gbdev/rgbds
1
0
Fork 0
mirror of https://github.com/gbdev/rgbds.git synced 2025-11-20 10:12:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

Implement warning diagnostic flags for RGBLINK (#1754)

Browse Source
This commit is contained in:
Rangi
2025-07-17 12:54:29 -04:00
committed by GitHub
parent 529989bde5
commit 5eb093f13e
15 changed files with 332 additions and 55 deletions
Download Patch File Download Diff File Expand all files Collapse all files

119
src/link/patch.cpp
View File

@@ -33,6 +33,25 @@ static void pushRPN(int32_t value, bool comesFromError) {
// has popped any values with the error flag set.
static bool isError = false;
#define diagnosticAt(patch, id, ...) \
do { \
bool errorDiag = warnings.getWarningBehavior(id) == WarningBehavior::ERROR; \
if (!isError || !errorDiag) { \
warningAt(patch, id, __VA_ARGS__); \
} \
if (errorDiag) { \
isError = true; \
} \
} while (0)
#define firstErrorAt(...) \
do { \
if (!isError) { \
errorAt(__VA_ARGS__); \
isError = true; \
} \
} while (0)
static int32_t popRPN(Patch const &patch) {
if (rpnStack.empty()) {
fatalAt(patch, "Internal error, RPN stack empty");
@@ -98,23 +117,26 @@ static int32_t computeRPNExpr(Patch const &patch, std::vector<Symbol> const &fil
case RPN_DIV:
value = popRPN(patch);
if (value == 0) {
if (!isError) {
errorAt(patch, "Division by 0");
isError = true;
}
firstErrorAt(patch, "Division by 0");
popRPN(patch);
value = INT32_MAX;
value = 0;
} else if (int32_t lval = popRPN(patch); lval == INT32_MIN && value == -1) {
diagnosticAt(
patch,
WARNING_DIV,
"Division of %" PRId32 " by -1 yields %" PRId32,
INT32_MIN,
INT32_MIN
);
value = INT32_MIN;
} else {
value = op_divide(popRPN(patch), value);
value = op_divide(lval, value);
}
break;
case RPN_MOD:
value = popRPN(patch);
if (value == 0) {
if (!isError) {
errorAt(patch, "Modulo by 0");
isError = true;
}
firstErrorAt(patch, "Modulo by 0");
popRPN(patch);
value = 0;
} else {
@@ -127,10 +149,7 @@ static int32_t computeRPNExpr(Patch const &patch, std::vector<Symbol> const &fil
case RPN_EXP:
value = popRPN(patch);
if (value < 0) {
if (!isError) {
errorAt(patch, "Exponent by negative value %" PRId32, value);
isError = true;
}
firstErrorAt(patch, "Exponent by negative value %" PRId32, value);
popRPN(patch);
value = 0;
} else {
@@ -204,14 +223,49 @@ static int32_t computeRPNExpr(Patch const &patch, std::vector<Symbol> const &fil
case RPN_SHL:
value = popRPN(patch);
if (value < 0) {
diagnosticAt(
patch, WARNING_SHIFT_AMOUNT, "Shifting left by negative amount %" PRId32, value
);
}
if (value >= 32) {
diagnosticAt(
patch, WARNING_SHIFT_AMOUNT, "Shifting left by large amount %" PRId32, value
);
}
value = op_shift_left(popRPN(patch), value);
break;
case RPN_SHR:
case RPN_SHR: {
value = popRPN(patch);
value = op_shift_right(popRPN(patch), value);
int32_t lval = popRPN(patch);
if (lval < 0) {
diagnosticAt(patch, WARNING_SHIFT, "Shifting right negative value %" PRId32, lval);
}
if (value < 0) {
diagnosticAt(
patch, WARNING_SHIFT_AMOUNT, "Shifting right by negative amount %" PRId32, value
);
}
if (value >= 32) {
