[dev.link] cmd/link: fix hash collision check

For content-addressable symbols, we build its content hash based
on the symbol data and relocations. When the compiler builds the
symbol data, it may not always include the trailing zeros, e.g.
the data of [10]int64{1,2,3} is only the first 24 bytes.
Therefore, we may end up with symbols with the same contents
(thus same hash) but different sizes. This is not actually a hash
collision. In this case, we can deduplicate them and keep the one
with the larger size.

Change-Id: If6834542d7914cc00f917d7db151955e5aee6f30
Reviewed-on: https://go-review.googlesource.com/c/go/+/243718
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Jeremy Faller <jeremy@golang.org>

1 files changed, 33 insertions, 0 deletions

diff --git a/src/cmd/link/testdata/testHashedSyms/p.go b/src/cmd/link/testdata/testHashedSyms/p.go
new file mode 100644
index 0000000000..87dddcfcac
--- /dev/null
+++ b/src/cmd/link/testdata/testHashedSyms/p.go
@@ -0,0 +1,33 @@
+// Copyright 2020 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// This test case contains two static temps (the array literals)
+// with same contents but different sizes. The linker should not
+// report a hash collision. The linker can (and actually does)
+// dedup the two symbols, by keeping the larger symbol. The dedup
+// is not a requirement for correctness and not checked in this test.
+// We do check the emitted symbol contents are correct, though.
+
+package main
+
+func main() {
+	F([10]int{1, 2, 3, 4, 5, 6}, [20]int{1, 2, 3, 4, 5, 6})
+}
+
+//go:noinline
+func F(x, y interface{}) {
+	x1 := x.([10]int)
+	y1 := y.([20]int)
+	for i := range y1 {
+		if i < 6 {
+			if x1[i] != i+1 || y1[i] != i+1 {
+				panic("FAIL")
+			}
+		} else {
+			if (i < len(x1) && x1[i] != 0) || y1[i] != 0 {
+				panic("FAIL")
+			}
+		}
+	}
+}