[SR-10137] “class” and “final class” between swift 4.2 and swift 5 (performance) #52539
Comments
|
In the non-final code, SILGen uses a generalized accessor to modify the array: %148 = class_method %0 : $TestGlobalArr, #TestGlobalArr.arr!modify.1 : (TestGlobalArr) -> () -> (), $@yield_once @convention(method) (@guaranteed TestGlobalArr) -> @yields @inout Array<Int> // user: %149
(%149, %150) = begin_apply %148(%0) : $@yield_once @convention(method) (@guaranteed TestGlobalArr) -> @yields @inout Array<Int> // users: %152, %156
// function_ref Array.subscript.modify
%151 = function_ref @$sSayxSiciM : $@yield_once @convention(method) <τ_0_0> (Int, @inout Array<τ_0_0>) -> @yields @inout τ_0_0 // user: %152
(%152, %153) = begin_apply %151<Int>(%143, %149) : $@yield_once @convention(method) <τ_0_0> (Int, @inout Array<τ_0_0>) -> @yields @inout τ_0_0 // users: %154, %155
store %146 to %152 : $*Int // id: %154
end_apply %153 // id: %155
end_apply %150 // id: %156When that generalized accessor is inlined, we end up with an extra retain_value: bb8: // Preds: bb7
%48 = load %21 : $*Int // user: %54
%49 = ref_element_addr %0 : $TestGlobalArr, #TestGlobalArr.arr // user: %50
%50 = load %49 : $*Array<Int> // users: %54, %58, %51
retain_value %50 : $Array<Int> // id: %51That retain_value prevents the COWArray pass from hoisting the array's It seems there are multiple points at which the optimizations fall 1. Why does SILGen require the generalized accessor in WMO mode. 2. Why does using the generalized accessor (TestGlobalArr.arr!modify) cause an extra retain. |
|
(Just a drive by comment, but we may want to add this as a benchmark. |
|
Comment by Oleh (JIRA) @belkadan thanks for editing the tags and components of this issue |
Attachment: Download
Environment
Linux 156974-vds-admin.gmhost.pp.ua 4.9.0-3-amd64 #1 SMP Debian 4.9.30-2+deb9u5 (2017-09-19) x86_64 GNU/Linux
Swift version 5.0-dev (LLVM 3207a50965, Clang 107de882f4, Swift 4d6e741)
Target: x86_64-unknown-linux-gnu
Additional Detail from JIRA
md5: 731c650729f8698e9aed68f26ae13d74
Issue Description:
Sorry, my english is very bad...
Please tell me why this code(without "final class") executing with so big difference like 4 vs 25 seconds between versions 4.2 and 5 ?
swiftc -Ounchecked -whole-module-optimization -Xcc -O2
https://github.com/woodcrust/puffin_bench/blob/master/source/2-sortGlobalArray/swift2_final_class.swift 18
https://github.com/woodcrust/puffin_bench/blob/master/source/2-sortGlobalArray/swift2_class.swift 13
I saw article about Increasing Performance by Reducing Dynamic Dispatch
https://developer.apple.com/swift/blog/?id=27
But 4 vs 25 seconds on different versions ... looks like performance regression
For repeat this issue on linux
run.sh contains compile commands and execute test code
https://github.com/woodcrust/puffin_bench/blob/master/run.sh
The text was updated successfully, but these errors were encountered: