Re: stable regression: tda18271_read_regs: [1-0060|M] ERROR: i2c_transfer returned: -19

[Mauro Carvalho Chehab <m.chehab@samsung.com>]

Em Fri, 13 Dec 2013 22:19:39 +0100
Frederik Himpe <fhimpe@telenet.be> escreveu:

> [My excuses for multiposting, it seems gmane does not permit posting to all
> the relevant lists]
> 
> Since upgrading my system from Linux 3.12 to 3.12.3, my PCTV Systems
> nanoStick T2 290e does not work anymore.
> 
> This happens with 3.12.3:
> 
> [    3.778817] em28174 #0: i2c eeprom 0000: 26 00 01 00 02 09 d8 85 80 80 e5 80 f4 f5 94 90
> [    3.779741] em28174 #0: i2c eeprom 0010: 78 0d e4 f0 f5 46 12 00 5a c2 eb c2 e8 30 e9 03
> [    3.780643] em28174 #0: i2c eeprom 0020: 12 09 de 30 eb 03 12 09 10 30 ec f1 12 07 72 80
> [    3.781562] em28174 #0: i2c eeprom 0030: ec 00 60 00 e5 f5 64 01 60 09 e5 f5 64 09 60 03
> [    3.782473] em28174 #0: i2c eeprom 0040: c2 c6 22 e5 f7 b4 03 13 e5 f6 b4 87 03 02 09 92
> [    3.783406] em28174 #0: i2c eeprom 0050: e5 f6 b4 93 03 02 07 e6 c2 c6 22 c2 c6 22 12 09
> [    3.784314] em28174 #0: i2c eeprom 0060: cf 02 06 19 1a eb 67 95 13 20 4f 02 c0 13 6b 10
> [    3.785213] em28174 #0: i2c eeprom 0070: a0 1a ba 14 ce 1a 39 57 00 5c 18 00 00 00 00 00
> [    3.786140] em28174 #0: i2c eeprom 0080: 00 00 00 00 44 36 00 00 f0 10 02 00 00 00 00 00
> [    3.787057] em28174 #0: i2c eeprom 0090: 5b 23 c0 00 00 00 20 40 20 80 02 20 01 01 00 00
> [    3.787970] em28174 #0: i2c eeprom 00a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [    3.788879] em28174 #0: i2c eeprom 00b0: c6 40 00 00 00 00 a7 00 00 00 00 00 00 00 00 00
> [    3.789790] em28174 #0: i2c eeprom 00c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 38 32
> [    3.790709] em28174 #0: i2c eeprom 00d0: 34 31 30 31 31 36 36 30 31 37 31 32 32 31 46 4b
> [    3.791625] em28174 #0: i2c eeprom 00e0: 4a 31 00 4f 53 49 30 30 33 30 38 44 30 31 31 30
> [    3.792531] em28174 #0: i2c eeprom 00f0: 46 4b 4a 31 00 00 00 00 00 00 00 00 00 00 31 30
> [    3.793444] em28174 #0: i2c eeprom 0100: ... (skipped)
> [    3.793502] em28174 #0: EEPROM ID = 26 00 01 00, EEPROM hash = 0xfcf432bb
> [    3.793559] em28174 #0: EEPROM info:
> [    3.793616] em28174 #0: 	microcode start address = 0x0004, boot configuration = 0x01
> [    3.804741] scsi 8:0:0:0: Direct-Access     Generic  Ultra HS-SD/MMC  1.82 PQ: 0 ANSI: 0
> [    3.805345] sd 8:0:0:0: Attached scsi generic sg3 type 0
> [    3.818139] em28174 #0: 	No audio on board.
> [    3.818194] em28174 #0: 	500mA max power
> [    3.818247] em28174 #0: 	Table at offset 0x39, strings=0x1aa0, 0x14ba, 0x1ace
> [    3.818318] em28174 #0: Identified as PCTV nanoStick T2 290e (card=78)
> [    3.818374] em28174 #0: v4l2 driver version 0.2.0
> [    3.821522] sd 8:0:0:0: [sdc] Attached SCSI removable disk
> [    3.823606] em28174 #0: V4L2 video device registered as video0
> [    3.823662] em28174 #0: dvb set to isoc mode.
> [    3.823972] usbcore: registered new interface driver em28xx
> [    3.844020] tda18271 1-0060: creating new instance
> [    3.868422] tda18271_read_regs: [1-0060|M] ERROR: i2c_transfer returned: -19
> [    3.868492] Error reading device ID @ 1-0060, bailing out.
> [    3.868548] tda18271_attach: [1-0060|M] error -5 on line 1285
> [    3.868603] tda18271 1-0060: destroying instance
> [    3.868666] Em28xx: Initialized (Em28xx dvb Extension) extension
> [    3.894687] Registered IR keymap rc-pinnacle-pctv-hd
> [    3.894819] input: em28xx IR (em28174 #0) as /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.7/rc/rc0/input23
> [    3.894979] rc0: em28xx IR (em28174 #0) as /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.7/rc/rc0
> [    3.895570] Em28xx: Initialized (Em28xx Input Extension) extension
> 
> I see the same problem reported here:
> https://github.com/Hexxeh/rpi-firmware/issues/38 where it is mentioned
> that this regression also appeared in 3.10 stable series recently.
> 
> I noticed upstream commit 8393796dfa4cf5dffcceec464c7789bec3a2f471
> (media: dvb-frontends: Don't use dynamic static allocation)
> entered both 3.10.22 (which is the first version introducing the
> regression in 3.10 stable according to the linked bug), and 3.12.3.
> This file contains stuff related to tda18271. Could this be the 
> culprit?

I'll analyze it better tomorrow, but commit 8393796dfa4cf5 touches only
on tda18271c2dd driver. The only em28xx device that uses this driver 
variant is Terratec H5, which is not your case.

Besides that, with commit 8393796dfa4c it should be returning -EINVAL
and printing a msg that it would look like:

	tda18271: i2c wr reg=0000: len=65 is too big!

However, on your report:

> [    3.868422] tda18271_read_regs: [1-0060|M] ERROR: i2c_transfer returned: -19

it is returning ENODEV:
	#define	ENODEV		19	/* No such device */

So, it seems that the regression was caused by some other patch.

are you using vanilla Kernel 3.12.3?

-- 

Cheers,
Mauro