you can probably make an unrolling impl by recursion + inlining

M

Maxine•10/22/23, 1:57 AM

my solution, since this method is only meant to handle scalar lists of length 1 to 4, is to hardcode the equality checks

M

Maxine•10/22/23, 1:57 AM

god help me

D

dfkeenan🍍🍕•10/22/23, 2:00 AM

congrats on getting v 2.18 out

D

dfkeenan🍍🍕•10/22/23, 2:01 AM

time to update "all" my things

PPerksey you can probably make an unrolling impl by recursion + inlining

M

Maxine•10/22/23, 2:15 AM

hard to tell, but i think this impl is just recursing

M

Maxine•10/22/23, 2:15 AM

maybe i'm just doing the pattern slightly wrong and the JIT doesn't get it

M

Maxine•10/22/23, 2:26 AM

should we have interfaces for the static methods on primitives? i guess i can implement them explicitly to discourage direct use

D

dfkeenan🍍🍕•10/22/23, 2:50 AM

Had no dramas updating. (Though nuget restore went wonky and I had to update twice.

)

M

Maxine•10/22/23, 3:12 AM

another optimization dead-end:

static unsafe bool IsCanonical<T>(Vector3D<T> value) where T : INumberBase<T> => FastAndDelegateForScalarEntries<Vector3D<T>, T>(value, &T.IsCanonical);

internal static unsafe bool FastAndDelegateForScalarEntries<T, TElem>(in T self, delegate*<TElem, bool> toCall)
{
  ... etc
}

static unsafe bool IsCanonical<T>(Vector3D<T> value) where T : INumberBase<T> => FastAndDelegateForScalarEntries<Vector3D<T>, T>(value, &T.IsCanonical);

internal static unsafe bool FastAndDelegateForScalarEntries<T, TElem>(in T self, delegate*<TElem, bool> toCall)
{
  ... etc
}

static unsafe bool IsCanonical<T>(Vector3D<T> value) where T : INumberBase<T> => FastAndDelegateForScalarEntries<Vector3D<T>, T>(value, &T.IsCanonical);

internal static unsafe bool FastAndDelegateForScalarEntries<T, TElem>(in T self, delegate*<TElem, bool> toCall)
{
  ... etc
}

static unsafe bool IsCanonical<T>(Vector3D<T> value) where T : INumberBase<T> => FastAndDelegateForScalarEntries<Vector3D<T>, T>(value, &T.IsCanonical);

internal static unsafe bool FastAndDelegateForScalarEntries<T, TElem>(in T self, delegate*<TElem, bool> toCall)
{
  ... etc
}

for T=

int

int

, JIT attemps to lower it, but doesn't realize that int.IsCanonical always returns true, so the entire method should become

mov eax, 1; ret 0xc

mov eax, 1; ret 0xc

mov eax, 1; ret 0xc

mov eax, 1; ret 0xc

M

Maxine•10/22/23, 3:13 AM

tanner pls fix /j

M

Maxine•10/22/23, 3:27 AM

are static virtual interface methods meant to not be callable from inside the derived type?

K

khitiara•10/22/23, 3:32 AM

i think they should work?

K

khitiara•10/22/23, 3:32 AM

not 100% sure

P

PerkseyOP•10/22/23, 3:35 AM

@grifon.petre how much bandwidth do you have atm

P

PerkseyOP•10/22/23, 3:36 AM

i want to nerdsnipe someone else into writing a competing solution to the overload problem

P

PerkseyOP•10/22/23, 3:36 AM

so that we have multiple prototypes to try out in a short timeframe

Kkhitiara i think they should work?

M

Maxine•10/22/23, 3:37 AM

it don't work!

K

khitiara•10/22/23, 3:37 AM

huh

P

PerkseyOP•10/22/23, 3:37 AM

you need a TSelf

K

khitiara•10/22/23, 3:37 AM

that almost feels like a bug lol

M

Maxine•10/22/23, 3:37 AM

yeah its such a strange design decision

K

khitiara•10/22/23, 3:37 AM

the static abstracts can be called from the member type but also i guess those have a separate declaration on that type

M

Maxine•10/22/23, 3:37 AM

yeah

M

Maxine•10/22/23, 3:38 AM

i wanted to use interfaces to reuse code

M

Maxine•10/22/23, 3:38 AM

that ain't happening now

M

Maxine•10/22/23, 3:38 AM

if c# had templates this shit would be done in minutes

PPerksey i want to nerdsnipe someone else into writing a competing solution to the overlo...

P

PerkseyOP•10/22/23, 3:39 AM

so the glass in my mind has been shattered a little bit

P

PerkseyOP•10/22/23, 3:39 AM

and... it's workable but it's making me realise just how ridiculous it is what we're doing

P

PerkseyOP•10/22/23, 3:39 AM

i still abstain that this is the only way without excessive overloads to achieve the level of flexibility we want

P

PerkseyOP•10/22/23, 3:39 AM

but okay here it goes

P

PerkseyOP•10/22/23, 3:39 AM

the Ptr type

P

PerkseyOP•10/22/23, 3:39 AM

how do you store one?

P

PerkseyOP•10/22/23, 3:40 AM

like in a class

M

Maxine•10/22/23, 3:40 AM

not meant to

P

PerkseyOP•10/22/23, 3:40 AM

but

P

PerkseyOP•10/22/23, 3:40 AM

then we have no safe storage medium

M

Maxine•10/22/23, 3:40 AM

you store either a T* or a T[], and that gets cast to Ptr<T> implicitly

P

PerkseyOP•10/22/23, 3:40 AM

ok but consider glfwCreateWindow

P

PerkseyOP•10/22/23, 3:40 AM

it returns a pointer

M

Maxine•10/22/23, 3:41 AM

it should instead return some sort of handle to GLFWwindow

M

Maxine•10/22/23, 3:41 AM

that can be cast to ptr<T> as well

P

PerkseyOP•10/22/23, 3:42 AM

yeah that's where i'm at. making a

StoredPtr

StoredPtr

StoredPtr

StoredPtr type (or add a

Untracked

Untracked

Untracked

Untracked variant of each type so you can just do

PtrPtr<T>.Untracked

PtrPtr<T>.Untracked

PtrPtr<T>.Untracked

PtrPtr<T>.Untracked)

P

PerkseyOP•10/22/23, 3:42 AM

but that's kinda crap

P

PerkseyOP•10/22/23, 3:42 AM

and you'd also have to assume that every return pointer is a stored pointer and not one we need to track

P

PerkseyOP•10/22/23, 3:42 AM

which is a reasonable assumption

M

Maxine•10/22/23, 3:42 AM

Ptr<T> is meant to be an input type not an output type, because the return value of glfwCreateWindow will never be a T[]

P

PerkseyOP•10/22/23, 3:42 AM

but it just feels meh

M

Maxine•10/22/23, 3:42 AM

in fact, you could just as well return T* directly

P

PerkseyOP•10/22/23, 3:42 AM

but then we'd get all the unsafe whiners back on our case

M

Maxine•10/22/23, 3:43 AM

i thought the point was not safety but the inconvenience of having to get a pointer to T[] to call a method that takes T*

you can probably make an unrolling impl by recursion + inlining

Similar Threads