enum op_class {
OPC_UNSPEC,
- OPC_RMD, /* removed or optimized out */
- OPC_DATA, /* data processing */
- OPC_DATA_FLAGS, /* data processing + sets flags */
- OPC_JMP, /* .. and call */
- OPC_JCC,
+ OPC_RMD, /* removed or optimized out */
+ OPC_DATA, /* data processing */
+ OPC_DATA_FLAGS, /* data processing + sets flags */
+ OPC_JMP, /* .. and call */
+ OPC_JCC, /* conditional jump */
+ OPC_SCC, /* conditionel set */
};
enum op_op {
OP_MOV,
OP_RET,
OP_ADD,
+ OP_SUB,
OP_TEST,
OP_CMP,
OP_CALL,
};
enum opr_lenmod {
- OPRM_UNSPEC,
- OPRM_BYTE,
- OPRM_WORD,
- OPRM_DWORD,
+ OPLM_UNSPEC,
+ OPLM_BYTE,
+ OPLM_WORD,
+ OPLM_DWORD,
};
#define MAX_OPERANDS 2
struct parsed_opr {
- enum opr_type type;
- enum opr_lenmod lmod;
- unsigned int val;
- char name[256];
+ enum opr_type type;
+ enum opr_lenmod lmod;
+ int reg;
+ unsigned int val;
+ char name[256];
};
struct parsed_op {
- enum op_class cls;
- enum op_op op;
- struct parsed_opr operand[MAX_OPERANDS];
- int operand_cnt;
+ enum op_class cls;
+ enum op_op op;
+ struct parsed_opr operand[MAX_OPERANDS];
+ int operand_cnt;
+ int regmask; // all referensed regs
+ void *datap;
+};
+
+struct parsed_equ {
+ char name[64];
+ enum opr_lenmod lmod;
+ int offset;
};
#define MAX_OPS 1024
static struct parsed_op ops[MAX_OPS];
-static char labels[MAX_OPS][32];
+static struct parsed_equ *g_eqs;
+static int g_eqcnt;
+static char g_labels[MAX_OPS][32];
+static struct parsed_proto g_func_pp;
+static char g_func[256];
+static char g_comment[256];
+static int g_bp_frame;
+static int g_bp_stack;
+#define ferr(op_, fmt, ...) do { \
+ printf("error:%s:#%ld: " fmt, g_func, (op_) - ops, ##__VA_ARGS__); \
+ exit(1); \
+} while (0)
+
+#define MAX_REGS 8
+
+const char *regs_r32[] = { "eax", "ebx", "ecx", "edx", "esi", "edi", "ebp", "esp" };
+const char *regs_r16[] = { "ax", "bx", "cx", "dx", "si", "di", "bp", "sp" };
+const char *regs_r8l[] = { "al", "bl", "cl", "dl" };
+const char *regs_r8h[] = { "ah", "bh", "ch", "dh" };
+
+enum x86_regs { xUNSPEC = -1, xAX, xBX, xCX, xDX, xSI, xDI, xBP, xSP };
+
+static int char_array_i(const char *array[], size_t len, const char *s)
+{
+ int i;
-const char *main_regs[] = { "eax", "ebx", "ecx", "edx", "esi", "edi", "ebp", "esp" };
+ for (i = 0; i < len; i++)
+ if (IS(s, array[i]))
+ return i;
-static int parse_operand(struct parsed_opr *opr,
+ return -1;
+}
+
+static void parse_reg(struct parsed_opr *opr, int *regmask,
+ char *s, int strict)
+{
+ char w[16];
+ int reg = xUNSPEC;
+ int c = 0;
+
+ while (*s != 0) {
+ while (my_isblank(*s) || my_issep(*s))
+ s++;
+ s = next_idt(w, sizeof(w), s);
+ if (w[0] == 0)
+ break;
+ c++;
+ reg = char_array_i(regs_r32, ARRAY_SIZE(regs_r32), w);
+ if (reg >= 0) {
+ opr->lmod = OPLM_DWORD;
+ *regmask |= 1 << reg;
+ continue;
+ }
+ reg = char_array_i(regs_r16, ARRAY_SIZE(regs_r16), w);
+ if (reg >= 0) {
+ opr->lmod = OPLM_WORD;
+ *regmask |= 1 << reg;
+ continue;
+ }
