | asylumed | cr2: i have to leave in on for now :/ |
| 20:46.55 | cr2 | hmm. |
| 20:47.17 | cr2 | i'll look at the 0x6 gpios. |
| 20:47.29 | asylumed | cr2: thx |
| 20:47.42 | cr2 | the problem is already 1 year old |
| 20:48.24 | asylumed | cr2: i can switch it off and on while the device is on... |
| 20:48.46 | asylumed | cr2: but it's got to do with the ordering i think... i got it right once by chance... but messed it up :( |
| 20:50.56 | lkcl | i messed it up, too. |
| 20:54.27 | lkcl | arse! write protect is on, on my laptop. |
| 20:54.32 | lkcl | on the usb_storage. |
| 20:54.56 | lkcl | arse! |
| 20:55.23 | lkcl | also i think i'm going for a record, on the number of vi sessions open and suspended: over 80! |
| 21:02.56 | lkcl | brb folks |
| 21:27.35 | *** join/#htc-linux lkcl (n=lkcl@host81-157-198-228.range81-157.btcentralplus.com) |
| 21:29.18 | BHSPitMonkey | would it be terribly off-topic to ask about an unmodified htc in here |
| 21:30.43 | BHSPitMonkey | ok, I recently got a "Star Trek" smartphone, and I'm noticing that the keys never light up. Even if I turn off the lights, or cover up the light sensor, nothing. |
| 21:30.47 | lkcl | probably not, BHSPit |
| 21:30.49 | lkcl | oooo cool! |
| 21:31.09 | BHSPitMonkey | Once I did see them light up really quickly, but then go away... I was just playing with the phone when that happened, though |
| 21:31.38 | BHSPitMonkey | is it safe to declare it as defective at this point? or is there some better way I can test out the backlight, e.g. activate them directly |
| 21:32.10 | lkcl | run gnuharet and there's a command for switching on the backlight. |
| 21:32.19 | lkcl | pwrbkl 0 or something |
| 21:33.13 | lkcl | ask cr2 when he's around or download the source and look at scripts.cpp |
| 21:33.30 | lkcl | arse. my backlight code has caused the kernel to crash |
| 21:33.43 | BHSPitMonkey | gnuharet is a windows (mobile) app? |
| 21:34.43 | lkcl | yes - it's an application compiled on linux using the arm-gnu-wince-gcc cross-compiler and the gnu wince project header files etc. |
| 21:35.16 | BHSPitMonkey | standalone? |
| 21:36.20 | BHSPitMonkey | I'll give it a look, thanks |
| 21:36.35 | lkcl | use google to find a precompiled version |
| 21:36.46 | lkcl | oh - wait - jornada820.sf.net/files i think |
| 21:37.06 | lkcl | http://jornada820.sourceforge.net/files/haret/ |
| 21:39.05 | lkcl | oh you poor thing, the star trek phone has an OMAP processor |
| 21:41.22 | *** join/#htc-linux g3gg0__ (n=g3gg0@ppp-82-135-65-122.dynamic.mnet-online.de) |
| 21:46.13 | cr2 | asylumed: there are some wild things going on there. can you trace 'pwrlcd' ? |
| 21:46.26 | cr2 | i have found 3 more pxa gpios. |
| 21:54.08 | asylumed | sure... |
| 21:54.30 | asylumed | what do i need to do? |
| 21:56.07 | lkcl | telnet to your device _twice_ |
| 21:56.10 | lkcl | at the same time |
| 21:56.15 | lkcl | once on port 9998 |
| 21:56.18 | lkcl | once on port 9999 |
| 21:56.30 | lkcl | on one telnet, do a wasic3gpio 60 |
| 21:56.39 | lkcl | on the other, run pwrlcd 4 |
| 21:56.42 | lkcl | then pwrlcd 0 |
| 21:56.47 | lkcl | or maybe it is the other way round |
| 21:56.54 | lkcl | sorry pwrlcd 1 not 0 |
| 21:56.57 | lkcl | you'll find out |
| 21:57.35 | cr2 | asylumed: set iasic3gpio(11) 1 |
| 21:57.37 | lkcl | huh. backlight doesn't work. |
| 21:57.42 | asylumed | hang on... which haret? |
| 21:57.46 | cr2 | set iasic3gpio(12) 1 |
| 21:57.52 | cr2 | set igpio(88) 1 |
| 21:57.58 | cr2 | set igpio(87) 1 |
| 21:58.02 | cr2 | set igpio(86) 1 |
| 21:58.25 | cr2 | latest http://jornada820.sourceforge.net/files/haret |
