-
SPIR-V has:
- add-with-carry OpIAddCarry
- sub-with-borrow OpISubBorrow
- unsigned multiply (1 word * 1 word produces 2 words)
- signed multiply (1 word * 1 word produces 2 words)
MSL …
-
Currently, `AlgebraicValue::F32/F64` uses [`decorum::Total`](https://docs.rs/decorum/latest/decorum/type.Total.html) and `Total` as a way to get `Ord` and `PartialEq` implementations for floats. The `…
-
x87 (instructions list: [current](https://en.wikipedia.org/wiki/X86_instruction_listings#x87_floating-point_instructions), [obsolete Intel/IIT/Cyrix](https://en.wikipedia.org/wiki/List_of_discontinued…
-
This issue is meant to be a reminder for myself, if the idea takes in effect in mega65-core. More information: https://github.com/MEGA65/mega65-core/issues/826
In nutshell: a new proposal to allow …
-
The current tagged integer operations are kind of slow, especially things like addition and subtraction. I believe that we can make them significantly faster.
Here are a few ideas:
* Inline the fa…
-
Statement in question:
> ["But in practice, at least on the architectures I’ve tested, everything above the 48th bit in a pointer is always zero."](https://github.com/munificent/craftinginterpreter…
-
### Description
Currently optimization of array bounds checks involving a bitwise AND only seems to work when one operand is a constant, which limits the applicability, and leaves out an important …
-
In one of my cryptographic applications of Scryer Prolog, the bottleneck turns out to be **formatting large integers** in base 16, using `library(format)`.
We can use the `~Nr` format control …
-
When disassembling M68K code, I ran into problems within `Reko.Core.Expressions.ExpressionTypeAscenderBase.PullDiffDataType()` during the "Reconstruct data types" phase.
First, I ran into problem…
-
We've been bitten in the past (see [a comment thread on #3232](https://github.com/spinalcordtoolbox/spinalcordtoolbox/pull/3944#discussion_r1021709189), #3943, #3944, and many issues linked from these…