Improved endianess detection

git-svn-id: https://lz4.googlecode.com/svn/trunk@86 650e7d94-2a16-8b24-b05c-7c0b3f6821cd

4 files changed, 69 insertions, 18 deletions

diff --git a/fuzzer.c b/fuzzer.c
index 9ba79d4..a5f9e2b 100644
--- a/fuzzer.c
+++ b/fuzzer.c
@@ -92,6 +92,29 @@ int test_canary(unsigned char *buf) {
         return 1;

 }

 

+int FUZ_SecurityTest()

+{

+  char* output;

+  char* input;

+  int i, r;

+

+  printf("Starting security tests...");

+  input = (char*) malloc (20<<20);

+  output = (char*) malloc (20<<20);

+  input[0] = 0x0F;

+  input[1] = 0x00;

+  input[2] = 0x00;

+  for(i = 3; i < 16840000; i++)

+    input[i] = 0xff;

+  r = LZ4_uncompress(input, output, 20<<20);

+

+  free(input);

+  free(output);

+  printf(" Completed (r=%i)\n",r);

+  return 0;

+}

+

+

 //int main(int argc, char *argv[]) {

 int main() {

         unsigned long long bytes = 0;

@@ -116,7 +139,9 @@ int main() {
 		}

 		printf("Seed = %u\n", seed);

 

-        for (i = 0; i < 2048; i++)

+		FUZ_SecurityTest();

+

+		for (i = 0; i < 2048; i++)

                 cbuf[FUZ_avail + i] = cbuf[FUZ_avail + 2048 + i] = FUZ_rand(&seed) >> 16;

 

         for (i = 0; i < NB_ATTEMPTS; i++) {

diff --git a/lz4.c b/lz4.c
index 00d6f95..a35f12b 100644
--- a/lz4.c
+++ b/lz4.c
@@ -68,8 +68,18 @@
 #endif

 

 // Little Endian or Big Endian ?

-// Note : overwrite the below #define if you know your architecture endianess

-#if (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN) || defined(_ARCH_PPC) || defined(__PPC__) || defined(__PPC) || defined(PPC) || defined(__powerpc__) || defined(__powerpc) || defined(powerpc) || ((defined(__BYTE_ORDER__)&&(__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__))) )

+// Overwrite the #define below if you know your architecture endianess

+#if defined (__GLIBC__)

+#  include <endian.h>

+#  if (__BYTE_ORDER == __BIG_ENDIAN)

+#     define LZ4_BIG_ENDIAN 1

+#  endif

+#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))

+#  define LZ4_BIG_ENDIAN 1

+#elif defined(__sparc) || defined(__sparc__) \

+   || defined(__ppc__) || defined(_POWER) || defined(__powerpc__) || defined(_ARCH_PPC) || defined(__PPC__) || defined(__PPC) || defined(PPC) || defined(__powerpc__) || defined(__powerpc) || defined(powerpc) \

+   || defined(__hpux)  || defined(__hppa) \

+   || defined(_MIPSEB) || defined(__s390__)

 #  define LZ4_BIG_ENDIAN 1

 #else

 // Little Endian assumed. PDP Endian and other very rare endian format are unsupported.

@@ -77,7 +87,7 @@
 

 // Unaligned memory access is automatically enabled for "common" CPU, such as x86.

 // For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected

-// If you know your target CPU supports unaligned memory access, you may want to force this option manually to improve performance

+// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance

 #if defined(__ARM_FEATURE_UNALIGNED)

 #  define LZ4_FORCE_UNALIGNED_ACCESS 1

 #endif

@@ -721,7 +731,7 @@ int LZ4_uncompress(const char* source,
         cpy = op+length;

         if unlikely(cpy>oend-COPYLENGTH)

         {

-            if (cpy != oend) goto _output_error;         // Error : we must necessarily stand at EOF

+            if (cpy != oend) goto _output_error;         // Error : not enough place for another match (min 4) + 5 literals

             memcpy(op, ip, length);

             ip += length;

             break;                                       // EOF

@@ -743,13 +753,13 @@ int LZ4_uncompress(const char* source,
 #else

             const int dec64 = 0;

 #endif

-			op[0] = ref[0];

+            op[0] = ref[0];

             op[1] = ref[1];

             op[2] = ref[2];

             op[3] = ref[3];

             op += 4, ref += 4; ref -= dec32table[op-ref];

             A32(op) = A32(ref); 

-			op += STEPSIZE-4; ref -= dec64;

