-
Notifications
You must be signed in to change notification settings - Fork 33
[RFC] Backports for v0.8.5 #293
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Closed
Closed
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
1863142
to
f132105
Compare
f132105
to
da29edc
Compare
This uses the type aliases for the message and avoids showing huge numbers. This also reduces the compilation time, i.e. the CI time.
This abstracts the selection of the types to be tested and makes it more systematic. As a result, we can reduce impractical test cases (e.g. for the 128-bit types) without significantly compromising the test quality. This also reorders some test sets.
da29edc
to
d37f075
Compare
This prevents `rand` from returning a `ReinterpretArray` to avoid the performance problem with `ReinterpretArray` . This also supports specifying the RNG option.
This displays the fixed point type of the input in the message, not the rawtype.
This adds `wrapping_mul` and `checked_mul` binary operations for `Normed`. This replaces most of Normed's implementation of multiplication with integer operations. This improves the speed in many cases and the accuracy in some cases.
)" This specializes most of the multiplication for `Fixed` and avoids floating point operations. A major change is that the rounding mode is changed from `RoundNearestTiesUp` to `RoundNearest`. The existing `RoundNearestTiesUp` and `RoundDown` modes are now supported by the new unexported function `mul_with_rounding`. This also improves `rem`. Unlike multiplication for `Normed`, the wrapping arithmetic is the default for `Fixed`.
…th#223) This reduces the environment-/optimization-dependent instability of the results for `NaN`. This also adds tests for `NaN`/`Inf`. This clarifies the behavior regarding `NaN` and `Inf`, which had not been explicitly defined.
d37f075
to
82e6199
Compare
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## release-0.8 #293 +/- ##
===============================================
+ Coverage 89.80% 96.98% +7.17%
===============================================
Files 6 7 +1
Lines 461 630 +169
===============================================
+ Hits 414 611 +197
+ Misses 47 19 -28 ☔ View full report in Codecov by Sentry. |
Closed
…or div/cld/fld (JuliaMath#226)" The wrapping and saturating `div`/`cld`/`fld` are not backported. The default arithmetic for `div` is still checked arithmetic. This changes the error type for overflow from `DivideError` to `OverflowError`. This also adds the support for `cld` and 3-arg `div`.
This removes `start`, `next` and `done`. They are no longer used in Julia v1 and were left as private methods of `FixedPointNumbers`.
The test runs on a docker container with QEMU. This also updates TagBot.yml.
This forces the `BigFloat`-->`Float16` conversion to go through `Float32` instead of `Float64`.
Co-authored-by: Johnny Chen <[email protected]> Co-authored-by: kimikage <[email protected]>
use Random explicitly
* Use ubuntu-latest for Arm CI (host and container) * Fix download URL of nightly build binary * Use julia-actions/setup-julia@v2 for the host and use the same version as the target
f77cad3
to
295b8d1
Compare
I am going to give this up. |
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 be backported
NotBiggerThanInt64
,nbitsint
,showtype
(#189
)throw_converterror
#205widen1(T::Type)
, tests (#207
)StableRNG
in tests #278unsafe_trunc
forBigFloat
#212Fixed
#218Fixed
#220% Normed{UInt32}
on ARM and ImproveNaN % FixedPoint
#223_unsafe_trunc
to reduce the likelihood of arbitrary values #291#222
)trunc
(#235
)QuoteNode
,return nothing
, tests(#243
)promote_type
#280StableRNG
in tests #278include
s of "test/common.jl." #286unsafe_trunc
returning arbitrary values #289_unsafe_trunc
to reduce the likelihood of arbitrary values #291workflow-dispatch
trigger to UnitTest.yml #295Not to be backported