Prevent random floats from occasionally being greater than 1. #2939
Conversation
This file contains hidden or 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
Previously, gen_f64 could generate numbers as high as 1.0000000002328306 in the case that u3 was 4294967295.0f64 and u2 was nonzero. This change divides the random numbers by 2**32 instead, effectively concatenating their bits as apparently intended. (Bonus fix: const.) The comments are updated to be more specific than "random float"; note that this can still generate 1.0f64 (P = 2**-54) due to rounding.
|
Passes |
|
I shudder to think the sort of bug this might have manifested in. |
|
Not actually encountered outside the microbenchmark thing I wrote in response to finding it, thankfully; I found it when reading the library source to see if the vague “a random float” meant [0,1] or what. (Although there was a can't-happen case that I handled instead of |
catamorphism
added a commit
that referenced
this pull request
Jul 20, 2012
Prevent random floats from occasionally being greater than 1.
|
Merged, thanks! |
celinval
added a commit
to celinval/rust-dev
that referenced
this pull request
Jun 4, 2024
This one was fairly straight forward. It still uses a few internal APIs for user friendly errors, for type layout, SIMD information and intrinsic validity.
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Outdated suggestions cannot be applied.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.
Previously, gen_f64 could generate numbers as high as 1.0000000002328306
in the case that u3 was 4294967295.0f64 and u2 was nonzero. This change
divides the random numbers by 2**32 instead, effectively concatenating
their bits as apparently intended. (Bonus fix: const.)
The comments are updated to be more specific than "random float"; note
that this can still generate 1.0f64 (P = 2**-54) due to rounding.
See also: https://gist.github.com/3127268