[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#584273: previous dmesg logs were incomplete



On Fri, 2010-06-04 at 11:55 -0300, Rajiv Andrade wrote:
> What's happening here is that this TPM seems to not be setting a proper
> status flag after a command is sent to it, so the device driver keeps
> waiting for such status flag to be set and then timeouts the attempt.
> 
> The TPM has three timeout values, and the longest one is by default
> 2min, given the TPM can take a good amount of time to perform asymmetric
> operations. If the command being sent to the TPM isn't recognized by the
> device driver, it assumes the timeout should be the maximum.
[...]

Blocking for up to 2 minutes in a probe function is totally
unacceptable, let alone the current 6 minutes.  If you really think the
time-out needs to be that high, then this initialisation should be moved
to a work item that is scheduled by the probe function.

Ben.

-- 
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: