Lines Matching full:loops
343 \&\-falign\-labels[=\fR\fIn\fR\fB] \-falign\-loops[=\fR\fIn\fR\fB] \-fassociative\-math
370 \&\-fpeel\-loops \-fpredictive\-commoning \-fprefetch\-loop\-arrays
376 \&\-frerun\-cse\-after\-loop \-freschedule\-modulo\-scheduled\-loops
383 \&\-fsel\-sched\-pipelining \-fsel\-sched\-pipelining\-outer\-loops
393 \&\-ftree\-parallelize\-loops=\fR\fIn\fR \fB\-ftree\-pre \-ftree\-reassoc
396 \&\-funit\-at\-a\-time \-funroll\-all\-loops \-funroll\-loops
397 \&\-funsafe\-loop\-optimizations \-funsafe\-math\-optimizations \-funswitch\-loops
622 \&\-malign\-loops \-mno\-align\-loops
1533 may become inline copy loops. The resulting code is often both smaller
3797 execute in different places because they were moved out of loops.
4623 Dump each function after applying vectorization of loops. The file name is
4643 and the total number of loops that got vectorized.
4644 If \fIn\fR=2 the vectorizer also reports non-vectorized loops that passed
4646 inner\-most, single\-bb, single\-entry/exit loops. This is the same verbosity
4649 reported loop, or same amount of information reported for more loops:
4653 If \fIn\fR=5, the vectorizer reports also non-vectorized inner-most loops
4656 If \fIn\fR=6, the vectorizer reports also non-vectorized nested loops.
4879 \&\-falign\-loops \-falign\-labels
4905 \&\fB\-funswitch\-loops\fR, \fB\-fpredictive\-commoning\fR,
4918 \&\fB\-falign\-functions \-falign\-jumps \-falign\-loops
5099 pass. This pass looks at innermost loops and reorders their
5225 stores out of loops. When used in conjunction with \fB\-fgcse\-lm\fR,
5226 loops containing a load/store sequence can be changed to a load before
5245 overflow, and that the loops with nontrivial exit condition are not
5330 \&\f(CW\*(C`one\*(C'\fR. The first value means using all loops as register
5332 all loops except for loops with small register pressure as the
5455 .IP "\fB\-freschedule\-modulo\-scheduled\-loops\fR" 4
5456 .IX Item "-freschedule-modulo-scheduled-loops"
5470 Enable software pipelining of innermost loops during selective scheduling.
5473 .IP "\fB\-fsel\-sched\-pipelining\-outer\-loops\fR" 4
5474 .IX Item "-fsel-sched-pipelining-outer-loops"
5475 When pipelining loops during selective scheduling, also pipeline outer loops.
5625 Perform loop interchange transformations on loops. Interchanging two
5626 nested loops switches the inner and outer loops. For example, given a
5657 Perform loop strip mining transformations on loops. Strip mining
5658 splits a loop into two nested loops. The outer loop has strides
5684 Perform loop blocking transformations on loops. Blocking strip mines
5686 element loops
5748 nontrivial sequences of insns). With \fB\-funswitch\-loops\fR it also moves
5762 .IP "\fB\-ftree\-parallelize\-loops=n\fR" 4
5763 .IX Item "-ftree-parallelize-loops=n"
5764 Parallelize loops, i.e., split their iteration space to run in n threads.
5765 This is only possible for loops whose iterations are independent
5767 profitable on multiprocessor machines, for loops that are CPU\-intensive,
5818 .IP "\fB\-funroll\-loops\fR" 4
5819 .IX Item "-funroll-loops"
5820 Unroll loops whose number of iterations can be determined at compile
5821 time or upon entry to the loop. \fB\-funroll\-loops\fR implies
5824 .IP "\fB\-funroll\-all\-loops\fR" 4
5825 .IX Item "-funroll-all-loops"
5826 Unroll all loops, even if their number of iterations is uncertain when
5828 \&\fB\-funroll\-all\-loops\fR implies the same options as
5829 \&\fB\-funroll\-loops\fR,
5850 iterations of loops.
5856 memory to improve the performance of loops that access large arrays.
5859 dependent on the structure of loops within the source code.
6047 If \fB\-falign\-loops\fR or \fB\-falign\-jumps\fR are applicable and
6054 .IP "\fB\-falign\-loops\fR" 4
6055 .IX Item "-falign-loops"
6057 .IP "\fB\-falign\-loops=\fR\fIn\fR" 4
6058 .IX Item "-falign-loops=n"
6060 Align loops to a power-of-two boundary, skipping up to \fIn\fR bytes
6065 \&\fB\-fno\-align\-loops\fR and \fB\-falign\-loops=1\fR are
6066 loops will not be aligned.
6083 equivalent and mean that loops will not be aligned.
6116 Enabled by default with \fB\-funroll\-loops\fR.
6177 \&\f(CW\*(C`\-funroll\-loops\*(C'\fR, \f(CW\*(C`\-fpeel\-loops\*(C'\fR
6446 Enabled by default with \fB\-funroll\-loops\fR.
6454 .IP "\fB\-funroll\-loops\fR" 4
6455 .IX Item "-funroll-loops"
6456 Unroll loops whose number of iterations can be determined at compile time or
6457 upon entry to the loop. \fB\-funroll\-loops\fR implies
6459 It also turns on complete loop peeling (i.e. complete removal of loops with
6464 .IP "\fB\-funroll\-all\-loops\fR" 4
6465 .IX Item "-funroll-all-loops"
6466 Unroll all loops, even if their number of iterations is uncertain when
6468 \&\fB\-funroll\-all\-loops\fR implies the same options as
6469 \&\fB\-funroll\-loops\fR.
6470 .IP "\fB\-fpeel\-loops\fR" 4
6471 .IX Item "-fpeel-loops"
6472 Peels the loops for that there is enough information that they do not
6474 (i.e. complete removal of loops with small constant number of iterations).
6481 .IP "\fB\-funswitch\-loops\fR" 4
6482 .IX Item "-funswitch-loops"
6513 thus hoisting loads out of loops and doing inter-block scheduling needs
6814 value may set to be larger for hot loops.
6824 may set to be larger for hot loops.
6835 set to be larger for hot loops.
6853 The induction variable optimizations give up on loops that contain more
7193 .IP "\fBira-max-loops-num\fR" 4
7194 .IX Item "ira-max-loops-num"
7196 contains loops more than number given by the parameter, only at most
7197 given number of the most frequently executed loops will form regions
7212 in amount of needed compile time memory, with very large loops. Loops
11210 .IP "\fB\-malign\-loops\fR" 4
11211 .IX Item "-malign-loops"
11212 Align all loops to a 32\-byte boundary.
11213 .IP "\fB\-mno\-align\-loops\fR" 4
11214 .IX Item "-mno-align-loops"
11215 Do not enforce a 32\-byte alignment for loops. This is the default.
