diff options
Diffstat (limited to 'libc/sysdeps/x86_64/multiarch/memcpy-ssse3.S')
| -rw-r--r-- | libc/sysdeps/x86_64/multiarch/memcpy-ssse3.S | 254 |
1 files changed, 125 insertions, 129 deletions
diff --git a/libc/sysdeps/x86_64/multiarch/memcpy-ssse3.S b/libc/sysdeps/x86_64/multiarch/memcpy-ssse3.S index 9642ceecd..0cedab244 100644 --- a/libc/sysdeps/x86_64/multiarch/memcpy-ssse3.S +++ b/libc/sysdeps/x86_64/multiarch/memcpy-ssse3.S @@ -31,10 +31,6 @@ # define MEMCPY_CHK __memcpy_chk_ssse3 #endif -#ifndef ALIGN -# define ALIGN(n) .p2align n -#endif - #define JMPTBL(I, B) I - B /* Branch to an entry in a jump table. TABLE is a jump table with @@ -80,7 +76,7 @@ L(copy_forward): jmp *%r9 ud2 - ALIGN (4) + .p2align 4 L(80bytesormore): #ifndef USE_AS_MEMMOVE cmp %dil, %sil @@ -113,7 +109,7 @@ L(80bytesormore): #endif BRANCH_TO_JMPTBL_ENTRY (L(shl_table), %r9, 4) - ALIGN (4) + .p2align 4 L(copy_backward): movdqu -16(%rsi, %rdx), %xmm0 add %rdx, %rsi @@ -144,7 +140,7 @@ L(copy_backward): #endif BRANCH_TO_JMPTBL_ENTRY (L(shl_table_bwd), %r9, 4) - ALIGN (4) + .p2align 4 L(shl_0): sub $16, %rdx movdqa (%rsi), %xmm1 @@ -172,7 +168,7 @@ L(shl_0_less_64bytes): add %rdx, %rdi BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_0_gobble): #ifdef DATA_CACHE_SIZE_HALF cmp $DATA_CACHE_SIZE_HALF, %RDX_LP @@ -228,7 +224,7 @@ L(shl_0_cache_less_64bytes): add %rdx, %rdi BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_0_gobble_mem_loop): prefetcht0 0x1c0(%rsi) prefetcht0 0x280(%rsi) @@ -287,7 +283,7 @@ L(shl_0_mem_less_32bytes): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_0_bwd): sub $16, %rdx movdqa -0x10(%rsi), %xmm1 @@ -313,7 +309,7 @@ L(shl_0_bwd): L(shl_0_less_64bytes_bwd): BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_0_gobble_bwd): #ifdef DATA_CACHE_SIZE_HALF cmp $DATA_CACHE_SIZE_HALF, %RDX_LP @@ -367,7 +363,7 @@ L(shl_0_gobble_bwd_loop): L(shl_0_gobble_bwd_less_64bytes): BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_0_gobble_mem_bwd_loop): prefetcht0 -0x1c0(%rsi) prefetcht0 -0x280(%rsi) @@ -423,7 +419,7 @@ L(shl_0_mem_bwd_less_64bytes): L(shl_0_mem_bwd_less_32bytes): BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_1): lea (L(shl_1_loop_L1)-L(shl_1))(%r9), %r9 cmp %rcx, %rdx @@ -466,7 +462,7 @@ L(shl_1_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_1_bwd): lea (L(shl_1_bwd_loop_L1)-L(shl_1_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -508,7 +504,7 @@ L(shl_1_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_2): lea (L(shl_2_loop_L1)-L(shl_2))(%r9), %r9 cmp %rcx, %rdx @@ -551,7 +547,7 @@ L(shl_2_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_2_bwd): lea (L(shl_2_bwd_loop_L1)-L(shl_2_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -593,7 +589,7 @@ L(shl_2_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_3): lea (L(shl_3_loop_L1)-L(shl_3))(%r9), %r9 cmp %rcx, %rdx @@ -636,7 +632,7 @@ L(shl_3_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_3_bwd): lea (L(shl_3_bwd_loop_L1)-L(shl_3_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -678,7 +674,7 @@ L(shl_3_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_4): lea (L(shl_4_loop_L1)-L(shl_4))(%r9), %r9 cmp %rcx, %rdx @@ -721,7 +717,7 @@ L(shl_4_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_4_bwd): lea (L(shl_4_bwd_loop_L1)-L(shl_4_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -763,7 +759,7 @@ L(shl_4_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_5): lea (L(shl_5_loop_L1)-L(shl_5))(%r9), %r9 cmp %rcx, %rdx @@ -806,7 +802,7 @@ L(shl_5_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_5_bwd): lea (L(shl_5_bwd_loop_L1)-L(shl_5_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -848,7 +844,7 @@ L(shl_5_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_6): lea (L(shl_6_loop_L1)-L(shl_6))(%r9), %r9 cmp %rcx, %rdx @@ -891,7 +887,7 @@ L(shl_6_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_6_bwd): lea (L(shl_6_bwd_loop_L1)-L(shl_6_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -933,7 +929,7 @@ L(shl_6_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_7): lea (L(shl_7_loop_L1)-L(shl_7))(%r9), %r9 cmp %rcx, %rdx @@ -976,7 +972,7 @@ L(shl_7_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_7_bwd): lea (L(shl_7_bwd_loop_L1)-L(shl_7_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1018,7 +1014,7 @@ L(shl_7_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_8): lea (L(shl_8_loop_L1)-L(shl_8))(%r9), %r9 cmp %rcx, %rdx @@ -1051,7 +1047,7 @@ L(shl_8_loop_L1): movaps %xmm5, -0x10(%rdi) jmp *%r9 ud2 - ALIGN (4) + .p2align 4 L(shl_8_end): lea 64(%rdx), %rdx movaps %xmm4, -0x20(%rdi) @@ -1061,7 +1057,7 @@ L(shl_8_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_8_bwd): lea (L(shl_8_bwd_loop_L1)-L(shl_8_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1103,7 +1099,7 @@ L(shl_8_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_9): lea (L(shl_9_loop_L1)-L(shl_9))(%r9), %r9 cmp %rcx, %rdx @@ -1146,7 +1142,7 @@ L(shl_9_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_9_bwd): lea (L(shl_9_bwd_loop_L1)-L(shl_9_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1188,7 +1184,7 @@ L(shl_9_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_10): lea (L(shl_10_loop_L1)-L(shl_10))(%r9), %r9 cmp %rcx, %rdx @@ -1231,7 +1227,7 @@ L(shl_10_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_10_bwd): lea (L(shl_10_bwd_loop_L1)-L(shl_10_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1273,7 +1269,7 @@ L(shl_10_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_11): lea (L(shl_11_loop_L1)-L(shl_11))(%r9), %r9 cmp %rcx, %rdx @@ -1316,7 +1312,7 @@ L(shl_11_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_11_bwd): lea (L(shl_11_bwd_loop_L1)-L(shl_11_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1358,7 +1354,7 @@ L(shl_11_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_12): lea (L(shl_12_loop_L1)-L(shl_12))(%r9), %r9 cmp %rcx, %rdx @@ -1401,7 +1397,7 @@ L(shl_12_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_12_bwd): lea (L(shl_12_bwd_loop_L1)-L(shl_12_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1443,7 +1439,7 @@ L(shl_12_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_13): lea (L(shl_13_loop_L1)-L(shl_13))(%r9), %r9 cmp %rcx, %rdx @@ -1486,7 +1482,7 @@ L(shl_13_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_13_bwd): lea (L(shl_13_bwd_loop_L1)-L(shl_13_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1528,7 +1524,7 @@ L(shl_13_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_14): lea (L(shl_14_loop_L1)-L(shl_14))(%r9), %r9 cmp %rcx, %rdx @@ -1571,7 +1567,7 @@ L(shl_14_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_14_bwd): lea (L(shl_14_bwd_loop_L1)-L(shl_14_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1613,7 +1609,7 @@ L(shl_14_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_15): lea (L(shl_15_loop_L1)-L(shl_15))(%r9), %r9 cmp %rcx, %rdx @@ -1656,7 +1652,7 @@ L(shl_15_end): add %rdx, %rsi BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(shl_15_bwd): lea (L(shl_15_bwd_loop_L1)-L(shl_15_bwd))(%r9), %r9 cmp %rcx, %rdx @@ -1698,7 +1694,7 @@ L(shl_15_bwd_end): movdqu %xmm0, (%r8) BRANCH_TO_JMPTBL_ENTRY(L(table_less_80bytes), %rdx, 4) - ALIGN (4) + .p2align 4 L(write_72bytes): movdqu -72(%rsi), %xmm0 movdqu -56(%rsi), %xmm1 @@ -1716,7 +1712,7 @@ L(write_72bytes): mov %rcx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_64bytes): movdqu -64(%rsi), %xmm0 mov -48(%rsi), %rcx @@ -1734,7 +1730,7 @@ L(write_64bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_56bytes): movdqu -56(%rsi), %xmm0 mov -40(%rsi), %r8 @@ -1750,7 +1746,7 @@ L(write_56bytes): mov %rcx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_48bytes): mov -48(%rsi), %rcx mov -40(%rsi), %r8 @@ -1766,7 +1762,7 @@ L(write_48bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_40bytes): mov -40(%rsi), %r8 mov -32(%rsi), %r9 @@ -1780,7 +1776,7 @@ L(write_40bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_32bytes): mov -32(%rsi), %r9 mov -24(%rsi), %r10 @@ -1792,7 +1788,7 @@ L(write_32bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_24bytes): mov -24(%rsi), %r10 mov -16(%rsi), %r11 @@ -1802,7 +1798,7 @@ L(write_24bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_16bytes): mov -16(%rsi), %r11 mov -8(%rsi), %rdx @@ -1810,14 +1806,14 @@ L(write_16bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_8bytes): mov -8(%rsi), %rdx mov %rdx, -8(%rdi) L(write_0bytes): ret - ALIGN (4) + .p2align 4 L(write_73bytes): movdqu -73(%rsi), %xmm0 movdqu -57(%rsi), %xmm1 @@ -1837,7 +1833,7 @@ L(write_73bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_65bytes): movdqu -65(%rsi), %xmm0 movdqu -49(%rsi), %xmm1 @@ -1855,7 +1851,7 @@ L(write_65bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_57bytes): movdqu -57(%rsi), %xmm0 mov -41(%rsi), %r8 @@ -1873,7 +1869,7 @@ L(write_57bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_49bytes): movdqu -49(%rsi), %xmm0 mov -33(%rsi), %r9 @@ -1889,7 +1885,7 @@ L(write_49bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_41bytes): mov -41(%rsi), %r8 mov -33(%rsi), %r9 @@ -1905,7 +1901,7 @@ L(write_41bytes): mov %dl, -1(%rdi) ret - ALIGN (4) + .p2align 4 L(write_33bytes): mov -33(%rsi), %r9 mov -25(%rsi), %r10 @@ -1919,7 +1915,7 @@ L(write_33bytes): mov %dl, -1(%rdi) ret - ALIGN (4) + .p2align 4 L(write_25bytes): mov -25(%rsi), %r10 mov -17(%rsi), %r11 @@ -1931,7 +1927,7 @@ L(write_25bytes): mov %dl, -1(%rdi) ret - ALIGN (4) + .p2align 4 L(write_17bytes): mov -17(%rsi), %r11 mov -9(%rsi), %rcx @@ -1941,7 +1937,7 @@ L(write_17bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_9bytes): mov -9(%rsi), %rcx mov -4(%rsi), %edx @@ -1949,13 +1945,13 @@ L(write_9bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_1bytes): mov -1(%rsi), %dl mov %dl, -1(%rdi) ret - ALIGN (4) + .p2align 4 L(write_74bytes): movdqu -74(%rsi), %xmm0 movdqu -58(%rsi), %xmm1 @@ -1975,7 +1971,7 @@ L(write_74bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_66bytes): movdqu -66(%rsi), %xmm0 movdqu -50(%rsi), %xmm1 @@ -1995,7 +1991,7 @@ L(write_66bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_58bytes): movdqu -58(%rsi), %xmm1 mov -42(%rsi), %r8 @@ -2013,7 +2009,7 @@ L(write_58bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_50bytes): movdqu -50(%rsi), %xmm0 mov -34(%rsi), %r9 @@ -2029,7 +2025,7 @@ L(write_50bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_42bytes): mov -42(%rsi), %r8 mov -34(%rsi), %r9 @@ -2045,7 +2041,7 @@ L(write_42bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_34bytes): mov -34(%rsi), %r9 mov -26(%rsi), %r10 @@ -2059,7 +2055,7 @@ L(write_34bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_26bytes): mov -26(%rsi), %r10 mov -18(%rsi), %r11 @@ -2071,7 +2067,7 @@ L(write_26bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_18bytes): mov -18(%rsi), %r11 mov -10(%rsi), %rcx @@ -2081,7 +2077,7 @@ L(write_18bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_10bytes): mov -10(%rsi), %rcx mov -4(%rsi), %edx @@ -2089,13 +2085,13 @@ L(write_10bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_2bytes): mov -2(%rsi), %dx mov %dx, -2(%rdi) ret - ALIGN (4) + .p2align 4 L(write_75bytes): movdqu -75(%rsi), %xmm0 movdqu -59(%rsi), %xmm1 @@ -2115,7 +2111,7 @@ L(write_75bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_67bytes): movdqu -67(%rsi), %xmm0 movdqu -59(%rsi), %xmm1 @@ -2135,7 +2131,7 @@ L(write_67bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_59bytes): movdqu -59(%rsi), %xmm0 mov -43(%rsi), %r8 @@ -2153,7 +2149,7 @@ L(write_59bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_51bytes): movdqu -51(%rsi), %xmm0 mov -35(%rsi), %r9 @@ -2169,7 +2165,7 @@ L(write_51bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_43bytes): mov -43(%rsi), %r8 mov -35(%rsi), %r9 @@ -2185,7 +2181,7 @@ L(write_43bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_35bytes): mov -35(%rsi), %r9 mov -27(%rsi), %r10 @@ -2199,7 +2195,7 @@ L(write_35bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_27bytes): mov -27(%rsi), %r10 mov -19(%rsi), %r11 @@ -2211,7 +2207,7 @@ L(write_27bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_19bytes): mov -19(%rsi), %r11 mov -11(%rsi), %rcx @@ -2221,7 +2217,7 @@ L(write_19bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_11bytes): mov -11(%rsi), %rcx mov -4(%rsi), %edx @@ -2229,7 +2225,7 @@ L(write_11bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_3bytes): mov -3(%rsi), %dx mov -2(%rsi), %cx @@ -2237,7 +2233,7 @@ L(write_3bytes): mov %cx, -2(%rdi) ret - ALIGN (4) + .p2align 4 L(write_76bytes): movdqu -76(%rsi), %xmm0 movdqu -60(%rsi), %xmm1 @@ -2257,7 +2253,7 @@ L(write_76bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_68bytes): movdqu -68(%rsi), %xmm0 movdqu -52(%rsi), %xmm1 @@ -2275,7 +2271,7 @@ L(write_68bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_60bytes): movdqu -60(%rsi), %xmm0 mov -44(%rsi), %r8 @@ -2293,7 +2289,7 @@ L(write_60bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_52bytes): movdqu -52(%rsi), %xmm0 mov -36(%rsi), %r9 @@ -2309,7 +2305,7 @@ L(write_52bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_44bytes): mov -44(%rsi), %r8 mov -36(%rsi), %r9 @@ -2325,7 +2321,7 @@ L(write_44bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_36bytes): mov -36(%rsi), %r9 mov -28(%rsi), %r10 @@ -2339,7 +2335,7 @@ L(write_36bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_28bytes): mov -28(%rsi), %r10 mov -20(%rsi), %r11 @@ -2351,7 +2347,7 @@ L(write_28bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_20bytes): mov -20(%rsi), %r11 mov -12(%rsi), %rcx @@ -2361,7 +2357,7 @@ L(write_20bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_12bytes): mov -12(%rsi), %rcx mov -4(%rsi), %edx @@ -2369,13 +2365,13 @@ L(write_12bytes): mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_4bytes): mov -4(%rsi), %edx mov %edx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_77bytes): movdqu -77(%rsi), %xmm0 movdqu -61(%rsi), %xmm1 @@ -2395,7 +2391,7 @@ L(write_77bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_69bytes): movdqu -69(%rsi), %xmm0 movdqu -53(%rsi), %xmm1 @@ -2413,7 +2409,7 @@ L(write_69bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_61bytes): movdqu -61(%rsi), %xmm0 mov -45(%rsi), %r8 @@ -2431,7 +2427,7 @@ L(write_61bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_53bytes): movdqu -53(%rsi), %xmm0 mov -45(%rsi), %r8 @@ -2448,7 +2444,7 @@ L(write_53bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_45bytes): mov -45(%rsi), %r8 mov -37(%rsi), %r9 @@ -2464,7 +2460,7 @@ L(write_45bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_37bytes): mov -37(%rsi), %r9 mov -29(%rsi), %r10 @@ -2478,7 +2474,7 @@ L(write_37bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_29bytes): mov -29(%rsi), %r10 mov -21(%rsi), %r11 @@ -2490,7 +2486,7 @@ L(write_29bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_21bytes): mov -21(%rsi), %r11 mov -13(%rsi), %rcx @@ -2500,7 +2496,7 @@ L(write_21bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_13bytes): mov -13(%rsi), %rcx mov -8(%rsi), %rdx @@ -2508,7 +2504,7 @@ L(write_13bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_5bytes): mov -5(%rsi), %edx mov -4(%rsi), %ecx @@ -2516,7 +2512,7 @@ L(write_5bytes): mov %ecx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_78bytes): movdqu -78(%rsi), %xmm0 movdqu -62(%rsi), %xmm1 @@ -2536,7 +2532,7 @@ L(write_78bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_70bytes): movdqu -70(%rsi), %xmm0 movdqu -54(%rsi), %xmm1 @@ -2554,7 +2550,7 @@ L(write_70bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_62bytes): movdqu -62(%rsi), %xmm0 mov -46(%rsi), %r8 @@ -2572,7 +2568,7 @@ L(write_62bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_54bytes): movdqu -54(%rsi), %xmm0 mov -38(%rsi), %r9 @@ -2588,7 +2584,7 @@ L(write_54bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_46bytes): mov -46(%rsi), %r8 mov -38(%rsi), %r9 @@ -2604,7 +2600,7 @@ L(write_46bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_38bytes): mov -38(%rsi), %r9 mov -30(%rsi), %r10 @@ -2618,7 +2614,7 @@ L(write_38bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_30bytes): mov -30(%rsi), %r10 mov -22(%rsi), %r11 @@ -2630,7 +2626,7 @@ L(write_30bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_22bytes): mov -22(%rsi), %r11 mov -14(%rsi), %rcx @@ -2640,7 +2636,7 @@ L(write_22bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_14bytes): mov -14(%rsi), %rcx mov -8(%rsi), %rdx @@ -2648,7 +2644,7 @@ L(write_14bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_6bytes): mov -6(%rsi), %edx mov -4(%rsi), %ecx @@ -2656,7 +2652,7 @@ L(write_6bytes): mov %ecx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(write_79bytes): movdqu -79(%rsi), %xmm0 movdqu -63(%rsi), %xmm1 @@ -2676,7 +2672,7 @@ L(write_79bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_71bytes): movdqu -71(%rsi), %xmm0 movdqu -55(%rsi), %xmm1 @@ -2694,7 +2690,7 @@ L(write_71bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_63bytes): movdqu -63(%rsi), %xmm0 mov -47(%rsi), %r8 @@ -2712,7 +2708,7 @@ L(write_63bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_55bytes): movdqu -55(%rsi), %xmm0 mov -39(%rsi), %r9 @@ -2728,7 +2724,7 @@ L(write_55bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_47bytes): mov -47(%rsi), %r8 mov -39(%rsi), %r9 @@ -2744,7 +2740,7 @@ L(write_47bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_39bytes): mov -39(%rsi), %r9 mov -31(%rsi), %r10 @@ -2758,7 +2754,7 @@ L(write_39bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_31bytes): mov -31(%rsi), %r10 mov -23(%rsi), %r11 @@ -2770,7 +2766,7 @@ L(write_31bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_23bytes): mov -23(%rsi), %r11 mov -15(%rsi), %rcx @@ -2780,7 +2776,7 @@ L(write_23bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_15bytes): mov -15(%rsi), %rcx mov -8(%rsi), %rdx @@ -2788,7 +2784,7 @@ L(write_15bytes): mov %rdx, -8(%rdi) ret - ALIGN (4) + .p2align 4 L(write_7bytes): mov -7(%rsi), %edx mov -4(%rsi), %ecx @@ -2796,7 +2792,7 @@ L(write_7bytes): mov %ecx, -4(%rdi) ret - ALIGN (4) + .p2align 4 L(large_page_fwd): movdqu (%rsi), %xmm1 lea 16(%rsi), %rsi @@ -2859,7 +2855,7 @@ L(large_page_less_64bytes): BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) #ifdef USE_AS_MEMMOVE - ALIGN (4) + .p2align 4 L(ll_cache_copy_fwd_start): prefetcht0 0x1c0(%rsi) prefetcht0 0x200(%rsi) @@ -2906,7 +2902,7 @@ L(large_page_ll_less_fwd_64bytes): BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) #endif - ALIGN (4) + .p2align 4 L(large_page_bwd): movdqu -0x10(%rsi), %xmm1 lea -16(%rsi), %rsi @@ -2966,7 +2962,7 @@ L(large_page_less_bwd_64bytes): BRANCH_TO_JMPTBL_ENTRY (L(table_less_80bytes), %rdx, 4) #ifdef USE_AS_MEMMOVE - ALIGN (4) + .p2align 4 L(ll_cache_copy_bwd_start): prefetcht0 -0x1c0(%rsi) prefetcht0 -0x200(%rsi) @@ -3014,7 +3010,7 @@ L(large_page_ll_less_bwd_64bytes): END (MEMCPY) .section .rodata.ssse3,"a",@progbits - ALIGN (3) + .p2align 3 L(table_less_80bytes): .int JMPTBL (L(write_0bytes), L(table_less_80bytes)) .int JMPTBL (L(write_1bytes), L(table_less_80bytes)) @@ -3097,7 +3093,7 @@ L(table_less_80bytes): .int JMPTBL (L(write_78bytes), L(table_less_80bytes)) .int JMPTBL (L(write_79bytes), L(table_less_80bytes)) - ALIGN (3) + .p2align 3 L(shl_table): .int JMPTBL (L(shl_0), L(shl_table)) .int JMPTBL (L(shl_1), L(shl_table)) @@ -3116,7 +3112,7 @@ L(shl_table): .int JMPTBL (L(shl_14), L(shl_table)) .int JMPTBL (L(shl_15), L(shl_table)) - ALIGN (3) + .p2align 3 L(shl_table_bwd): .int JMPTBL (L(shl_0_bwd), L(shl_table_bwd)) .int JMPTBL (L(shl_1_bwd), L(shl_table_bwd)) |