diff --git a/examples/0045-basic-string-eq-short-circuit.sx b/examples/0045-basic-string-eq-short-circuit.sx new file mode 100644 index 0000000..5f0ed2b --- /dev/null +++ b/examples/0045-basic-string-eq-short-circuit.sx @@ -0,0 +1,53 @@ +// String `==`/`!=` as an operand of a short-circuit `and`/`or`. +// +// A string compare lowers to its own multi-block memcmp sub-CFG, so the +// operand finishes in a later basic block than the one the short-circuit +// started in. The `and`/`or` merge PHI must take that actual block as the +// incoming predecessor. +// +// Regression (issue 0078): combining string equality with `and`/`or` used to +// emit invalid LLVM (`PHI node entries do not match predecessors!`). +#import "modules/std.sx"; + +Json :: enum { + str: string; + int_: s64; + null_; +} + +main :: () { + a := "k"; + b := "v"; + + // string == on both sides of `and` + and_tt := a == "k" and b == "v"; + and_tf := a == "k" and b == "x"; + and_ft := a == "z" and b == "v"; + print("and: {} {} {}\n", and_tt, and_tf, and_ft); + + // string == on both sides of `or` + or_ff := a == "z" or b == "x"; + or_tf := a == "k" or b == "x"; + or_ft := a == "z" or b == "v"; + print("or: {} {} {}\n", or_ff, or_tf, or_ft); + + // string == feeding both `and` and `or` in one expression + mixed := a == "k" and b == "v" or a == "z"; + print("mixed: {}\n", mixed); + + // string `!=` operands too + ne := a != "z" and b != "z"; + print("ne: {}\n", ne); + + // the larger shape: a match-expression value plus an enum-payload string + // == combined under `and`/`or`. + v : Json = .str("v"); + kind := if v == { + case .str: 1; + case .int_: 2; + case .null_: 3; + } + ok := kind == 1 and v.str == "v"; + bad := kind == 2 or v.str == "x"; + print("payload: {} {}\n", ok, bad); +} diff --git a/examples/expected/0045-basic-string-eq-short-circuit.exit b/examples/expected/0045-basic-string-eq-short-circuit.exit new file mode 100644 index 0000000..573541a --- /dev/null +++ b/examples/expected/0045-basic-string-eq-short-circuit.exit @@ -0,0 +1 @@ +0 diff --git a/examples/expected/0045-basic-string-eq-short-circuit.stderr b/examples/expected/0045-basic-string-eq-short-circuit.stderr new file mode 100644 index 0000000..8b13789 --- /dev/null +++ b/examples/expected/0045-basic-string-eq-short-circuit.stderr @@ -0,0 +1 @@ + diff --git a/examples/expected/0045-basic-string-eq-short-circuit.stdout b/examples/expected/0045-basic-string-eq-short-circuit.stdout new file mode 100644 index 0000000..1c9fcca --- /dev/null +++ b/examples/expected/0045-basic-string-eq-short-circuit.stdout @@ -0,0 +1,5 @@ +and: true false false +or: false true true +mixed: true +ne: true +payload: true false diff --git a/issues/0078-string-eq-operand-of-short-circuit-and-invalid-phi.md b/issues/0078-string-eq-operand-of-short-circuit-and-invalid-phi.md index e5139cf..352a595 100644 --- a/issues/0078-string-eq-operand-of-short-circuit-and-invalid-phi.md +++ b/issues/0078-string-eq-operand-of-short-circuit-and-invalid-phi.md @@ -1,3 +1,21 @@ +# 0078 — string `==` as an `and`/`or` operand emits an invalid PHI + +> **RESOLVED.