Lines Matching full:loop
178 \&\fB\-fmove\-loop\-invariants\fR, \fB\-Wformat\fR and so on. Most of
265 \&\-Winvalid\-pch \-Wlarger\-than=\fR\fIlen\fR \fB\-Wunsafe\-loop\-optimizations
363 \&\-fmodulo\-sched\-allow\-regmoves \-fmove\-loop\-invariants \-fmudflap
370 \&\-fpeel\-loops \-fpredictive\-commoning \-fprefetch\-loop\-arrays
376 \&\-frerun\-cse\-after\-loop \-freschedule\-modulo\-scheduled\-loops
389 \&\-ftree\-dominator\-opts \-ftree\-dse \-ftree\-fre \-ftree\-loop\-im
390 \&\-ftree\-loop\-distribution
391 \&\-ftree\-loop\-ivcanon \-ftree\-loop\-linear \-ftree\-loop\-optimize
395 \&\-ftree\-ter \-ftree\-vect\-loop\-version \-ftree\-vectorize \-ftree\-vrp
397 \&\-funsafe\-loop\-optimizations \-funsafe\-math\-optimizations \-funswitch\-loops
1746 a \fIfor-init-statement\fR is limited to the \fBfor\fR loop itself,
3128 undefined signed overflow when estimating how many iterations a loop
3129 will require, in particular when determining whether a loop will be
3310 .IP "\fB\-Wunsafe\-loop\-optimizations\fR" 4
3311 .IX Item "-Wunsafe-loop-optimizations"
3312 Warn if the loop cannot be optimized because the compiler could not
3313 assume anything on the bounds of the loop indices. With
3314 \&\fB\-funsafe\-loop\-optimizations\fR warn if the compiler made
4252 loop optimization passes.
4546 Dump each function after copying loop headers. The file name is made by
4642 If \fIn\fR=1 the vectorizer reports each loop that got vectorized,
4649 reported loop, or same amount of information reported for more loops:
4891 \&\-frerun\-cse\-after\-loop
4920 \&\-fprefetch\-loop\-arrays \-ftree\-vect\-loop\-version\fR
4954 instructions and checks if the result can be simplified. If loop unrolling
4956 loop unrolling.
5196 .IP "\fB\-frerun\-cse\-after\-loop\fR" 4
5197 .IX Item "-frerun-cse-after-loop"
5198 Re-run common subexpression elimination after loop optimizations has been
5217 allows a loop containing a load/store sequence to be changed to a load outside
5218 the loop, and a copy/store within the loop.
5227 the loop and a store after the loop.
5242 .IP "\fB\-funsafe\-loop\-optimizations\fR" 4
5243 .IX Item "-funsafe-loop-optimizations"
5244 If given, the loop optimizer will assume that loop indices do not
5246 infinite. This enables a wider range of loop optimizations even if
5247 the loop optimizer itself cannot prove that these assumptions are valid.
5248 Using \fB\-Wunsafe\-loop\-optimizations\fR, the compiler will warn you
5249 if it finds this kind of loop.
5457 The modulo scheduling comes before the traditional scheduling, if a loop
5607 Perform loop header copying on trees. This is beneficial since it increases
5615 .IP "\fB\-ftree\-loop\-optimize\fR" 4
5616 .IX Item "-ftree-loop-optimize"
5617 Perform loop optimizations on trees. This flag is enabled by default
5619 .IP "\fB\-ftree\-loop\-linear\fR" 4
5620 .IX Item "-ftree-loop-linear"
5621 Perform linear loop transformations on tree. This flag can improve cache
5622 performance and allow further loop optimizations to take place.
5625 Perform loop interchange transformations on loops. Interchanging two
5627 loop like:
5637 loop interchange will transform the loop as if the user had written:
5649 contiguously by column, and the original loop iterates over rows,
5654 Graphite loop transformation infrastructure.
5657 Perform loop strip mining transformations on loops. Strip mining
5658 splits a loop into two nested loops. The outer loop has strides
5659 equal to the strip size and the inner loop has strides of the
5660 original loop within a strip. For example, given a loop like:
5668 loop strip mining will transform the loop as if the user had written:
5681 enable the Graphite loop transformation infrastructure.
5684 Perform loop blocking transformations on loops. Blocking strip mines
5685 each loop in the loop nest such that the memory accesses of the
5686 loop like:
5696 loop blocking will transform the loop as if the user had written:
5711 because the innermost loop will iterate over a smaller amount of data
5715 and \fB\-\-with\-cloog\fR to enable the Graphite loop transformation
5721 .IP "\fB\-ftree\-loop\-distribution\fR" 4
5722 .IX Item "-ftree-loop-distribution"
5723 Perform loop distribution. This flag can improve cache performance on
5724 big loop bodies and allow further loop optimizations, like
5725 parallelization or vectorization, to take place. For example, the loop
5744 .IP "\fB\-ftree\-loop\-im\fR" 4
5745 .IX Item "-ftree-loop-im"
5746 Perform loop invariant motion on trees. This pass moves only invariants that
5749 operands of conditions that are invariant out of the loop, so that we can use
5750 just trivial invariantness analysis in loop unswitching. The pass also includes
5752 .IP "\fB\-ftree\-loop\-ivcanon\fR" 4
5753 .IX Item "-ftree-loop-ivcanon"
5754 Create a canonical counter for number of iterations in the loop for that
5791 Perform loop vectorization on trees. This flag is enabled by default at
5793 .IP "\fB\-ftree\-vect\-loop\-version\fR" 4
5794 .IX Item "-ftree-vect-loop-version"
5795 Perform loop versioning when doing loop vectorization on trees. When a loop
5798 the loop are generated along with runtime checks for alignment or dependence
5821 time or upon entry to the loop. \fB\-funroll\-loops\fR implies
5822 \&\fB\-frerun\-cse\-after\-loop\fR. This option makes code larger,
5827 the loop is entered. This usually makes programs run more slowly.
