By far, one of the most common things to check for is whether or not a
value is zero, as it typically allows folding away unnecesary
operations (other close contenders that can help with eliding operations are 1 and -1).
So instead of requiring a check for an immediate and then actually
retrieving the integral value and checking it, we can wrap it within a
function to make it more convenient.
This'll make it slightly nicer to do basic constant folding for 32-bit
and 64-bit variants of the same IR opcode type. By that, I mean it's
possible to inspect immediate values without a bunch of conditional
checks beforehand to verify that it's possible to call GetU32() or
GetU64, etc.
This is useful when we wish to know if a contained value is something
like 0xFFFFFFFF, as this helps perform constant folding. For example the
operation: x & 0xFFFFFFFF can be folded to just x in the 32-bit case.
An arithmetic shift is by definition a signed shift, and a logical shift is by definition an unsigned shift.
- Rename VectorLogicalVShiftS* -> VectorArithmeticVShift*
- Rename VectorLogicalVShiftU* -> VectorLogicalVShift*
Fixes decoding for 64-bit instructions
Does not help/apply to any currently supported ARM versions (since
all are 32-bit length or below), it's for future-proofing should
such an arch be supported.
Makes the comparison interface consistent by providing all of the
relevant members. This also modifies the comparison operators to take
the Imm instance by value, as it's really only a u32 under the covers,
and it's cheaper to shuffle around a u32 than a 64-bit pointer address.
Moves the declaration of datasize to the scope that it's used within.
This also takes the opportunity to apply const where applicable, and
make early-exits all vertically consistent with one another.
* Return both the upper and lower parts of the multiply if required
* SSE2 does not support the pmuldq instruction, do sign correction to an unsigned result instead
* Improve port utilisation where possible (punpck instructions were a bottleneck)
We can optimize these cases further for with the use of a fair bit of
shuffling via pshufb and the use of masks, but given the uncommon use of
this instruction, I wouldn't consider it to be beneficial in terms of
amount of code to be worth it over a simple manageable naive solution
like this.
If we ever do hit a case where vectorized CLZ happens to be a
bottleneck, then we can revisit this. At least with AVX-512CD, this can
be done with a single instruction for the 32-bit word case.
We can simply modify the general function made for SMULL{2} and
UMULL{2}'s by-element variants to also handle the other multiply-based
by-element variants.
