|
|
1 | TODO: naming of cutnions, index_to_coord, xy_to_s, xy_to_linear/seq, 32/64 |
|
|
2 | TODO: returning packed coordinates, really |
|
|
3 | TODO: support order=0? |
|
|
4 | 0x0001000c |
|
|
5 | - added ecb_hilbert2d function family. |
|
|
6 | - use __builtin_popcountl on 64 bit clang and gcc. |
|
|
7 | - use 64 bit builtins, not 128 bit builtins, for ctz/clz on 64 bit cpus. |
|
|
8 | - add ecb_clz* functions, |
|
|
9 | - renamed ecb_gray* functions to have trailing bit width. |
|
|
10 | - use ecb_function_ consistently for newly added functions. |
|
|
11 | - say goodbye to gnu-style function definitions - they were not |
|
|
12 | used consistently anyway, and imho make the code harder to read. |
|
|
13 | |
|
|
14 | 0x0001000b |
|
|
15 | - added ecb_gray*_{de,en}code functions. |
|
|
16 | |
|
|
17 | 0x0001000a |
|
|
18 | - added ecb_ptrmix. |
|
|
19 | |
|
|
20 | 0x00010009 |
|
|
21 | - added ecb_i2a family of functions. |
|
|
22 | - added ECB_64BIT_NATIVE. |
|
|
23 | |
|
|
24 | TODO: |
|
|
25 | 08:30:06 <b_jonas> I think it could be worth to add a macro that works like alignof or _Alignof on sane |
|
|
26 | compilers, and like __alignof on MS compilers that support it, see |
|
|
27 | http://msdn.microsoft.com/en-us/library/45t0s5f4.aspx |
|
|
28 | 08:30:24 <b_jonas> even if you can't support it on all the old compilers |
|
|
29 | 08:31:17 <b_jonas> I'd also like a macro for alignas, but sadly, that seems impossible in general, because |
|
|
30 | the MS compiler only has some half-attempt to do something similar but with different and |
|
|
31 | more broken semantics, see http://msdn.microsoft.com/en-us/library/83ythb65.aspx |
|
|
32 | 08:31:35 <b_jonas> but I wonder if some special case could still be worth to support |
|
|
33 | 08:32:23 <b_jonas> probably not, because it would just account to making a union with a highly aligned type, |
|
|
34 | which is something I can do on any compiler portably |
|
|
35 | |
|
|
36 | TODO: #define ECB_IS_INTEGRAL(x) !((1 ? 1 : (x)) / 2) |
|
|
37 | #define ECB_IS_INTEGRAL(x) (sizeof ((x) + 1.0f) != sizeof((x) + 1ULL)) |
|
|
38 | |
|
|
39 | TODO: ecb_minpot, either using bit tricks or ecb_ldXX |
|
|
40 | |
|
|
41 | TODO: __builtin_popcountll exists... |
|
|
42 | |
|
|
43 | TODO: __builtin_powi |
|
|
44 | |
|
|
45 | TODO: https://gustedt.wordpress.com/2010/06/08/detect-empty-macro-arguments/ |
|
|
46 | |
| 1 | implement is_constant for c11: https://gustedt.wordpress.com/2013/08/22/testing-compile-time-constness-and-null-pointers-with-c11s-_generic/ |
47 | implement is_constant for c11: https://gustedt.wordpress.com/2013/08/22/testing-compile-time-constness-and-null-pointers-with-c11s-_generic/ |
| 2 | |
48 | |
| 3 | #ifdef _MSC_VER |
49 | #ifdef _MSC_VER |
| 4 | |
50 | |
| 5 | #include <stdlib.h> |
51 | #include <stdlib.h> |
| … | |
… | |
| 44 | |
90 | |
| 45 | #include <byteswap.h> |
91 | #include <byteswap.h> |
| 46 | |
92 | |
| 47 | #endif |
93 | #endif |
| 48 | |
94 | |
|
|
95 | TODO: generic poprcount etc., also fast_t for them? |
|
|
96 | |
|
|
97 | - allow any rotate count in rcb_rot*, at the expense |
|
|
98 | of bad optimisation results on some platforms or with |
|
|
99 | some word sizes. |
|
|
100 | - try to use inttypes.h on mingw. |
|
|
101 | |
|
|
102 | 0x00010008 |
|
|
103 | - aligned/unaligned load/store, bswap, host order |
|
|
104 | conversion. |
|
|
105 | - generic C++ ecb_rot[lr], ecb_popcount, ecb_ctz, |
|
|
