"disagrees about version of symbol module_layout" is saying the gslx680 module is not for the same kernel build as the kernel.

Try using:

https://github.com/jabjoe/linux-sunxi/tree/sunxi_3.4_gslx680

(Just add it as a remote, do a fetch, then checkout the sunxi_3.4_gslx680 branch, that should avoid pulling all of git repo down, just my bits)

and adding CONFIG_TOUCHSCREEN_GSLX680_TS  to your build config.

It's a very rough branch as it doesn't take any of the history.

Joe


On 30/05/14 12:18, [email protected] wrote:
Can't attach files here for some reason:
Dmesg log of the TS and the LCD panel
[   71.988222] ===========================gslx680_ts_init=====================
[   71.990707] _fetch_sysconfig_para.
[   71.994277] gslx680 firmware 420TPC.fw.
[   72.003405] _fetch_sysconfig_para: after: ctp_twi_addr is 0x40, 
dirty_addr_buf: 0x40. dirty_addr_buf[1]: 0xfffe
[   72.007980] _fetch_sysconfig_para: ctp_twi_id is 1.
[   72.012163] _fetch_sysconfig_para: screen_max_x = 480.
[   72.016639] _fetch_sysconfig_para: screen_max_y = 272.
[   72.020732] _fetch_sysconfig_para: revert_x_flag = 0.
[   72.025096] _fetch_sysconfig_para: revert_y_flag = 0.
[   72.029553] _fetch_sysconfig_para: exchange_x_y_flag = 0.
[   72.038401] i2c-core: driver [gslx680] using legacy suspend method
[   72.043973] i2c-core: driver [gslx680] using legacy resume method
[   72.049655] ctp_detect: Detected chip gslx680 at adapter 1, address 0x40
[   72.055382] ====gslx680_ts_probe begin=====.
[   72.057404] ==kzalloc success=
[   72.060192] [GSLX680] Enter gsl_ts_init_ts
[   72.072400] gslx680: disagrees about version of symbol module_layout
[   72.077220] ctp_set_irq_mode: config gpio to int mode.
[   72.083418] ctp_set_irq_mode, 854: gpio_int_info, port = 7, port_num = 11.
[   72.086606]  INTERRUPT CONFIG
[   72.088189] ==CTP_IRQ_NO=
[   72.096139] input: gslx680 as 
/devices/platform/sunxi-i2c.1/i2c-1/1-0040/input/input5
[   72.120945] incomplete xfer (0xff)
[   72.123925] reset_chip: gsl_ts_write 1 fail!
[   72.127072] init_chip: reset_chip fail: -70
[   72.130301] gslx680 1-0040: init_chip failed
[   72.135248] gslx680: probe of 1-0040 failed with error -70

*****************************************************************
[    3.131863] [DISP] ***
[    3.135430] [DISP] *** LCD Panel Parameter Check
[    3.136756] [DISP] ***
[    3.140323] [DISP] ***             by dulianping
[    3.141642] [DISP] ***
[    3.148328] [DISP] 
*****************************************************************
[    3.149653] [DISP] ***
[    3.151747] [DISP] *** Interface:
[    3.154621] [DISP] *** Parallel HV Panel
[    3.157507] [DISP] *** Lcd Frm to RGB666
[    3.158822] [DISP] ***
[    3.160828] [DISP] *** Timing:
[    3.163529] [DISP] *** lcd_x:      480
[    3.166229] [DISP] *** lcd_y:      272
[    3.168944] [DISP] *** lcd_ht:     525
[    3.171472] [DISP] *** lcd_hbp:    2
[    3.174173] [DISP] *** lcd_vt:     572
[    3.176708] [DISP] *** lcd_vbp:    2
[    3.179236] [DISP] *** lcd_hspw:   0
[    3.181768] [DISP] *** lcd_vspw:   0
[    3.184902] [DISP] *** lcd_frame_frq:  79Hz
[    3.186224] [DISP] ***
[    3.190659] [DISP] *** WRN03: Recommend "lcd_dclk_frq = 9"
[    3.191979] [DISP] ***
[    3.195892] [DISP] *** LCD Panel Parameter Check End
[    3.202590] [DISP] 
*****************************************************************

I still have no idea why it disagrees about module_layout and perhaps a manual 
hex edit is needed. I have compiled the driver against my kernel so that should 
not be there.


On Friday, 30 May 2014 05:57:45 UTC-4, Joe Burmeister  wrote:
On 30/05/14 10:49, [email protected] wrote:

Ok, the driver is compiled now.
loads without force modprobe.
It creates an input device, but after:
i2c state isn't idle (0xf8)
STOP failed!
incomplete xfer (0xff)
reset_chip: gsl_ts_write 1 fail!
init_chip: reset_chip fail: -70
gslx680 1-0040: init_chip failed
The incomplete xfer is not 0x20 so the twi_id is 1 and adapter address correct.
Not sure what the 0xf8 / 0xff codes are but
I'm guessing something is being read sequentially from START. I think it wants 
early suspend in the kernel source so the touchscreen is idle in low powered 
state.
Still trying.


Can you give me the dmesg?



Joe

--
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to