Hi Wolfgang,
On Fri, Apr 3, 2020 at 5:02 PM Wolfgang Wallner wolfgang.wallner@br-automation.com wrote:
Hi Andy,
-----"Andy Shevchenko" andriy.shevchenko@linux.intel.com schrieb: -----
The commit 720f9e1fdb0c ("Move PCI access from ofdata_to_platdata() to probe()") while doing formally a right thing, actually brings a regression to the drivers that would like to pre-initialize hardware before calling ns16550_serial_probe(). In particular, the code, which gets moved out, is responsible for getting the address of the hardware.
The commit breaks serial console on the Intel Edison, for example.
Since we are very close to the release, the quick fix is to revert the culprit commit 720f9e1fdb0c92d3fd16e1bfc25bcbd35612675c.
Note, it doesn't affect SoCFPGA case.
Cc: Wolfgang Wallner wolfgang.wallner@br-automation.com Signed-off-by: Andy Shevchenko andriy.shevchenko@linux.intel.com
v3: drop wrong patch, better explanation in the commit message drivers/serial/ns16550.c | 40 ++++++++++++---------------------------- 1 file changed, 12 insertions(+), 28 deletions(-)
Reviewed-by: Wolfgang Wallner wolfgang.wallner@br-automation.com
We can't revert as that will put PCI based ns16550 in a broken state again.
I've sent a patch to fix it. Please have a try.
Regards, Bin