Lines Matching full:loops
324 \&\fB\-falign\-labels=\fR\fIn\fR \fB\-falign\-loops=\fR\fIn\fR
356 \&\-fsched2\-use\-traces \-fsee \-freschedule\-modulo\-scheduled\-loops
360 \&\-funroll\-all\-loops \-funroll\-loops \-fpeel\-loops
361 \&\-fsplit\-ivs\-in\-unroller \-funswitch\-loops
574 \&\-malign\-loops \-mno\-align\-loops
1255 may become inline copy loops. The resulting code is often both smaller
3291 execute in different places because they were moved out of loops.
3991 Dump each function after applying vectorization of loops. The file name is
4011 and the total number of loops that got vectorized.
4012 If \fIn\fR=2 the vectorizer also reports non-vectorized loops that passed
4014 inner\-most, single\-bb, single\-entry/exit loops. This is the same verbosity
4017 reported loop, or same amount of information reported for more loops:
4021 If \fIn\fR=5, the vectorizer reports also non-vectorized inner-most loops
4024 If \fIn\fR=6, the vectorizer reports also non-vectorized nested loops.
4245 \&\-falign\-loops \-falign\-labels
4259 \&\fB\-funswitch\-loops\fR and \fB\-fgcse\-after\-reload\fR options.
4270 \&\fB\-falign\-functions \-falign\-jumps \-falign\-loops
4455 pass. This pass looks at innermost loops and reorders their
4571 stores out of loops. When used in conjunction with \fB\-fgcse\-lm\fR,
4572 loops containing a load/store sequence can be changed to a load before
4591 overflow, and that the loops with nontrivial exit condition are not
4729 .IP "\fB\-freschedule\-modulo\-scheduled\-loops\fR" 4
4730 .IX Item "-freschedule-modulo-scheduled-loops"
4818 nontrivial sequences of insns). With \fB\-funswitch\-loops\fR it also moves
4880 .IP "\fB\-funroll\-loops\fR" 4
4881 .IX Item "-funroll-loops"
4882 Unroll loops whose number of iterations can be determined at compile
4883 time or upon entry to the loop. \fB\-funroll\-loops\fR implies
4886 .IP "\fB\-funroll\-all\-loops\fR" 4
4887 .IX Item "-funroll-all-loops"
4888 Unroll all loops, even if their number of iterations is uncertain when
4890 \&\fB\-funroll\-all\-loops\fR implies the same options as
4891 \&\fB\-funroll\-loops\fR,
4911 memory to improve the performance of loops that access large arrays.
4914 dependent on the structure of loops within the source code.
5087 If \fB\-falign\-loops\fR or \fB\-falign\-jumps\fR are applicable and
5094 .IP "\fB\-falign\-loops\fR" 4
5095 .IX Item "-falign-loops"
5097 .IP "\fB\-falign\-loops=\fR\fIn\fR" 4
5098 .IX Item "-falign-loops=n"
5100 Align loops to a power-of-two boundary, skipping up to \fIn\fR bytes
5105 \&\fB\-fno\-align\-loops\fR and \fB\-falign\-loops=1\fR are
5106 equivalent and mean that loops will not be aligned.
5123 equivalent and mean that loops will not be aligned.
5180 Enabled by default with \fB\-funroll\-loops\fR.
5213 \&\f(CW\*(C`\-funroll\-loops\*(C'\fR, \f(CW\*(C`\-fpeel\-loops\*(C'\fR, \f(CW\*(C`\-ftracer\*(C'\fR
5404 Enabled by default with \fB\-funroll\-loops\fR.
5412 .IP "\fB\-funroll\-loops\fR" 4
5413 .IX Item "-funroll-loops"
5414 Unroll loops whose number of iterations can be determined at compile time or
5415 upon entry to the loop. \fB\-funroll\-loops\fR implies
5417 It also turns on complete loop peeling (i.e. complete removal of loops with
5422 .IP "\fB\-funroll\-all\-loops\fR" 4
5423 .IX Item "-funroll-all-loops"
5424 Unroll all loops, even if their number of iterations is uncertain when
5426 \&\fB\-funroll\-all\-loops\fR implies the same options as
5427 \&\fB\-funroll\-loops\fR.
5428 .IP "\fB\-fpeel\-loops\fR" 4
5429 .IX Item "-fpeel-loops"
5430 Peels the loops for that there is enough information that they do not
5432 (i.e. complete removal of loops with small constant number of iterations).
5439 .IP "\fB\-funswitch\-loops\fR" 4
5440 .IX Item "-funswitch-loops"
5471 thus hoisting loads out of loops and doing inter-block scheduling needs
5754 The induction variable optimizations give up on loops that contain more
9521 .IP "\fB\-malign\-loops\fR" 4
9522 .IX Item "-malign-loops"
9523 Align all loops to a 32\-byte boundary.
9524 .IP "\fB\-mno\-align\-loops\fR" 4
9525 .IX Item "-mno-align-loops"
9526 Do not enforce a 32\-byte alignment for loops. This is the default.
12148 iteration count on the C3x is 2^{23 + 1} (but who iterates loops more than
12197 for innermost loops that do not call functions or jump across the loop
12198 boundaries. There is no advantage having nested \s-1RPTB\s0 loops due to the