+ reg = char_array_i(regs_r8h, ARRAY_SIZE(regs_r8h), w);
+ if (reg >= 0) {
+ opr->lmod = OPLM_BYTE;
+ *regmask |= 1 << reg;
+ continue;
+ }
+ reg = char_array_i(regs_r8l, ARRAY_SIZE(regs_r8l), w);
+ if (reg >= 0) {
+ opr->lmod = OPLM_BYTE;
+ *regmask |= 1 << reg;
+ continue;
+ }
+
+ if (strict)
+ aerr("bad reg: '%s'\n", w);
+ }
+
+ if (c == 1)
+ opr->reg = reg;
+}
+
+static long parse_number(const char *number)
+{
+ int len = strlen(number);
+ const char *p = number;
+ char *endp = NULL;
+ int neg = 0;
+ int bad;
+ long ret;
+
+ if (*p == '-') {
+ neg = 1;
+ p++;
+ }
+ if (len > 1 && *p == '0')
+ p++;
+ if (number[len - 1] == 'h') {
+ ret = strtol(p, &endp, 16);
+ bad = (*endp != 'h');
+ }
+ else {
+ ret = strtol(p, &endp, 10);
+ bad = (*endp != 0);
+ }
+ if (bad)
+ aerr("number parsing failed\n");
+ if (neg)
+ ret = -ret;
+ return ret;
+}
+
+static int parse_operand(struct parsed_opr *opr, int *regmask,
char words[16][256], int wordc, int w, enum op_class cls)
{
- char *endp = NULL;
int ret, len;
int i;
if (w >= wordc)
aerr("parse_operand w %d, wordc %d\n", w, wordc);
+ opr->reg = xUNSPEC;
+
for (i = w; i < wordc; i++) {
len = strlen(words[i]);
if (words[i][len - 1] == ',') {
if (wordc - w >= 3) {
if (IS(words[w + 1], "ptr")) {
if (IS(words[w], "dword"))
- opr->lmod = OPRM_DWORD;
+ opr->lmod = OPLM_DWORD;
else if (IS(words[w], "word"))
- opr->lmod = OPRM_WORD;
+ opr->lmod = OPLM_WORD;
else if (IS(words[w], "byte"))
- opr->lmod = OPRM_BYTE;
+ opr->lmod = OPLM_BYTE;
else
aerr("type parsing failed\n");
w += 2;
if (words[w][0] == '[') {
opr->type = OPT_REGMEM;
- ret = sscanf(words[w], "[%256s]", opr->name);
+ ret = sscanf(words[w], "[%[^]]]", opr->name);
if (ret != 1)
aerr("[] parse failure\n");
+ parse_reg(opr, regmask, opr->name, 0);
return wordc;
}
else if (('0' <= words[w][0] && words[w][0] <= '9')
|| words[w][0] == '-')
{
- opr->type = OPT_CONST;
- i = 0;
- if (len > 1 && words[w][0] == '0')
- i = 1;
- if (words[w][len - 1] == 'h') {
- words[w][len - 1] = 0;
- opr->val = strtoul(&words[w][i], &endp, 16);
- }
- else {
- opr->val = strtoul(&words[w][i], &endp, 10);
- }
- if (*endp != 0)
- aerr("const parse failed\n");
- return wordc;
+ opr->type = OPT_CONST;
+ opr->val = (unsigned int)parse_number(words[w]);
+ return wordc;
}
- strcpy(opr->name, words[w]);
opr->type = OPT_REG;
+ strcpy(opr->name, words[w]);
+ parse_reg(opr, regmask, opr->name, 1);
+
return wordc;
}
static void parse_op(struct parsed_op *op, char words[16][256], int wordc)
{
- int w = 1;
- int opr;
- int i;
+ int opr = 0;
+ int w = 1;
+ int i;
- for (i = 0; i < ARRAY_SIZE(op_table); i++) {
- if (!IS(words[0], op_table[i].name))
- continue;
+ for (i = 0; i < ARRAY_SIZE(op_table); i++) {
+ if (!IS(words[0], op_table[i].name))
+ continue;
- for (opr = 0; opr < op_table[i].minopr; opr++) {
- w = parse_operand(&op->operand[opr],
- words, wordc, w, op_table[i].cls);
- }
+ op->regmask = 0;
- for (; w < wordc && opr < op_table[i].maxopr; opr++) {
