version 1.2.2

Changelog

@@ -1,7 +1,26 @@
+version 1.2.2-release
+
+	- added support for neg: Two's Complement Negation
+
+	- fixed a bug where shift and rotate instruction produce the wrong opcode 
+	  when immediate is zero 
+
+	- fixed a bug when using sib addressing with no base and a single operand
+
+	- asmline can now count the number of instructions that breaks a chunk boundary
+
+	- the number of instructions that break a chunk boundary can be counted when 
+	  assembling a file. 
+
+	- adding `cpuid`-instruction
+
+
+
 version 1.2.1-release (2022-02-23)
 
 	- Fixed bugs regarding -imm/8
 
+
 version 1.2.0-release (2022-02-19)
 
 	- binary file can now be generated from libassemblyline 

Makefile.am

@@ -70,7 +70,7 @@ include_HEADERS = src/assemblyline.h
 # 3. Programs may need to be changed, recompiled, and relinked in order to use the new version. Bump current, set revision and age to 0.
 
 # in short, if only patch from configure.ac is bumped, bump the middle number below. If more is changed, read the above
-libassemblyline_la_LDFLAGS = -version-info 3:1:2
+libassemblyline_la_LDFLAGS = -version-info 3:2:2
 ##############################################################################################
 ##############################################################################################
 ###############################      BINS       ##############################################
@@ -127,10 +127,12 @@ TAP_LOG_COMPILER = ./test/tap/compiler.sh
 
 # add TAP tests here
 TEST_TAP = \
-		test/tap/misc.tap \
-		test/tap/lea.tap \
+		test/tap/cmp.tap \
 		test/tap/imul.tap \
-		test/tap/nasm_incompatible.tap
+		test/tap/lea.tap \
+		test/tap/misc.tap \
+		test/tap/nasm_incompatible.tap \
+		test/tap/xor.tap
 
 # those are the tests, where asmline is expected to fail, for the suite to be ok.
 # eaf stands for 'expected asmline to fail'
@@ -174,6 +176,7 @@ TEST_ASM = \
 		  test/bzhi.asm \
 		  test/clc.asm \
 		  test/cmp.asm \
+		  test/cpuid.asm \
 		  test/high_low_xmm.asm \
 		  test/imul.asm \
 		  test/lea.asm \
@@ -187,6 +190,7 @@ TEST_ASM = \
 		  test/movq.asm \
 		  test/movzx.asm \
 		  test/mulx.asm \
+		  test/neg.asm \
 		  test/negative_mem_disp.asm \
 		  test/new_instruction.asm \
 		  test/no_ptr.asm \

configure.ac

@@ -2,7 +2,7 @@
 # Process this file with autoconf to produce a configure script.
 
 AC_PREREQ([2.69])
-AC_INIT([assemblyline],[1.2.1],[[email protected]])
+AC_INIT([assemblyline],[1.2.2],[[email protected]])
 AC_CONFIG_HEADERS([config.h])
 AC_CONFIG_SRCDIR([src/assemblyline.c])
 AC_CONFIG_AUX_DIR([build-aux])

man/asmline.1

@@ -36,6 +36,11 @@ The corresponding machine code will be printed to \fIFILENAME\fR in binary form.
 .BR \-c ", " \-\-chunk " " \fICHUNK_SIZE>1
 Sets a given \fICHUNK_SIZE\fR boundary in bytes. Nop padding will be used to ensure no instruction opcode will cross the specified \fICHUNK_SIZE\fR boundary.
 
