diff --git a/common.mak b/common.mak
index ec706480d0..821e84c4c4 100644
--- a/common.mak
+++ b/common.mak
@@ -29,7 +29,8 @@ CFLAGS     += $(ECFLAGS)
 CCFLAGS     = $(CPPFLAGS) $(CFLAGS)
 ASFLAGS    := $(CPPFLAGS) $(ASFLAGS)
 CXXFLAGS   += $(CPPFLAGS) $(CFLAGS)
-YASMFLAGS  += $(IFLAGS) -I$(SRC_PATH)/libavutil/x86/ -Pconfig.asm
+YASMFLAGS  += $(IFLAGS:%=%/) -I$(SRC_PATH)/libavutil/x86/ -Pconfig.asm
+
 HOSTCCFLAGS = $(IFLAGS) $(HOSTCFLAGS)
 LDFLAGS    := $(ALLFFLIBS:%=-Llib%) $(LDFLAGS)
 
diff --git a/configure b/configure
index ef4298a21f..98444d9758 100755
--- a/configure
+++ b/configure
@@ -1220,6 +1220,7 @@ HAVE_LIST="
     closesocket
     cmov
     cpuid
+    cpunop
     dcbzl
     dev_bktr_ioctl_bt848_h
     dev_bktr_ioctl_meteor_h
@@ -3229,6 +3230,7 @@ EOF
             die "yasm not found, use --disable-yasm for a crippled build"
         check_yasm "vextractf128 xmm0, ymm0, 0" || disable avx
         check_yasm "vfmaddps ymm0, ymm1, ymm2, ymm3" || disable fma4
+        check_yasm "CPU amdnop" && enable cpunop
     fi
 
     case "$cpu" in
diff --git a/libavcodec/x86/deinterlace.asm b/libavcodec/x86/deinterlace.asm
index a09473bdae..bcc275be89 100644
--- a/libavcodec/x86/deinterlace.asm
+++ b/libavcodec/x86/deinterlace.asm
@@ -39,7 +39,7 @@ cglobal deinterlace_line_mmx,         7,7,7, dst, lum_m4, lum_m3, lum_m2, lum_m1
 %endif
     pxor  mm7, mm7
     movq  mm6, [pw_4]
-.nextrow
+.nextrow:
     movd  mm0, [lum_m4q]
     movd  mm1, [lum_m3q]
     movd  mm2, [lum_m2q]
diff --git a/libavcodec/x86/dsputil_yasm.asm b/libavcodec/x86/dsputil_yasm.asm
index 06d2027c69..19884a36a8 100644
--- a/libavcodec/x86/dsputil_yasm.asm
+++ b/libavcodec/x86/dsputil_yasm.asm
@@ -1143,7 +1143,7 @@ VECTOR_CLIP_INT32 6, 1, 0, 0
 cglobal vector_fmul_reverse, 4,4,2, dst, src0, src1, len
     lea       lenq, [lend*4 - 2*mmsize]
 ALIGN 16
-.loop
+.loop:
 %if cpuflag(avx)
     vmovaps     xmm0, [src1q + 16]
     vinsertf128 m0, m0, [src1q], 1
@@ -1182,7 +1182,7 @@ VECTOR_FMUL_REVERSE
 cglobal vector_fmul_add, 5,5,2, dst, src0, src1, src2, len
     lea       lenq, [lend*4 - 2*mmsize]
 ALIGN 16
-.loop
+.loop:
     mova    m0,   [src0q + lenq]
     mova    m1,   [src0q + lenq + mmsize]
     mulps   m0, m0, [src1q + lenq]
@@ -1313,7 +1313,7 @@ cglobal bswap32_buf, 3,4,5
     add      r0, 4
     dec      r2
     jnz      .loop2
-.end
+.end:
     RET
 
 ; %1 = aligned/unaligned
diff --git a/libavcodec/x86/dsputilenc_yasm.asm b/libavcodec/x86/dsputilenc_yasm.asm
index 1be359d667..2d805c7da8 100644
--- a/libavcodec/x86/dsputilenc_yasm.asm
+++ b/libavcodec/x86/dsputilenc_yasm.asm
@@ -184,7 +184,7 @@ cglobal hadamard8_diff16_%1, 5, 6, %2
     call hadamard8x8_diff_%1
     add            r5d, eax
 
-.done
+.done:
     mov            eax, r5d
 %ifndef m8
     ADD            rsp, pad
@@ -288,7 +288,7 @@ cglobal sse16_sse2, 5, 5, 8
     pxor      m0, m0         ; mm0 = 0
     pxor      m7, m7         ; mm7 holds the sum
 
-.next2lines ; FIXME why are these unaligned movs? pix1[] is aligned
+.next2lines: ; FIXME why are these unaligned movs? pix1[] is aligned
     movu      m1, [r1   ]    ; mm1 = pix1[0][0-15]
     movu      m2, [r2   ]    ; mm2 = pix2[0][0-15]
     movu      m3, [r1+r3]    ; mm3 = pix1[1][0-15]
diff --git a/libavcodec/x86/fft_mmx.asm b/libavcodec/x86/fft_mmx.asm
index faa27a01c6..79c7c183b9 100644
--- a/libavcodec/x86/fft_mmx.asm
+++ b/libavcodec/x86/fft_mmx.asm
@@ -608,7 +608,7 @@ cglobal fft_calc, 2,5,8
     add     rcx, 3
     shl     r2, cl
     sub     r4, r2
-.loop
+.loop:
 %if mmsize == 8
     PSWAPD  m0, [r4 + r2 + 4]
     mova [r4 + r2 + 4], m0
diff --git a/libavcodec/x86/fmtconvert.asm b/libavcodec/x86/fmtconvert.asm
index 7368b8f518..21e0cce3b9 100644
--- a/libavcodec/x86/fmtconvert.asm
+++ b/libavcodec/x86/fmtconvert.asm
@@ -404,7 +404,7 @@ cglobal float_interleave2_%1, 3,4,%2, dst, src, len, src1
     mov     src1q, [srcq+gprsize]
     mov      srcq, [srcq        ]
     sub     src1q, srcq
-.loop
+.loop:
     MOVPS      m0, [srcq             ]
     MOVPS      m1, [srcq+src1q       ]
     MOVPS      m3, [srcq      +mmsize]
diff --git a/libavcodec/x86/h264_chromamc.asm b/libavcodec/x86/h264_chromamc.asm
index 4155947e58..bee833051f 100644
--- a/libavcodec/x86/h264_chromamc.asm
+++ b/libavcodec/x86/h264_chromamc.asm
@@ -69,7 +69,7 @@ SECTION .text
 