** Root cause was in the LLVM emitter, not the `and`/`or` +> lowering: `fixupPhiNodes` wired each short-circuit merge PHI's incoming +> edge to `block_map[ir_block]` — the LLVM block the IR block *started* as. +> But a single IR instruction can expand into its own sub-CFG during +> emission (string `==`'s `str.memcmp`/`str.merge` blocks; a value `match`'s +> arm blocks), leaving the builder in a later block. The terminator — and +> therefore the real predecessor edge — lands in that later block, so the +> recorded predecessor was stale (`%entry`/`%and.rhs.0` instead of +> `%str.merge`). Fix: in `src/ir/emit_llvm.zig`, record the builder's +> *actual* insertion block after emitting each IR block's instructions +> (`term_block_map`, captured via `LLVMGetInsertBlock`) and use that as the +> PHI predecessor in `fixupPhiNodes`. General — corrects the incoming block +> for ANY operand that emitted intermediate basic blocks, not just string +> `==`. Mirrors the issue-0066 "stale PHI incoming-block after an operand +> emits new blocks" shape. Regression: `examples/0045-basic-string-eq-short-circuit.sx`. + # Symptom A string equality (`a == "x"`) used as an operand of a short-circuit diff --git a/src/ir/emit_llvm.zig b/src/ir/emit_llvm.zig index 183892c..e246e59 100644 --- a/src/ir/emit_llvm.zig +++ b/src/ir/emit_llvm.zig @@ -124,6 +124,13 @@ pub const LLVMEmitter = struct { // Maps (func_idx, block_idx) → LLVM BasicBlock block_map: std.AutoHashMap(u64, c.LLVMBasicBlockRef), + // For each IR block, the LLVM block its terminator was actually emitted + // into. Usually equals `block_map[block]`, but an instruction can expand + // into its own sub-CFG (string `==`'s memcmp blocks, a value `match`'s + // arm blocks) and leave the builder in a later block, so the terminator — + // and therefore the PHI predecessor edge — lands there instead. Keyed the + // same way as `block_map`. + term_block_map: std.AutoHashMap(u64, c.LLVMBasicBlockRef), // Cached LLVM types cached_i1: c.LLVMTypeRef, @@ -278,6 +285,7 @@ pub const LLVMEmitter = struct { .func_map = std.AutoHashMap(u32, c.LLVMValueRef).init(alloc), .global_map = std.AutoHashMap(u32, c.LLVMValueRef).init(alloc), .block_map = std.AutoHashMap(u64, c.LLVMBasicBlockRef).init(alloc), + .term_block_map = std.AutoHashMap(u64, c.LLVMBasicBlockRef).init(alloc), .pending_phis = std.ArrayList(PendingPhi).empty, .cached_i1 = c.LLVMInt1TypeInContext(ctx), .cached_i8 = c.LLVMInt8TypeInContext(ctx), @@ -311,6 +319,7 @@ pub const LLVMEmitter = struct { self.jni_slots.deinit(); self.global_map.deinit(); self.block_map.deinit(); + self.term_block_map.deinit(); self.di_files.deinit(); var fsc_it = self.frame_str_cache.keyIterator(); while (fsc_it.next()) |k| self.alloc.free(k.*); @@ -1270,6 +1279,7 @@ pub const LLVMEmitter = struct { // Clear pending phis for this function self.pending_phis.clearRetainingCapacity(); + self.term_block_map.clearRetainingCapacity(); // Emit instructions for each block — use first_ref to sync ref numbering for (func.blocks.items, 0..) |block, bi| { @@ -1285,6 +1295,12 @@ pub const LLVMEmitter = struct { _ = inst_i; self.emitInst(&instruction, func_idx); } + + // The terminator may have landed in a later LLVM block than `bb` + // if an instruction in this IR block expanded into its own sub-CFG. + // Record where the builder actually is so PHI predecessors point at + // the block that holds the branch, not the block we started in. + self.term_block_map.put(block_key, c.LLVMGetInsertBlock(self.builder)) catch unreachable; } // Fixup PHI nodes: scan all blocks for branches that pass args @@ -1300,7 +1316,10 @@ pub const LLVMEmitter = struct { for (func.blocks.items, 0..) |block, bi| { const src_key = makeBlockKey(func_idx, @intCast(bi)); - const src_bb = self.block_map.get(src_key) orelse continue; + // Predecessor is the block the terminator was emitted into, which + // differs from `block_map[bi]` when an instruction expanded the + // block into a sub-CFG (string `==`, value `match`, …). + const src_bb = self.term_block_map.get(src_key) orelse continue; for (block.insts.items) |instruction| { switch (instruction.op) {