- w = parse_operand(&op->operand[opr],
- words, wordc, w, op_table[i].cls);
- }
+ for (opr = 0; opr < op_table[i].minopr; opr++) {
+ w = parse_operand(&op->operand[opr], &op->regmask,
+ words, wordc, w, op_table[i].cls);
+ }
- goto done;
- }
+ for (; w < wordc && opr < op_table[i].maxopr; opr++) {
+ w = parse_operand(&op->operand[opr], &op->regmask,
+ words, wordc, w, op_table[i].cls);
+ }
- aerr("unhandled op: '%s'\n", words[0]);
+ goto done;
+ }
-done:
- if (w < wordc)
- aerr("parse_op %s incomplete: %d/%d\n",
- words[0], w, wordc);
+ aerr("unhandled op: '%s'\n", words[0]);
- op->cls = op_table[i].cls;
- op->op = op_table[i].op;
- return;
+done:
+ if (w < wordc)
+ aerr("parse_op %s incomplete: %d/%d\n",
+ words[0], w, wordc);
+
+ op->cls = op_table[i].cls;
+ op->op = op_table[i].op;
+ op->operand_cnt = opr;
+ return;
}
-int gen_func(FILE *fout, FILE *fhdr, const char *funcn, int opcnt)
+static const char *opr_name(struct parsed_op *po, int opr_num)
{
- struct parsed_proto pp;
- int ret;
- int i;
-
- ret = proto_parse(fhdr, funcn, &pp);
- if (ret)
- return ret;
-
- fprintf(fout, "%s %s(", pp.ret_type, funcn);
- for (i = 0; i < pp.argc; i++) {
- if (i > 0)
- fprintf(fout, ", ");
- fprintf(fout, "%s a%d", pp.arg[i].type, i);
- }
- fprintf(fout, ")\n{\n");
-
-
-
- fprintf(fout, "}\n\n");
- proto_release(&pp);
- return 0;
+ if (opr_num >= po->operand_cnt)
+ ferr(po, "opr OOR: %d/%d\n", opr_num, po->operand_cnt);
+ return po->operand[opr_num].name;
}
-int main(int argc, char *argv[])
+static unsigned int opr_const(struct parsed_op *po, int opr_num)
{
- FILE *fout, *fasm, *fhdr;
- char line[256];
- char words[16][256];
- char func[256];
- int in_func = 0;
- int pi = 0;
- int len;
- char *p;
- int wordc;
-
- if (argc != 4) {
- printf("usage:\n%s <.c> <.asm> <hdrf>\n",
- argv[0]);
- return 1;
- }
-
- hdrfn = argv[3];
- fhdr = fopen(hdrfn, "r");
- my_assert_not(fhdr, NULL);
-
- asmfn = argv[2];
- fasm = fopen(asmfn, "r");
- my_assert_not(fasm, NULL);
-
- fout = fopen(argv[1], "w");
- my_assert_not(fout, NULL);
-
-
- while (fgets(line, sizeof(line), fasm))
- {
- asmln++;
-
- p = sskip(line);
- if (*p == 0 || *p == ';')
- continue;
-
- memset(words, 0, sizeof(words));
- for (wordc = 0; wordc < 16; wordc++) {
- p = sskip(next_word(words[wordc], sizeof(words[0]), p));
- if (*p == 0 || *p == ';') {
- wordc++;
- break;
- }
- }
+ if (opr_num >= po->operand_cnt)
+ ferr(po, "opr OOR: %d/%d\n", opr_num, po->operand_cnt);
+ if (po->operand[opr_num].type != OPT_CONST)
+ ferr(po, "opr %d: const expected\n", opr_num);
+ return po->operand[opr_num].val;
+}
- if (wordc == 0) {
- // shouldn't happen
- awarn("wordc == 0?\n");
- continue;
- }
+static const char *opr_reg_p(struct parsed_op *po, struct parsed_opr *popr)
+{
+ if ((unsigned int)popr->reg >= MAX_REGS)
+ ferr(po, "invalid reg: %d\n", popr->reg);
+ return regs_r32[popr->reg];
+}
- // don't care about this:
- if (words[0][0] == '.'
