diff options
author | Alexander Monakov <amonakov@ispras.ru> | 2018-04-28 16:56:07 +0300 |
---|---|---|
committer | Wolfram Sang <wsa@the-dreams.de> | 2018-05-15 10:42:19 +0200 |
commit | 9f4659ba384780f8bc5b18717865026d83d455ce (patch) | |
tree | 36c570dbf4a689c24272cc5e564b3ab7cbbfd67c /drivers/message | |
parent | e6218bf390a9992de86a535bcb227f75e2c20cf9 (diff) |
i2c: designware: refactor low-level enable/disable
Low-level controller enable function __i2c_dw_enable is overloaded to
also handle disabling. What's worse, even though the documentation
requires polling the IC_ENABLE_STATUS register when disabling, this
is not done: polling needs to be requested specifically by calling
__i2c_dw_enable_and_wait, which can also poll on enabling, but that
doesn't work if the IC_ENABLE_STATUS register is not implemented.
This is quite confusing if not in fact backwards.
Especially since the documentation says that disabling should be
followed by polling, the driver should be using a separate function
where it does one-shot disables to make the optimization stand out.
This refactors the two functions so that requested status is given
in the name rather than in a boolean argument. Specifically:
- __i2c_dw_enable: enable without polling (in accordance with docs)
- __i2c_dw_disable: disable and do poll (also as suggested by docs)
- __i2c_dw_disable_nowait: disable without polling (Linux-specific)
No functional change.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
[wsa: fixed blank lines in header file]
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Diffstat (limited to 'drivers/message')
0 files changed, 0 insertions, 0 deletions