lazyweb, have you ever been in a situation with I2C where you needed to do a read, repeated start, then a write, and it had to be a repeated start specifically, not a stop + start? write then read is common (EEPROMs), read then write I've never seen before
-
-
that's exactly why I'm asking: I have a master that can't do it (FX2), at least not in a documented way (I -did- manage to exploit a race condition in their I2C peripheral to do it, but, ew), and I'm designing a software API right now
-
I'm curious what device demands this kind of transfer and for what purpose. Seems like one of those things where someone either found a really clever solution to a hard problem, or was just accidentally holding their copy of the spec upside-down at the time.
- Još 1 odgovor
Novi razgovor -
-
-
(Linux's i2c_msg interface _might_ be able to express it, I'm not sure)
-
yeah, I think you just put I2C_M_RD in the first message instead of second
Kraj razgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.