- || IS(words[0], "include")
- || IS(words[0], "assume") || IS(words[1], "segment")
- || IS(words[0], "align"))
- {
- continue;
- }
+static void bg_frame_access(struct parsed_op *po, char *buf,
+ size_t buf_size, const char *bp_arg, int is_src)
+{
+ struct parsed_equ *eq;
+ int i, arg_i, arg_s;
+
+ snprintf(g_comment, sizeof(g_comment), "%s", bp_arg);
+
+ for (i = 0; i < g_eqcnt; i++)
+ if (IS(g_eqs[i].name, bp_arg))
+ break;
+ if (i >= g_eqcnt)
+ ferr(po, "unresolved bp_arg: '%s'\n", bp_arg);
+ eq = &g_eqs[i];
+
+ if (eq->offset >= 0) {
+ arg_i = eq->offset / 4 - 2;
+ if (arg_i < 0 || arg_i >= g_func_pp.argc_stack)
+ ferr(po, "offset %d doesn't map to any arg\n", eq->offset);
+
+ for (i = arg_s = 0; i < g_func_pp.argc; i++) {
+ if (g_func_pp.arg[i].reg != NULL)
+ continue;
+ if (arg_s == arg_i)
+ break;
+ arg_s++;
+ }
+ if (i == g_func_pp.argc)
+ ferr(po, "arg %d not in prototype?\n", arg_i);
+ snprintf(buf, buf_size, "%sa%d", is_src ? "(u32)" : "", i + 1);
+ }
+ else {
+ if (g_bp_stack == 0)
+ ferr(po, "bp_stack access after it was not detected\n");
+ ferr(po, "TODO\n");
+ }
+}
- if (IS(words[1], "proc")) {
- if (in_func)
- aerr("proc '%s' while in_func '%s'?\n",
- words[0], func);
- strcpy(func, words[0]);
- in_func = 1;
- continue;
- }
+static char *out_src_opr(char *buf, size_t buf_size,
+ struct parsed_op *po, struct parsed_opr *popr)
+{
+ switch (popr->type) {
+ case OPT_REG:
+ switch (popr->lmod) {
+ case OPLM_DWORD:
+ snprintf(buf, buf_size, "%s", opr_reg_p(po, popr));
+ break;
+ default:
+ ferr(po, "invalid src lmod: %d\n", popr->lmod);
+ }
+ break;
+ case OPT_REGMEM:
+ if (g_bp_frame && !strncmp(popr->name, "ebp+", 4)) {
+ bg_frame_access(po, buf, buf_size, popr->name + 4, 1);
+ break;
+ }
+ ferr(po, "unhandled OPT_REGMEM variation\n");
+ break;
+ case OPT_LABEL:
+ snprintf(buf, buf_size, "%s", popr->name);
+ break;
+ case OPT_CONST:
+ snprintf(buf, buf_size, popr->val < 10 ? "%u" : "0x%02x", popr->val);
+ break;
+ default:
+ ferr(po, "invalid src type: %d\n", popr->type);
+ }
+
+ return buf;
+}
- if (IS(words[1], "endp")) {
- if (!in_func)
- aerr("endp '%s' while not in_func?\n", words[0]);
- if (!IS(func, words[0]))
- aerr("endp '%s' while in_func '%s'?\n",
- words[0], func);
- gen_func(fout, fhdr, func, pi);
- in_func = 0;
- func[0] = 0;
- if (pi != 0) {
- memset(&ops, 0, pi * sizeof(ops[0]));
- memset(labels, 0, pi * sizeof(labels[0]));
- pi = 0;
- }
- exit(1);
- continue;
- }
+static char *out_dst_opr(char *buf, size_t buf_size,
+ struct parsed_op *po, struct parsed_opr *popr)
+{
+ switch (popr->type) {
+ case OPT_REG:
+ switch (popr->lmod) {
+ case OPLM_DWORD:
+ snprintf(buf, buf_size, "%s", opr_reg_p(po, popr));
+ break;
+ default:
+ ferr(po, "invalid dst lmod: %d\n", popr->lmod);
+ }
+ break;
+ default:
+ ferr(po, "invalid dst type: %d\n", popr->type);
+ }
+
+ return buf;
+}
- if (IS(words[1], "="))
- // lots of work will be have to be done here, but for now..
