On 04/22/2015 06:30 AM, Alexandre Belloni wrote: Apologies on a tardy response, got dragged into another issue and got cooped up in lab whole day.
> On 21/04/2015 at 20:59:15 -0500, Nishanth Menon wrote : >>>> Why is that so? when set alarm is requested for time X, you want >>>> interrupt at time X, not an interrupt for previous configured RTC >>>> alarm time! >>>> >>> >>> You expect at least an interrupt. >> >> And you will get an interrupt if the event occurs before the i2c burst >> starts. Once the i2cburst does start, you are committing to the new time. >> > > You mean that even if ALM0EN is set after ALM0IF was set to 1, it will > trigger the interrupt? I had a look at the MFP output block diagram > would let me think that this is the case. I was thinking otherwise > before. If that is so, then indeed, your patch is OK. At least based on my testing it seems to be the case, and as you can see in the block diagram, the ALM0EN mux comes after ALM0IF.. agreed that I am not sure the mux to have some internal buffers/latches etc.. dont have access to rtl to make more comments. > > My concern was about the time between ds1307->write_block_data() and > i2c_smbus_write_byte_data() which actually calls cond_sched(). > > I fully agree that your patch doesn't change the behaviour for the other > cases you presented and further clean up is to be done in a separate set > of patches. > Can I take this as an Acked-by? -- Regards, Nishanth Menon -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
