interupt PINOUT arduino MEGA is wrong

[naghoss]

in first page for example INT4 is pin2 on arduino but in second page says is pin19 on arduino.....realy pin 19 === INT4....pinout is wrong...is not it???

https://www.arduino.cc/en/uploads/Hacking/PinMap2560big.png
https://www.arduino.cc/en/Reference/AttachInterrupt

[naghoss]

how can i report it to ARDUINO for Edit it???

[matthijskooijman]

I think you're confused by the interrupt numbering. "Interrupt 4" in the reference means "Pass 4 to attachInterrupt, which unfortunately does not correspond to INT4 in the atmega chip reference (attachInterrupt does some remapping of its own, presumably for backward compatibility to keep "interrupt 0" and "interrupt 1" on pin 2 & 3).

It might be good to add a remark about this to the reference though. For example, just above the table with pin numbers, add something like:

Note that in the table below, the interrupt numbers refer to the number to be passed to attachInterrupt(). For historical reasons, this numbering does not always correspond directly to the interrupt numbering on the atmega chip (e.g. int.0 correspons to INT4 on the Atmega2560 chip).

[naghoss]

so you say INT4 in atmega chip reference is equal to INT0 on attachInterrupt??
and what about T0 & T1 ,...(Timer Interrupt)? they are same in atmega chip and arduino board?

[naghoss]

and is a picture that "show" interrupt pin and timer interrupt pin?

[naghoss]

i got it tanks about your reply  M.B.S mobile num :09177610424 On Thursday, January 12, 2017 1:50 PM, Matthijs Kooijman <notifications@github.com> wrote: I think you're confused by the interrupt numbering. "Interrupt 4" in the reference means "Pass 4 to attachInterrupt, which unfortunately does not correspond to INT4 in the atmega chip reference (attachInterrupt does some remapping of its own, presumably for backward compatibility to keep "interrupt 0" and "interrupt 1" on pin 2 & 3).It might be good to add a remark about this to the reference though. For example, just above the table with pin numbers, add something like: Note that in the table below, the interrupt numbers refer to the number to be passed to attachInterrupt(). For historical reasons, this numbering does not always correspond directly to the interrupt numbering on the atmega chip (e.g. int.0 correspons to INT4 on the Atmega2560 chip). — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

[oqibidipo]

Here is the pin to interrupt mapping:

pin	INTx	digitalPinToInterrupt(pin)
		
3  	INT5	1
2  	INT4	0		
18 TX1	INT3	5
19 RX1	INT2	4
20 SDA	INT1	3
21 SCL	INT0	2

[CSTaylorJr]

Given the above pin to interrupt mapping, how come the digitalPinToPCICR(p) macro in pins_arduino.h (under the directory .\Arduino\hardware\arduino\avr\variants\mega) lists the following:

#define digitalPinToPCICR(p) ( (((p) >= 10) && ((p) <= 13)) ||
(((p) >= 50) && ((p) <= 53)) ||
(((p) >= 62) && ((p) <= 69)) ? (&PCICR) : ((uint8_t *)0) )

Shouldn't it be:

#define digitalPinToPCICR(p) ( (((p) >= 2) && ((p) <= 3)) ||
(((p) >= 18) && ((p) <= 21)) ? (&PCICR) : ((uint8_t *)0) )

This appears to be a bug?

[oqibidipo]

No error there (other than the omission of pins 0, 14 and 15*). Pin change interrupts (PCINT) are not the same as INTx. Up to 8 pins share one PCINT, and as the name suggests, it is triggered by any change to any pin that has PCINT enabled.

* The digitalPintoPCI* macros were originally part of the SoftwareSerial library, which explains this comment in variants/mega/pins_arduino.h

// (I've deliberately left out pin mapping to the Hardware USARTs - seems senseless to me)

[CSTaylorJr]

Ah! Got it!

So SoftwareSerial only uses PCINTs? It doesn't use INTx ints?

[oqibidipo]

That's correct.

[CSTaylorJr]

Good to know. Thanks! :)

[agdl]

I added the sentence @matthijskooijman suggested. Since doubts are cleared, can I close this issue?