issues with SD card when using RPIO about rpio

[bpowers]

can you post the relevant part of dmesg/syslog that has info about the DMA errors?

[raspitoaster]

the last try resulted in

[Sun Apr 28 18:01:41 2013] mmc0: Timeout waiting for hardware interrupt - cmd25.
[Sun Apr 28 18:01:41 2013] mmc0: resetting ongoing cmd 25DMA before 4096/4096 [1]/[1] complete
[Sun Apr 28 18:01:41 2013] mmcblk0: error -110 transferring data, sector 110240, nr 8, cmd response 0x900, card status 0xc00
[Sun Apr 28 18:01:41 2013] end_request: I/O error, dev mmcblk0, sector 110240
[Sun Apr 28 18:01:41 2013] Buffer I/O error on device mmcblk0p2, logical block 468

[Sun Apr 28 18:01:41 2013] lost page write due to I/O error on mmcblk0p2

system is still running but I had much more weird effects before - resulting in system crash:

Apr 28 15:56:21 localhost kernel: [ 290.686107] mmc0: Timeout waiting for hardware interrupt - cmd25.
Apr 28 15:56:21 localhost kernel: [ 290.692270] mmc0: resetting ongoing cmd 25DMA before 8192/8192 [1]/[1] complete
Apr 28 15:56:21 localhost kernel: [ 290.694635] mmcblk0: error -110 transferring data, sector 106496, nr 16, cmd response 0x900, card status 0xc00
Apr 28 15:56:21 localhost kernel: [ 290.708460] end_request: I/O error, dev mmcblk0, sector 106496
Apr 28 15:56:21 localhost kernel: [ 290.714351] Buffer I/O error on device mmcblk0p2, logical block 0
Apr 28 15:56:21 localhost kernel: [ 290.722330] lost page write due to I/O error on mmcblk0p2
Apr 28 15:56:21 localhost kernel: [ 290.722391] end_request: I/O error, dev mmcblk0, sector 106504
Apr 28 15:56:21 localhost kernel: [ 290.730531] Buffer I/O error on device mmcblk0p2, logical block 1
Apr 28 15:56:21 localhost kernel: [ 290.738024] lost page write due to I/O error on mmcblk0p2
Apr 28 15:57:01 localhost kernel: [ 330.727189] mmc0: Timeout waiting for hardware interrupt - cmd25.
Apr 28 15:57:01 localhost kernel: [ 330.733365] mmc0: resetting ongoing cmd 25DMA before 4096/4096 [1]/[1] complete
Apr 28 15:57:01 localhost kernel: [ 330.736721] mmcblk0: error -110 transferring data, sector 200392, nr 8, cmd response 0x900, card status 0xc00
Apr 28 15:57:01 localhost kernel: [ 330.749926] end_request: I/O error, dev mmcblk0, sector 200392
Apr 28 15:57:01 localhost kernel: [ 330.755832] Buffer I/O error on device mmcblk0p2, logical block 11737
Apr 28 15:57:01 localhost kernel: [ 330.763777] JBD2: Detected IO errors while flushing file data on mmcblk0p2-8
Apr 28 15:57:01 localhost kernel: [ 330.763843] EXT4-fs warning (device mmcblk0p2): ext4_end_bio:250: I/O error writing to inode 4458 (offset 143360 size 4096 starting block 25050)
Apr 28 15:57:01 localhost kernel: [ 330.780691] ------------[ cut here ]------------
Apr 28 15:57:01 localhost kernel: [ 330.780762] WARNING: at fs/buffer.c:1106 mark_buffer_dirty+0xb8/0xdc()
Apr 28 15:57:01 localhost kernel: [ 330.780774] Modules linked in: bnep bluetooth evbug rfkill autofs4
Apr 28 15:57:01 localhost kernel: [ 330.780847] from
Apr 28 15:57:01 localhost kernel: [ 330.780875] from
Apr 28 15:57:01 localhost kernel: [ 330.780941] from
Apr 28 15:57:01 localhost kernel: [ 330.780975] from
Apr 28 15:57:01 localhost kernel: [ 330.781006] from
Apr 28 15:57:01 localhost kernel: [ 330.781038] from
Apr 28 15:57:01 localhost kernel: [ 330.781066] from
Apr 28 15:57:01 localhost kernel: [ 330.781100] from
Apr 28 15:57:01 localhost kernel: [ 330.781113] ---[ end trace 079fdde119f36658 ]---
Apr 28 15:57:06 localhost kernel: [ 336.038042] JBD2: Detected IO errors while flushing file data on mmcblk0p2-8

Apr 28 15:57:12 localhost kernel: [ 342.033191] JBD2: Detected IO errors while flushing file data on mmcblk0p2-8

the errors occur when playing around with the servo for longer that approx 30s or when using them e.g. only once and then leaving the system untouched (of course, having used "servo.stop_servo" before).

[bpowers]

I should have asked before, but what GPIO pins are you using? and what does your circuit look like? I.E. are you trying to drive the servo directly from the IO pin?

[raspitoaster]

I tried different ones - with same result (also with piblaster). Currently GPIO 4 is used.
Yes, servo control is directly connected to the pin P1-07 (only a resistor is between) - ground is common with the PI, power is separate.

[itdaniher]

I've also had issues with using DMA Ch0 on Fedora resulting in system crashes, most recently:

kernel:[ 1371.332993] journal commit I/O error

I'm willing to spend some time trying to get to the bottom of this if there's still interest.

[devbence]

Sounds to be power source problem. Have you tried metering?

[itdaniher]

The issues I had with DMA Ch0 crashing fedora manifested when I had a current-limited LED attached, so power source issues seem less likely. That said, I'm pretty much out of ideas, so I'll check it out.

[itdaniher]

Bump? Still open issue.