- continue;
-
- if (pi >= ARRAY_SIZE(ops))
- aerr("too many ops\n");
-
- p = strchr(words[0], ':');
- if (p != NULL) {
- len = p - words[0];
- if (len > sizeof(labels[0]) - 1)
- aerr("label too long: %d\n", len);
- if (labels[pi][0] != 0)
- aerr("dupe label?\n");
- memcpy(labels[pi], words[0], len);
- labels[pi][len] = 0;
- continue;
- }
+static void split_cond(struct parsed_op *po, enum op_op *op, int *is_neg)
+{
+ *is_neg = 0;
+
+ switch (*op) {
+ case OP_JNO:
+ *op = OP_JO;
+ *is_neg = 1;
+ break;
+ case OP_JNC:
+ *op = OP_JC;
+ *is_neg = 1;
+ break;
+ case OP_JNZ:
+ *op = OP_JZ;
+ *is_neg = 1;
+ break;
+ case OP_JNS:
+ *op = OP_JS;
+ *is_neg = 1;
+ break;
+ case OP_JNP:
+ *op = OP_JP;
+ *is_neg = 1;
+ break;
+ case OP_JO:
+ case OP_JC:
+ case OP_JZ:
+ case OP_JS:
+ case OP_JP:
+ //
+ case OP_JBE:
+ case OP_JA:
+ case OP_JL:
+ case OP_JGE:
+ case OP_JLE:
+ case OP_JG:
+ break;
+ default:
+ ferr(po, "split_cond: bad op %d\n", *op);
+ break;
+ }
+}
- parse_op(&ops[pi], words, wordc);
- pi++;
+static void out_test_for_cc(char *buf, size_t buf_size,
+ struct parsed_op *po, enum opr_lenmod lmod, const char *expr)
+{
+ enum op_op op = po->op;
+ int is_neg = 0;
+
+ split_cond(po, &op, &is_neg);
+ switch (op) {
+ case OP_JZ:
+ switch (lmod) {
+ case OPLM_DWORD:
+ snprintf(buf, buf_size, "(%s %s 0)", expr, is_neg ? "!=" : "==");
+ break;
+ default:
+ ferr(po, "%s: unhandled lmod for JZ: %d\n", __func__, lmod);
+ }
+ break;
+ default:
+ ferr(po, "%s: unhandled op: %d\n", __func__, op);
+ }
+}
- (void)proto_parse;
- }
+static void propagete_lmod(struct parsed_op *po, struct parsed_opr *popr1,
+ struct parsed_opr *popr2)
+{
+ if (popr1->lmod == OPLM_UNSPEC && popr2->lmod == OPLM_UNSPEC)
+ ferr(po, "missing lmod for both operands\n");
+
+ if (popr1->lmod == OPLM_UNSPEC)
+ popr1->lmod = popr2->lmod;
+ else if (popr2->lmod == OPLM_UNSPEC)
+ popr2->lmod = popr1->lmod;
+ else if (popr1->lmod != popr2->lmod)
+ ferr(po, "conflicting lmods: %d vs %d\n", popr1->lmod, popr2->lmod);
+}
- fclose(fout);
- fclose(fasm);
- fclose(fhdr);
+static void gen_func(FILE *fout, FILE *fhdr, const char *funcn, int opcnt)
+{
+ struct parsed_op *delayed_op = NULL, *tmp_op;
+ char buf1[256], buf2[256], buf3[256];
+ struct parsed_proto *pp;
+ const char *tmpname;
+ int had_decl = 0;
+ int regmask_arg = 0;
+ int regmask = 0;
+ int no_output;
+ int arg;
+ int i, j;
+ int reg;
+ int ret;
+
+ g_bp_frame = g_bp_stack = 0;
+
+ ret = proto_parse(fhdr, funcn, &g_func_pp);
+ if (ret)
+ ferr(ops, "proto_parse failed for '%s'\n", funcn);
+
+ fprintf(fout, "%s %s(", g_func_pp.ret_type, funcn);
+ for (i = 0; i < g_func_pp.argc; i++) {
+ if (i > 0)
+ fprintf(fout, ", ");
+ fprintf(fout, "%s a%d", g_func_pp.arg[i].type, i + 1);
+ }
+ fprintf(fout, ")\n{\n");
+
+ // pass1:
+ // - handle ebp frame, remove ops related to it
+ if (ops[0].op == OP_PUSH && IS(opr_name(&ops[0], 0), "ebp")
+ && ops[1].op == OP_MOV
