PPCGeeks - [TESTING] - NAND OMGB (1.2.3)

PPCGeeks (http://forum.ppcgeeks.com/index.php)

- Android On TP2 Development (http://forum.ppcgeeks.com/forumdisplay.php?f=319)

- - [TESTING] - NAND OMGB (1.2.3) (http://forum.ppcgeeks.com/showthread.php?t=147260)

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by [ACL] (Post 2162661)

ahh ok, well it may be just my changes that messed up the power down then.

The files are unchanged but im not exactly sure if they are from the 300 i have or the 400. Thing is this whole acoustic change was made to fix the gsm rhod mic not working issue. But i'm not exactly sure which ones i packaged up. We need to see if all the rhods carry diff files or not.

also updating LK will wipe almost everything. Since "almost" is not good enough when doing updates. I recommend doing a regular backup and restore when updating LK. The only real change was the ability to reboot into android without getting stuck on fastboot. So you can always wait for the next full release since this is just a test we have now.

Thanks for the feedback.

let me know if you need the 400 files

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by arrrghhh (Post 2162650)

Either the CSV's or the acoustic code in general needs to be updated...

Power down as in hold the power button and choose shut down.

IMHO ACL just needs to make sure he's using the newest acoustic stuff, emwe did a lot of tweaking and fine tuning. Haven't had audio die on HaRET acoustic in a while in other words ;).

Yup.. checked with emwe and i missed his commits. Either way i'll need someone to harvest all the csvs and do a compare.

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by mbellot (Post 2162657)

Just checked and my Rhod400 powers down just fine (and went into charge mode since it's connected right now).

Power Down is working fine for me too.

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by [ACL] (Post 2162670)

Yup.. checked with emwe and i missed his commits. Either way i'll need someone to harvest all the csvs and do a compare.

I'm already doing this for HaRET :thumbleft:

Re: [TESTING] - NAND OMGB (1.2.3)

Can I ignore the BAD BLOCK error message, or it will cause some unknown problem.

Quote:

<4>[ 2.217590] #MTD# parts in atag = 6
<5>[ 2.217651] Creating 6 MTD partitions on "msm_nand":
<5>[ 2.217712] 0x03100000-0x036c0000 : "recovery"
<5>[ 2.222656] 0x036c0000-0x03720000 : "misc"
<5>[ 2.224243] 0x03720000-0x03b20000 : "boot"
<5>[ 2.228149] 0x03b20000-0x0fb20000 : "system"
<4>[ 2.319091] ###BAD BLOCK 1594 0C740000###
<5>[ 2.353820] 0x0fb20000-0x1e000000 : "userdata"
<5>[ 2.514495] 0x1e000000-0x20000000 : "cache"

Kernel from git server, source update to Dec 30, hope usefully

Quote:

<4>[ 8768.268005] sdio_reset_comm():
<4>[ 8768.269073] ------------[ cut here ]------------
<4>[ 8768.269104] WARNING: at drivers/mmc/core/core.c:222 mmc_wait_for_cmd+0x34/0x7c()
<4>[ 8768.269134] Modules linked in: bcm4329

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by muziling (Post 2162700)

Can I ignore the BAD BLOCK error message, or it will cause some unknown problem.

As long as there aren't a ton of bad blocks, it should be OK. They will be ignored... Almost every device ships with 1-2 bad blocks, I have 2 myself on each RHOD400 I own...

Re: [TESTING] - NAND OMGB (1.2.3)

Found Division by zero in kernel.

htc_rhod_batt_corr() in file htc_battery_smem.c

Code:

    BATT("htc_get_batt_info 888\n");

        /* TODO: temp algo is not ok..if someone have the timte -> must look into the driver */

    buffer->batt_tempRAW = ( buffer->batt_tempRAW * 5200 ) / htc_adc_range;

    BATT("htc_get_batt_info 999\n");

    av_index = ( buffer->batt_tempRAW * 18 ) / ( 2600 - buffer->batt_tempRAW );

    BATT("htc_get_batt_info aaa\n");

    buffer->batt_tempRAW = htc_battery_temperature_lut( av_index );

    if ( buffer->batt_tempRAW < 250 ) /* fixed til so long we can get constant values - otherwise the algo wouln'T work good */

            buffer->batt_tempRAW = 250;

  BATT("htc_get_batt_info bbb\n");

    fix_batt_values(buffer);

  BATT("htc_get_batt_info ccc\n");

    htc_battery_level_compute( buffer );

