#4178 Floating expressions initialize constexpr integers without a diagnostic

[Christopher Bazley]

1: Sample code that reproduces the problem.

constexpr int object = 1.0;

2: Exact command used to run SDCC on this sample code.

sdcc -mz80 --std=c2y --nostdinc -S bug.c -o bug.asm

3: SDCC version tested.

SDCC 4.6.4 #17064 (Linux), tested 21 October 2026.

4: Observed and expected behaviour.

SDCC exits successfully and emits no diagnostic. A diagnostic is required by
WG14 N3886,
6.7.2 paragraph 6.

An explicit initialiser for a constexpr integer object or sub-object must be an integer constant expression. SDCC accepts 1.0, 1.5 and a floating expression initialising an integer member. The 1.0 example isolates this requirement from representability: its numerical value fits int exactly, but its expression is not an integer constant expression. The integer expression 1 is a valid case.

5: Additional checks.

The sample is silently accepted on Z80, MCS51 default, STM8 default, PDK14 and MOS65C02 in C2Y mode. These are translation and diagnostic checks; no runtime result is claimed.

Compiler Explorer: SDCC 4.6.0, SDCC trunk and GCC 16.2. All three panes use the sample above. The saved source, compiler selections and options were round-trip verified on 21 October 2026. Both SDCC panes use -mz80 --std=c2y; GCC uses -std=c2y -pedantic-errors -fsyntax-only. The public trunk build is r17013, distinct from the local r17064 retest.

CE compiler	Observed result
SDCC 4.6.0	Accepted silently
SDCC trunk r17013	Accepted silently
GCC 16.2	Rejected with diagnostics

Case	Expected	Current SDCC results
constexpr-integer-control.c	Valid	Accepted silently on all five targets
constexpr-integer-fraction.c	Diagnostic required	Accepted silently on all five targets
constexpr-integer-integral-float.c	Diagnostic required	Accepted silently on all five targets
constexpr-integer-member.c	Diagnostic required	Accepted silently on all five targets

[Christopher Bazley]

Please consider the attached candidate bugfix patch authored with Codex. It introduces a new diagnostic message: E_CONSTEXPR_INTEGER: "initialiser for constexpr integer object is not an integer constant expression".
The neighbouring TODO about loss of range or precision in constExprValue appears to be more closely related to [bugs:#4191] than this bug.
"warning 158: overflow in implicit constant conversion" is already emitted for integer constant expressions that exceed the range of the constexpr being initialised. Missing diagnostics for constexpr initialisers for bool type were reported separately as [bugs:#4230].