5833 of the unrolled loop using the value in the first iteration. This breaks
5837 same effect. However in cases the loop body is more complicated than
5845 local variables when unrolling a loop which can result in superior code.
5853 .IP "\fB\-fprefetch\-loop\-arrays\fR" 4
5854 .IX Item "-fprefetch-loop-arrays"
6061 like \fB\-falign\-functions\fR. The hope is that the loop will be
6112 passes, such as \s-1CSE\s0, loop optimizer and trivial dead code remover. It can,
6457 upon entry to the loop. \fB\-funroll\-loops\fR implies
6458 \&\fB\-frerun\-cse\-after\-loop\fR, \fB\-fweb\fR and \fB\-frename\-registers\fR.
6459 It also turns on complete loop peeling (i.e. complete removal of loops with
6467 the loop is entered. This usually makes programs run more slowly.
6473 roll much (from profile feedback). It also turns on complete loop peeling
6477 .IP "\fB\-fmove\-loop\-invariants\fR" 4
6478 .IX Item "-fmove-loop-invariants"
6479 Enables the loop invariant motion pass in the \s-1RTL\s0 loop optimizer. Enabled
6483 Move branches with loop invariant conditions out of the loop, with duplicates
6484 of the loop on both branches (modified according to result of the condition).
6777 .IP "\fBmin-vect-loop-bound\fR" 4
6778 .IX Item "min-vect-loop-bound"
6779 The minimum number of iterations under which a loop will not get vectorized
6785 The maximum number of instructions that a loop should have if that loop
6786 is unrolled, and if the loop is unrolled, it determines how many times
6787 the loop code is unrolled.
6791 that a loop should have if that loop is unrolled, and if the loop is unrolled,
6792 it determines how many times the loop code is unrolled.
6795 The maximum number of unrollings of a single loop.
6798 The maximum number of instructions that a loop should have if that loop
6799 is peeled, and if the loop is peeled, it determines how many times
6800 the loop code is peeled.
6803 The maximum number of peelings of a single loop.
6810 The maximum number of insns of a completely peeled loop.
6813 feedback is available and the loop is hot. Because of the real profiles, this
6821 The maximum number of insns of a peeled loop that rolls only once.
6823 is available and the loop is hot. Because of the real profiles, this value
6831 The maximum number of iterations of a loop to be suitable for complete peeling.
6834 is available and the loop is hot. Because of the real profiles, this value may
6838 The maximum number of insns of an unswitched loop.
6841 The maximum number of branches unswitched in a single loop.
6844 The minimum cost of an expensive expression in the loop invariant motion.
6895 doing loop versioning for alignment in the vectorizer. See option
6896 ftree-vect-loop-version for more information.
6900 doing loop versioning for alias in the vectorizer. See option
6901 ftree-vect-loop-version for more information.
6904 The maximum number of iterations of a loop the brute force algorithm
6905 for analysis of # of iterations of the loop tries to evaluate.
6916 The maximum number of loop iterations we predict statically. This is useful
6917 in cases where function contain single loop with known bound and other loop
6920 loop without bounds would appear artificially cold relative to the other one.
6925 .IP "\fBalign-loop-iterations\fR" 4
6926 .IX Item "align-loop-iterations"
6927 A loop expected to iterate at lest the selected number of iterations will get
7209 .IP "\fBloop-invariant-max-bbs-in-loop\fR" 4
7210 .IX Item "loop-invariant-max-bbs-in-loop"
7211 Loop invariant motion can be very expensive, both in compile time and
7213 with more basic blocks than this parameter won't have loop invariant
7228 (in the number of gimple statements) of a basic block in a loop is
10773 of specified size, \f(CW\*(C`byte_loop\*(C'\fR, \f(CW\*(C`loop\*(C'\fR, \f(CW\*(C`unrolled_loop\*(C'\fR for
10774 expanding inline loop, \f(CW\*(C`libcall\*(C'\fR for always expanding library call.
13797 use a \f(CW\*(C`mvc\*(C'\fR loop instead. This is the default unless optimizing for
14114 ptrel before a branch, or hoist it out of a loop. For example,
14119 the loop comes to the end of the list, the program crashes because ptabs
14789 treatment of auto-aligned instructions like \f(CW\*(C`LOOP\*(C'\fR, which the
