This is the mail archive of the ecos-patches@sourceware.org mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: i2c interface polled/IT driven transfer



Personally I'm not sure there's a need. The driver can automatically adapt between polled and interrupt-driven mode with a test of whether global interrupts are enabled or not. If only one or other mode is desired than per-driver CDL can eliminate unwanted code. This is what I've done in some of my own (non-public) drivers.


It seems better to me for the decision between polled vs interrupt driven not to have to be explicitly chosen by higher level code, particularly since that higher layer code may itself be generic (not part of the application) and be even less aware of global interrupt state.


OK, checking global interrupt is OK for me, if you think. At least, docu writing can be skipped :)
Peter



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]