 %macro mv0_pixels_mc8 0
     lea           r4, [r2*2 ]
-.next4rows
+.next4rows:
     movq         mm0, [r1   ]
     movq         mm1, [r1+r2]
     add           r1, r4
@@ -117,7 +117,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7 + extra_regs, 0
     mv0_pixels_mc8
     REP_RET
 
-.at_least_one_non_zero
+.at_least_one_non_zero:
 %ifidn %2, rv40
 %if ARCH_X86_64
     mov           r7, r5
@@ -145,7 +145,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7 + extra_regs, 0
     test         r4d, r4d
     mov           r6, r2        ; dxy = x ? 1 : stride
     jne .both_non_zero
-.my_is_zero
+.my_is_zero:
     ; mx == 0 XOR my == 0 - 1 dimensional filter only
     or           r4d, r5d       ; x + y
 
@@ -166,7 +166,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7 + extra_regs, 0
     pxor          m7, m7
     psubw         m4, m5        ; mm4 = A = 8-x
 
-.next1drow
+.next1drow:
     movq          m0, [r1   ]   ; mm0 = src[0..7]
     movq          m2, [r1+r6]   ; mm1 = src[1..8]
 
@@ -197,7 +197,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7 + extra_regs, 0
     jne .next1drow
     REP_RET
 
-.both_non_zero ; general case, bilinear
+.both_non_zero: ; general case, bilinear
     movd          m4, r4d         ; x
     movd          m6, r5d         ; y
 %ifidn %2, rv40
@@ -232,7 +232,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7 + extra_regs, 0
 
     movq          m0, [r1  ]      ; mm0 = src[0..7]
     movq          m1, [r1+1]      ; mm1 = src[1..8]
-.next2drow
+.next2drow:
     add           r1, r2
 
     movq          m2, m0
@@ -330,7 +330,7 @@ cglobal %1_%2_chroma_mc4_%3, 6, 6 + extra_regs, 0
     pmullw        m6, m2
     paddw         m6, m0
 
-.next2rows
+.next2rows:
     movd          m0, [r1  ]
     movd          m1, [r1+1]
     add           r1, r2
@@ -397,7 +397,7 @@ cglobal %1_%2_chroma_mc2_%3, 6, 7, 0
     punpcklbw     m2, m7
     pshufw        m2, m2, 0x94    ; mm0 = src[0,1,1,2]
 
-.nextrow
+.nextrow:
     add           r1, r2
     movq          m1, m2
     pmaddwd       m1, m5          ; mm1 = A * src[0,1] + B * src[1,2]
@@ -474,7 +474,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7, 8
     mv0_pixels_mc8
     REP_RET
 
-.at_least_one_non_zero
+.at_least_one_non_zero:
     test         r5d, r5d
     je .my_is_zero
     test         r4d, r4d
@@ -501,7 +501,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7, 8
     movlhps       m7, m7
     movlhps       m6, m6
 
-.next2rows
+.next2rows:
     movq          m1, [r1+r2*1   ]
     movq          m2, [r1+r2*1+1]
     movq          m3, [r1+r2*2  ]
@@ -535,7 +535,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7, 8
     jg .next2rows
     REP_RET
 
-.my_is_zero
+.my_is_zero:
     mov          r5d, r4d
     shl          r4d, 8
     add           r4, 8
@@ -545,7 +545,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7, 8
     pshuflw       m7, m7, 0
     movlhps       m7, m7
 
-.next2xrows
+.next2xrows:
     movq          m0, [r1     ]
     movq          m1, [r1   +1]
     movq          m2, [r1+r2  ]
@@ -572,7 +572,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7, 8
     jg .next2xrows
     REP_RET
 
-.mx_is_zero
+.mx_is_zero:
     mov          r4d, r5d
     shl          r5d, 8
     add           r5, 8
@@ -582,7 +582,7 @@ cglobal %1_%2_chroma_mc8_%3, 6, 7, 8
     pshuflw       m7, m7, 0
     movlhps       m7, m7
 
-.next2yrows
+.next2yrows:
     movq          m0, [r1     ]
     movq          m1, [r1+r2  ]
     movdqa        m2, m1
@@ -632,7 +632,7 @@ cglobal %1_%2_chroma_mc4_%3, 6, 7, 0
     punpcklbw     m0, [r1+1]
     pshufw        m6, m6, 0
 
-.next2rows
+.next2rows:
     movd          m1, [r1+r2*1  ]
     movd          m3, [r1+r2*2  ]
     punpcklbw     m1, [r1+r2*1+1]
diff --git a/libavcodec/x86/h264_chromamc_10bit.asm b/libavcodec/x86/h264_chromamc_10bit.asm
index 370c7b5a46..f8a2cff68f 100644
--- a/libavcodec/x86/h264_chromamc_10bit.asm
+++ b/libavcodec/x86/h264_chromamc_10bit.asm
@@ -38,7 +38,7 @@ SECTION .text
 %macro MV0_PIXELS_MC8 0
     lea           r4, [r2*3   ]
     lea           r5, [r2*4   ]
-.next4rows
+.next4rows:
     movu          m0, [r1     ]
     movu          m1, [r1+r2  ]
     CHROMAMC_AVG  m0, [r0     ]
@@ -72,14 +72,14 @@ cglobal %1_h264_chroma_mc8_10, 6,7,8
     MV0_PIXELS_MC8
     REP_RET
 
-.at_least_one_non_zero
+.at_least_one_non_zero:
     mov          r6d, 2
     test         r5d, r5d
     je .x_interpolation
     mov           r6, r2        ; dxy = x ? 1 : stride
     test         r4d, r4d
     jne .xy_interpolation
-.x_interpolation
+.x_interpolation:
     ; mx == 0 XOR my == 0 - 1 dimensional filter only
     or           r4d, r5d       ; x + y
     movd          m5, r4d
@@ -88,7 +88,7 @@ cglobal %1_h264_chroma_mc8_10, 6,7,8
     SPLATW        m5, m5        ; mm5 = B = x
     psubw         m4, m5        ; mm4 = A = 8-x
 
-.next1drow
+.next1drow:
     movu          m0, [r1   ]   ; mm0 = src[0..7]
     movu          m2, [r1+r6]   ; mm2 = src[1..8]
 
@@ -107,7 +107,7 @@ cglobal %1_h264_chroma_mc8_10, 6,7,8
     jne .next1drow
     REP_RET
 
-.xy_interpolation ; general case, bilinear
+.xy_interpolation: ; general case, bilinear
     movd          m4, r4m         ; x
     movd          m6, r5m         ; y
 
@@ -125,7 +125,7 @@ cglobal %1_h264_chroma_mc8_10, 6,7,8
 
     movu          m0, [r1  ]      ; mm0 = src[0..7]
     movu          m1, [r1+2]      ; mm1 = src[1..8]
-.next2drow
+.next2drow:
     add           r1, r2
 
     pmullw        m2, m0, m4
@@ -192,7 +192,7 @@ cglobal %1_h264_chroma_mc4_10, 6,6,7
     pmullw        m6, m2
     paddw         m6, m0
 
-.next2rows
+.next2rows:
     MC4_OP m0, m6
     MC4_OP m6, m0
     sub   r3d, 2
@@ -221,7 +221,7 @@ cglobal %1_h264_chroma_mc2_10, 6,7
     pxor          m7, m7
     pshufw        m2, [r1], 0x94    ; mm0 = src[0,1,1,2]
 
-.nextrow
+.nextrow:
     add           r1, r2
     movq          m1, m2
     pmaddwd       m1, m5          ; mm1 = A * src[0,1] + B * src[1,2]
diff --git a/libavcodec/x86/h264_deblock_10bit.asm b/libavcodec/x86/h264_deblock_10bit.asm
index d625eee4a4..8a6bbb9d9e 100644
--- a/libavcodec/x86/h264_deblock_10bit.asm
+++ b/libavcodec/x86/h264_deblock_10bit.asm
@@ -625,7 +625,7 @@ cglobal deblock_v_luma_intra_10, 4,7,16
     shl    r2d, 2
     shl    r3d, 2
     LOAD_AB aa, bb, r2d, r3d
-.loop
+.loop:
     mova    p2, [r4+r1]
     mova    p1, [r4+2*r1]
     mova    p0, [r4+r5]
@@ -676,7 +676,7 @@ cglobal deblock_h_luma_intra_10, 4,7,16
     mova    m0, [pw_2]
     shl    r2d, 2
     shl    r3d, 2
-.loop
+.loop:
     movu    q3, [r0-8]
     movu    q2, [r0+r1-8]
     movu    q1, [r0+r1*2-8]
diff --git a/libavcodec/x86/h264_idct.asm b/libavcodec/x86/h264_idct.asm
index edd603b793..7003a14add 100644
--- a/libavcodec/x86/h264_idct.asm
+++ b/libavcodec/x86/h264_idct.asm
@@ -308,7 +308,7 @@ cglobal h264_idct_add16_8_mmx, 5, 7 + npicregs, 0, dst, block_offset, block, str
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -316,7 +316,7 @@ cglobal h264_idct_add16_8_mmx, 5, 7 + npicregs, 0, dst, block_offset, block, str
     mov         r6d, dword [r1+r5*4]
     lea          r6, [r0+r6]
     IDCT4_ADD    r6, r2, r3
-.skipblock
+.skipblock:
     inc          r5
     add          r2, 32
     cmp          r5, 16
@@ -333,7 +333,7 @@ cglobal h264_idct8_add4_8_mmx, 5, 7 + npicregs, 0, dst, block_offset, block, str
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -347,7 +347,7 @@ cglobal h264_idct8_add4_8_mmx, 5, 7 + npicregs, 0, dst, block_offset, block, str
     mov         r6d, dword [r1+r5*4]
     lea          r6, [r0+r6+4]
     IDCT8_ADD_MMX_END   r6  , rsp+8, r3
-.skipblock
+.skipblock:
     add          r5, 4
     add          r2, 128
     cmp          r5, 16
@@ -362,7 +362,7 @@ cglobal h264_idct_add16_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block, s
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -388,11 +388,11 @@ cglobal h264_idct_add16_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block, s
     cmp          r5, 16
     jl .nextblock
     REP_RET
-.no_dc
+.no_dc:
     mov         r6d, dword [r1+r5*4]
     add          r6, r0
     IDCT4_ADD    r6, r2, r3
-.skipblock
+.skipblock:
     inc          r5
     add          r2, 32
     cmp          r5, 16
@@ -406,7 +406,7 @@ cglobal h264_idct_add16intra_8_mmx, 5, 7 + npicregs, 0, dst, block_offset, block
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     or          r6w, word [r2]
@@ -415,7 +415,7 @@ cglobal h264_idct_add16intra_8_mmx, 5, 7 + npicregs, 0, dst, block_offset, block
     mov         r6d, dword [r1+r5*4]
     add          r6, r0
     IDCT4_ADD    r6, r2, r3
-.skipblock
+.skipblock:
     inc          r5
     add          r2, 32
     cmp          r5, 16
@@ -429,7 +429,7 @@ cglobal h264_idct_add16intra_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, blo
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -442,7 +442,7 @@ cglobal h264_idct_add16intra_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, blo
     cmp          r5, 16
     jl .nextblock
     REP_RET
-.try_dc
+.try_dc:
     movsx        r6, word [r2]
     test         r6, r6
     jz .skipblock
@@ -457,7 +457,7 @@ cglobal h264_idct_add16intra_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, blo
 %if ARCH_X86_64 == 0
     mov          r1, r1m
 %endif
-.skipblock
+.skipblock:
     inc          r5
     add          r2, 32
     cmp          r5, 16
@@ -474,7 +474,7 @@ cglobal h264_idct8_add4_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block, s
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -504,7 +504,7 @@ cglobal h264_idct8_add4_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block, s
 
     ADD         rsp, pad
     RET
-.no_dc
+.no_dc:
     mov         r6d, dword [r1+r5*4]
     add          r6, r0
     add   word [r2], 32
@@ -514,7 +514,7 @@ cglobal h264_idct8_add4_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block, s
     mov         r6d, dword [r1+r5*4]
     lea          r6, [r0+r6+4]
     IDCT8_ADD_MMX_END   r6  , rsp+8, r3
-.skipblock
+.skipblock:
     add          r5, 4
     add          r2, 128
     cmp          r5, 16
@@ -531,7 +531,7 @@ cglobal h264_idct8_add4_8_sse2, 5, 8 + npicregs, 10, dst1, block_offset, block,
 %ifdef PIC
     lea     picregq, [scan8_mem]
 %endif
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -560,7 +560,7 @@ INIT_MMX
     cmp          r5, 16
     jl .nextblock
     REP_RET
-.no_dc
+.no_dc:
 INIT_XMM
     mov       dst2d, dword [r1+r5*4]
     add       dst2q, r0
@@ -568,7 +568,7 @@ INIT_XMM
 %if ARCH_X86_64 == 0
     mov          r1, r1m
 %endif
-.skipblock
+.skipblock:
     add          r5, 4
     add          r2, 128
     cmp          r5, 16
@@ -577,7 +577,7 @@ INIT_XMM
 
 INIT_MMX
 h264_idct_add8_mmx_plane:
-.nextblock
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     or          r6w, word [r2]
@@ -592,7 +592,7 @@ h264_idct_add8_mmx_plane:
     add          r0, dword [r1+r5*4]
 %endif
     IDCT4_ADD    r0, r2, r3
-.skipblock
+.skipblock:
     inc          r5
     add          r2, 32
     test         r5, 3
@@ -621,8 +621,8 @@ cglobal h264_idct_add8_8_mmx, 5, 8 + npicregs, 0, dst1, block_offset, block, str
     call         h264_idct_add8_mmx_plane
     RET
 
-h264_idct_add8_mmx2_plane
-.nextblock
+h264_idct_add8_mmx2_plane:
+.nextblock:
     movzx        r6, byte [scan8+r5]
     movzx        r6, byte [r4+r6]
     test         r6, r6
@@ -641,7 +641,7 @@ h264_idct_add8_mmx2_plane
     test         r5, 3
     jnz .nextblock
     rep ret
-.try_dc
+.try_dc:
     movsx        r6, word [r2]
     test         r6, r6
     jz .skipblock
@@ -655,7 +655,7 @@ h264_idct_add8_mmx2_plane
     add          r0, dword [r1+r5*4]
 %endif
     DC_ADD_MMX2_OP movh, r0, r3, r6
-.skipblock
+.skipblock:
     inc          r5
     add          r2, 32
     test         r5, 3
@@ -734,7 +734,7 @@ h264_add8x4_idct_sse2:
     add         r0, r0m
 %endif
     call        h264_add8x4_idct_sse2
-.cycle%1end
+.cycle%1end:
 %if %1 < 7
     add         r2, 64
 %endif
@@ -770,7 +770,7 @@ cglobal h264_idct_add16_8_sse2, 5, 5 + ARCH_X86_64, 8
 %endif
     call        h264_add8x4_idct_sse2
     jmp .cycle%1end
-.try%1dc
+.try%1dc:
     movsx       r0, word [r2   ]
     or         r0w, word [r2+32]
     jz .cycle%1end
@@ -781,7 +781,7 @@ cglobal h264_idct_add16_8_sse2, 5, 5 + ARCH_X86_64, 8
     add         r0, r0m
 %endif
     call        h264_idct_dc_add8_mmx2
-.cycle%1end
+.cycle%1end:
 %if %1 < 7
     add         r2, 64
 %endif
@@ -817,7 +817,7 @@ cglobal h264_idct_add16intra_8_sse2, 5, 7 + ARCH_X86_64, 8
 %endif
     call        h264_add8x4_idct_sse2
     jmp .cycle%1end
-.try%1dc
+.try%1dc:
     movsx       r0, word [r2   ]
     or         r0w, word [r2+32]
     jz .cycle%1end
@@ -830,7 +830,7 @@ cglobal h264_idct_add16intra_8_sse2, 5, 7 + ARCH_X86_64, 8
     add         r0, dword [r1+(%1&1)*8+64*(1+(%1>>1))]
 %endif
     call        h264_idct_dc_add8_mmx2
-.cycle%1end
+.cycle%1end:
 %if %1 == 1
     add         r2, 384+64
 %elif %1 < 3
diff --git a/libavcodec/x86/h264_idct_10bit.asm b/libavcodec/x86/h264_idct_10bit.asm
index 67bc68ec5c..3d4ac5fb71 100644
--- a/libavcodec/x86/h264_idct_10bit.asm
+++ b/libavcodec/x86/h264_idct_10bit.asm
@@ -225,7 +225,7 @@ IDCT8_DC_ADD
 ; h264_idct_add16intra(pixel *dst, const int *block_offset, dctcoef *block, int stride, const uint8_t nnzc[6*8])
 ;-----------------------------------------------------------------------------
 %macro AC 1
-.ac%1
+.ac%1:
     mov  r5d, [r1+(%1+0)*4]
     call add4x4_idct %+ SUFFIX
     mov  r5d, [r1+(%1+1)*4]
diff --git a/libavcodec/x86/h264_intrapred.asm b/libavcodec/x86/h264_intrapred.asm
index fd78456294..609cb2303d 100644
--- a/libavcodec/x86/h264_intrapred.asm
+++ b/libavcodec/x86/h264_intrapred.asm
@@ -484,7 +484,7 @@ cglobal pred16x16_plane_%1, 2,9,7
 %endif
 
     mov          r4, 8
-.loop
+.loop:
     mova         m3, m0           ; b[0..7]
     mova         m4, m2           ; b[8..15]
     psraw        m3, 5
@@ -680,7 +680,7 @@ cglobal pred8x8_plane, 2,9,7
 
     mov          r4, 4
 ALIGN 16
-.loop
+.loop:
 %if mmsize == 16
     mova         m3, m0           ; b[0..7]
     paddw        m0, m1
@@ -1045,7 +1045,7 @@ cglobal pred8x8l_top_dc_%1, 4,4
     psrlq       mm5, 56
     psllq       mm5, 56
     pxor        mm1, mm5
-.body
+.body:
     PRED4x4_LOWPASS mm0, mm2, mm1, mm3, mm5
     psadbw   mm7, mm0
     paddw    mm7, [pw_4]
@@ -1141,7 +1141,7 @@ cglobal pred8x8l_dc_%1, 4,5
     jz .fix_lt_2
     test         r2, r2
     jz .fix_tr_1
-.body
+.body:
     lea          r1, [r0+r3*2]
     PRED4x4_LOWPASS mm6, mm2, mm1, mm3, mm5
     pxor        mm0, mm0
@@ -1276,7 +1276,7 @@ cglobal pred8x8l_vertical_%1, 4,4
     psrlq       mm5, 56
     psllq       mm5, 56
     pxor        mm1, mm5
-.body
+.body:
     PRED4x4_LOWPASS mm0, mm2, mm1, mm3, mm5
 %rep 3
     movq [r0+r3*1], mm0
@@ -1576,7 +1576,7 @@ cglobal pred8x8l_down_right_mmxext, 4,5
     psllq       mm5, 56
     pxor        mm1, mm5
     jmp .do_top
-.body
+.body:
     lea         r1, [r0+r3*2]
     movq       mm1, mm7
     movq       mm7, mm5
@@ -1822,7 +1822,7 @@ cglobal pred8x8l_vertical_right_mmxext, 4,5
     jz .fix_lt_2
     test         r2, r2
     jz .fix_tr_1
-.do_top
+.do_top:
     PRED4x4_LOWPASS mm6, mm2, mm1, mm3, mm5
     lea         r1, [r0+r3*2]
     movq       mm2, mm6
@@ -1931,7 +1931,7 @@ cglobal pred8x8l_vertical_right_%1, 4,5,7
     jz .fix_lt_2
     test         r2, r2
     jz .fix_tr_1
-.do_top
+.do_top:
     PRED4x4_LOWPASS mm6, mm2, mm1, mm3, mm5
     lea           r1, [r0+r3*2]
     movq2dq     xmm4, mm6
diff --git a/libavcodec/x86/h264_qpel_10bit.asm b/libavcodec/x86/h264_qpel_10bit.asm
index 788d715d61..f380cfc77b 100644
--- a/libavcodec/x86/h264_qpel_10bit.asm
+++ b/libavcodec/x86/h264_qpel_10bit.asm
@@ -264,7 +264,7 @@ cglobal_mc %1, %2, mc20, %3, 3,4,9
 %else
     %define p16 [pw_16]
 %endif
-.nextrow
+.nextrow:
 %if %0 == 4
     movu     m2, [r1-4]
     movu     m3, [r1-2]
@@ -330,7 +330,7 @@ MC_CACHE MC30
 %macro MC10 3-4
 cglobal_mc %1, %2, mc10, %3, 3,5,9
     mov      r4, r1
-.body
+.body:
     mov     r3d, %3
     mova     m1, [pw_pixel_max]
 %if num_mmregs > 8
@@ -339,7 +339,7 @@ cglobal_mc %1, %2, mc10, %3, 3,5,9
 %else
     %define p16 [pw_16]
 %endif
-.nextrow
+.nextrow:
 %if %0 == 4
     movu     m2, [r1-4]
     movu     m3, [r1-2]
@@ -446,7 +446,7 @@ MC MC02
 %macro MC01 3
 cglobal_mc %1, %2, mc01, %3, 3,5,8
     mov      r4, r1
-.body
+.body:
     PRELOAD_V
 
     sub      r4, r2
@@ -535,7 +535,7 @@ SWAP 0,1,2,3,4,5
 ; this REALLY needs x86_64
 cglobal_mc %1, %2, mc11, %3, 3,6,8
     mov      r4, r1
-.body
+.body:
     PRELOAD_V
 
     sub      r0, r2
@@ -778,7 +778,7 @@ cglobal_mc %1, %2, mc12, %3, 3,7,12
     call put_hv%3_10_%1
 
     xor       r4d, r4d
-.body
+.body:
     mov       r3d, %3
     pxor       m0, m0
     mova       m7, [pw_pixel_max]
@@ -837,7 +837,7 @@ put_h%2_10_%1:
     mov       r3d, %2
     xor       r4d, r4d
     mova       m6, [pad20]
-.nextrow
+.nextrow:
     movu       m2, [r5-4]
     movu       m3, [r5-2]
     movu       m4, [r5+0]
@@ -864,7 +864,7 @@ H_NRD sse2  , 8
 %macro MC21 3
 cglobal_mc %1, %2, mc21, %3, 3,7,12
     mov   r5, r1
-.body
+.body:
 %define PAD mmsize*8*3*2   ; SIZE*16*4*sizeof(pixel)
     mov   r6, rsp          ; backup stack pointer
     and  rsp, ~(mmsize-1)  ; align stack
diff --git a/libavcodec/x86/h264_weight.asm b/libavcodec/x86/h264_weight.asm
index 26233bf080..2ad5a8f458 100644
--- a/libavcodec/x86/h264_weight.asm
+++ b/libavcodec/x86/h264_weight.asm
@@ -73,7 +73,7 @@ SECTION .text
 INIT_MMX
 cglobal h264_weight_16_mmx2, 6, 6, 0
     WEIGHT_SETUP
-.nextrow
+.nextrow:
     WEIGHT_OP 0,  4
     mova     [r0  ], m0
     WEIGHT_OP 8, 12
@@ -86,7 +86,7 @@ cglobal h264_weight_16_mmx2, 6, 6, 0
 %macro WEIGHT_FUNC_MM 3
 cglobal h264_weight_%1_%3, 6, 6, %2
     WEIGHT_SETUP
-.nextrow
+.nextrow:
     WEIGHT_OP 0, mmsize/2
     mova     [r0], m0
     add        r0, r1
@@ -105,7 +105,7 @@ cglobal h264_weight_%1_%3, 6, 6, %2
     WEIGHT_SETUP
     sar       r2d, 1
     lea        r3, [r1*2]
-.nextrow
+.nextrow:
     WEIGHT_OP 0, r1
     movh     [r0], m0
 %if mmsize == 16
@@ -178,7 +178,7 @@ INIT_MMX
 cglobal h264_biweight_16_mmx2, 7, 8, 0
     BIWEIGHT_SETUP
     movifnidn r3d, r3m
-.nextrow
+.nextrow:
     BIWEIGHT_STEPA 0, 1, 0
     BIWEIGHT_STEPA 1, 2, 4
     BIWEIGHT_STEPB
@@ -197,7 +197,7 @@ cglobal h264_biweight_16_mmx2, 7, 8, 0
 cglobal h264_biweight_%1_%3, 7, 8, %2
     BIWEIGHT_SETUP
     movifnidn r3d, r3m
-.nextrow
+.nextrow:
     BIWEIGHT_STEPA 0, 1, 0
     BIWEIGHT_STEPA 1, 2, mmsize/2
     BIWEIGHT_STEPB
@@ -220,7 +220,7 @@ cglobal h264_biweight_%1_%3, 7, 8, %2
     movifnidn r3d, r3m
     sar        r3, 1
     lea        r4, [r2*2]
-.nextrow
+.nextrow:
     BIWEIGHT_STEPA 0, 1, 0
     BIWEIGHT_STEPA 1, 2, r2
     BIWEIGHT_STEPB
@@ -288,7 +288,7 @@ cglobal h264_biweight_16_ssse3, 7, 8, 8
     BIWEIGHT_SSSE3_SETUP
     movifnidn r3d, r3m
 
-.nextrow
+.nextrow:
     movh       m0, [r0]
     movh       m2, [r0+8]
     movh       m3, [r1+8]
@@ -309,7 +309,7 @@ cglobal h264_biweight_8_ssse3, 7, 8, 8
     sar        r3, 1
     lea        r4, [r2*2]
 
-.nextrow
+.nextrow:
     movh       m0, [r0]
     movh       m1, [r1]
     movh       m2, [r0+r2]
diff --git a/libavcodec/x86/h264_weight_10bit.asm b/libavcodec/x86/h264_weight_10bit.asm
index dec9aba461..24386f882e 100644
--- a/libavcodec/x86/h264_weight_10bit.asm
+++ b/libavcodec/x86/h264_weight_10bit.asm
@@ -40,7 +40,7 @@ SECTION .text
 ;                  int weight, int offset);
 ;-----------------------------------------------------------------------------
 %macro WEIGHT_PROLOGUE 0
-.prologue
+.prologue:
     PROLOGUE 0,6,8
     movifnidn  r0, r0mp
     movifnidn r1d, r1m
@@ -93,7 +93,7 @@ SECTION .text
 cglobal h264_weight_16_10_%1
     WEIGHT_PROLOGUE
     WEIGHT_SETUP %1
-.nextrow
+.nextrow:
     WEIGHT_OP %1,  0
     mova [r0   ], m5
     WEIGHT_OP %1, 16
@@ -113,7 +113,7 @@ WEIGHT_FUNC_DBL sse4
 cglobal h264_weight_8_10_%1
     WEIGHT_PROLOGUE
     WEIGHT_SETUP %1
-.nextrow
+.nextrow:
     WEIGHT_OP  %1, 0
     mova     [r0], m5
     add        r0, r1
@@ -133,7 +133,7 @@ cglobal h264_weight_4_10_%1
     sar         r2d, 1
     WEIGHT_SETUP %1
     lea         r3, [r1*2]
-.nextrow
+.nextrow:
     WEIGHT_OP   %1, 0, r1
     movh      [r0], m5
     movhps [r0+r1], m5
@@ -159,7 +159,7 @@ DECLARE_REG_TMP 7
 %endif
 
 %macro BIWEIGHT_PROLOGUE 0
-.prologue
+.prologue:
     PROLOGUE 0,8,8
     movifnidn  r0, r0mp
     movifnidn  r1, r1mp
@@ -221,7 +221,7 @@ DECLARE_REG_TMP 7
 cglobal h264_biweight_16_10_%1
     BIWEIGHT_PROLOGUE
     BIWEIGHT_SETUP %1
-.nextrow
+.nextrow:
     BIWEIGHT  %1,  0
     mova [r0   ], m0
     BIWEIGHT  %1, 16
@@ -241,7 +241,7 @@ BIWEIGHT_FUNC_DBL sse4
 cglobal h264_biweight_8_10_%1
     BIWEIGHT_PROLOGUE
     BIWEIGHT_SETUP %1
-.nextrow
+.nextrow:
     BIWEIGHT %1, 0
     mova   [r0], m0
     add      r0, r2
@@ -261,7 +261,7 @@ cglobal h264_biweight_4_10_%1
     BIWEIGHT_SETUP %1
     sar        r3d, 1
     lea        r4, [r2*2]
-.nextrow
+.nextrow:
     BIWEIGHT    %1, 0, r2
     movh   [r0   ], m0
     movhps [r0+r2], m0
diff --git a/libavcodec/x86/vp56dsp.asm b/libavcodec/x86/vp56dsp.asm
index 21ff041d61..bc09903575 100644
--- a/libavcodec/x86/vp56dsp.asm
+++ b/libavcodec/x86/vp56dsp.asm
@@ -139,7 +139,7 @@ cglobal vp6_filter_diag4, 5, 7, 8
 
     mov          r3, rsp
     mov          r6, 11
-.nextrow
+.nextrow:
     DIAG4        r1, -1, 0, 1, 2, r3
     add          r3, 8
     add          r1, r2
@@ -151,7 +151,7 @@ cglobal vp6_filter_diag4, 5, 7, 8
 
     lea          r3, [rsp+8]
     mov          r6, 8
-.nextcol
+.nextcol:
     DIAG4        r3, -8, 0, 8, 16, r0
     add          r3, 8
     add          r0, r2
diff --git a/libavcodec/x86/vp8dsp.asm b/libavcodec/x86/vp8dsp.asm
index 75c0b56f94..19853c494f 100644
--- a/libavcodec/x86/vp8dsp.asm
+++ b/libavcodec/x86/vp8dsp.asm
@@ -189,7 +189,7 @@ cglobal put_vp8_epel%1_h6, 6, 6 + npicregs, 8, dst, dststride, src, srcstride, h
     mova      m6, [sixtap_filter_hb+mxq*8-32]
     mova      m7, [sixtap_filter_hb+mxq*8-16]
 
-.nextrow
+.nextrow:
     movu      m0, [srcq-2]
     mova      m1, m0
     mova      m2, m0
@@ -229,7 +229,7 @@ cglobal put_vp8_epel%1_h4, 6, 6 + npicregs, 7, dst, dststride, src, srcstride, h
     mova      m5, [fourtap_filter_hb+mxq-16] ; set up 4tap filter in bytes
     mova      m6, [fourtap_filter_hb+mxq]
 
-.nextrow
+.nextrow:
     movu      m0, [srcq-1]
     mova      m1, m0
     pshufb    m0, m3
@@ -264,7 +264,7 @@ cglobal put_vp8_epel%1_v4, 7, 7, 8, dst, dststride, src, srcstride, height, picr
     movh      m2, [srcq+2*srcstrideq]
     add     srcq, srcstrideq
 
-.nextrow
+.nextrow:
     movh      m3, [srcq+2*srcstrideq]      ; read new row
     mova      m4, m0
     mova      m0, m1
@@ -304,7 +304,7 @@ cglobal put_vp8_epel%1_v6, 7, 7, 8, dst, dststride, src, srcstride, height, picr
     movh      m3, [srcq]
     movh      m4, [srcq+srcstrideq]
 
-.nextrow
+.nextrow:
     movh      m5, [srcq+2*srcstrideq]      ; read new row
     mova      m6, m0
     punpcklbw m6, m5
@@ -350,7 +350,7 @@ cglobal put_vp8_epel4_h4, 6, 6 + npicregs, 0, dst, dststride, src, srcstride, he
     movq      mm7, [pw_64]
     pxor      mm6, mm6
 
-.nextrow
+.nextrow:
     movq      mm1, [srcq-1]                ; (ABCDEFGH) load 8 horizontal pixels
 
     ; first set of 2 pixels
@@ -399,7 +399,7 @@ cglobal put_vp8_epel4_h6, 6, 6 + npicregs, 0, dst, dststride, src, srcstride, he
     movq      mm7, [pw_64]
     pxor      mm3, mm3
 
-.nextrow
+.nextrow:
     movq      mm1, [srcq-2]                ; (ABCDEFGH) load 8 horizontal pixels
 
     ; first set of 2 pixels
@@ -459,7 +459,7 @@ cglobal put_vp8_epel8_h4, 6, 6 + npicregs, 10, dst, dststride, src, srcstride, h
     mova      m8, [mxq+32]
     mova      m9, [mxq+48]
 %endif
-.nextrow
+.nextrow:
     movq      m0, [srcq-1]
     movq      m1, [srcq-0]
     movq      m2, [srcq+1]
@@ -510,7 +510,7 @@ cglobal put_vp8_epel8_h6, 6, 6 + npicregs, 14, dst, dststride, src, srcstride, h
     mova     m12, [mxq+64]
     mova     m13, [mxq+80]
 %endif
-.nextrow
+.nextrow:
     movq      m0, [srcq-2]
     movq      m1, [srcq-1]
     movq      m2, [srcq-0]
@@ -577,7 +577,7 @@ cglobal put_vp8_epel%1_v4, 7, 7, 8, dst, dststride, src, srcstride, height, picr
     punpcklbw m1, m7
     punpcklbw m2, m7
 
-.nextrow
+.nextrow:
     ; first calculate negative taps (to prevent losing positive overflows)
     movh      m4, [srcq+2*srcstrideq]      ; read new row
     punpcklbw m4, m7
@@ -635,7 +635,7 @@ cglobal put_vp8_epel%1_v6, 7, 7, 8, dst, dststride, src, srcstride, height, picr
     punpcklbw m3, m7
     punpcklbw m4, m7
 
-.nextrow
+.nextrow:
     ; first calculate negative taps (to prevent losing positive overflows)
     mova      m5, m1
     pmullw    m5, [myq+16]
@@ -689,7 +689,7 @@ cglobal put_vp8_bilinear%1_v, 7, 7, 7, dst, dststride, src, srcstride, height, p
     mova      m5, [bilinear_filter_vw+myq-1*16]
     neg      myq
     mova      m4, [bilinear_filter_vw+myq+7*16]
-.nextrow
+.nextrow:
     movh      m0, [srcq+srcstrideq*0]
     movh      m1, [srcq+srcstrideq*1]
     movh      m3, [srcq+srcstrideq*2]
@@ -733,7 +733,7 @@ cglobal put_vp8_bilinear%1_h, 6, 6 + npicregs, 7, dst, dststride, src, srcstride
     mova      m5, [bilinear_filter_vw+mxq-1*16]
     neg      mxq
     mova      m4, [bilinear_filter_vw+mxq+7*16]
-.nextrow
+.nextrow:
     movh      m0, [srcq+srcstrideq*0+0]
     movh      m1, [srcq+srcstrideq*0+1]
     movh      m2, [srcq+srcstrideq*1+0]
@@ -783,7 +783,7 @@ cglobal put_vp8_bilinear%1_v, 7, 7, 5, dst, dststride, src, srcstride, height, p
 %endif
     pxor      m4, m4
     mova      m3, [bilinear_filter_vb+myq-16]
-.nextrow
+.nextrow:
     movh      m0, [srcq+srcstrideq*0]
     movh      m1, [srcq+srcstrideq*1]
     movh      m2, [srcq+srcstrideq*2]
@@ -820,7 +820,7 @@ cglobal put_vp8_bilinear%1_h, 6, 6 + npicregs, 5, dst, dststride, src, srcstride
     pxor      m4, m4
     mova      m2, [filter_h2_shuf]
     mova      m3, [bilinear_filter_vb+mxq-16]
-.nextrow
+.nextrow:
     movu      m0, [srcq+srcstrideq*0]
     movu      m1, [srcq+srcstrideq*1]
     pshufb    m0, m2
@@ -1488,7 +1488,7 @@ cglobal vp8_%1_loop_filter_simple, 3, %2, 8, dst, stride, flim, cntr
 %endif
 
 %if mmsize == 8 ; mmx / mmxext
-.next8px
+.next8px:
 %endif
 %ifidn %1, v
     ; read 4 half/full rows of pixels
diff --git a/libavresample/x86/audio_mix.asm b/libavresample/x86/audio_mix.asm
index 8aeac8b242..bab4292e13 100644
--- a/libavresample/x86/audio_mix.asm
+++ b/libavresample/x86/audio_mix.asm
@@ -361,7 +361,7 @@ cglobal mix_%1_to_%2_%3_flt, 3,in_channels+2,needed_mmregs+matrix_elements_mm, s
     mov         src0q, [src0q]
     add         src0q, lenq
     neg          lenq
-.loop
+.loop:
 ; for x86-32 with 7-8 channels we do not have enough gp registers for all src
 ; pointers, so we have to load some of them from the stack each time
 %define copy_src_from_stack ARCH_X86_32 && in_channels >= 7 && %%i >= 5
diff --git a/libavutil/x86/float_dsp.asm b/libavutil/x86/float_dsp.asm
index 7a18a20aca..ce823302f6 100644
--- a/libavutil/x86/float_dsp.asm
+++ b/libavutil/x86/float_dsp.asm
@@ -32,7 +32,7 @@ SECTION .text
 cglobal vector_fmul, 4,4,2, dst, src0, src1, len
     lea       lenq, [lend*4 - 2*mmsize]
 ALIGN 16
-.loop
+.loop:
     mova      m0,   [src0q + lenq]
     mova      m1,   [src0q + lenq + mmsize]
     mulps     m0, m0, [src1q + lenq]
@@ -74,7 +74,7 @@ cglobal vector_fmac_scalar, 4,4,3, dst, src, mul, len
 %endif
 %endif
     lea    lenq, [lend*4-2*mmsize]
-.loop
+.loop:
     mulps    m1, m0, [srcq+lenq       ]
     mulps    m2, m0, [srcq+lenq+mmsize]
     addps    m1, m1, [dstq+lenq       ]
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index d29740f278..01be0f8fc0 100644
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -110,12 +110,14 @@
     default rel
 %endif
 
+%macro CPUNOP 1
+    %if HAVE_CPUNOP
+        CPU %1
+    %endif
+%endmacro
+
 ; Always use long nops (reduces 0x90 spam in disassembly on x86_32)
-; Not supported by NASM (except via smartalign package + ALIGNMODE k8,
-; however that fails when used together with the -M option)
-%ifdef __YASM_VER__
-CPU amdnop
-%endif
+CPUNOP amdnop
 
 ; Macros to eliminate most code duplication between x86_32 and x86_64:
 ; Currently this works only for leaf functions which load all their arguments
@@ -522,22 +524,8 @@ DECLARE_ARG 7, 8, 9, 10, 11, 12, 13, 14
 ; Applies any symbol mangling needed for C linkage, and sets up a define such that
 ; subsequent uses of the function name automatically refer to the mangled version.
 ; Appends cpuflags to the function name if cpuflags has been specified.
-%macro cglobal 1-2+ ; name, [PROLOGUE args]
-%if %0 == 1
-    ; HACK: work around %+ broken with empty SUFFIX for nasm 2.09.10
-    %ifndef cpuname
-    cglobal_internal %1
-    %else
-    cglobal_internal %1 %+ SUFFIX
-    %endif
-%else
-    ; HACK: work around %+ broken with empty SUFFIX for nasm 2.09.10
-    %ifndef cpuname
-    cglobal_internal %1, %2
-    %else
+%macro cglobal 1-2+ "" ; name, [PROLOGUE args]
     cglobal_internal %1 %+ SUFFIX, %2
-    %endif
-%endif
 %endmacro
 %macro cglobal_internal 1-2+
     %ifndef cglobaled_%1
@@ -555,7 +543,7 @@ DECLARE_ARG 7, 8, 9, 10, 11, 12, 13, 14
     %1:
     RESET_MM_PERMUTATION ; not really needed, but makes disassembly somewhat nicer
     %assign stack_offset 0
-    %if %0 > 1
+    %ifnidn %2, ""
         PROLOGUE %2
     %endif
 %endmacro
@@ -622,9 +610,7 @@ SECTION .note.GNU-stack noalloc noexec nowrite progbits
 ; All subsequent functions (up to the next INIT_CPUFLAGS) is built for the specified cpu.
 ; You shouldn't need to invoke this macro directly, it's a subroutine for INIT_MMX &co.
 %macro INIT_CPUFLAGS 0-2
-%ifdef __YASM_VER__
-    CPU amdnop
-%endif
+    CPUNOP amdnop
     %if %0 >= 1
         %xdefine cpuname %1
         %assign cpuflags cpuflags_%1
@@ -648,7 +634,7 @@ SECTION .note.GNU-stack noalloc noexec nowrite progbits
         %endif
         %ifdef __YASM_VER__
         %if notcpuflag(mmx2)
-            CPU basicnop
+            CPUNOP basicnop
         %endif
         %endif
     %else
@@ -826,18 +812,13 @@ INIT_XMM
 
 ; Append cpuflags to the callee's name iff the appended name is known and the plain name isn't
 %macro call 1
-    ; HACK: work around %+ broken with empty SUFFIX for nasm 2.09.10
-    %ifndef cpuname
-    call_internal %1, %1
-    %else
-    call_internal %1, %1 %+ SUFFIX
-    %endif
+    call_internal %1 %+ SUFFIX, %1
 %endmacro
 %macro call_internal 2
-    %xdefine %%i %1
-    %ifndef cglobaled_%1
-        %ifdef cglobaled_%2
-            %xdefine %%i %2
+    %xdefine %%i %2
+    %ifndef cglobaled_%2
+        %ifdef cglobaled_%1
+            %xdefine %%i %1
         %endif
     %endif
     call %%i