diagnosticAt(
patch, WARNING_SHIFT_AMOUNT, "Shifting right by large amount %" PRId32, value
);
}
value = op_shift_right(lval, value);
break;
}
case RPN_USHR:
value = popRPN(patch);
if (value < 0) {
diagnosticAt(
patch, WARNING_SHIFT_AMOUNT, "Shifting right by negative amount %" PRId32, value
);
}
if (value >= 32) {
diagnosticAt(
patch, WARNING_SHIFT_AMOUNT, "Shifting right by large amount %" PRId32, value
);
}
value = op_shift_right_unsigned(popRPN(patch), value);
break;
@@ -324,14 +378,13 @@ static int32_t computeRPNExpr(Patch const &patch, std::vector<Symbol> const &fil
case RPN_HRAM:
value = popRPN(patch);
if (value < 0 || (value > 0xFF && value < 0xFF00) || value > 0xFFFF) {
if (!isError) {
errorAt(patch, "Address $%" PRIx32 " for LDH is not in HRAM range", value);
isError = true;
}
firstErrorAt(patch, "Address $%" PRIx32 " for LDH is not in HRAM range", value);
value = 0;
} else if (value >= 0 && value <= 0xFF) {
warningAt(
patch, "LDH is deprecated with values from $00 to $FF; use $FF00 to $FFFF"
patch,
WARNING_OBSOLETE,
"LDH is deprecated with values from $00 to $FF; use $FF00 to $FFFF"
);
}
value &= 0xFF;
@@ -341,10 +394,7 @@ static int32_t computeRPNExpr(Patch const &patch, std::vector<Symbol> const &fil
value = popRPN(patch);
// Acceptable values are 0x00, 0x08, 0x10, ..., 0x38
if (value & ~0x38) {
if (!isError) {
errorAt(patch, "Value $%" PRIx32 " is not a RST vector", value);
isError = true;
}
firstErrorAt(patch, "Value $%" PRIx32 " is not a RST vector", value);
value = 0;
}
value |= 0xC7;
@@ -355,10 +405,7 @@ static int32_t computeRPNExpr(Patch const &patch, std::vector<Symbol> const &fil
int32_t mask = getRPNByte(expression, size, patch);
// Acceptable values are 0 to 7
if (value & ~0x07) {
if (!isError) {
errorAt(patch, "Value $%" PRIx32 " is not a bit index", value);
isError = true;
}
firstErrorAt(patch, "Value $%" PRIx32 " is not a bit index", value);
value = 0;
}
value = mask | (value << 3);
@@ -437,6 +484,7 @@ void patch_CheckAssertions() {
case ASSERT_WARN:
warningAt(
assert.patch,
WARNING_ASSERT,
"%s",
!assert.message.empty() ? assert.message.c_str() : "assert failure"
);
@@ -486,8 +534,8 @@ static void applyFilePatches(Section &section, Section &dataSection) {
uint16_t address = patch.pcSection->org + patch.pcOffset + 2;
int16_t jumpOffset = value - address;
if (!isError && (jumpOffset < -128 || jumpOffset > 127)) {
errorAt(
if (jumpOffset < -128 || jumpOffset > 127) {
firstErrorAt(
patch,
"JR target must be between -128 and 127 bytes away, not %" PRId16
"; use JP instead",
@@ -497,12 +545,13 @@ static void applyFilePatches(Section &section, Section &dataSection) {
dataSection.data[offset] = jumpOffset & 0xFF;
} else {
// Patch a certain number of bytes
if (!isError && (value < type.min || value > type.max)) {
errorAt(
if (value < type.min || value > type.max) {
diagnosticAt(
patch,
"Value %" PRId32 "%s is not %u-bit",
WARNING_TRUNCATION,
"Value $%" PRIx32 "%s is not %u-bit",
value,
value < 0 ? " (maybe negative?)" : "",
value < 0 ? " (may be negative?)" : "",
type.size * 8U
);
}