Fix compilation of primitive types #98
Merged
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
This addresses the issue where e.g. 'long' and 'long long' are both 64 bit, but we only overload for int64_t which is either long or 'long long', meaning template instantiation will fail with the other type. Fixes #96
Instead of the libdivide namespace, use the libdivide::detail namespace, to limit what we inject.
|
@adbancroft if possible please check whether this works with your AVR setup, thank you |
The tests passed on both simulator & real hardware (AT Mega 2560) |
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Outdated suggestions cannot be applied.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.
To restate #96: currently libdivide "dispatches" based on whether its type is
uint32_t,uint64_t, etc. However these types are mapped to a primitive type likeunsigned int; if sayunsigned longhas the same width thendivider<unsigned long>will fail to compile since there is no specialization for it.The idea here is to go back to a dispatch based on the size and signedness of the type, so that
unsigned longandunsigned intwill be handled the same. To detect the signedness we avoidtype_traitswhich is apparently unavailable in AVR; instead we use the following trick:which will be true only for signed types. We can prevent floating point by throwing in a shift:
An inelegant hack for an uncivilized language but it does the job. Fixes #96.