Tim Kientzle
1d23852f2c
Float16 support only on Swift 5.3+
2025-10-10 17:22:04 -07:00
Tim Kientzle
7b55ac3094
Remove availability restrictions from these internal functions
...
This is possible because this replaces
* `InlineArray` with `internal _InlineArray`
* `UInt128` with `internal _UInt128`
And the replacements are not availability-restricted.
We don't need to do the same thing with `MutableSpan`
because that back-deploys to Swift 5.0, which is as
far back as the stdlib builds anyway.
2025-10-10 14:07:44 -07:00
Tim Kientzle
f0679b363e
Re-apply PR #82750 : Reimplement floating-point description implementation in Swift.
...
This reverts PR #84576 , which was a revert of PR #82750
It reverts commit 4ac18aa32eb46eddbabd
2025-10-07 07:11:57 -07:00
Tim Kientzle
7d85d75f78
Revert "Merge pull request #82750 from tbkka/tbkka-swift-floatingpointtostring"
...
This reverts commit 54627fb49bdda4608a84
2025-09-29 15:13:19 -07:00
Tim Kientzle
b761d8f743
Another try at fixing the ABI change
2025-09-24 09:37:50 -07:00
Tim Kientzle
9635ef51c5
Try to fix up availability of legacy shim functions
2025-09-24 07:29:56 -07:00
Tim Kientzle
f8d9b52e7d
another attempt at fixing the float16 ABI break
2025-09-23 15:11:38 -07:00
Tim Kientzle
7a4bd46d4c
Adjust availability checks for float16
2025-09-23 08:07:04 -07:00
Tim Kientzle
a337e727ab
Try to fix up the availability of the Swift and C projections of the legacy ABI
2025-09-22 12:40:12 -07:00
Tim Kientzle
e943878f9a
Prefill the work buffer with '0' characters
...
For example, this avoids the need to do any explicit byte-by-byte
writes when expanding "123" out to "123000000.0".
This also required reworking the "back out extra digits" process
for Float64 to ensure the unused digits get written as '0' characters
instead of null bytes.
2025-09-22 12:40:12 -07:00
Tim Kientzle
1a3a40003c
Minor safety and performance tweaks
2025-09-22 12:40:11 -07:00
Tim Kientzle
ae75afbb88
Try to fix a crash only seen so far on WASI32
2025-09-22 12:40:11 -07:00
Tim Kientzle
31930123e5
Remove extra semicolons
2025-09-22 12:40:11 -07:00
Tim Kientzle
acd43ad355
Fix Float32 formatting on 32-bit architectures
2025-09-22 12:40:11 -07:00
Tim Kientzle
a6fd1aa955
Fix build when Int is 16 bits
2025-09-22 12:40:11 -07:00
Tim Kientzle
2bf38bf578
Remove a layer of indirection
2025-09-22 12:40:11 -07:00
Tim Kientzle
de7c280e9c
Try to match stdlib coding conventions
2025-09-22 12:40:10 -07:00
Tim Kientzle
cf552ea9ce
Match up conditional for Float80 support
2025-09-22 12:40:10 -07:00
Tim Kientzle
8e40ac8568
Fix availability
2025-09-22 12:40:10 -07:00
Tim Kientzle
d9907324a1
Comment updates
2025-09-22 12:40:09 -07:00
Tim Kientzle
e3990c6085
Float80 performance
2025-09-22 12:40:09 -07:00
Tim Kientzle
7be098f839
Initial Float80 support + some general reorganization and commenting
2025-09-22 12:40:09 -07:00
Tim Kientzle
175e1d78b3
Float32/64 safety improvements
2025-09-22 12:40:09 -07:00
Tim Kientzle
b50a2e6ead
Float16 tweaks: balancing safety and performance
2025-09-22 12:40:09 -07:00
Tim Kientzle
070fd6006e
Add Float16 support to the Swift implementation
2025-09-22 12:40:08 -07:00
Tim Kientzle
4cf03a47bc
Reimplement floating-point description implementation in Swift.
...
Note: This is still incomplete. I need to finish porting the Float16
and Float80 support before it can fully replace the existing
C implementation.
2025-09-22 12:40:08 -07:00
