New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
System: UCD driver is claiming device but raw access required for workarounds #751
Comments
This may not be an bug. May be related with #653 |
This is normal system behaviour when a driver is loaded and bound to the address on the bus. As the kernel driver is in control of the device, you see UU at the address when using i2cdetect. We need to provide Paul with documentation on how to use the pmbus sysfs files to perform the tests that he is doing. What are these tests? Which bits of information is he looking for? |
Here are the UCD i2c commands that are failing. How do we do this with UCD driver installed?
|
@paulgrobertson I know a way to do the test, but it could be a bit complicated: remove |
The purpose of the above sequence of i2c command is to reconfigure the UCD to delete the VCSA and VCSB voltage rails from sequencer, reducing total rails from 16 to 14. Does the ucd9xxx driver even have the functional capability to do this type of reconfiguration? How? |
I think you're correct that the driver does not provide the functionality we need. While I work through the issue, you can unbind the driver from the device. This allows you to use your i2cset command again. I assume once the host has been brought up you could re-bind the driver. To unbind:
To re-bind once you're finished with the sequence:
I could not test this, so I am guessing at the |
I emailed @paulgrobertson with these details |
unbind/bind work around verified successfully. |
Once we provide official power on support in the code, will this i2c command be required anymore (i.e. is there anything to do with this bug or do we just need to get official power on support in)? |
Yes. VCS problem not fixed until dd2.0 |
A solution has been in place for the past several months |
Paul Robertson is seeing this issue:
Having problems with some devices on I2C12 after loading latest BMC image. Being unable to talk to UCD chip, dev 0x64, causing problems with power on script.
Following is the output (correct one) with older driver:
The text was updated successfully, but these errors were encountered: