Compile fixes for pre-ARMv6T2 and/or PIC

diff --git a/common/arm/asm.S b/common/arm/asm.S
index cbb28064aca3b4e7074022e057eb9c3ea38e1094..f7b9f14124e507e4dc432f73d0d8e7485aa230b5 100644 (file)
--- a/common/arm/asm.S
+++ b/common/arm/asm.S
@@ -46,6 +46,17 @@
 #endif
         .endm
 
+.macro movconst rd, val
+#ifdef HAVE_ARMV6T2
+    movw        \rd, #:lower16:\val
+.if \val >> 16
+    movt        \rd, #:upper16:\val
+.endif
+#else
+    ldr         \rd, =\val
+#endif
+.endm
+
 #define FENC_STRIDE 16
 #define FDEC_STRIDE 32
 

diff --git a/common/arm/deblock-a.S b/common/arm/deblock-a.S
index da1a3161dd58bf6a807bb192e347d16cf37bacff..6d60242c2b9ed003698ce08463c9bc991bb0e3e1 100644 (file)
--- a/common/arm/deblock-a.S
+++ b/common/arm/deblock-a.S
@@ -22,6 +22,8 @@
 
 #include "asm.S"
 
+.fpu neon
+
 .macro h264_loop_filter_start
     ldr             ip,  [sp]
     ldr             ip,  [ip]

diff --git a/common/arm/mc-a.S b/common/arm/mc-a.S
index 33b8dadf09980f0506b8296379392b05c60015f7..afd881c7493d95cf8c88ae7ec543ea3016469687 100644 (file)
--- a/common/arm/mc-a.S
+++ b/common/arm/mc-a.S
@@ -135,6 +135,7 @@ memcpy_table:
 .word memcpy_aligned_8_8_neon
 .text
 
+.ltorg
 
 // void x264_memzero_aligned( void *dst, size_t n )
 function x264_memzero_aligned_neon, export=1

diff --git a/common/arm/pixel-a.S b/common/arm/pixel-a.S
index d9854875292763bdd0b6532b0779950547601575..effe9395f6eb59a2bf94e89725e5c5b4271a2a98 100644 (file)
--- a/common/arm/pixel-a.S
+++ b/common/arm/pixel-a.S
@@ -200,12 +200,6 @@ SAD_FUNC  16, 16, _aligned, ,:128
 
 .macro SAD_FUNC_DUAL w, h
 function x264_pixel_sad_aligned_\w\()x\h\()_neon_dual, export=1
-.if \w == 16
-    .set r, \h / 2 - 2
-.else
-    .set r, \h / 2 - 1
-.endif
-
     SAD_DUAL_START_\w
 .rept \h / 2 - \w / 8
     SAD_DUAL_\w
@@ -1209,9 +1203,8 @@ function x264_pixel_ssim_end4_neon, export=1
     vshl.s32    q2,  q2,  #6
     vadd.s32    q1,  q8,  q8
 
-    mov         r3, #416        // ssim_c1= .01*.01*255*255*64
-    movw        ip, #39355      // ssim_c2= .03*.03*255*255*64*63 - 3<<16
-    movt        ip, #3
+    mov         r3, #416        // ssim_c1 = .01*.01*255*255*64
+    movconst    ip, 235963      // ssim_c2 = .03*.03*255*255*64*63
     vdup.32     q14, r3
     vdup.32     q15, ip
 

diff --git a/common/arm/predict-a.S b/common/arm/predict-a.S
index 46e687b353ca22045fdaca47a1cf6a1bcc013d14..8ff61a23183193f844c23558d5524c3ac9d18df3 100644 (file)
--- a/common/arm/predict-a.S
+++ b/common/arm/predict-a.S
@@ -102,7 +102,7 @@ function x264_predict_4x4_ddr_armv6, export=1
     add     r4, r4, r3, lsl #8
     add     r5, r5, r4, lsl #8
     add     r6, r6, r5, lsl #8
-    ldr     ip, pb_1
+    ldr     ip, =0x01010101
     PRED4x4_LOWPASS r1, r2, r3, r4, r5, r6, ip
     str     r1, [r0, #0*FDEC_STRIDE]
     lsl     r2, r1, #8
@@ -118,8 +118,6 @@ function x264_predict_4x4_ddr_armv6, export=1
     pop     {r4-r6,pc}
 .endfunc
 
-pb_1: .word 0x01010101
-
 function x264_predict_4x4_ddl_neon, export=1
     sub         r0, #FDEC_STRIDE
     mov         ip, #FDEC_STRIDE

diff --git a/common/arm/quant-a.S b/common/arm/quant-a.S
index 81ec1b18ec994041d34be5443dfd290046409a46..f348e5fdd677f8e9230b3b5da6592d2077399d87 100644 (file)
--- a/common/arm/quant-a.S
+++ b/common/arm/quant-a.S
@@ -124,7 +124,7 @@ function x264_quant_8x8_neon, export=1
 .endfunc
 
 .macro DEQUANT_START mf_size offset dc=no
-    movw        r3,  #0x2b
+    mov         r3,  #0x2b
     mul         r3,  r3,  r2
     lsr         r3,  r3,  #8            // i_qbits = i_qp / 6
     add         ip,  r3,  r3,  lsl #1

diff --git a/configure b/configure
index 767570a8e542c9197b7463bd7dc308732187657f..c39f67c2b9fae4a6913548e9c2aff6f57f6e2f0c 100755 (executable)
--- a/configure
+++ b/configure
@@ -305,7 +305,7 @@ esac
 
 cc_check || die "No working C compiler found."
 
-if [ $shared = yes -a \( $ARCH = "X86_64" -o $ARCH = "PPC" -o $ARCH = "ALPHA" \) ] ; then
+if [ $shared = yes -a \( $ARCH = "X86_64" -o $ARCH = "PPC" -o $ARCH = "ALPHA" -o $ARCH = "ARM" \) ] ; then
     pic="yes"
 fi
 
@@ -332,12 +332,12 @@ if [ $asm = yes -a $ARCH = ARM ] ; then
     # set flags so neon is built by default
     echo $CFLAGS | grep -Eq '(-mcpu|-march|-mfpu|-mfloat-abi)' || CFLAGS="$CFLAGS -mcpu=cortex-a8 -mfpu=neon -mfloat-abi=softfp"
 
-    if  cc_check '' '' 'asm("rev r0, r0");' ; then      CFLAGS="$CFLAGS -DHAVE_ARMV6"
+    if  cc_check '' '' 'asm("rev ip, ip");' ; then      CFLAGS="$CFLAGS -DHAVE_ARMV6"
         cc_check '' '' 'asm("movt r0, #0");'         && CFLAGS="$CFLAGS -DHAVE_ARMV6T2"
         cc_check '' '' 'asm("vadd.i16 q0, q0, q0");' && CFLAGS="$CFLAGS -DHAVE_NEON"
         ASFLAGS="$ASFLAGS $CFLAGS -c"
     else
-        echo "You specified a pre-ARMv6 CPU in your CFLAGS."
+        echo "You specified a pre-ARMv6 or Thumb-1 CPU in your CFLAGS."
         echo "If you really want to run on such a CPU, configure with --disable-asm."
         exit 1
     fi
@@ -437,6 +437,10 @@ fi
 
 if [ "$debug" = "yes" ]; then
     CFLAGS="-O1 -g $CFLAGS"
+elif [ $ARCH = ARM ]; then
+    # arm-gcc-4.2 produces incorrect output with -ffast-math
+    # and it doesn't save any speed anyway on 4.4, so disable it
+    CFLAGS="-O4 -fno-fast-math $CFLAGS"
 else
     CFLAGS="-O4 -ffast-math $CFLAGS"
 fi