+ && IS(opr_name(&ops[1], 0), "ebp")
+ && IS(opr_name(&ops[1], 1), "esp"))
+ {
+ g_bp_frame = 1;
+ ops[0].cls = OPC_RMD;
+ ops[1].cls = OPC_RMD;
+
+ if (ops[2].op == OP_SUB && IS(opr_name(&ops[2], 0), "esp")) {
+ g_bp_stack = opr_const(&ops[2], 1);
+ ops[2].cls = OPC_RMD;
+ }
+
+ i = 2;
+ do {
+ for (; i < opcnt; i++)
+ if (ops[i].op == OP_RET)
+ break;
+ if (ops[i - 1].op != OP_POP || !IS(opr_name(&ops[i - 1], 0), "ebp"))
+ ferr(&ops[i - 1], "'pop ebp' expected\n");
+ ops[i - 1].cls = OPC_RMD;
+
+ if (g_bp_stack != 0) {
+ if (ops[i - 2].op != OP_MOV
+ || !IS(opr_name(&ops[i - 2], 0), "esp")
+ || !IS(opr_name(&ops[i - 2], 1), "ebp"))
+ {
+ ferr(&ops[i - 2], "esp restore expected\n");
+ }
+ ops[i - 2].cls = OPC_RMD;
+ }
+ i++;
+ } while (i < opcnt);
+ }
+
+ // pass2:
+ // - scan for all used registers
+ // - process calls
+ for (i = 0; i < opcnt; i++) {
+ if (ops[i].cls == OPC_RMD)
+ continue;
+ regmask |= ops[i].regmask;
+
+ if (ops[i].op == OP_CALL) {
+ pp = malloc(sizeof(*pp));
+ my_assert_not(pp, NULL);
+ tmpname = opr_name(&ops[i], 0);
+ ret = proto_parse(fhdr, tmpname, pp);
+ if (ret)
+ ferr(&ops[i], "proto_parse failed for '%s'\n", tmpname);
+
+ for (arg = 0; arg < pp->argc; arg++)
+ if (pp->arg[arg].reg == NULL)
+ break;
+
+ for (j = i - 1; j >= 0 && arg < pp->argc; j--) {
+ if (ops[j].cls == OPC_RMD)
+ continue;
+ if (ops[j].op != OP_PUSH)
+ continue;
+
+ pp->arg[arg].datap = &ops[j];
+ ops[j].cls = OPC_RMD;
+ for (arg++; arg < pp->argc; arg++)
+ if (pp->arg[arg].reg == NULL)
+ break;
+ }
+ if (arg < pp->argc)
+ ferr(&ops[i], "arg collect failed for '%s'\n", tmpname);
+ ops[i].datap = pp;
+ }
+ }
+
+ // instantiate arg-registers
+ for (i = 0; i < g_func_pp.argc; i++) {
+ if (g_func_pp.arg[i].reg != NULL) {
+ reg = char_array_i(regs_r32,
+ ARRAY_SIZE(regs_r32), g_func_pp.arg[i].reg);
+ if (reg < 0)
+ ferr(ops, "arg '%s' is not a reg?\n", g_func_pp.arg[i].reg);
+
+ regmask_arg |= 1 << reg;
+ fprintf(fout, " u32 %s = (u32)a%d;\n",
+ g_func_pp.arg[i].reg, i);
+ had_decl = 1;
+ }
+ }
+
+ // instantiate other regs - special case for eax
+ if (!((regmask | regmask_arg) & 1) && !IS(g_func_pp.ret_type, "void")) {
+ fprintf(fout, " u32 eax = 0;\n");
+ had_decl = 1;
+ }
+
+ regmask &= ~regmask_arg;
+ if (g_bp_frame)
+ regmask &= ~(1 << xBP);
+ if (regmask) {
+ for (reg = 0; reg < 8; reg++) {
+ if (regmask & (1 << reg)) {
+ fprintf(fout, " u32 %s;\n", regs_r32[reg]);
+ had_decl = 1;
+ }
+ }
+ }
+
+ if (had_decl)
+ fprintf(fout, "\n");
+
+ // output ops
+ for (i = 0; i < opcnt; i++) {
+ if (g_labels[i][0] != 0)
+ fprintf(fout, "\n%s:\n", g_labels[i]);
+
+ if (ops[i].cls == OPC_RMD)
+ continue;
+
+ no_output = 0;
+
+ #define internal_error() \
+ ferr(&ops[i], "%s:%d: ICE\n", __FILE__, __LINE__)
+ #define assert_operand_cnt(n_) \
+ if (ops[i].operand_cnt != n_) \
+ ferr(&ops[i], "operand_cnt is %d/%d\n", ops[i].operand_cnt, n_)
+
+ // conditional op?
