Lines Matching full:optimization
91 * Tree SSA:: Analysis and optimization of the tree representation.
2199 the optimization will lead to code being optimized away (for example,
2216 without optimization. When all calls to a built-in function should
2522 optimization options.
2545 with multiple optimization options, since sometimes invalid code
2546 crashes the compiler with optimization.
2553 broken easily. These tests are run with multiple optimization
2555 optimization levels belong here. This also contains tests to
2568 different combinations of optimization options. All warnings are
2695 a branch can differ depending on the target or the optimization level.
2716 optimization. This testing requires that a test program be built and
2722 To check that the optimization actually generated better code, a test
2729 about a specific optimization:
3063 This chapter is dedicated to giving an overview of the optimization and
3072 * Pass manager:: Sequencing the optimization passes.
3216 The following briefly describes the tree optimization passes that are
3315 optimization. In the first pass we only warn for uses that are
3326 is run multiple times throughout the optimization process. It is
3333 run multiple times throughout the optimization process. It it
3341 multiple times throughout the optimization process. It is located
3367 * May-alias optimization
3385 compiler on a subsequent run so as to allow optimization based on
3399 rewritten into SSA form, which allows subsequent optimization
3438 * Loop optimization
3606 * Return value optimization
3612 optimization applied to GIMPLE. The pass is located in
3615 * Return slot optimization
3640 they are all geared to the optimization of data locality in array
3670 * Optimization of `stdarg' functions
3689 The following briefly describes the rtl generation and optimization
3690 passes that are run after tree optimization.
3719 the "jump optimization pass". The bulk of the code for this pass
3737 loops--that is left to the loop optimization pass. If MR PRE
3748 * Loop optimization
6259 10 Analysis and Optimization of GIMPLE Trees
6283 and programming constructs needed to implement optimization passes for
6335 GIMPLE is a simplified subset of GENERIC for use in optimization. The
6395 in the future if someone writes an optimization pass which would work
6709 pass which runs before most of the optimization passes eliminates these
7004 variables during the optimization process. For instance, we need to
7029 Data flow analysis and optimization is done on all tree nodes
7033 of optimization, we need to distinguish between references to local
7331 optimization can manipulate the stmt when all the uses have been
7349 list is up to date, as well as checking that an optimization didn't
7388 bring them up to date. This must be done before the optimization is
7476 Some optimization passes make changes to the function that invalidate
7918 updated only by the optimization passes that deal with loops, but some
7920 optimization passes.
7932 it in the middle of an optimization pass, you must be able to deal with
9662 references is so that jump optimization can distinguish them.
9741 optimization phases. These represent locations in the stack frame
10689 Peephole optimization, which takes place together with final
10694 in compilation, but it is ok then because no further optimization
10969 `JUMP_LABEL' which is defined once jump optimization has completed.
11022 The field `LABEL_NUSES' is only defined once the jump optimization
11117 note may be deleted by jump optimization.
11266 presence of this note allows jump optimization to be aware that OP
11267 is, in fact, being used, and flow optimization to build an
11277 the optimization to partition basic blocks into hot and cold
11333 common subexpression elimination and loop optimization) to tell
11341 `REG_EQUIV' note and are not useful to the early optimization
11364 Loop optimization uses this note to treat such a sequence as a
11931 During optimization, the frequency of the most frequent basic block can
11933 cross-jumping optimization), so scaling sometimes has to be performed
11941 Most optimization passes can use only the frequency information of a
12016 without any `tree' optimization enabled. When the `tree' optimizations
12081 not transparent and each optimization pass is required to do so
12113 While debugging the optimization pass, an `verify_flow_info' function
12251 Once the insn list is generated, various optimization passes convert,
15072 tested in the C condition. In the example, the optimization is applied
15376 patterns for `divM3' and `modM3'. This allows optimization in the
16470 They are necessary because jump optimization can turn direct-conditional
16562 if optimization is not requested. So you must use one of the solutions
16572 scheduler cannot perform this optimization if it is not permitted to
17330 consecutive insns. The optimization applies to a sequence of insns when
17338 generation in an unoptimized compilation, or at various optimization
17359 to perform the optimization (we do so if the expression is nonzero). If
17361 optimization is applied to every sequence of insns that matches the
17381 Applying the optimization means replacing the sequence of insns with
17416 The effect of this optimization is to change
19539 performed for various optimization levels. This macro, if
19540 defined, is executed once just after the optimization level is
19545 LEVEL is the optimization level specified; 2 if `-O2' is
19552 optimization level on all supported machines. Use this macro to
23074 True if it is ok to do sibling call optimization for the specified
23083 successful sibling call optimization may vary greatly between
23879 the results of peephole optimization: insns whose patterns are
27245 It is used in conjunction with an optimization that partitions
27254 memory. It is used in conjunction with an optimization that
27343 divisions that should be there for GCC to perform the optimization
27851 instruction stream. The compiler will run it at all optimization
28008 Branch target register optimization will by default exclude
30043 * Doug Evans for much of the global optimization framework, arc,
30068 multiply- and divide-by-constant optimization, improved long long
30127 * Falk Hueffner for working on C and optimization bug reports.
30774 * Simon Kitching for `String' cleanups and optimization suggestions.
33163 * Optimization infrastructure for GIMPLE: Tree SSA. (line 6)
33219 * peephole optimization, RTL representation: Side Effects. (line 235)
