Date: Mon, 2 Jun 2025 01:31:05 +0000
It tells compiler to change entry point of function in need (if we know but compiler doesn't know). Compiler can call function without known check-ups, or inline it, or just do nothing (sometimes assure doesn't useful), and if paraments follow the constacts, the result is same as the original. Optimization is up to compiler, compiler can ignore assure.
To use assure, we should know how a function works so that we can use assure correctly.
Anyway, thanks for your advice :)
________________________________
发件人: Std-Proposals <std-proposals-bounces_at_lists.isocpp.org> 代表 Jeremy Rifkin via Std-Proposals <std-proposals_at_[hidden]>
发送时间: 2025年6月2日 2:54
收件人: std-proposals_at_[hidden] <std-proposals_at_[hidden]>
抄送: Jeremy Rifkin <rifkin.jer_at_[hidden]>
主题: Re: [std-proposals] A Proposal about A New Keyword assure
This isn't something that would make sense for standardization. First it would need more compiler research. Just my two cents: This is approaching the problem from the wrong angle. Ideally you'd want inlining to address things like this, but also look into things like function cloning.
Cheers,
Jeremy
On Sun, Jun 1, 2025 at 8:19 AM SD SH via Std-Proposals <std-proposals_at_[hidden]<mailto:std-proposals_at_[hidden]>> wrote:
This proposal introduces a new contextual-keyword assure that allows callers to declare parament Constraints at function call sites. can utilizing these Constraints to optimize function calling. (e.g. removing branches) Violations result in undefined behavior. Analogous to [[assume]].
Motivation
Check-ups of function definition sites is unnessary at some call sites, but compilers may preserve check-ups for them. Sometimes we need more performance optimizations but we can't do many optimizations such as built-in functions and third-party libraries.
e.g. Lighting computing will call sqrt many times.
If sqrt(double _X) like this:
double __fastcall sqrt(double _X)
{
if(_X < 0.0)
{
return std::numeric_limits<double>::infinity();
}
else
{
// ...
}
}
// asm:
/*
_sqrt:
pxor xmm1, xmm1
ucomisd xmm0, xmm1
jl _sqrt__Xl0
; ...
_sqrt__Xl0:
movsd xmm0, 0x7FF8000000000000
ret
*/
In lighting computing, _X is always equals zero or greater than zero, so branch of if(_X < 0.0) is never performed there, but compiler may not remove it.
assure will slove this issue.
sqrt(d assure(d >= 0)) may generates
_sqrt:
pxor xmm1, xmm1
ucomisd xmm0, xmm1
jl _sqrt__Xl0
_sqrt_sometime:
; ...
ret
_sqrt__Xl0:
movsd xmm0, 0x7FF8000000000000
ret
;...
movsd xmm0, [somewhere]
call _sqrt_simetime
or inline call or a new function of sqrt, we needn't program another sqrt for this call site and compiler can optimize program better.
Design Decisions
Syntax
func(para1 assure(const-bool-expr(INPUT: { para1, const paras... })), paras...)
Rejected Alternatives:
func(paras...) assure(const-bool-expr(INPUT: const paras...))
makes compiler implementation difficult.
Semantics:
Constraints are purely caller-to-compiler optimization hints.
Expressions must be side-effect-free and dependent only on caller-visible state.
Expressions should be true, or function calling has undefined behavior
Alternative Considered:
Runtime assertions (for security) rejected due to performance impact.
Ecosystem Impact
User: Performance gains with constraints
Implementers: Medium or low implementation effort
STL: Greater room for optimization
View the full text (R0) on GitHub: CPP-Proposals/assure.txt<https://github.com/TheNameofSH/CPP-Proposals/blob/main/assure.txt>
Thank you!
E. S. Himers <Z5515zwy_at_[hidden]<mailto:Z5515zwy_at_[hidden]>>
--
Std-Proposals mailing list
Std-Proposals_at_[hidden]<mailto:Std-Proposals_at_lists.isocpp.org>
https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals
To use assure, we should know how a function works so that we can use assure correctly.
Anyway, thanks for your advice :)
________________________________
发件人: Std-Proposals <std-proposals-bounces_at_lists.isocpp.org> 代表 Jeremy Rifkin via Std-Proposals <std-proposals_at_[hidden]>
发送时间: 2025年6月2日 2:54
收件人: std-proposals_at_[hidden] <std-proposals_at_[hidden]>
抄送: Jeremy Rifkin <rifkin.jer_at_[hidden]>
主题: Re: [std-proposals] A Proposal about A New Keyword assure
This isn't something that would make sense for standardization. First it would need more compiler research. Just my two cents: This is approaching the problem from the wrong angle. Ideally you'd want inlining to address things like this, but also look into things like function cloning.
Cheers,
Jeremy
On Sun, Jun 1, 2025 at 8:19 AM SD SH via Std-Proposals <std-proposals_at_[hidden]<mailto:std-proposals_at_[hidden]>> wrote:
This proposal introduces a new contextual-keyword assure that allows callers to declare parament Constraints at function call sites. can utilizing these Constraints to optimize function calling. (e.g. removing branches) Violations result in undefined behavior. Analogous to [[assume]].
Motivation
Check-ups of function definition sites is unnessary at some call sites, but compilers may preserve check-ups for them. Sometimes we need more performance optimizations but we can't do many optimizations such as built-in functions and third-party libraries.
e.g. Lighting computing will call sqrt many times.
If sqrt(double _X) like this:
double __fastcall sqrt(double _X)
{
if(_X < 0.0)
{
return std::numeric_limits<double>::infinity();
}
else
{
// ...
}
}
// asm:
/*
_sqrt:
pxor xmm1, xmm1
ucomisd xmm0, xmm1
jl _sqrt__Xl0
; ...
_sqrt__Xl0:
movsd xmm0, 0x7FF8000000000000
ret
*/
In lighting computing, _X is always equals zero or greater than zero, so branch of if(_X < 0.0) is never performed there, but compiler may not remove it.
assure will slove this issue.
sqrt(d assure(d >= 0)) may generates
_sqrt:
pxor xmm1, xmm1
ucomisd xmm0, xmm1
jl _sqrt__Xl0
_sqrt_sometime:
; ...
ret
_sqrt__Xl0:
movsd xmm0, 0x7FF8000000000000
ret
;...
movsd xmm0, [somewhere]
call _sqrt_simetime
or inline call or a new function of sqrt, we needn't program another sqrt for this call site and compiler can optimize program better.
Design Decisions
Syntax
func(para1 assure(const-bool-expr(INPUT: { para1, const paras... })), paras...)
Rejected Alternatives:
func(paras...) assure(const-bool-expr(INPUT: const paras...))
makes compiler implementation difficult.
Semantics:
Constraints are purely caller-to-compiler optimization hints.
Expressions must be side-effect-free and dependent only on caller-visible state.
Expressions should be true, or function calling has undefined behavior
Alternative Considered:
Runtime assertions (for security) rejected due to performance impact.
Ecosystem Impact
User: Performance gains with constraints
Implementers: Medium or low implementation effort
STL: Greater room for optimization
View the full text (R0) on GitHub: CPP-Proposals/assure.txt<https://github.com/TheNameofSH/CPP-Proposals/blob/main/assure.txt>
Thank you!
E. S. Himers <Z5515zwy_at_[hidden]<mailto:Z5515zwy_at_[hidden]>>
--
Std-Proposals mailing list
Std-Proposals_at_[hidden]<mailto:Std-Proposals_at_lists.isocpp.org>
https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals
Received on 2025-06-02 01:31:17