+ if (ops[i].cls == OPC_JCC || ops[i].cls == OPC_SCC) {
+ if (delayed_op == NULL)
+ ferr(&ops[i], "no delayed_op before cond op\n");
+ if (delayed_op->op == OP_TEST) {
+ if (IS(opr_name(delayed_op, 0), opr_name(delayed_op, 1))) {
+ out_dst_opr(buf3, sizeof(buf3), delayed_op,
+ &delayed_op->operand[0]);
+ }
+ else {
+ out_dst_opr(buf1, sizeof(buf1), delayed_op,
+ &delayed_op->operand[0]);
+ out_src_opr(buf2, sizeof(buf2), delayed_op,
+ &delayed_op->operand[1]);
+ snprintf(buf3, sizeof(buf3), "(%s & %s)", buf1, buf2);
+ }
+ out_test_for_cc(buf1, sizeof(buf1), &ops[i],
+ delayed_op->operand[0].lmod, buf3);
+ if (ops[i].cls == OPC_JCC)
+ fprintf(fout, " if %s\n", buf1);
+ else {
+ out_dst_opr(buf2, sizeof(buf2), &ops[i], &ops[i].operand[0]);
+ fprintf(fout, " %s = %s;", buf2, buf1);
+ }
+ }
+ else {
+ ferr(&ops[i], "TODO\n");
+ }
+ }
+
+ switch (ops[i].op)
+ {
+ case OP_MOV:
+ assert_operand_cnt(2);
+ propagete_lmod(&ops[i], &ops[i].operand[0], &ops[i].operand[1]);
+ fprintf(fout, " %s = %s;",
+ out_dst_opr(buf1, sizeof(buf1), &ops[i], &ops[i].operand[0]),
+ out_src_opr(buf2, sizeof(buf2), &ops[i], &ops[i].operand[1]));
+ break;
+
+ case OP_TEST:
+ case OP_CMP:
+ propagete_lmod(&ops[i], &ops[i].operand[0], &ops[i].operand[1]);
+ delayed_op = &ops[i];
+ no_output = 1;
+ break;
+
+ // note: we reuse OP_Jcc for SETcc, only cls differs
+ case OP_JO ... OP_JG:
+ if (ops[i].cls == OPC_JCC)
+ fprintf(fout, " goto %s;", ops[i].operand[0].name);
+ break;
+
+ case OP_CALL:
+ pp = ops[i].datap;
+ if (pp == NULL)
+ ferr(&ops[i], "NULL pp\n");
+
+ fprintf(fout, " ");
+ if (!IS(pp->ret_type, "void")) {
+ fprintf(fout, "eax = ");
+ if (strchr(pp->ret_type, '*'))
+ fprintf(fout, "(u32)");
+ }
+ fprintf(fout, "%s(", opr_name(&ops[i], 0));
+ for (arg = 0; arg < pp->argc; arg++) {
+ if (arg > 0)
+ fprintf(fout, ", ");
+ if (pp->arg[arg].reg != NULL) {
+ fprintf(fout, "%s", pp->arg[i].reg);
+ continue;
+ }
+
+ // stack arg
+ tmp_op = pp->arg[arg].datap;
+ if (tmp_op == NULL)
+ ferr(&ops[i], "parsed_op missing for arg%d\n", arg);
+ fprintf(fout, "%s",
+ out_src_opr(buf1, sizeof(buf1), tmp_op, &tmp_op->operand[0]));
+ }
+ fprintf(fout, ");");
+ break;
+
+ case OP_RET:
+ if (IS(g_func_pp.ret_type, "void"))
+ fprintf(fout, " return;");
+ else
+ fprintf(fout, " return eax;");
+ break;
+
+ case OP_PUSH:
+ ferr(&ops[i], "push encountered\n");
+ break;
+
+ case OP_POP:
+ ferr(&ops[i], "pop encountered\n");
+ break;
+
+ default:
+ no_output = 1;
+ ferr(&ops[i], "unhandled op type %d, cls %d\n",
+ ops[i].op, ops[i].cls);
+ break;
+ }
+
+ if (g_comment[0] != 0) {
+ fprintf(fout, " // %s", g_comment);
+ g_comment[0] = 0;
+ no_output = 0;
+ }
+ if (!no_output)
+ fprintf(fout, "\n");
+ }
+
+ fprintf(fout, "}\n\n");
+
+ // cleanup
+ for (i = 0; i < opcnt; i++) {
+ if (ops[i].op == OP_CALL) {
+ pp = ops[i].datap;
+ if (pp) {
+ proto_release(pp);
+ free(pp);
+ }
+ }
+ }
+ proto_release(&g_func_pp);
+}
- return 0;
+int main(int argc, char *argv[])
+{
+ FILE *fout, *fasm, *fhdr;
+ char line[256];
+ char words[16][256];
+ int in_func = 0;
+ int eq_alloc;
+ int pi = 0;
+ int len;
+ char *p;
+ int wordc;
+
+ if (argc != 4) {
+ printf("usage:\n%s <.c> <.asm> <hdrf>\n",
+ argv[0]);
+ return 1;
+ }
+
+ hdrfn = argv[3];
+ fhdr = fopen(hdrfn, "r");
+ my_assert_not(fhdr, NULL);
+
+ asmfn = argv[2];
+ fasm = fopen(asmfn, "r");
+ my_assert_not(fasm, NULL);
+
+ fout = fopen(argv[1], "w");
+ my_assert_not(fout, NULL);
+
+ eq_alloc = 128;
+ g_eqs = malloc(eq_alloc * sizeof(g_eqs[0]));
+ my_assert_not(g_eqs, NULL);
+
+ while (fgets(line, sizeof(line), fasm))
+ {
+ asmln++;
+
+ p = sskip(line);
+ if (*p == 0 || *p == ';')
+ continue;
+
+ memset(words, 0, sizeof(words));
+ for (wordc = 0; wordc < 16; wordc++) {
+ p = sskip(next_word(words[wordc], sizeof(words[0]), p));
+ if (*p == 0 || *p == ';') {
+ wordc++;
+ break;
+ }
+ }
+
+ if (wordc == 0) {
+ // shouldn't happen
+ awarn("wordc == 0?\n");
+ continue;
+ }
+
+ // don't care about this:
+ if (words[0][0] == '.'
