Add support for the rest of the megaAVR 0-series, the tinyAVR 0/1-series, and the new AVR Dx-series, make interrupt safe. #94
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Instead of checking for the single 0-series megaAVR that is most common, check for all tinyAVR 0/1-series, megaAVR 0-series (__AVR_ARCH__103) as well as the Dx series (__AVR_ARCH__=104), since they use the same kind of peripherals
Use the VPORTx registers - should give slightly smaller, faster, atomic code. Previous implementation used the PORTx registers; those are not in the I/O space, so |= and &= were not atomic! They still may not be in all cases (depending on what the compiler is able to optimize away) - but this will transparently take advantage of updates to the macros supplied with the cores which do, if/when they come.
Fix -fpermissive warning
|
Just discovered something: AVR_ARCH is 102, 103, or 104 for the new Dx series parts, not always 104. 103 which is what it is on the tinyAVR/megaAVR 0/1-series as well as 32k Dx parts; these can map all of their flash in the data space at once. 102 is for 64k (two sections), and 104 is for 128k parts (which have to split their flash into 4 sections) Updated pr to account for this discovery. |
|
Please add this support for the new AVRs in the next version. Thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
These all have the same patterns in register locations, so can use the same formulae for them.
This also switches to using the VPORT registers instead of the PORT registers, which makes setting or clearing a single bit into an atomic, single word, single cycle operation (SBI/CBI), as opposed to requiring a read-modify-write cycle like the PORT.OUT and PORT.DIR registers do, making those operations interrupt safe like they were on classic AVRs, as well as reducing sketch size and executing faster. Taking advantage of the consistent numbering of ports returned by digitalPinToPort() among the cores supporting these parts to get the base register address shrinks the compiled sketch size a touch too.
Non-pins passed to digitalPinToPort() will get back NOT_A_PORT, which is defined as 255, would give base reg of 0x03FC, which falls into one of the many gaps in the register space on these parts as well - so it won't fiddle with some random other peripheral somewhere (as it happens, it's right before where the PORT registers start at 0x0400, which seems fitting for "port -1")