  BATT("htc_get_batt_info ddd\n");

    return 0;

Quote:

<6>[ 4413.083618] [BATT] htc_get_batt_info 444
<6>[ 4413.083679] [BATT] htc_get_batt_info 555
<6>[ 4413.083770] [BATT] htc_get_batt_info 666
<6>[ 4413.083801] [BATT] htc_get_batt_info 888
<6>[ 4413.083862] [BATT] htc_get_batt_info 999
<4>[ 4413.083892] Division by zero in kernel.
<4>[ 4413.083923] [<c0281e48>] (dump_stack+0x0/0x14) from [<c002df94>] (__div0+0x18/0x20)
<4>[ 4413.084075] [<c002df7c>] (__div0+0x0/0x20) from [<c0142f4c>] (Ldiv0+0x8/0x10)
<4>[ 4413.084197] [<c00568ec>] (htc_get_batt_info+0x0/0x9a4) from [<c0057d30>] (htc_battery_thread+0x54/0xa0)
<4>[ 4413.084350] [<c0057cdc>] (htc_battery_thread+0x0/0xa0) from [<c0060f8c>] (do_exit+0x0/0x7a8)
<4>[ 4413.084472] r7:00000000 r6:00000000 r5:00000000 r4:00000000
<6>[ 4413.084533] [BATT] htc_get_batt_info aaa
<6>[ 4413.084564] [BATT] htc_get_batt_info bbb
<6>[ 4413.084625] [BATT] htc_get_batt_info ccc
<6>[ 4413.084655] [BATT] htc_get_batt_info ddd
<6>[ 4413.084686] [BATT] htc_get_batt_info 777
<6>[ 4423.102691] [BATT] htc_get_batt_info 111

HeroC battery driver, many different, https://raw.github.com/TeamICS/heroc.../htc_battery.c
Our battery chip is ds2746, same as HD2?

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by muziling (Post 2162764)

Found Division by zero in kernel.

htc_rhod_batt_corr() in file htc_battery_smem.c

Code:

    BATT("htc_get_batt_info 888\n");

        /* TODO: temp algo is not ok..if someone have the timte -> must look into the driver */

    buffer->batt_tempRAW = ( buffer->batt_tempRAW * 5200 ) / htc_adc_range;

    BATT("htc_get_batt_info 999\n");

    av_index = ( buffer->batt_tempRAW * 18 ) / ( 2600 - buffer->batt_tempRAW );

    BATT("htc_get_batt_info aaa\n");

    buffer->batt_tempRAW = htc_battery_temperature_lut( av_index );

    if ( buffer->batt_tempRAW < 250 ) /* fixed til so long we can get constant values - otherwise the algo wouln'T work good */

            buffer->batt_tempRAW = 250;

  BATT("htc_get_batt_info bbb\n");

    fix_batt_values(buffer);

  BATT("htc_get_batt_info ccc\n");

    htc_battery_level_compute( buffer );

  BATT("htc_get_batt_info ddd\n");

    return 0;

HeroC battery driver, many different, https://raw.github.com/TeamICS/heroc.../htc_battery.c

Do you know exactly what line ? I mean this could only happen if buffer->batt_tempRAW is 0. Can you printK to see if thats the case ?

Re: [TESTING] - NAND OMGB (1.2.3)

Quote:

Originally Posted by [ACL] (Post 2162670)

Yup.. checked with emwe and i missed his commits. Either way i'll need someone to harvest all the csvs and do a compare.

Rhod400 csv files. I'd attach others, but I only have the 400. LOL

Re: [TESTING] - NAND OMGB (1.2.3)

Whoo hoo! I finally am able to join the land of nand! Replacement TP2 after the previous one decided to perpetually flash up the Video Out screen in WM. First time I tried (before x-mas) LK kept thinking the end key was pressed. So I gave up. Tried again today from scratch, following the wiki (which by the way to which I added a comment about a minor correction); deleted EVERYTHING, made sure to erase all of the partitions multiple times, flashed recovery 4x times before rebooting. Boom. IT WORKS!

I have to say you guys are awesome. This runs lightyears more smoothly than FRX. Swype was always a bit laggy in FRX. Now it runs smooth as butter.

Now I just need to find the post where you put the calibration apk. because mine is off pretty wicked (it appears to be missing the top ~10-20 pixels and the bottom 10-20 pixels plus about 10 on the right side).