| 21:59.00 | cr2 | lkcl: why doesn't your backlight work? |
| 21:59.42 | cr2 | i thought universal LCD power was bad. blueangel is not better ;) |
| 22:00.19 | cr2 | the gpio that is set to 1 on 0x5 is set to 0 on 0x6. |
| 22:00.26 | lkcl | dunno dude. where is it in /sys? |
| 22:00.59 | cr2 | <PROTECTED> |
| 22:01.26 | cr2 | asylumed: set iasic3gpioa(12) 1 |
| 22:01.35 | cr2 | set iasic3gpioa(11) 1 |
| 22:01.47 | lkcl | there's actually two GPIOs for the backlight - i think one is keyboard, the other is screen |
| 22:01.48 | cr2 | otherwise you will be spammed :) |
| 22:02.01 | cr2 | lkcl: do you read wiki ? |
| 22:02.04 | lkcl | yep |
| 22:02.05 | lkcl | where? |
| 22:02.24 | cr2 | <PROTECTED> |
| 22:02.34 | cr2 | <PROTECTED> |
| 22:02.43 | cr2 | GPIOD12 0x1000 backlight lcd |
| 22:02.45 | lkcl | ah ha! |
| 22:02.50 | cr2 | http://wiki.xda-developers.com/index.php?pagename=SableASIC3 |
| 22:03.07 | lkcl | ok i just echoed 1023 > /sys/class/corgi_bl/brightness |
| 22:03.09 | lkcl | and it worked. |
| 22:03.16 | lkcl | so it's opie that's buggered. |
| 22:03.20 | lkcl | ok i'm happy with that. |
| 22:03.34 | lkcl | whoo hoo! |
| 22:03.35 | cr2 | opie does not know about your phone ? |
| 22:03.58 | lkcl | not exactly... |
| 22:03.59 | cr2 | it's time to dehardcode "universal" where appropriate. |
| 22:04.10 | lkcl | ack :) |
| 22:04.12 | cr2 | odevice_htc.cpp in opie. |
| 22:04.15 | lkcl | ok. |
| 22:04.21 | cr2 | we need 1 opie image. |
| 22:04.32 | lkcl | i'll work on it later. |
| 22:04.35 | cr2 | universal should be replaced with ${MACHINE} |
| 22:04.39 | lkcl | oh right. |
| 22:04.49 | cr2 | in the current OE universal setup. |
| 22:04.54 | lkcl | you got the todo list in your head? :) |
| 22:05.28 | asylumed | cr2: ? |
| 22:05.31 | cr2 | don't care about userspace right now :) |
| 22:05.51 | cr2 | asylumed: gpioa 11 and 12 are pwm0 and pwm1. |
| 22:05.53 | lkcl | bugger. crashed again. |
| 22:05.57 | lkcl | can't power up. |
| 22:06.18 | lkcl | oh. must try brightness of 0 of course. |
| 22:06.36 | cr2 | lkcl: it's just 1 gpio and 1 cken :) |
| 22:06.50 | lkcl | two gpios, dude! |
| 22:07.06 | lkcl | well. one for keyboard, anyway... |
| 22:07.13 | cr2 | well. i think it's userspace/opie issue. |
| 22:07.42 | lkcl | yep. |
| 22:07.52 | lkcl | anyway. |
| 22:08.15 | lkcl | what now. |
| 22:08.28 | lkcl | suspend/resume |
| 22:10.20 | lkcl | btw cr2 what's the address of that IPL> |
| 22:10.28 | lkcl | from memory map, right? |
| 22:10.38 | cr2 | lkcl: 0xa0040000 |
| 22:10.44 | cr2 | like on universal and apache. |
| 22:10.51 | lkcl | a9100000 00000000 00000001 IPL |
| 22:10.52 | cr2 | lkcl: what about the keyboard ? |
| 22:10.52 | asylumed | GPIOB4 S 0 ? 954390 |
| 22:10.57 | asylumed | GPIOB5 S 0 ? 954406 |
| 22:10.57 | asylumed | GPIOB6 S 0 954421 |
| 22:10.57 | asylumed | GPIOB7 S 0 ? 954436 |
| 22:10.57 | asylumed | GPIOC14 S 0 954452 |
| 22:10.57 | asylumed | GPIOC15 S 0 ? 954468 |
| 22:11.20 | asylumed | woops... and a GPIOB3 S 0 ? 954369 at the beginning |
| 22:11.34 | cr2 | ok, powerup. |
| 22:11.43 | cr2 | 6 gpios :) |
| 22:11.51 | cr2 | 7! |
| 22:12.14 | lkcl | good grief |
| 22:12.54 | lkcl | ok 0xa0040000 is in the main SDRAM memory... that's a physical address, isn't it. |
| 22:12.58 | cr2 | lkcl: there are 3+irq more on pxa. |
| 22:13.07 | cr2 | lkcl: yes. |
| 22:14.22 | lkcl | hm. |
| 22:14.31 | lkcl | how the hell do i debug this stuff????? |
| 22:14.39 | asylumed | GPIOB3 S 1 ? 1176599 |
| 22:14.42 | asylumed | GPIOB4 S 1 ? 1176690 |
