馃 AI text below 馃
Parent: #2250
Scope
Audit MLIR tests and diagnostics for isolated in-process GoogleTest coverage, meaningful valid and invalid cases, semantic IR assertions, actionable locations and messages, and genuine CLI-only subprocess tests.
Verify that the documented mqt-cc debugging workflow covers reduced inputs, single-pass pipelines, generic printing, IR before and after failures, diagnostic stack traces, disabled threading, dialect-conversion tracing, and crash reproducers. Mark debug-build-only options accurately.
Delivery rules
Report critical gaps before v4 and fix behavioral defects with focused tests. Do not introduce lit or FileCheck, and do not broadly restyle existing tests.
馃 AI text below 馃
Parent: #2250
Scope
Audit MLIR tests and diagnostics for isolated in-process GoogleTest coverage, meaningful valid and invalid cases, semantic IR assertions, actionable locations and messages, and genuine CLI-only subprocess tests.
Verify that the documented
mqt-ccdebugging workflow covers reduced inputs, single-pass pipelines, generic printing, IR before and after failures, diagnostic stack traces, disabled threading, dialect-conversion tracing, and crash reproducers. Mark debug-build-only options accurately.Delivery rules
Report critical gaps before v4 and fix behavioral defects with focused tests. Do not introduce
litor FileCheck, and do not broadly restyle existing tests.