2 date: Sun 01 Nov 2020 02:41:20 PM CET
4 - minor improvements to coalescing
5 - use build compiler to build extract_interface
6 - add some convenience functions
7 - ignore parameters in isl_union_* hash tables
10 date: Sun Jan 12 10:48:18 CET 2020
15 date: Fri Nov 1 18:39:30 CET 2019
17 - require C++11 to generate bindings
19 - scheduler fix involving fixed dimensions
20 - accept ranges in tuples during parsing
21 - add some convenience functions
24 date: Sat Mar 9 15:25:29 CET 2019
26 - preliminary C++ bindings
27 - use incremental scheduler by default
28 - introduce isl_size type
29 - rename isl_ast_op_type to isl_ast_expr_op_type
31 - use isl_bool to return extra boolean argument
34 date: Sat Jul 21 18:10:08 CEST 2018
36 - keep track of domain in 0D isl_multi_pw_aff and isl_multi_union_pw_aff
37 - add isl_aff_eval and isl_pw_aff_eval
38 - add fixed-size rectangular box hull
41 date: Sat Mar 3 10:44:49 CET 2018
43 - minor improvements to coalescing
44 - minor improvement to parametric integer programming
45 - try harder to avoid large coefficients in scheduler
46 - support kill accesses in dependence analysis
47 - drop deprecated isl_int
48 - drop deprecated band forests
49 - drop deprecated functions
52 date: Sun Dec 18 11:01:58 CET 2016
54 - improve elimination of redundant existentially quantified variables
56 - improve parametric integer programming
57 - preserve isolate option in isl_schedule_node_band_split
58 - print AST nodes in YAML format
59 - minor improvements to Python bindings
62 date: Fri May 6 12:02:48 CEST 2016
64 - fix bug in coalescing treatment
67 date: Tue May 3 14:26:43 CEST 2016
69 - optionally combine SCCs incrementally in scheduler
70 - optionally maximize coincidence in scheduler
71 - optionally avoid loop coalescing in scheduler
72 - fix handling of nested integer divisions
73 - optionally detect min/max expressions during AST generation
74 - minor AST generator improvements
75 - simplify stride constraints
76 - improve support for expansions in schedule trees
79 date: Thu Jan 14 18:08:06 CET 2016
81 - fix bug in simplification
84 date: Tue Jan 12 09:56:16 CET 2016
86 - add 32 bit integer optimization for IMath
87 - minor AST generator improvements
88 - add isl_union_flow_get_full_{may,must}_dependence
89 - minor improvements to Python bindings
90 - minor improvements to set and map printing
93 date: Thu Jun 11 12:45:33 CEST 2015
96 - add isl_union_access_info_compute_flow
97 - add mark nodes in AST
98 - add isl_union_pw_aff and isl_multi_union_pw_aff
100 - deprecate band forests
101 - deprecate separation_class AST generation option
102 - introduce isl_bool and isl_stat types
105 date: Thu Apr 9 12:57:23 CEST 2015
107 - fix bug in affine expression normalization
108 - fix handling of conditional validity constraints
111 date: Sat Oct 25 16:08:47 CEST 2014
113 - support IMath as an optional replacement for GMP
114 - minor AST generator improvements
117 date: Mon Apr 14 11:08:45 CEST 2014
120 - improved support for multi piecewise quasi-affine expressions
121 - allow the user to impose a bound on the number of low-level operations
122 - add isl_id_to_ast_expr and isl_id_to_pw_aff
123 - add isl_schedule_constraints
124 - hide internal structure of isl_vec
125 - remove support for piplib
128 date: Sun Jan 12 12:09:46 CET 2014
130 - MinGW-w64 build fix
131 - fix simplification bug
134 date: Wed Jul 24 12:54:46 CEST 2013
136 - handle malloc returning NULL on zero-size allocation
137 - fix regression in AST generator
140 date: Sun Jun 23 20:23:05 CEST 2013
142 - add isl_val abstraction
145 date: Tue Apr 9 18:45:10 CEST 2013
147 - make code generation output the same on Solaris
148 - fix some hard to trigger bugs
151 date: Mon Dec 10 11:55:30 CET 2012
153 - add LICENSE file to distribution
154 - make code generation output independent of endianness
157 date: Mon Dec 3 08:17:18 CET 2012
159 - change license from LGPL 2.1 to MIT
160 - add support for multi piecewise quasi-affine expressions
161 - add code generation
162 - various minor bug fixes
165 date: Sun Jun 3 18:00:16 CEST 2012
167 - support for interaction with dependence analysis
168 - add public API for vectors
169 - improved support for (piecewise) multi quasi-affine expressions
170 - various minor bug fixes
173 date: Sat Dec 17 18:19:26 CET 2011
175 - improved argument parsing
176 - hide internal structure of isl_options
177 - improved support for parameter sets
178 - configurable scheduling
181 date: Fri Oct 21 12:36:20 CEST 2011
184 - drop isl_div abstraction
185 - rename isl_dim to isl_space
187 explicitly differentiate between spaces of maps,
188 sets and parameter sets
189 - add support for identifiers
190 - add support for (piecewise) multi quasi-affine expressions
191 - preliminary Python bindings
194 date: Tue Jul 12 19:34:51 CEST 2011
196 - hide internal structures of isl_div and isl_constraint
197 - preliminary scheduling
198 - add support for local spaces and (piecewise) quasi-affine expressions
201 date: Fri Mar 18 15:59:16 CET 2011
204 - consistency changes in API
205 - hide internal structure of isl_ctx
208 date: Wed Jan 5 10:21:42 CET 2011
210 - fix simple symmetry detection in parametric integer programming
213 date: Thu Dec 23 17:03:14 CET 2010
215 - rename header files from isl_header.h to isl/header.h
216 - add higher level interface for dependence analysis
217 - improved argument parsing
218 - optionally triangulate domains during Bernstein expansion
219 - support extended PolyLib format
220 - hide internal structure of some data types
221 - improved coalescing
222 - add simple symmetry detection in parametric integer programming
225 date: Fri Sep 10 12:57:50 CEST 2010
227 - rename isl_pw_qpolynomial_fold_add
228 - add isl_map_apply_pw_qpolynomial_fold
229 - support named and nested spaces
230 - support union sets and maps
231 - add public API for matrices
234 date: Tue Jun 29 13:16:46 CEST 2010
236 - new printing functions
237 - support for "may" accesses in dependence analysis
238 - improved coalescing
239 - improved transitive closure
240 - fix several hard to trigger bugs
241 - improved argument parsing
242 - support parametric vertex enumeration for barvinok
243 - optionally use Bernstein expansion to compute bounds