| 22:14.42 | asylumed | GPIOB5 S 1 ? 1176703 |
| 22:14.42 | asylumed | GPIOB6 S 1 1176716 |
| 22:14.42 | asylumed | GPIOB7 S 1 ? 1176730 |
| 22:14.42 | asylumed | GPIOC14 S 1 1176745 |
| 22:14.43 | asylumed | GPIOC15 S 1 ? 1176762 |
| 22:16.42 | asylumed | k... so B7 and B6 are phone keys and backlight at least... |
| 22:35.34 | lkcl | ok B14 is switched on when usb is connected. |
| 22:35.42 | lkcl | could that be battery charge? |
| 22:36.46 | cr2 | lkcl: yes. |
| 22:36.55 | lkcl | ok. |
| 22:36.57 | cr2 | do you have AC power supply ? |
| 22:37.57 | lkcl | what the _heck_ is that ds1wm stuff?? |
| 22:37.58 | lkcl | not with me |
| 22:38.27 | lkcl | ok abandon charge for now |
| 22:39.03 | cr2 | :) |
| 22:39.16 | lkcl | it might be b6. |
| 22:39.19 | cr2 | lkcl: can you check the AC gpio ? |
| 22:39.29 | lkcl | it's probably b6 |
| 22:39.58 | lkcl | AC gpio is an input, right? |
| 22:40.25 | lkcl | ok keyboard. |
| 22:40.36 | lkcl | ahh this is going to be fun. |
| 22:42.19 | lkcl | what's with all the ALT_FN stuff in htcapache gpio .h |
| 22:42.20 | lkcl | ?? |
| 22:46.49 | cr2 | switch to keyboard from normal gpio |
| 22:47.20 | cr2 | lkcl: found gpiob8 for charge-related things. |
| 22:47.25 | lkcl | cool. |
| 22:57.04 | asylumed | cr2: the times next to each of those traces... in ms? |
| 22:59.44 | cr2 | yes. |
| 22:59.57 | asylumed | damn... |
| 23:00.21 | asylumed | tried initializing then correctly as far as i can see... |
| 23:00.50 | asylumed | the interesting part is that i can blank the screen while the device is on... and it "resumes" properly from off... |
| 23:00.52 | cr2 | i see 100 and 600 |
| 23:01.17 | cr2 | as time delays. |
| 23:01.58 | lkcl | um - btw, i am having unreliability issues on my htcsable now that i have an lcd device |
| 23:02.17 | lkcl | if i rmmod htcsable_lcd it goes _really_ weird - processes crashing |
| 23:02.21 | lkcl | not starting |
| 23:02.22 | lkcl | etc. |
| 23:02.28 | lkcl | so it might not be you, asylumed. |
| 23:02.33 | asylumed | hmm... i see delays shorter than that... what am i missing? |
| 23:03.30 | asylumed | lkcl: processes crashing? ag, that's a pain... |
| 23:03.35 | lkcl | yep.. |
| 23:05.01 | lkcl | ok - for these keys, i'm literally pressing them all and measuring the gpios that go up/down, right? |
| 23:08.07 | cr2 | you can name them 1x1 1x2 ... in the table, and log them. |
| 23:11.51 | lkcl | oh god. |
| 23:20.50 | lkcl | ok i don't understand this. |
| 23:23.53 | lkcl | ok... KPAS. |
| 23:26.36 | cr2 | look at dump gpio. how much MKIN/MKOUT do you have ? |
| 23:26.42 | lkcl | 8 each |
| 23:26.48 | lkcl | MKIN 8 MKOUT 8 |
| 23:27.14 | lkcl | yeh but how do i work out the relationship between these MKIN/OUT and the actual keys? |
| 23:27.42 | lkcl | i think i need to boot into linux and put some printk into pxa27x_keyboard.c |
| 23:27.56 | lkcl | then press keys there |
| 23:28.42 | lkcl | ok what's that program - scankeys? |
| 23:29.16 | cr2 | hmm. don't know. |
| 23:29.24 | cr2 | you need to dump raw scancodes. |
| 23:29.33 | cr2 | universal also has 8x8 |
| 23:30.16 | lkcl | ack ok. |
| 23:30.27 | cr2 | on my notebook there is kbd_mode -s |
| 23:34.27 | lkcl | ok |
| 23:36.28 | cr2 | maybe pxa27x_keyboard printks is not a bad idea :) |
| 23:36.59 | lkcl | wheeeee |
| 23:39.26 | cr2 | it's faster than anything else now. |
| 23:43.56 | *** join/#htc-linux RoEn_PC (n=roen@p54A67A38.dip.t-dialin.net) |
| 23:46.36 | cr2 | lkcl: here |
| 23:46.39 | cr2 | <PROTECTED> |
| 23:46.39 | cr2 | <PROTECTED> |
| 23:46.39 | cr2 | <PROTECTED> |
| 23:53.48 | lkcl | yep - got it |