bugfix: FFI C parsers could not parse some C constructs like __attribute((aligned(N))) and #pragma.

Decoupled hash functions used in comparison (hardcoded) and string table.

This bug had first appeared in v2.1-20170405 (or OpenResty 1.11.2.3).

Signed-off-by: Yichun Zhang (agentzh) <[email protected]>

Author: Shuxin Yang

src/lib_ffi.c

@@ -727,7 +727,7 @@ LJLIB_CF(ffi_abi)	LJLIB_REC(.)
 {
   GCstr *s = lj_lib_checkstr(L, 1);
   int b = 0;
-  switch (s->hash) {
+  switch (lj_str_indep_hash(s)) {
 #if LJ_64
   case H_(849858eb,ad35fd06): b = 1; break;  /* 64bit */
 #else

src/lj_cparse.c

@@ -1059,7 +1059,7 @@ static void cp_decl_gccattribute(CPState *cp, CPDecl *decl)
     if (cp->tok == CTOK_IDENT) {
       GCstr *attrstr = cp->str;
       cp_next(cp);
-      switch (attrstr->hash) {
+      switch (lj_str_indep_hash(attrstr)) {
       case H_(64a9208e,8ce14319): case H_(8e6331b2,95a282af):  /* aligned */
 	cp_decl_align(cp, decl);
 	break;
@@ -1128,7 +1128,7 @@ static void cp_decl_msvcattribute(CPState *cp, CPDecl *decl)
   while (cp->tok == CTOK_IDENT) {
     GCstr *attrstr = cp->str;
     cp_next(cp);
-    switch (attrstr->hash) {
+    switch (lj_str_indep_hash(attrstr)) {
     case H_(bc2395fa,98f267f8):  /* align */
       cp_decl_align(cp, decl);
       break;
@@ -1718,16 +1718,16 @@ static void cp_pragma(CPState *cp, BCLine pragmaline)
 {
   cp_next(cp);
   if (cp->tok == CTOK_IDENT &&
-      cp->str->hash == H_(e79b999f,42ca3e85))  {  /* pack */
+      (lj_str_indep_hash(cp->str)) == H_(e79b999f,42ca3e85))  {  /* pack */
     cp_next(cp);
     cp_check(cp, '(');
     if (cp->tok == CTOK_IDENT) {
-      if (cp->str->hash == H_(738e923c,a1b65954)) {  /* push */
+      if (lj_str_indep_hash(cp->str) == H_(738e923c,a1b65954)) {  /* push */
 	if (cp->curpack < CPARSE_MAX_PACKSTACK) {
 	  cp->packstack[cp->curpack+1] = cp->packstack[cp->curpack];
 	  cp->curpack++;
 	}
-      } else if (cp->str->hash == H_(6c71cf27,6c71cf27)) {  /* pop */
+      } else if (lj_str_indep_hash(cp->str) == H_(6c71cf27,6c71cf27)) {  /* pop */
 	if (cp->curpack > 0) cp->curpack--;
       } else {
 	cp_errmsg(cp, cp->tok, LJ_ERR_XSYMBOL);
@@ -1777,12 +1777,12 @@ static void cp_decl_multi(CPState *cp)
 	cp_line(cp, hashline);
 	continue;
       } else if (tok == CTOK_IDENT &&
-		 cp->str->hash == H_(187aab88,fcb60b42)) { /* line */
+		 lj_str_indep_hash(cp->str) == H_(187aab88,fcb60b42)) { /* line */
 	if (cp_next(cp) != CTOK_INTEGER) cp_err_token(cp, tok);
 	cp_line(cp, hashline);
 	continue;
       } else if (tok == CTOK_IDENT &&
-	  cp->str->hash == H_(f5e6b4f8,1d509107)) { /* pragma */
+	  lj_str_indep_hash(cp->str) == H_(f5e6b4f8,1d509107)) { /* pragma */
 	cp_pragma(cp, hashline);
 	continue;
       } else {

src/lj_str.c

@@ -128,11 +128,6 @@ void lj_str_resize(lua_State *L, MSize newmask)
   g->strhash = newhash;
 }
 
