[PATCH] i2c: omap: fix OOPS in omap_i2c_unidle() during probe

2010-03-23 Thread Mika Westerberg
Commit d84d3ea317ce0db89ce0903b4037f800c5d4c477 added register shift to allow also 16-bit register access. However, omap_i2c_unidle() is called before these are set which causes the following OOPS: Unhandled fault: alignment exception (0x801) at 0xfa070009 Internal error: : 801 [#1]

Re: i2c probing

2010-03-23 Thread Jean Delvare
Hi Matthieu, On Sun, 21 Mar 2010 15:22:00 +0100, matthieu castet wrote: Do you know why there is 2 methods of probing i2c [1] and [2], with different quirks for eeprom. Historical reasons, mainly. i2c_detect() is an old thing, it is used by device drivers which want to automatically probe for

Re: [PATCH]I2C device - release cleanup

2010-03-23 Thread Jean Delvare
On Tue, 23 Mar 2010 13:02:41 +, jhautb...@gmail.com wrote: Hi Jean, snip Did you test your patch? I am very skeptical that calling single_release() out of the blue is the right thing to do. My instinct tells me that single_release() is only meant for callers of

[PATCH] i2c-imx: fix error handling

2010-03-23 Thread Arnaud Patard
- Return -ETIMEDOUT on bus busy error - Fix timeout test time_after(jiffies, orig_jiffies + HZ / 1000) : By default, HZ=100 on arm. This means that this test has no chances to work and may result in a dead loop. Set timeout to 500ms. - Don't try to send a new message if we failed to transmit