From 794a49e9389b9f817cdf2828884c7202d7135bb7 Mon Sep 17 00:00:00 2001 From: agra Date: Tue, 19 May 2026 18:51:56 +0300 Subject: [PATCH] =?UTF-8?q?ffi=201.9:=204=C3=97f64=20HFA=20round-trip=20th?= =?UTF-8?q?rough=20#objc=5Fcall=20(UIEdgeInsets=20shape)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 105/105 regression tests pass (+ffi-objc-call-07-fp-hfa-return). Same round-trip pattern as 1.8 — register an Obj-C class at runtime with class_addMethod, IMP returns specific non-zero values, #objc_call reads them back — but for an all-double 32 B HFA instead of a 24 B int aggregate. Locks in the f32-vs-f64 landmine that bit us when we first wrote safeAreaInsets in uikit.sx: the homogeneous-float-aggregate ABI routes 1..4 f32 or f64 fields through v0..v3 (AAPCS64) / xmm0..xmm3 (SysV AMD64) WITHOUT integer coercion. As long as the LLVM call-site function type carries the precise struct (which our `objc_msg_send` arm does), the backend lowers it correctly. This is the smaller cousin of 1.8 — 1.8 needed an emit_llvm code change to make the sret transform work; 1.9 needs no codegen change because HFAs of any size up to v0..v3 stay register-resident. The test just pins that path with a real, value-bearing IMP so a future ABI-rule shake-up has a regression net. --- examples/ffi-objc-call-07-fp-hfa-return.sx | 46 +++++++++++++++++++ .../ffi-objc-call-07-fp-hfa-return.exit | 1 + .../ffi-objc-call-07-fp-hfa-return.txt | 2 + 3 files changed, 49 insertions(+) create mode 100644 examples/ffi-objc-call-07-fp-hfa-return.sx create mode 100644 tests/expected/ffi-objc-call-07-fp-hfa-return.exit create mode 100644 tests/expected/ffi-objc-call-07-fp-hfa-return.txt diff --git a/examples/ffi-objc-call-07-fp-hfa-return.sx b/examples/ffi-objc-call-07-fp-hfa-return.sx new file mode 100644 index 0000000..585683d --- /dev/null +++ b/examples/ffi-objc-call-07-fp-hfa-return.sx @@ -0,0 +1,46 @@ +// Phase 1 step 1.9 (PLAN-FFI.md): all-double HFA returns through +// `#objc_call`. 4×f64 = 32 B, stays in v0..v3 on AAPCS64 and +// xmm0..xmm3 on SysV AMD64 — same shape as UIEdgeInsets / NSRect / +// CGRect, the f32-vs-f64 landmine that bit us when we first wrote +// `safeAreaInsets` in uikit.sx. +// +// Nominally covered by ffi-objc-call-05's nil-recv NSRect case, +// but that only checks that zeros come back. Here we install a +// real IMP that returns specific non-zero values and verify each +// field comes through the float-register file intact. + +#import "modules/std.sx"; +#import "modules/compiler.sx"; +#import "modules/std/objc.sx"; + +UIEdgeInsets :: struct { + top: f64; + left: f64; + bottom: f64; + right: f64; +} + +insets_imp :: (self: *void, _cmd: *void) -> UIEdgeInsets callconv(.c) { + UIEdgeInsets.{ top = 1.5, left = 2.5, bottom = 3.5, right = 4.5 }; +} + +main :: () -> s32 { + inline if OS == .macos { + ns_object := objc_getClass("NSObject".ptr); + my_cls := objc_allocateClassPair(ns_object, "SxInsetsProbe".ptr, 0); + sel := sel_registerName("safeAreaInsets".ptr); + // Method type encoding: {UIEdgeInsets=dddd}@: → returns 4×f64, + // implicit (self: id, _cmd: SEL). `d` = double. + ok := class_addMethod(my_cls, sel, xx insets_imp, "{UIEdgeInsets=dddd}@:".ptr); + print("addMethod = {}\n", ok); + objc_registerClassPair(my_cls); + + instance := class_createInstance(my_cls, 0); + ins := #objc_call(UIEdgeInsets)(instance, "safeAreaInsets"); + print("insets = ({}, {}, {}, {})\n", ins.top, ins.left, ins.bottom, ins.right); + } + inline if OS != .macos { + print("skipped (not macos)\n"); + } + 0; +} diff --git a/tests/expected/ffi-objc-call-07-fp-hfa-return.exit b/tests/expected/ffi-objc-call-07-fp-hfa-return.exit new file mode 100644 index 0000000..573541a --- /dev/null +++ b/tests/expected/ffi-objc-call-07-fp-hfa-return.exit @@ -0,0 +1 @@ +0 diff --git a/tests/expected/ffi-objc-call-07-fp-hfa-return.txt b/tests/expected/ffi-objc-call-07-fp-hfa-return.txt new file mode 100644 index 0000000..3b9ef56 --- /dev/null +++ b/tests/expected/ffi-objc-call-07-fp-hfa-return.txt @@ -0,0 +1,2 @@ +addMethod = true +insets = (1.500000, 2.500000, 3.500000, 4.500000)