1This file is the short summary of the API changes: 2 305.03.2014 - Backward compatible 4 The sljit_set_target now supports those jumps, which 5 does not created with SLJIT_REWRITABLE_JUMP flag. 6 Reason: sljit_emit_ijump does not support conditional 7 jumps. 8 903.03.2014 - Non-backward compatible 10 SLJIT_MOV_UI cannot be combined with SLJIT_INT_OP. 11 Reason: SLJIT_INT_OP flag is not recommended to use 12 directly, and SLJIT_IMOV has no sign bit. 13 1429.01.2014 - Backward compatible 15 Bits assigned to SLJIT_MEM and SLJIT_IMM flags are changed. 16 Reason: the most common cases are fits into one byte now, 17 and more registers can be supported in the future. 18 1908.11.2012 - Non-backward compatible 20 SLJIT_TEMPORARY_REGx registers are renamed to SLJIT_SCRATCH_REGx. 21 2207.11.2012 - Non-backward compatible 23 sljit_emit_cond_value is renamed to sljit_emit_op_flags. An 24 extra source argument is added which will be used in the future. 25 2605.11.2012 - Backward compatible 27 sljit_emit_cond_value now supports SLJIT_AND and SLJIT_INT_OP 28 flags, which makes this function complete. 29 3001.11.2012 - Non-backward compatible 31 SLJIT_F* opcodes are renamed to SLJIT_*D to show that 32 they are double precision operators. Furthermore 33 SLJIT_*S single precision opcodes are added. 34 3501.11.2012 - Non-backward compatible 36 Register arguments of operations with SLJIT_INT_OP flag 37 must be computed by another operation with SLJIT_INT_OP flag. 38 The same way as SLJIT_SINGLE_OP flag works with floating point 39 numbers. See the description of SLJIT_INT_OP. 40 4101.11.2012 - Backward compatible 42 All operations whose support the SLJIT_INT_OP flag, have an 43 alternate name now, which includes the SLJIT_INT_OP. These 44 names starting with I. 45 4631.10.2012 - Non-backward compatible 47 Renaming sljit_w to sljit_sw, sljit_i to sljit_si, sljit_h 48 to sljit_sh, and sljit_b to sljit_sb. Reason: their sign 49 bit is part of the type now. 50 5120.10.2012 - Non-backward compatible 52 Renaming SLJIT_C_FLOAT_NAN to SLJIT_C_FLOAT_UNORDERED. 53 Reason: all architectures call these unordered comparions. 54