[jogibear9988]

Indexer reads with a try/finally argument currently return null from strict compilation. Use a small operand-emission helper to spill the receiver and indices when the existing collection metadata marks that index expression.

Receiver and index evaluation order, cleanup, null receivers, throwing indices and supported compiler flags. The parity test now expects strict compilation for reads; assignment fallback expectations remain. 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	1170.7 → 1155.5	2.4 → 2.4	872.0 → 872.0	0.0 → 0.0
Standard / list_read	1755.9 → 1893.1	2.3 → 2.2	952.0 → 952.0	0.0 → 0.0
Standard / list_read_unrelated_try	2477.4 → 2709.9	2.2 → 2.2	1552.0 → 1552.0	0.0 → 0.0
Light / add_control	1293.2 → 1527.3	2.4 → 2.5	872.0 → 872.0	0.0 → 0.0
Light / list_read	1720.3 → 1747.9	2.7 → 2.5	952.0 → 952.0	0.0 → 0.0
Light / list_read_unrelated_try	2216.8 → 2218.8	2.2 → 2.3	1552.0 → 1552.0	0.0 → 0.0

net10.0 results (master → change)

API / workload	Compile ns	Invoke ns	Compile B/op	Invoke B/op
Standard / add_control	3763.3 → 3480.0	2.3 → 2.2	872.0 → 872.0	0.0 → 0.0
Standard / list_read	4320.7 → 4379.8	2.3 → 1.9	952.0 → 952.0	0.0 → 0.0
Standard / list_read_unrelated_try	5810.0 → 6270.3	2.0 → 2.0	1552.1 → 1552.1	0.0 → 0.0
Light / add_control	2906.3 → 2832.6	2.2 → 2.1	872.0 → 872.0	0.0 → 0.0
Light / list_read	3260.9 → 3404.7	1.9 → 1.9	952.0 → 952.0	0.0 → 0.0
Light / list_read_unrelated_try	3541.4 → 3794.5	1.8 → 1.8	1552.0 → 1552.0	0.0 → 0.0

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