+.TP
+.BR \-b ", " \-\-breaks " " \fICHUNK_BOUNDARY>1
+
+Given a \fICHUNK_BOUNDARY\fR asmline will count the number of instructions where their opcode crosses the specified \fICHUNK_BOUNDARY\fR size in bytes.
+
 .TP
 .BR \-o ", " \-\-object " " \fIFILENAME
 Generates a binary file from path/to/file.asm called \fIFILENAME\fR.bin in the current directory.

man/libassemblyline.3

@@ -21,25 +21,30 @@ Allocates an instance of assemblyline_t and attaches a pointer to a memory \fIbu
 
 .TP
 .BI "int asm_destroy_instance(assemblyline_t " instance );
-Frees all memory associated with \fIinstance\fR.
+Frees all memory associated with \fIinstance\fR. Returns EXIT_SUCCESS or EXIT_FAILURE.
 
 .TP
 .BI "int asm_assemble_str(assemblyline_t " al ", const char *" assembly_str );
-Assembles the given string \fIassembly_str\fR containing valid x64 assembly code with instance \fIal\fR. It writes the corresponding machine code to the memory location specified by al->buffer.
+Assembles the given string \fIassembly_str\fR containing valid x64 assembly code with instance \fIal\fR. It writes the corresponding machine code to the memory location specified by the buffer associated with \fIal\fR. Returns EXIT_SUCCESS or EXIT_FAILURE.
 
 .TP
 .BI "int asm_assemble_file(assemblyline_t " al ", char *" asm_file );
-Assembles the given file path \fIasm_file\fR containing valid x64 assembly code with instance \fIal\fR. It writes the corresponding machine code to the memory location specified by al->buffer.
+Assembles the given file path \fIasm_file\fR containing valid x64 assembly code with instance \fIal\fR. It writes the corresponding machine code to the memory location specified by the buffer associated with \fIal\fR. Returns EXIT_SUCCESS or EXIT_FAILURE.
 
 .TP
 .BI "int asm_assemble_string_counting_chunks(assemblyline_t " al ", char *" str ", int " chunk_size ", int *" dest );
-Assembles the given null-terminated @param string with instance \fIal\fR. It counts the number of instructions that break the chunk boundary of size \fIchunk_size\fR and saves it to \fIdest\fR. It does not nop-pad by default, depends on instance \fIal\fR (you can nop-pad and count different chunk breaks).
+Assembles the given null-terminated string \fIstr\fR with instance \fIal\fR. It counts the number of instructions that break the chunk boundary of size \fIchunk_size\fR and saves it to \fIdest\fR. It does not nop-pad by default, depends on instance \fIal\fR (you can nop-pad and count different chunk breaks). Returns EXIT_SUCCESS or EXIT_FAILURE.
+
 .br
 \fBNOTE:\fR you cannot pass const char* as \fIstr\fR, it will segfault, because string will be altered.
 
+.TP
+.BI "int asm_assemble_file_counting_chunks(assemblyline_t " al ", char *" asm_file ", int " chunk_size ", int *" dest );
+Assembles the given file \fIasm_file\fR with instance \fIal\fR. It counts the number of instructions that break the chunk boundary of size \fIchunk_size\fR and saves it to \fIdest\fR. It does not nop-pad by default, depends on instance \fIal\fR (you can nop-pad and count different chunk breaks).
+
 .TP
 .BI "void asm_set_chunk_size(assemblyline_t " al ", size_t " chunk_size );
-Sets a given chunk size boundary \fIchunk_size\fR in bytes with instance \fIal\fR. When called before assemble_str() or assemble_file() assemblyline will ensure no instruction opcode will cross the specified  \fIchunk_size\fR boundary via nop padding.
+Sets a given chunk size boundary \fIchunk_size\fR in bytes with instance \fIal\fR. When called before assemble_str() or assemble_file() assemblyline will ensure no instruction opcode will cross the specified \fIchunk_size\fR boundary via nop padding.
 .br
 \fBNOTE:\fR \fIchunk_size\fR must be greater than 2 in order to be classified as a valid memory chunk boundary size.
 