+ || IS(words[0], "include")
+ || IS(words[0], "assume") || IS(words[1], "segment")
+ || IS(words[0], "align"))
+ {
+ continue;
+ }
+
+ if (IS(words[1], "proc")) {
+ if (in_func)
+ aerr("proc '%s' while in_func '%s'?\n",
+ words[0], g_func);
+ strcpy(g_func, words[0]);
+ in_func = 1;
+ continue;
+ }
+
+ if (IS(words[1], "endp")) {
+ if (!in_func)
+ aerr("endp '%s' while not in_func?\n", words[0]);
+ if (!IS(g_func, words[0]))
+ aerr("endp '%s' while in_func '%s'?\n",
+ words[0], g_func);
+ gen_func(fout, fhdr, g_func, pi);
+ in_func = 0;
+ g_func[0] = 0;
+ if (pi != 0) {
+ memset(&ops, 0, pi * sizeof(ops[0]));
+ memset(g_labels, 0, pi * sizeof(g_labels[0]));
+ pi = 0;
+ }
+ g_eqcnt = 0;
+ exit(1);
+ continue;
+ }
+
+ if (IS(words[1], "=")) {
+ if (wordc != 5)
+ aerr("unhandled equ, wc=%d\n", wordc);
+ if (g_eqcnt >= eq_alloc) {
+ eq_alloc *= 2;
+ g_eqs = realloc(g_eqs, eq_alloc * sizeof(g_eqs[0]));
+ my_assert_not(g_eqs, NULL);
+ }
+
+ len = strlen(words[0]);
+ if (len > sizeof(g_eqs[0].name) - 1)
+ aerr("equ name too long: %d\n", len);
+ strcpy(g_eqs[g_eqcnt].name, words[0]);
+
+ if (!IS(words[3], "ptr"))
+ aerr("unhandled equ\n");
+ if (IS(words[2], "dword"))
+ g_eqs[g_eqcnt].lmod = OPLM_DWORD;
+ else if (IS(words[2], "word"))
+ g_eqs[g_eqcnt].lmod = OPLM_WORD;
+ else if (IS(words[2], "byte"))
+ g_eqs[g_eqcnt].lmod = OPLM_BYTE;
+ else
+ aerr("bad lmod: '%s'\n", words[2]);
+
+ g_eqs[g_eqcnt].offset = parse_number(words[4]);
+ g_eqcnt++;
+ continue;
+ }
+
+ if (pi >= ARRAY_SIZE(ops))
+ aerr("too many ops\n");
+
+ p = strchr(words[0], ':');
+ if (p != NULL) {
+ len = p - words[0];
+ if (len > sizeof(g_labels[0]) - 1)
+ aerr("label too long: %d\n", len);
+ if (g_labels[pi][0] != 0)
+ aerr("dupe label?\n");
+ memcpy(g_labels[pi], words[0], len);
+ g_labels[pi][len] = 0;
+ continue;
+ }
+
+ parse_op(&ops[pi], words, wordc);
+ pi++;
+
+ (void)proto_parse;
+ }
+
+ fclose(fout);
+ fclose(fasm);
+ fclose(fhdr);
+
+ return 0;
}
+
+// vim:ts=2:shiftwidth=2:expandtab
notaz.gp2x.de / ia32rtools.git / commitdiff