build: preserve main package identity for test builds

[visualfc]

Summary

• Preserve the main package identity during test builds instead of renaming it to main.test.
• Add an internal @fortest marker to distinguish test-main packages when resolving ABI paths.
• Keep c-shared test exports pointed at main.init and main.main.
• Add linkname coverage for same-package, external-test, and main-package references.
• Update the stdlib build-mode test runner for test binaries that retain the main package identity.
• Keep generated test library names package-specific, such as libtar.test.
• Link runners against main.init and main.main rather than package-path-derived test symbols.

Testing

• go test ./test/linkname
• LLGO=./dev/llgo.sh dev/test_std_buildmodes.sh ./test/std/archive/tar

[github-actions]

LLGo baseline benchmarks

6348e54e0e57 | workflow run | long-term charts

Program measurements

Platform	Workload	File size	vs base	Build	vs base	Run	vs base
Linux	cprintf	19288 B	0 B / +0.0%	338.411 ms	-8.723 ms / -2.5% (better)	1.420 ms	+53.45 us / +3.9% (worse)
Linux	fmtprintf	1860168 B	0 B / +0.0%	2.716 s	-148.9 ms / -5.2% (better)	3.561 ms	-84.89 us / -2.3% (better)
Linux	println	68776 B	0 B / +0.0%	338.392 ms	-17.98 ms / -5.0% (better)	1.728 ms	+41 us / +2.4% (worse)
macOS	cprintf	84672 B	0 B / +0.0%	445.286 ms	+92.45 ms / +26.2% (worse)	3.151 ms	+908.5 us / +40.5% (worse)
macOS	fmtprintf	1892720 B	0 B / +0.0%	2.975 s	+349.5 ms / +13.3% (worse)	15.315 ms	+4.028 ms / +35.7% (worse)
macOS	println	121360 B	0 B / +0.0%	355.779 ms	+5.88 ms / +1.7% (worse)	3.464 ms	+189.5 us / +5.8% (worse)

Core language and compiler benchmarks

Platform	Benchmark	ns/op	vs base
Linux	BenchmarkLookupPCRandom	12.250 ns/op	-0.01 ns/op / -0.1% (better)
Linux	BenchmarkMergeCompilerFlags	144 ns/op	-3.5 ns/op / -2.4% (better)
Linux	BenchmarkMergeLinkerFlags	93.790 ns/op	-1.51 ns/op / -1.6% (better)
Linux	BenchmarkChannelBuffered	36.330 ns/op	-0.09 ns/op / -0.2% (better)
Linux	BenchmarkChannelHandoff	24506 ns/op	+369 ns/op / +1.5% (worse)
Linux	BenchmarkDefer	45.960 ns/op	-0.44 ns/op / -0.9% (better)
Linux	BenchmarkDirectCall	1.758 ns/op	+0.001 ns/op / +0.1% (worse)
Linux	BenchmarkGlobalRead	1.758 ns/op	0 ns/op / +0.0%
Linux	BenchmarkGlobalWrite	2.805 ns/op	0 ns/op / +0.0%
Linux	BenchmarkGoroutine	30571 ns/op	+285 ns/op / +0.9% (worse)
Linux	BenchmarkInterfaceCall	8.450 ns/op	+0.004 ns/op / +0.04736% (worse)
Linux	BenchmarkRuntimeGetG	2.112 ns/op	-0.001 ns/op / -0.04733% (better)
macOS	BenchmarkLookupPCRandom	11.840 ns/op	-0.2 ns/op / -1.7% (better)
macOS	BenchmarkMergeCompilerFlags	167.600 ns/op	+43.8 ns/op / +35.4% (worse)
macOS	BenchmarkMergeLinkerFlags	82.870 ns/op	+14.57 ns/op / +21.3% (worse)
macOS	BenchmarkChannelBuffered	21.360 ns/op	+0.86 ns/op / +4.2% (worse)
macOS	BenchmarkChannelHandoff	6574 ns/op	+849 ns/op / +14.8% (worse)
macOS	BenchmarkDefer	28.780 ns/op	+2.04 ns/op / +7.6% (worse)
macOS	BenchmarkDirectCall	0.981 ns/op	+0.0387 ns/op / +4.1% (worse)
macOS	BenchmarkGlobalRead	0.994 ns/op	+0.0177 ns/op / +1.8% (worse)
macOS	BenchmarkGlobalWrite	0.986 ns/op	+0.0441 ns/op / +4.7% (worse)
macOS	BenchmarkGoroutine	48831 ns/op	+25189 ns/op / +106.5% (worse)
macOS	BenchmarkInterfaceCall	4.511 ns/op	+0.419 ns/op / +10.2% (worse)
macOS	BenchmarkRuntimeGetG	1.961 ns/op	-0.251 ns/op / -11.3% (better)

Compared with 0314faeba69d measured in the same runner job.

[fennoai]

Review Summary

This PR reworks how //go:linkname resolves into test-compiled main packages. It replaces the previous approach (renaming the package to "main.test", which mutated the real package name) with a scope sentinel: an @ForTest marker constant is inserted into the package scope when a main package is built for test, and ssa/abi/abi.go PathOf() consults that marker to return the real import path instead of collapsing to "main".

The approach is sound and a genuine improvement over mutating Types.Name():

• The @ prefix is deliberately un-spellable as a Go identifier, so the sentinel cannot collide with, or be spoofed by, user code. No security issue.
• The added Scope().Lookup in PathOf is O(1) and guarded by Name() == "main" short-circuit, so it only runs for main packages. No performance concern.
• Removing the obsolete main.test rename (and its test) is correct hygiene; the surviving .test suffix checks operate on pkg.ID/pkg.PkgPath, not the removed rename, so they are unaffected.

Findings below are maintainability/robustness improvements, not blockers. The new test/linkname/ regression tests are a good addition.

[codecov]

Codecov Report

❌ Patch coverage is 81.81818% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
internal/build/build.go	85.71%	0 Missing and 1 partial ⚠️
test/linkname/main.go	66.66%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!