| commit | 2d01479c13a71fabaa6a2ffefe4a1d8713ef63ad | |
| author | Anthony Liguori <aliguori@us.ibm.com> | |
| Tue, 9 Nov 2010 22:57:49 +0000 (9 16:57 -0600) | ||
| committer | Anthony Liguori <aliguori@us.ibm.com> | |
| Tue, 9 Nov 2010 22:57:49 +0000 (9 16:57 -0600) | ||
| tree | 460855240baacfa8929a6f714f7702f79182f2cc | treesnapshot (tar.gz zip) |
| parent | cd8f2841cdd854b9f96b662c7416cad79f23f8ac | commitdiff |
mc146818a: be less clever in computing next period timeout
Most likely through evolution, the code programs the next period based on the
last period which effectively gives it a default behavior of
DRIFT_REINJECT_FAST. That said, this behavior is subtle and the code reads like
it actually implements DRIFT_DROP because current_time doesn't always refer to
the current time.
Refactor to make this more obvious and subsequently simplify in the process.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Most likely through evolution, the code programs the next period based on the
last period which effectively gives it a default behavior of
DRIFT_REINJECT_FAST. That said, this behavior is subtle and the code reads like
it actually implements DRIFT_DROP because current_time doesn't always refer to
the current time.
Refactor to make this more obvious and subsequently simplify in the process.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
| pc/mc146818a.cpp | diffblobblamehistory |