Goodix GT801NI_3R15_1AV ("2 in 1"?) touchscreen doesn't work about linux-sunxi HOT 18 OPEN

tried contacting this Jicky http://forum.linuxmce.org/index.php?action=profile;u=59920 ? it's like the best match Google gives

from linux-sunxi.

No, I haven't. I did however send an email to [email protected] earlier today, no answer yet.

from linux-sunxi.

After an ugly hack on the kernel (allwinner-dev-team@80752bd), disabling CONFIG_MODVERSIONS and editing the vermagic on the stock ROM module (goodix_touch_3F.ko), the binary module loads correctly, detects the touchscreen and works. This pretty much confirms it's not a bug on the i2c stack.

from linux-sunxi.

turl can you write some howto and upload working goodix_touch_3F.ko? i'd like to try it on gemei g9 tablet. thank you.

from linux-sunxi.

What I did is, basically:

1. Include the patch linked previously on your kernel and enable the new config option
2. Disable MODVERSIONS (see https://github.com/allwinner-dev-team/linux-allwinner/blob/allwinner-3.0/arch/arm/configs/zatab_defconfig#L167). I also disabled MODULE_SRCVERSION_ALL, not sure if it's needed.
3. Make sure LOCALVERSION is empty and LOCALVERSION_AUTO is unset (see https://github.com/allwinner-dev-team/linux-allwinner/blob/allwinner-3.0/arch/arm/configs/zatab_defconfig#L37)
4. Build and get the kernel on your device
5. insmod the modified module .ko, you can get it from here https://github.com/allwinner-dev-team/android_device_allwinner_zatab/blob/ics/goodix_touch_3F.ko
6. Check dmesg, getevent

from linux-sunxi.

where is the above mentioned patch? i get a 404

from linux-sunxi.

@simonckenyon: @turl's tree is rebased, so the hash changes every time. For the sake of the ticket I've pasted it into a gist. https://gist.github.com/3091730 https://gist.github.com/3093889

edit: correct patch

from linux-sunxi.

On 07/11/12 17:59, Alejandro Mery wrote:

@simonckenyon: @turl's tree is rebased, so the hash changes every time. For the sake of the ticket I've pasted it into a gist. https://gist.github.com/3091730

---

Reply to this email directly or view it on GitHub:
https://github.com/amery/linux-allwinner/issues/26#issuecomment-6912717

thank you.
unfortunately it doesn't work for me.
says there is an undefined method/call/whatever

this is so frustrating. all i want to do is compile a driver and i'm
having to go though this mess to achieve it.

thanks for all your help

simon

from linux-sunxi.

that patch should work fine in the 3.0-v2 branch. what are you building?

from linux-sunxi.

looks like @amery pasted the wrong patch, this is probably the one you're looking for https://github.com/allwinner-dev-team/linux-allwinner/commit/8e1968f86e15e442a9ae5944535ac8000356d2e1.patch

from linux-sunxi.

@turl: doh :< ... yes, with the conversation in IRC I got stuck with the wifi driver

from linux-sunxi.

I have been building that version.
My mail problem is that I bought a tablet from a vendor who will not provide any help. Not even an image.
So I am having to guess at what hardware is in it.
I eventally found an image that supports all the hardware and I have been using that as my backup.

My goal is to get cm9 with a self compiled kernel that I can then use for my project. That requires me to build a device driver for a digital persona fingerprint reader.

I am going to ask my business partners if I can put up a bounty. We have not gt funding yet, so money is very tight. But right now I have spent weeks on this and I can I'll afford to continue doing that.

Sorry for the rant but it has been a tough day.

Simon

Alejandro Mery [email protected] wrote:

that patch should work fine in the 3.0-v2 branch. what are you building?

---

Reply to this email directly or view it on GitHub:
https://github.com/amery/linux-allwinner/issues/26#issuecomment-6916978

from linux-sunxi.

He did but I found the correct one. Thanks.

turl [email protected] wrote:

looks like @amery pasted the wrong patch, this is probably the one you're looking for https://github.com/allwinner-dev-team/linux-allwinner/commit/8e1968f86e15e442a9ae5944535ac8000356d2e1.patch

---

Reply to this email directly or view it on GitHub:
https://github.com/amery/linux-allwinner/issues/26#issuecomment-6917061

from linux-sunxi.

Hallo,
May I start with a thank you and you guys rawk!