106 | ecb_is_pot32 and ecb_bitrev functions. |
|
|
107 | |
|
|
108 | 0x00010007 |
|
|
109 | - new ECB_OPTIMIZE_SIZE symbol. |
|
|
110 | |
|
|
111 | 0x00010006 |
|
|
112 | - new ECB_MEMORY_FENCE_RELAXED memory fence. |
|
|
113 | - use acquire/eelease memory barriers on sun workshop pro, not read/write. |
|
|
114 | - rely on c++ compiler barriers to do the right thing in gcc/clang. |
|
|
115 | - change release memory fence to memory barrier on ia32/ia64. |
|
|
116 | - apply ctz/ld patch for msc by Zsbán Ambrus. |
|
|
117 | - ECB_PTRSIZE erroneously was 8 on most 32bit systems ( |
|
|
118 | found by Zsbán Ambrus). |
|
|
119 | - improved compiletime detection of endianness, also, allow |
|
|
120 | runtime detection to indicate other-than-big/little endianness. |
|
|
121 | - no memory barrier neded on arm < 6. |
|
|
122 | |
|
|
123 | 0x00010005 |
|
|
124 | - improve ecb_binary16_to_float. |
|
|
125 | - add ecb_float_to_binary16. |
|
|
126 | - add ecb_binary16_to_binary32 and ecb_binary32_to_binary16 pair. |
|
|
127 | |
| 49 | 0x00010001 |
128 | 0x00010001 |
| 50 | - add ecb_is_pot32/64. |
129 | - add ecb_is_pot32/64. |
| 51 | - add intptr_t/uintptr_t. |
130 | - add intptr_t/uintptr_t. |
| 52 | - add ECB_PTRSIZE. |
131 | - add ECB_PTRSIZE. |
| 53 | - more macros for C/C++ version checks. |
132 | - more macros for C/C++ version checks. |
| … | |
… | |
| 55 | - support gcc-4.7 atomics for memory fences. |
134 | - support gcc-4.7 atomics for memory fences. |
| 56 | - support m68k, m88k and sh (patch by Miod Vallat). |
135 | - support m68k, m88k and sh (patch by Miod Vallat). |
| 57 | - add ecb_binary16_to_float. |
136 | - add ecb_binary16_to_float. |
| 58 | |
137 | |
| 59 | TODO: ecb_restrict_array etc. http://ue.tst.eu/5093eafd713ec5fda776d8065070aa4c.txt |
138 | TODO: ecb_restrict_array etc. http://ue.tst.eu/5093eafd713ec5fda776d8065070aa4c.txt |
| 60 | TODO: ffs/clz |
139 | TODO: ffs |
| 61 | 64 bit variants of everything |
140 | 64 bit variants of everything |
| 62 | TODO: examples from X for clz/ctz |
141 | TODO: examples from X for clz/ctz |
| 63 | TODO: arithmetic right shift |
142 | TODO: arithmetic right shift |
| 64 | TODO: template/generic functions for x32/x64 and so on |
143 | TODO: template/generic functions for x32/x64 and so on |
| 65 | TODO: #define ecb_integer_multiples_of(n,d) ((char (*)[d])(n) - (char (*)[d])0) |
144 | TODO: #define ecb_integer_multiples_of(n,d) ((char (*)[d])(n) - (char (*)[d])0) |
| … | |
… | |
| 70 | a = (v << x) & -(((unsigned int)x) < 32); |
149 | a = (v << x) & -(((unsigned int)x) < 32); |
| 71 | x = -x; |
150 | x = -x; |
| 72 | b = (v >> x) & -(((unsigned int)x) < 32); |
151 | b = (v >> x) & -(((unsigned int)x) < 32); |
| 73 | return a|b; |
152 | return a|b; |
| 74 | } |
153 | } |
|
|
154 | |
|
|
155 | TODO: export(=dllexport) & hidden |
|
|
156 | TODO: flatten |
|
|
157 | TODO: warning(msg) |
|
|
158 | TODO: error(msg) |
|
|
159 | TODO: leaf (uh), noclone (hmmm) |
|
|
160 | TODO: nonnull, returns_nonnull |
|
|
161 | TODO: nothrow |
|
|
162 | TODO: used |
|
|
163 | TODO: trap |
|
|
164 | TODO: http://llvm.org/docs/doxygen/html/Compiler_8h_source.html |
|
|
165 | |
|
|
166 | TODO: read/write unaligned macros |
|
|
167 | TODO: htonl and friends |
|
|
168 | TODO: macro to convert from unsigned to signed "the natural way" |
|
|
169 | TODO: ecb_static_assert, with message (just like boost), or somesuch, using array-declaration |
|
|
170 | TODO: alignof |
|
|
171 | |
|
|
172 | |
