Merge branch 'master' into dev

ChangeLog.md

@@ -73,6 +73,10 @@ the C API and `TJTransform.OPT_COPYNONE` in the Java API) that allows the
 copying of markers (including EXIF and ICC profile data) to be disabled for a
 particular transform.
 
+7. Fixed a NullPointerException in the TurboJPEG Java wrapper that occurred
+when using the YUVImage constructor that creates an instance backed by separate
+image planes and allocates memory for the image planes.
+
 
 1.5.2
 =====
@@ -140,6 +144,10 @@ timer, in order to improve the consistency of the results.  Furthermore, the
 `-warmup` option is now used to specify the amount of warmup time rather than
 the number of warmup iterations.
 
+11. Fixed an error (`short jump is out of range`) that occurred when assembling
+the 32-bit x86 SIMD extensions with NASM versions prior to 2.04.  This was a
+regression introduced by 1.5 beta1[12].
+
 
 1.5.1
 =====

java/org/libjpegturbo/turbojpeg/YUVImage.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (C)2014 D. R. Commander.  All Rights Reserved.
+ * Copyright (C)2014, 2017 D. R. Commander.  All Rights Reserved.
  * Copyright (C)2015 Viktor Szathmáry.  All Rights Reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -220,10 +220,13 @@ public class YUVImage {
       throw new IllegalArgumentException("Invalid argument in YUVImage::setBuf()");
 
     int nc = (subsamp == TJ.SAMP_GRAY ? 1 : 3);
-    if (planes.length != nc || (offsets != null && offsets.length != nc) ||
+    if ((planes != null && planes.length != nc) ||
+        (offsets != null && offsets.length != nc) ||
         (strides != null && strides.length != nc))
       throw new IllegalArgumentException("YUVImage::setBuf(): planes, offsets, or strides array is the wrong size");
 
+    if (planes == null)
+      planes = new byte[nc][];
     if (offsets == null)
       offsets = new int[nc];
     if (strides == null)

simd/CMakeLists.txt

@@ -284,6 +284,18 @@ endif()
 
 elseif(CPU_TYPE STREQUAL "powerpc")
 
+check_c_source_compiles("
+  #include <altivec.h>
+  int main(void) {
+    __vector int vi = { 0, 0, 0, 0 };
+    return vi[0];
+  }" HAVE_ALTIVEC)
+
+if(NOT HAVE_ALTIVEC)
+  simd_fail("SIMD extensions not available for this CPU (PowerPC SPE)")
+  return()
+endif()
+
 set(SIMD_SOURCES powerpc/jccolor-altivec.c powerpc/jcgray-altivec.c
   powerpc/jcsample-altivec.c powerpc/jdcolor-altivec.c
   powerpc/jdmerge-altivec.c powerpc/jdsample-altivec.c

simd/i386/jchuff-sse2.asm

@@ -1,7 +1,7 @@
 ;
 ; jchuff-sse2.asm - Huffman entropy encoding (SSE2)
 ;
-; Copyright (C) 2009-2011, 2014-2016, D. R. Commander.
+; Copyright (C) 2009-2011, 2014-2017, D. R. Commander.
 ; Copyright (C) 2015, Matthieu Darbois.
 ;
 ; Based on the x86 SIMD extension for IJG JPEG library
@@ -288,13 +288,13 @@ EXTN(jsimd_huff_encode_one_block_sse2):
 
 .BLOOP:
     bsf         ecx, edx                ; r = __builtin_ctzl(index);
-    jz          .ELOOP
+    jz          near .ELOOP
     lea         esi, [esi+ecx*2]        ; k += r;
     shr         edx, cl                 ; index >>= r;
     mov         DWORD [esp+temp3], edx
 .BRLOOP:
     cmp         ecx, 16                       ; while (r > 15) {
-    jl          .ERLOOP
+    jl          near .ERLOOP
     sub         ecx, 16                       ; r -= 16;
     mov         DWORD [esp+temp], ecx
     mov         eax, INT [ebp + 240 * 4]      ; code_0xf0 = actbl->ehufco[0xf0];
@@ -348,7 +348,7 @@ EXTN(jsimd_huff_encode_one_block_sse2):
     sub         eax, esi
     shr         eax, 1
     bsf         ecx, edx                ; r = __builtin_ctzl(index);
-    jz          .ELOOP2
+    jz          near .ELOOP2
     shr         edx, cl                 ; index >>= r;
     add         ecx, eax
     lea         esi, [esi+ecx*2]        ; k += r;
@@ -356,13 +356,13 @@ EXTN(jsimd_huff_encode_one_block_sse2):
     jmp         .BRLOOP2
 .BLOOP2:
     bsf         ecx, edx                ; r = __builtin_ctzl(index);
-    jz          .ELOOP2
+    jz          near .ELOOP2
     lea         esi, [esi+ecx*2]        ; k += r;
     shr         edx, cl                 ; index >>= r;
     mov         DWORD [esp+temp3], edx
 .BRLOOP2:
     cmp         ecx, 16                       ; while (r > 15) {
-    jl          .ERLOOP2
+    jl          near .ERLOOP2
     sub         ecx, 16                       ; r -= 16;
     mov         DWORD [esp+temp], ecx
     mov         eax, INT [ebp + 240 * 4]      ; code_0xf0 = actbl->ehufco[0xf0];