I think this is related:

I did a modprobe on goodix_touch

kern.log had this::

Jan 1 00:33:34 linaro-alip kernel: [ 2014.680000] ===========================goodix_ts_init=====================
Jan 1 00:33:34 linaro-alip kernel: [ 2014.690000] ctp_fetch_sysconfig_para.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.690000] ctp_fetch_sysconfig_para: after: ctp_twi_addr is 0x5d, dirty_addr_buf: 0
x5d. dirty_addr_buf[1]: 0xfffe
Jan 1 00:33:34 linaro-alip kernel: [ 2014.700000] ctp_fetch_sysconfig_para: ctp_twi_id is 2.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.710000] ctp_fetch_sysconfig_para: screen_max_x = 800.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.710000] ctp_fetch_sysconfig_para: screen_max_y = 480.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.720000] ctp_fetch_sysconfig_para: revert_x_flag = 1.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.720000] ctp_fetch_sysconfig_para: revert_y_flag = 0.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.730000] ctp_fetch_sysconfig_para: exchange_x_y_flag = 0.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.730000] goodix_ts_init: after fetch_sysconfig_para: normal_i2c: 0x5d. normal_i2
c[1]: 0xfffe
Jan 1 00:33:34 linaro-alip kernel: [ 2014.740000] ctp_reset.
Jan 1 00:33:34 linaro-alip kernel: [ 2014.770000] ctp_wakeup.
Jan 1 00:33:35 linaro-alip kernel: [ 2014.990000] i2c-core: driver [Goodix-TS] using legacy suspend method
Jan 1 00:33:35 linaro-alip kernel: [ 2014.990000] i2c-core: driver [Goodix-TS] using legacy resume method
Jan 1 00:33:35 linaro-alip kernel: [ 2015.000000] ctp_detect: Detected chip Goodix-TS at adapter 2, address 0x5d
Jan 1 00:33:35 linaro-alip kernel: [ 2015.010000] ===============================GT801 Probe===========================
Jan 1 00:33:35 linaro-alip kernel: [ 2015.010000] Begin goodix i2c test
Jan 1 00:33:35 linaro-alip kernel: [ 2015.010000] START can't sendout!
Jan 1 00:33:35 linaro-alip kernel: [ 2015.020000] Retrying transmission 2
Jan 1 00:33:35 linaro-alip kernel: [ 2015.030000] START can't sendout!
Jan 1 00:33:35 linaro-alip kernel: [ 2015.030000] Retrying transmission 1
Jan 1 00:33:35 linaro-alip kernel: [ 2015.030000] START can't sendout!
Jan 1 00:33:35 linaro-alip kernel: [ 2015.030000] Retrying transmission 0
Jan 1 00:33:35 linaro-alip kernel: [ 2015.050000] START can't sendout!
Jan 1 00:33:35 linaro-alip kernel: [ 2015.050000] Retrying transmission 2
Jan 1 00:33:35 linaro-alip kernel: [ 2015.050000] START can't sendout!
.
.
.
Jan 1 00:33:35 linaro-alip kernel: [ 2015.150000] Retrying transmission 0
Jan 1 00:33:35 linaro-alip kernel: [ 2015.170000] Warnning: I2C connection might be something wrong!
Jan 1 00:34:04 linaro-alip kernel: [ 2044.920000] Goodix-TS 2-005d: The driver is removing...
Jan 1 00:34:05 linaro-alip kernel: [ 2044.930000] ------------[ cut here ]------------
Jan 1 00:34:05 linaro-alip kernel: [ 2044.930000] WARNING: at kernel/irq/manage.c:1166 __free_irq+0xa8/0x1ac()
Jan 1 00:34:05 linaro-alip kernel: [ 2044.930000] Trying to free already-free IRQ 28
Jan 1 00:34:05 linaro-alip kernel: [ 2044.930000] Modules linked in: goodix_touch(-) disp_ump 8192cu mali_drm drm mali ump hdmi lcd disp cfbcopyarea cfbimgblt cfbfillrect

kern.log + backtrace is at http://pastebin.com/nqKD3NAB
syslog/dmesg spit out similar things..

Is this relevant to the current issue? I have a hunch it could be because of some device specific settings in script.bin.

tested on an A10_MID tablet...

from linux-sunxi.

btw the script.bin is the one from the device... no extra tweaks...

from linux-sunxi.

May be simular issue that I had
please send script.bin

from linux-sunxi.

The script.fex is here: http://pastebin.com/H4xc41yr

Once the kernel is loaded, why does the script.bin config disrupt the driver??

from linux-sunxi.

Could someone test this commit: 82bcb1a

from linux-sunxi.