ERROR::handle_hc_chhltd_intr_dma:2592: handle_hc_chhltd_intr_dma #372
Comments
|
Hi. Some more debug info would be nice. Please post the full output of sudo lsusb -v, and dmesg from boot until your breakage is experienced. Also, cgminer appears to be for ASICs (?) - how are you running it on an ARMv6? |
ghost
assigned 
|
cgminer ist for mining with USB Asic Block Erupter. comment was too long: >>> http://pastebin.com/MJiTqvQs |
|
I count no less than 17 usb-serial devices attached to 5 x 4-port hubs. I'm amazed that this even works. Can you post a lsusb -t as well? |
|
the 17 usb-serial devices are USB ASIC Miner. Meanwhile thi setup runs without any problems on my NSLU2 with Linux hostname 3.2.0-4-ixp4xx #1 Debian 3.2.46-1 armv5tel GNU/Linux; lsusb-t.txt with formating: http://pastebin.com/YhJ9fLtN |
|
Your NSLU2 uses a PCI-attached EHCI USB host. This is the same hardware you get on a PC. The Raspberry Pi has an OTG host which has significant hardware differences. Chief of these is that there can be only 8 active USB transactions at a single time - we cannot handle any more than these. Accessing 17 devices at once is likely to run out of resources. When you say the Pi freezes - how are you communicating with it? If USB dropped out then ethernet will drop out as well and SSH will fail. It could be that the Pi is perfectly fine, but has stopped talking to USB devices. The number of hubs you are using is a problem:
A better approach would be to buy 7-port hubs that actually have 7-port hub chips inside. That d-link hub contains 2x4-port hubs daisy chained together internally. In the meantime - flatten the tree as much as possible. Plug daisy-chained hubs only into the first 3 ports of the hubs to reduce the depth. |
|
Thanks for your fast replay. I don't know that the NIC ist conected via "USB" too. I setup a serial terminal and run it again... now at the and there are more informations. I know thats it is crazy to drive so many USB-devices ;). full output >> http://pastebin.com/8VpAbpTe << |
|
aha. So it is genuinely crashing. Why are there so many disconnect messages at 105 seconds? Are you physically unplugging the devices? The crash experienced is most likely due to memory corruption. The dwc_otg USB driver has been fairly well profiled in this regard - it's more probable that your serial port driver (cp210x) is doing something bad. Can you add |
With the option slub_debug=FPUZ nothing happend so far the system runs without any problems |
|
Is there anything in dmesg? (usually SLUB will be very noisy if something goes wrong that it fixes) Can you pastebin the contents of dmesg? |
|
There is still nothing special, i think. [ 0.000000] Booting Linux on physical CPU 0 |
|
Now sometinge happend: http://pastebin.com/Tq4ZuSm0 [ 100.181523] cp210x 1-1.2.2.1.1:1.0: device disconnected |
|
ThePi run the wohle night. This morning i try to connect via ssh, when it suddenly reboots :( [ 7974.816632] FIX kmalloc-256: Marking all objects used |
|
Odd. Reboots like that are most often power-related. Out of interest, are you able to successfully use all 17 mining devices simultaneously? |
|
Yes, that ist no problem. All 17 miner are running, until the pi reboots :) The Pi is powered by a Samsung Galaxy Charger about 1,2A, all hubs are powered, too |
|
Just out of interest, can you post the output of |
|
I'm having similar issue, 24 miners, runs < an hour. I can get the raspberry to lock within a minute if I try to stress it with a compile (http://www.instructables.com/id/Free-Landline-Using-a-Google-Voice/?ALLSTEPS) at same time. Hope this helps! |