+            op += STEPSIZE-4; ref -= dec64;

         } else { LZ4_COPYSTEP(ref,op); }

         cpy = op + length - (STEPSIZE-4);

         if (cpy>oend-COPYLENGTH)

@@ -832,13 +842,13 @@ int LZ4_uncompress_unknownOutputSize(
 #else

             const int dec64 = 0;

 #endif

-			op[0] = ref[0];

+            op[0] = ref[0];

             op[1] = ref[1];

             op[2] = ref[2];

             op[3] = ref[3];

             op += 4, ref += 4; ref -= dec32table[op-ref];

             A32(op) = A32(ref); 

-			op += STEPSIZE-4; ref -= dec64;

+            op += STEPSIZE-4; ref -= dec64;

         } else { LZ4_COPYSTEP(ref,op); }

         cpy = op + length - (STEPSIZE-4);

         if (cpy>oend-COPYLENGTH)

diff --git a/lz4demo.c b/lz4demo.c
index 067ac2e..8b9a0b9 100644
--- a/lz4demo.c
+++ b/lz4demo.c
@@ -59,7 +59,7 @@
 

 #if defined(_MSC_VER)    // Visual Studio

 #define swap32 _byteswap_ulong

-#elif GCC_VERSION >= 430

+#elif GCC_VERSION >= 403

 #define swap32 __builtin_bswap32

 #else

 static inline unsigned int swap32(unsigned int x) {

diff --git a/lz4hc.c b/lz4hc.c
index 7cd8bb4..200efda 100644
--- a/lz4hc.c
+++ b/lz4hc.c
@@ -37,23 +37,39 @@
 //**************************************

 // 32 or 64 bits ?

 #if (defined(__x86_64__) || defined(__x86_64) || defined(__amd64__) || defined(__amd64) || defined(__ppc64__) || defined(_WIN64) || defined(__LP64__) || defined(_LP64) )   // Detects 64 bits mode

-#define LZ4_ARCH64 1

+#  define LZ4_ARCH64 1

 #else

-#define LZ4_ARCH64 0

+#  define LZ4_ARCH64 0

 #endif

 

-// Little Endian or Big Endian ? 

-#if (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN) || defined(_ARCH_PPC) || defined(__PPC__) || defined(__PPC) || defined(PPC) || defined(__powerpc__) || defined(__powerpc) || defined(powerpc) || ((defined(__BYTE_ORDER__)&&(__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__))) )

-#define LZ4_BIG_ENDIAN 1

+// Little Endian or Big Endian ?

+// Overwrite the #define below if you know your architecture endianess

+#if defined (__GLIBC__)

+#  include <endian.h>

+#  if (__BYTE_ORDER == __BIG_ENDIAN)

+#     define LZ4_BIG_ENDIAN 1

+#  endif

+#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))

+#  define LZ4_BIG_ENDIAN 1

+#elif defined(__sparc) || defined(__sparc__) \

+   || defined(__ppc__) || defined(_POWER) || defined(__powerpc__) || defined(_ARCH_PPC) || defined(__PPC__) || defined(__PPC) || defined(PPC) || defined(__powerpc__) || defined(__powerpc) || defined(powerpc) \

+   || defined(__hpux)  || defined(__hppa) \

+   || defined(_MIPSEB) || defined(__s390__)

+#  define LZ4_BIG_ENDIAN 1

 #else

 // Little Endian assumed. PDP Endian and other very rare endian format are unsupported.

 #endif

 

 // Unaligned memory access is automatically enabled for "common" CPU, such as x86.

 // For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected

-// If you know your target CPU supports unaligned memory access, you may want to force this option manually to improve performance

+// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance

 #if defined(__ARM_FEATURE_UNALIGNED)

-#define LZ4_FORCE_UNALIGNED_ACCESS 1

+#  define LZ4_FORCE_UNALIGNED_ACCESS 1

+#endif

+

+// Define this parameter if your target system or compiler does not support hardware bit count

+#if defined(_MSC_VER) && defined(_WIN32_WCE)            // Visual Studio for Windows CE does not support Hardware bit count

+#  define LZ4_FORCE_SW_BITCOUNT

 #endif

 

 

@@ -63,7 +79,7 @@
 #if __STDC_VERSION__ >= 199901L    // C99

   /* "restrict" is a known keyword */

 #else

-#define restrict  // Disable restrict

+#  define restrict  // Disable restrict

 #endif

 

 #ifdef _MSC_VER