@@ -67,7 +72,7 @@ Returns the buffer associated with \fIal\fR (DEPRECATED: use \fBasm_get_code(3)\
 
 .TP
 .BI "int asm_create_bin_file(assemblyline_t " al ", char *" file_name );
-Generates a binary file \fIfile_name\fR from assembled machine code up to the memory offset of the current instance \fIal\fR.
+Generates a binary file \fIfile_name\fR from assembled machine code up to the memory offset of the current instance \fIal\fR. Returns EXIT_SUCCESS or EXIT_FAILURE.
 
 .TP
 .BI "void asm_mov_imm(assemblyline_t " al ", enum asm_opt "option );

src/assemblyline.c

@@ -33,7 +33,7 @@
  * INSTR_TABLE[] where the first occurrence of each letter of the alphabet to
  * instr_index_table for more efficient instruction lookup
  */
-static void asm_build_index_tables(assemblyline_t al) {
+static void asm_build_index_tables() {
   // INSTR_TABLE index starts at the SKIP entry
   size_t i = 2;
   char previous_char = 'a' - 1;
@@ -53,7 +53,6 @@ static void asm_build_index_tables(assemblyline_t al) {
       previous_char = OPD_FORMAT_TABLE[i].str[0];
     }
   }
-  al->table_built = true;
 }
 
 assemblyline_t asm_create_instance(uint8_t *buffer, int len) {
@@ -84,7 +83,7 @@ assemblyline_t asm_create_instance(uint8_t *buffer, int len) {
   al->chunk_size++;
   al->debug = false;
   al->finalized = false;
-  asm_build_index_tables(al);
+  asm_build_index_tables();
   return al;
 }
 
@@ -141,27 +140,44 @@ int asm_assemble_string_counting_chunks(assemblyline_t al, char *str,
   return EXIT_SUCCESS;
 }
 
-int assemble_file(assemblyline_t al, char *asm_file) {
-  return asm_assemble_file(al, asm_file);
-}
-
-int asm_assemble_file(assemblyline_t al, char *asm_file) {
+static void *asm_mmap_file(char *asm_file, size_t *str_len) {
   // open file for reading
   int fd = open(asm_file, O_RDONLY, S_IRUSR | S_IRUSR);
-  FAIL_SYS(fd == -1, "failed to open file\n");
+  FAIL_SYS(fd == -1, "failed to open file\n", MAP_FAILED);
   struct stat file_stat;
-  FAIL_SYS(fstat(fd, &file_stat), "failed to get file stats\n");
+  FAIL_SYS(fstat(fd, &file_stat), "failed to get file stats\n", MAP_FAILED);
   // map file contents to a string
-  size_t str_len = file_stat.st_size;
-  const char *assembly_str =
-      mmap(NULL, file_stat.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
+  *str_len = file_stat.st_size;
+  void *str = mmap(NULL, *str_len, PROT_READ, MAP_PRIVATE, fd, 0);
   close(fd);
-  FAIL_SYS(assembly_str == MAP_FAILED, "failed to allocate memory via mmap\n");
-  int exit_status = asm_assemble_str(al, assembly_str);
+  return str;
+}
+
+int asm_assemble_file_counting_chunks(assemblyline_t al, char *asm_file,
+                                      int chunk_size, int *dest) {
+
+  size_t str_len = 0;
+  char *str = asm_mmap_file(asm_file, &str_len);
+  FAIL_SYS(str == MAP_FAILED, "mmap failed to read file\n", EXIT_FAILURE);
+  int exit = asm_assemble_string_counting_chunks(al, str, chunk_size, dest);
+  // free mmap memory used for reading file
+  FAIL_SYS(munmap((void *)str, str_len) == -1, "munmap failed\n", EXIT_FAILURE);
+  return exit;
+}
+
+int assemble_file(assemblyline_t al, char *asm_file) {
+  return asm_assemble_file(al, asm_file);
+}
+
+int asm_assemble_file(assemblyline_t al, char *asm_file) {
+
+  size_t str_len = 0;
+  const char *str = asm_mmap_file(asm_file, &str_len);
+  FAIL_SYS(str == MAP_FAILED, "mmap failed to read file\n", EXIT_FAILURE);
+  int exit = asm_assemble_str(al, str);
   // free mmap memory used for reading file
-  FAIL_SYS(munmap((void *)assembly_str, str_len) == -1,
-           "Error: failed to free memory\n");
-  return exit_status;
+  FAIL_SYS(munmap((void *)str, str_len) == -1, "munmap failed\n", EXIT_FAILURE);
+  return exit;
 }
 
 void asm_set_chunk_size(assemblyline_t al, size_t chunk_size) {
@@ -188,6 +204,7 @@ uint8_t __attribute__((deprecated("use asm_get_code instead"))) *
 void *asm_get_code(assemblyline_t al) { return (void *)al->buffer; }
 
 int asm_create_bin_file(assemblyline_t al, const char *file_name) {
+
   void *buffer = asm_get_code(al);
   int len = asm_get_offset(al);
   FILE *write_ptr = fopen(file_name, "wb");

src/assemblyline.h

@@ -40,14 +40,16 @@ typedef struct assemblyline *assemblyline_t;
 assemblyline_t asm_create_instance(uint8_t *buffer, int len);
 
 /**
- * frees all memory associated with @param instance
+ * frees all memory associated with @param instance. Returns EXIT_SUCCESS or
+ * EXIT_FAILURE.
  */
 int asm_destroy_instance(assemblyline_t instance);
 
 /**
  * assembles the given string @param assembly_str containing valid x64 assembly
  * code with instance @param al It writes the corresponding machine code to the
- * memory location specified by al->buffer.
+ * memory location specified by the buffer attached to @param al. Returns
+ * EXIT_SUCCESS or EXIT_FAILURE.
  * (DEPRECATED: use asm_assemble_str() instead)
  */
 int __attribute__((deprecated("use asm_assemble_str instead")))
@@ -56,14 +58,16 @@ assemble_str(assemblyline_t al, const char *assembly_str);
 /**
  * assembles the given string @param assembly_str containing valid x64 assembly
  * code with instance @param al It writes the corresponding machine code to the
- * memory location specified by al->buffer.
+ * memory location specified by buffer attached to @param al. Returns
+ * EXIT_SUCCESS or EXIT_FAILURE.
  */
 int asm_assemble_str(assemblyline_t al, const char *assembly_str);
 
 /**
  * assembles the given file path @param asm_file containing valid x64 assembly
  * code with instance @param al It writes the corresponding machine code to the
- * memory location specified by al->buffer.
+ * memory location specified by buffer attached to @param al. Returns
+ * EXIT_SUCCESS or EXIT_FAILURE.
  * (DEPRECATED: use asm_assemble_file() instead)
  */
 int __attribute__((deprecated("use asm_assemble_file instead")))
@@ -72,7 +76,8 @@ assemble_file(assemblyline_t al, char *asm_file);
 /**
  * assembles the given file path @param asm_file containing valid x64 assembly
  * code with instance @param al It writes the corresponding machine code to the
- * memory location specified by al->buffer.
+ * memory location specified by buffer attached to @param al. Returns
+ * EXIT_SUCCESS or EXIT_FAILURE.
  */
 int asm_assemble_file(assemblyline_t al, char *asm_file);
 
@@ -81,7 +86,7 @@ int asm_assemble_file(assemblyline_t al, char *asm_file);
  * It counts the number of instructions that break the chunk boundary of size
  * @param chunk_size and saves it to @param dest It does not nop-pad
  * necessarily, depends on the @param al instance (you can nop-pad and count
- * different chunk breaks).
+ * different chunk breaks). Returns EXIT_SUCCESS or EXIT_FAILURE.
  * NOTE: you cannot pass const char* as @param string, it will segfault, because
  * string will be altered.
  * (DEPRECATED: use asm_assemble_string_counting_chunks() instead)
@@ -96,17 +101,27 @@ assemble_string_counting_chunks(assemblyline_t al, char *string, int chunk_size,
  * It counts the number of instructions that break the chunk boundary of size
  * @param chunk_size and saves it to @param dest It does not nop-pad
  * necessarily, depends on the @param al instance (you can nop-pad and count
- * different chunk breaks).
+ * different chunk breaks). Returns EXIT_SUCCESS or EXIT_FAILURE.
  * NOTE: you cannot pass const char* as @param string, it will segfault, because
  * string will be altered.
  */
 int asm_assemble_string_counting_chunks(assemblyline_t al, char *string,
                                         int chunk_size, int *dest);
 
+/**
+ * assembles the given @param asm_file with instance @param al.
+ * It counts the number of instructions that break the chunk boundary of size
+ * @param chunk_size and saves it to @param dest It does not nop-pad
+ * necessarily, depends on the @param al instance (you can nop-pad and count
+ * different chunk breaks). Returns EXIT_SUCCESS or EXIT_FAILURE.
+ */
+int asm_assemble_file_counting_chunks(assemblyline_t al, char *asm_file,
+                                      int chunk_size, int *dest);
+
 /**
  * sets a given chunk size boundary @param chunk_size in bytes with instance
  * @param al. When called before assemble_str() or assemble_file() assemblyline
- * will ensure no instruction opcode will cross the specified  @param chunk_size
+ * will ensure no instruction opcode will cross the specified @param chunk_size
  * boundary via nop padding.
  * NOTE: @param chunk_size must be greater than 2 in order to
  * be classified as a valid memory chunk boundary
@@ -147,7 +162,8 @@ void *asm_get_code(assemblyline_t al);
 
 /**
  * Generates a binary file @param file_name from assembled machine code up to
- * the memory offset of the current instance @param al
+ * the memory offset of the current instance @param al. Returns EXIT_SUCCESS or
+ * EXIT_FAILURE.
  */
 int asm_create_bin_file(assemblyline_t al, const char *file_name);
 

src/common.h

@@ -43,12 +43,15 @@
 #define NEG32BIT 0xffffffff00000000
 #define NEG8BIT 0xffffffffffffff00
 #define NEG80BIT 0xffffffffffffff80
+#define NEG80_32BIT 0xffffff80
 #define MAX_SIGNED_8BIT 0x7f
 #define MAX_UNSIGNED_8BIT 0xff
 #define MAX_UNSIGNED_16BIT 0xffff
 #define MAX_SIGNED_32BIT 0x7fffffff
 #define MAX_UNSIGNED_32BIT 0xffffffff
 #define NEG32BIT_CHECK 0x80000000
+// check if a number is at least 32 bits
+#define X32BIT_CHECK 0x10000000
 #define NEG8BIT_CHECK 0x80
 
 // set register length to 1 byte
@@ -134,8 +137,8 @@
 
 // used only in tokenizer
 #define IN_RANGE(var, lower, upper) ((var >= lower) && (var <= upper))
-#define DO_NOT_PAD(reduce, set)                                                \
-  reduce &= MAX_UNSIGNED_32BIT;                                                \
+#define DO_NOT_PAD(reduce, set, mask)                                          \
+  reduce &= mask;                                                              \
   set = true;
 
 // keyword length
@@ -178,11 +181,11 @@
     return EXIT_FAILURE;                                                       \
   }
 
-#define FAIL_SYS(EXP, MSG)                                                     \
+#define FAIL_SYS(EXP, MSG, RET)                                                \
   if (EXP) {                                                                   \
     fprintf(stderr, "assembyline: " MSG);                                      \
-    perror("error: ");                                                         \
-    return EXIT_FAILURE;                                                       \
+    perror("error ");                                                          \
+    return RET;                                                                \
   }
 
 #define FAIL_IF_ERR(EXP)                                                       \