|
I have this problem too. When I run cgminer on my raspberry pi it will randomly crash with a "Unable to handle kernel paging request at virtual address" error. I can see this error only if I plug the hdmi cable since the pi freezes and the only thing I can do is unplug it and plug it back in. I've tried 2 different pis, with 2 different SD cards, both with wired and wirless ethernet. I have two devices plugged in the USB ports and I've tried 2 different hubs in addition to plugging in the devices directly. The devices are 2 4-board bitburners setups (bitcoin mining ASICs). I at a loss as to how to fix this. |
|
output of: |
|
output of: Linux bluepi 3.6.11+ #557 PREEMPT Wed Oct 2 18:49:09 BST 2013 armv6l GNU/Linux vcgencmd version Oct 18 2013 16:07:43 |
|
can you test with the next tree too? |
|
CGminer still crashes the Pi. Always did. |
|
Possibly related, issue #205, I'm now seeing these "ERROR::handle_hc_chhltd_intr_dma" messages (together with NYET/NACK/ACK) after upgrading to 3.10.18+. |
|
Has anybody resolved this issue? I have the same problem with RPi with cgminer and USB bi*fury (2 bit fury asic's). It hangs randomly the pi with the Unable to handle kernel paging request at virtual address error. Is it cgminer error or what? |
|
I fixed it by adding "slub_debug=FP" to /boot/cmdline.txt . |
|
I already have slub_debug=FPUZ and nothing changed... Do you have an ethernet connection or wifi pn USB? Maybe this can be an issue? |
|
Ethernet here... |
|
I've tried both wifi and ethernet. Same result. On Sat, Nov 23, 2013 at 5:14 PM, Kafkamorph [email protected]:
|
|
Please retry with See http://www.raspberrypi.org/forum/viewtopic.php?f=28&t=70437 for info. |
|
Other Github issues reporting related problems note that they have gone away with fiq_fsm. No response from OP in 2 weeks: closing. |
Hi,
my Raspberry Pi frezzes when using cgminer after a few minutes. Without using cgminer it runs stabel for weeks. In order to get the Pi working agan i have to disconnect an reconnect power.
uname -a:
Linux enigmaRPi 3.6.11+ #538 PREEMPT Fri Aug 30 20:42:08 BST 2013 armv6l GNU/Linux
/var/log/kern.log:
Sep 6 18:52:53 enigmaRPi kernel: [ 457.067163] ERROR::handle_hc_chhltd_intr_dma:2592: handle_hc_chhltd_intr_dma: Channel 3, DMA Mode -- ChHltd set, but reason for halting is unknown, hcint 0x00000002, intsts 0x04600001
Sep 6 18:52:53 enigmaRPi kernel: [ 457.067163]
Sep 6 18:53:56 enigmaRPi kernel: [ 519.853693] ERROR::handle_hc_chhltd_intr_dma:2592: handle_hc_chhltd_intr_dma: Channel 1, DMA Mode -- ChHltd set, but rea
son for halting is unknown, hcint 0x00000002, intsts 0x04600001
Sep 6 18:53:56 enigmaRPi kernel: [ 519.853693]
Sep 6 18:59:45 enigmaRPi kernel: [ 868.280027] ERROR::handle_hc_chhltd_intr_dma:2592: handle_hc_chhltd_intr_dma: Channel 2, DMA Mode -- ChHltd set, but rea
son for halting is unknown, hcint 0x00000002, intsts 0x04600001
Sep 6 18:59:45 enigmaRPi kernel: [ 868.280027]
Sep 6 18:59:54 enigmaRPi kernel: [ 877.289319] ERROR::handle_hc_chhltd_intr_dma:2592: handle_hc_chhltd_intr_dma: Channel 5, DMA Mode -- ChHltd set, but rea
son for halting is unknown, hcint 0x00000002, intsts 0x04600001
Sep 6 18:59:54 enigmaRPi kernel: [ 877.289319]
Jan 1 01:00:20 enigmaRPi kernel: imklog 5.8.11, log source = /proc/kmsg started.
Jan 1 01:00:20 enigmaRPi kernel: [ 0.000000] Booting Linux on physical CPU 0
The text was updated successfully, but these errors were encountered: