[jogibear9988]

An ordered switch test containing try/finally can invalidate the switch comparison stack. Evaluate and store the test value before loading the saved switch operand, preserving test order and cleanup behavior.

Nullable and non-nullable switch values, matching and missing cases, throwing tests and cleanup order. Regressions use strict FEC compilation for both Standard and LightExpression.

Validation

Release solution build: zero warnings/errors.

Runtime	TestX groups	Legacy tests
.NET 10	443 passed	1,756 passed
.NET 8	443 passed	1,755 passed
.NET Framework 4.7.2	422 passed	1,639 passed

Measurements

Current master f1d2fb33 → this change, Release, Windows ARM64, .NET 8.0.28 / 10.0.9. Compilation uses prebuilt trees; runtime measures invocation through the same wrapper. Two fresh process rounds with reversed load order, seven samples per round, about 30 ms calibration, one logical CPU and tiered compilation disabled. Tables use the median of the two per-round medians.

The controls below produce correct results on master for their measured inputs. Newly supported trees have no valid before/after execution comparison because master rejects them, produces incorrect results or emits invalid IL. Timings include machine and JIT/code-placement variation; the unchanged add control helps show that variation. These observations do not prove universal performance neutrality.

net8.0 results (master → change)

API / workload	Compile ns	Invoke ns	Compile B/op	Invoke B/op
Standard / add_control	1191.2 → 1192.6	2.2 → 2.2	872.0 → 872.0	0.0 → 0.0
Standard / switch_runtime_int	1928.9 → 1802.1	2.2 → 2.3	1112.1 → 1112.1	0.0 → 0.0
Standard / switch_runtime_unrelated_try	2183.5 → 2435.1	2.1 → 2.2	1528.1 → 1528.1	0.0 → 0.0
Light / add_control	1345.7 → 1383.5	2.0 → 2.0	872.0 → 872.0	0.0 → 0.0
Light / switch_runtime_int	1573.4 → 1536.0	2.0 → 2.1	1112.0 → 1112.0	0.0 → 0.0
Light / switch_runtime_unrelated_try	2302.6 → 2631.2	2.5 → 2.5	1528.0 → 1528.0	0.0 → 0.0

net10.0 results (master → change)

API / workload	Compile ns	Invoke ns	Compile B/op	Invoke B/op
Standard / add_control	4120.7 → 4137.5	2.6 → 2.5	872.1 → 872.1	0.0 → 0.0
Standard / switch_runtime_int	6843.9 → 6672.4	2.4 → 2.4	1112.1 → 1112.1	0.0 → 0.0
Standard / switch_runtime_unrelated_try	5918.3 → 7581.3	2.1 → 2.1	1528.1 → 1528.1	0.0 → 0.0
Light / add_control	2629.6 → 3120.9	2.5 → 2.4	872.0 → 872.0	0.0 → 0.0
Light / switch_runtime_int	4156.4 → 4437.5	2.5 → 2.5	1112.0 → 1112.0	0.0 → 0.0
Light / switch_runtime_unrelated_try	4802.8 → 5268.4	2.1 → 2.0	1528.0 → 1528.0	0.0 → 0.0

Emitted IL differs in these measured workloads: none.
No measured allocation increase above 0.1 B/op in these controls.