Remove padding from some structs, add comment to cache.

2 files changed, 33 insertions, 24 deletions

diff --git a/minias.h b/minias.h
index adf12d6..42b3e62 100644
--- a/minias.h
+++ b/minias.h
@@ -203,21 +203,21 @@ typedef enum {
 
 typedef union Parsev Parsev;
 
-typedef struct {
+typedef struct Label {
   AsmKind kind;
   const char *name;
 } Label;
 
-typedef struct {
+typedef struct Globl {
   AsmKind kind;
   const char *name;
 } Globl;
 
-typedef struct {
+typedef struct DirSection {
   AsmKind kind;
+  int32_t type;
   const char *name;
   const char *flags;
-  int type;
 } DirSection;
 
 typedef struct {
@@ -225,53 +225,53 @@ typedef struct {
   const char *l;
 } Value;
 
-typedef struct {
+typedef struct Byte {
   AsmKind kind;
   Value value;
 } Byte;
 
-typedef struct {
+typedef struct Short {
   AsmKind kind;
   Value value;
 } Short;
 
-typedef struct {
+typedef struct Int {
   AsmKind kind;
   Value value;
 } Int;
 
-typedef struct {
+typedef struct Quad {
   AsmKind kind;
   Value value;
 } Quad;
 
-typedef struct {
+typedef struct Balign {
   AsmKind kind;
   uint64_t align;
 } Balign;
 
-typedef struct {
+typedef struct Fill {
   AsmKind kind;
   int32_t size;
   int32_t repeat;
   int64_t value;
 } Fill;
 
-typedef struct {
+typedef struct Imm {
   AsmKind kind;
-  uint8_t nbytes;
+  uint32_t nbytes;
   Value v;
 } Imm;
 
-typedef struct {
+typedef struct Memarg {
   AsmKind kind;
   AsmKind base;
   AsmKind index;
-  uint8_t scale;
+  uint32_t scale;
   Value disp;
 } Memarg;
 
-typedef struct {
+typedef struct String {
   AsmKind kind;
   size_t len;
   uint8_t *data;
@@ -279,24 +279,24 @@ typedef struct {
 typedef String Ascii;
 typedef String Asciiz;
 
-typedef struct {
+typedef struct Call {
   AsmKind kind;
-  uint8_t indirect;
+  uint32_t indirect;
   union {
     const Parsev *indirect;
     Value  direct;
   } target;
 } Call;
 
-typedef struct {
+typedef struct Jmp {
   AsmKind kind;
-  uint8_t variant;
+  uint32_t variant;
   const char *target;
 } Jmp;
 
-typedef struct {
+typedef struct Instr {
   AsmKind kind;
-  uint8_t variant;
+  uint32_t variant;
   const Parsev *arg1;
   const Parsev *arg2;
   const Parsev *arg3;
diff --git a/parse.c b/parse.c
index 591e8eb..b272121 100644
--- a/parse.c
+++ b/parse.c
@@ -2,9 +2,18 @@
 
 /* Maintain a direct mapped cache of Parsev*. */
 static const Parsev *internparsev(Parsev *p) {
-  /* An extremely simple direct mapped cache of *Parsev,
-     It relies on direct pointer comparison, which
-     itself only works because our pointers are interned. */
+  /*
+     A simple direct mapped cache that prevents our parser
+     from allocating duplicate values. Note that it uses memcmp
+     for equality, even on pointer values, this works because the
+     pointers themselves are also interned.
+
+     This simplicity somes with one big cost - Parsev variants with padding
+     can trigger a false positive on valgrind. It should still safe
+     because reading these undefined bytes do not change the behavior of the
+     program. The best fix is still to avoid the padding bytes in the Parsev
+     variant layout using a tool such as 'pahole'.
+  */
   size_t idx;
   const Parsev *interned;
   static const Parsev *cache[4096] = {0};