-#include "x64/src/lj_str_hash_x64.h"
-
-#if defined(LJ_ARCH_STR_HASH)
-#define LJ_STR_HASH LJ_ARCH_STR_HASH
-#else
 static MSize
 lj_str_original_hash(const char *str, size_t lenx) {
   MSize len = (MSize)lenx;
@@ -160,6 +155,17 @@ lj_str_original_hash(const char *str, size_t lenx) {
 
   return h;
 }
+
+MSize
+lj_str_indep_hash(GCstr *str) {
+  return lj_str_original_hash(strdata(str), str->len);
+}
+
+#include "x64/src/lj_str_hash_x64.h"
+
+#if defined(LJ_ARCH_STR_HASH)
+#define LJ_STR_HASH LJ_ARCH_STR_HASH
+#else
 #define LJ_STR_HASH lj_str_original_hash
 #endif
 

src/lj_str.h

@@ -24,4 +24,6 @@ LJ_FUNC void LJ_FASTCALL lj_str_free(global_State *g, GCstr *s);
 #define lj_str_newz(L, s)	(lj_str_new(L, s, strlen(s)))
 #define lj_str_newlit(L, s)	(lj_str_new(L, "" s, sizeof(s)-1))
 
+MSize lj_str_indep_hash(GCstr *str);
+
 #endif

src/x64/test/Makefile

@@ -24,6 +24,7 @@ CXXFLAGS := -O3 -MD -g -msse4.2 -Wall -I../src -I../../../src
 test: $(TEST_PROGRAM)
 	@echo "some unit test"
 	$(VALGRIND) ./$(TEST_PROGRAM)
+	./unit_test.sh
 
 	@echo "smoke test"
 	../../luajit test_str_comp.lua

src/x64/test/unit/ffi/test_abi.lua

@@ -0,0 +1,10 @@
+local ffi = require "ffi"
+
+-- TODO: test "gc64" and "win" parameters
+assert((ffi.abi("32bit") or ffi.abi("64bit"))
+        and ffi.abi("le")
+        and not ffi.abi("be")
+        and ffi.abi("fpu")
+        and not ffi.abi("softfp")
+        and ffi.abi("hardfp")
+        and not ffi.abi("eabi"))

src/x64/test/unit/ffi/test_line_directive.lua

@@ -0,0 +1,15 @@
+local x = [=[
+local ffi = require "ffi"
+
+ffi.cdef [[
+    #line 100
+    typedef Int xxx
+]]
+]=]
+
+local function foo()
+    loadstring(x)()
+end
+
+local r, e = pcall(foo)
+assert(string.find(e, "declaration specifier expected near 'Int' at line 100") ~= nil)

src/x64/test/unit/ffi/test_pragma_pack_pushpop.lua

@@ -0,0 +1,12 @@
+local ffi = require "ffi"
+
+ffi.cdef[[
+#pragma pack(push, 1)
+typedef struct {
+    char x;
+    double y;
+} foo;
+#pragma pack(pop)
+]]
+
+assert(ffi.sizeof("foo") == 9)

src/x64/test/unit/ffi/test_var_attribute.lua

@@ -0,0 +1,22 @@
+local ffi = require "ffi"
+
+ffi.cdef[[
+typedef struct { int a; char b; } __attribute__((packed)) myty1;
+typedef struct { int a; char b; } __attribute__((__packed__)) myty1_a;
+
+typedef struct { int a; char b; } __attribute__((aligned(16))) myty2_a;
+typedef struct { int a; char b; } __attribute__((__aligned__(16))) myty2;
+
+typedef int __attribute__ ((vector_size (32))) myty3;
+typedef int __attribute__ ((__vector_size__ (32))) myty3_a;
+
+typedef int __attribute__ ((mode(DI))) myty4;
+]]
+
+assert(ffi.sizeof("myty1") == 5 and
+       ffi.sizeof("myty1_a") == 5 and
+       ffi.alignof("myty2") == 16 and
+       ffi.alignof("myty2_a") == 16 and
+       ffi.sizeof("myty3") == 32 and
+       ffi.sizeof("myty3_a") == 32 and
+       ffi.sizeof("myty4") == 8)

src/x64/test/unit_test.sh

@@ -0,0 +1,22 @@
+#!/bin/sh
+DIR=$(cd $(dirname $0); pwd)
+cd $DIR
+
+LUAJIT=$DIR/../../luajit
+HASERR=0
+
+find $DIR/unit -name "*.lua" -print | while read x; do
+    $LUAJIT $x >/dev/null 2>/dev/null
+    if [ $? -eq 0 ]; then
+        echo "$x ok"
+    else
+        HASERR=1
+        echo "$x failed"
+    fi
+done
+
+if [ $HASERR -eq 0 ]; then
+    exit 0
+fi
+
+exit 1