On Friday 18 April 2008 04:15:53 Andy Walls wrote:
> Hans,
>
> In the following 4 messages I'm submitting I2C related patches for
> the cx18 driver. Some of these patches are cleaned up versions of
> ones I posted earlier to the list.
>
> The patches build upon each other and must be applied in order. They
> were broken up to facilitate inspection of one changeset at time.
>
> Summary of patches:
> Patch 1/4: Collapse per I2C bus callback functions into a common set
Good one.
> Patch 2/4: Force PCI MMIO posted writes to complete for time
> sensitive i2c bus line manipulation
I do not believe that this will make any difference. I have tried this
in the past, but without any improvement.
In particular I do not like the extra DEBUG_WARN code: this should
really never happen and AFAIK can never happen. It's OK for people to
test, but I do not agree with putting this in the final driver.
> Patch 3/4: Perform more extensive I2C bus normalization and slave
> reset
This is interesting. I'd be very curious to hear whether this will fix
things!
> Patch 4/4: Fix I2C timing constants
No problem.
> I have tested all of these and they don't introduce any problems for
> me. My hope is that patches 2/4 and 3/4 fix the I2C problems HVR-1600
> users have been experiencing, or at least shed some light on the
> symptoms with additional debug messages.
Everyone who has i2c problems: please try these patches and send
feedback!
BTW: it is likely that the cx18 driver will go into 2.6.26, even if
there are still some outstanding issues. So that is good news.
Hans
_______________________________________________
ivtv-devel mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-devel