Hey Jean, > I've finally renamed this to I801_INTR_EN for consistency. > > Ok > There's some redundancy between I801_HST_STS_MASK_NORM and > STATUS_FLAGS... It would be great to clean this up a bit, otherwise it > will become confusing. My impression is that your > I801_HST_STS_MASK_NORM and my STATUS_FLAGS are essentially the same. It > should be OK to handle SMBHSTSTS_BYTE_DONE in the interrupt handler. > That bit is set between the bytes of block transactions when the block > buffer isn't used. Your code doesn't enable interrupts for these > transactions at the moment, so you will never see that bit for now, but > it might happen later as we add support. > > So I think I would change the code as follows: > > /* Mask for events we normally handle */ > #define STATUS_FLAGS (SMBHSTSTS_BYTE_DONE | SMBHSTSTS_FAILED | \ > SMBHSTSTS_BUS_ERR | SMBHSTSTS_DEV_ERR | \ > SMBHSTSTS_INTR) > > /* Mask for all events */ > #define STATUS_FLAGS_ALL (STATUS_FLAGS | SMBHSTSTS_SMBALERT_STS) > > If that's OK with you, I'll do that change myself. > > Sure, less code and better readability for basically the same functionality. Thanks. > You obviously meant "use_irq", not "force". > > Ugh, yes. Glad you're paying more attention than I seem to ;p > All the rest looks fine to me now. I'll do some more testing now. > > Great and thanks. Am really curious about the stability of this code :) > Thanks, > Ivo
_______________________________________________ i2c mailing list [email protected] http://lists.lm-sensors.org/mailman/listinfo/i2c
