crenshaw/emul: added printing of data section (assuming it's following the code)

3 files changed, 31 insertions, 3 deletions

diff --git a/crenshaw/README b/crenshaw/README
index a64a6f5..201ffcd 100644
--- a/crenshaw/README
+++ b/crenshaw/README
@@ -133,3 +133,10 @@ Checking for LF feels hacky.
 Mentions now the lexer. The mean reason for a lexer is to keep the
 parser simple, for instance it can work on one lookahead 'character'
 like GREATER_EQUALS instead of individually '>=' and '=>'.
+
+tutor4
+
+Good point, even an expression parser can use an interpreter to
+simplify constant expressions before generating code for them.
+The other option is to let the programmer use the final constant
+and add a comment for it.
diff --git a/crenshaw/emul.c b/crenshaw/emul.c
index f1dcad9..727acdb 100644
--- a/crenshaw/emul.c
+++ b/crenshaw/emul.c
@@ -57,14 +57,27 @@ static void dump_stack( uc_engine *uc )
 {
 	uint32_t esp;
 	uint8_t mem[4];
+		
 	uc_reg_read( uc, UC_X86_REG_ESP, &esp );
 	
+	printf( "stack:\n" );
 	for( int i = esp; i < STACK_START; i += 4 ) {
 		uc_mem_read( uc, i, &mem, 4 );
 		printf( "%08X: %02X%02X%02X%02X\n", i, mem[3], mem[2], mem[1], mem[0] ); 
 	}
 }
 
+static void dump_memory( uc_engine *uc, uint64_t start_address, uint64_t end_address )
+{
+	uint8_t mem[4];
+
+	printf( "data:\n" );
+	for( uint64_t a = start_address; a < end_address; a += 4 ) {
+		uc_mem_read( uc, a, &mem, 4 );
+		printf( "%08X: %02X%02X%02X%02X\n", (uint32_t)a, mem[3], mem[2], mem[1], mem[0] ); 
+	}
+}
+
 static uint32_t mul_hash( uint64_t x, int p )
 {
 	uint32_t v = x * INT32_C( 2654435761 );
@@ -94,6 +107,7 @@ int main( int argc, char *argv[] )
 	cs_err cerr;
 	size_t nof_instrs;
 	cs_insn *instrs;
+	uint64_t data_start = 0;
 
 	uerr = uc_open( UC_ARCH_X86, UC_MODE_32, &uc );
 	if( uerr != UC_ERR_OK ) {
@@ -184,7 +198,14 @@ int main( int argc, char *argv[] )
 		for( int j = ( 16 - instrs[i].size ) * 2; j > 0; j-- ) {
 			printf( " " );
 		}
-		printf( "%s %s\n", instrs[i].mnemonic, instrs[i].op_str );
+		if( instrs[i].size == 2 && instrs[i].bytes[0] == 0 && instrs[i].bytes[1] == 0 ) {
+			if( data_start == 0 ) {
+				data_start = instrs[i].address;
+			}
+			printf( "data\n" );
+		} else {
+			printf( "%s %s\n", instrs[i].mnemonic, instrs[i].op_str );
+		}
 	}
 	
 	// remember address to instrs indexes so we can get the current
@@ -233,7 +254,6 @@ int main( int argc, char *argv[] )
 	uc_reg_write( uc, UC_X86_REG_ESP, &esp );
 	
 	uint64_t address = CODE_START;
-	//~ cs_insn *instr = cs_malloc( cs );	
 	bool terminate = false;
 	int iteration = 1;
 
@@ -278,6 +298,7 @@ int main( int argc, char *argv[] )
 		
 		dump_regs( uc );
 		dump_stack( uc );
+		dump_memory( uc, data_start, CODE_START + code_size );
 		
 		if( strcmp( instrs[n].mnemonic, "hlt" ) == 0 ) {
 			terminate = true;
diff --git a/crenshaw/main.pas b/crenshaw/main.pas
index 294d5a9..13499b0 100644
--- a/crenshaw/main.pas
+++ b/crenshaw/main.pas
@@ -283,7 +283,7 @@ begin
 	for i := 0 to nof_symbols-1 do begin
 		if symbols[i].sym_type = VariableType then begin
 			EmitLabel(symbols[i].name);
-			EmitLn('dw 0');
+			EmitLn('dd 0');
 		end;
 	end;
 end;