* [doc] cleanups for optimization options in invoke.texi @ 2007-10-18 19:13 Janis Johnson 2007-10-20 20:11 ` Gerald Pfeifer 0 siblings, 1 reply; 5+ messages in thread From: Janis Johnson @ 2007-10-18 19:13 UTC (permalink / raw) To: gcc-patches This patch cleans up several little things for optimization options in the GCC Manual. Tested with make dvi/info/make. OK for mainline? Should I backport it to 4.2, 4.1? I have more cleanups coming but want to get these out of the way first. 2007-10-18 Janis Johnson <janis187@us.ibm.com> * doc/invoke.texi (Optimization Options): In summary, show that falign-* and -fsched-stalled-insns options take an optional value and that -ftree-parallelize-loops takes a value; add opindex for several optimization options; fix types in opindex for several others; in description for -fsched-stalled-insns, show it with and without a value. Index: gcc/doc/invoke.texi =================================================================== --- gcc/doc/invoke.texi (revision 129450) +++ gcc/doc/invoke.texi (working copy) @@ -315,8 +315,8 @@ @item Optimization Options @xref{Optimize Options,,Options that Control Optimization}. -@gccoptlist{-falign-functions=@var{n} -falign-jumps=@var{n} @gol --falign-labels=@var{n} -falign-loops=@var{n} @gol +@gccoptlist{-falign-functions[=@var{n}] -falign-jumps[=@var{n}] @gol +-falign-labels[=@var{n}] -falign-loops[=@var{n}] @gol -fbounds-check -fmudflap -fmudflapth -fmudflapir @gol -fbranch-probabilities -fprofile-values -fvpt -fbranch-target-load-optimize @gol -fbranch-target-load-optimize2 -fbtr-bb-exclusive @gol @@ -347,7 +347,7 @@ -fschedule-insns -fschedule-insns2 @gol -fno-sched-interblock -fno-sched-spec -fsched-spec-load @gol -fsched-spec-load-dangerous @gol --fsched-stalled-insns=@var{n} -fsched-stalled-insns-dep=@var{n} @gol +-fsched-stalled-insns[=@var{n}] -fsched-stalled-insns-dep=@var{n} @gol -fsched2-use-superblocks @gol -fsched2-use-traces -fsee -freschedule-modulo-scheduled-loops @gol -fsection-anchors -fsignaling-nans -fsingle-precision-constant @gol @@ -358,7 +358,7 @@ -fvariable-expansion-in-unroller -ftree-reassoc @gol -ftree-pre -ftree-ccp -ftree-dce -ftree-loop-optimize @gol -ftree-loop-linear -ftree-loop-im -ftree-loop-ivcanon -fivopts @gol --fcheck-data-deps -ftree-parallelize-loops @gol +-fcheck-data-deps -ftree-parallelize-loops=@var{n} @gol -ftree-dominator-opts -ftree-dse -ftree-copyrename -ftree-sink @gol -ftree-ch -ftree-sra -ftree-ter -ftree-fre -ftree-vectorize @gol -ftree-vect-loop-version -fvect-cost-model -ftree-salias -fipa-pta -fweb @gol @@ -5275,6 +5275,7 @@ optimization is turned on, use the @option{-fno-keep-static-consts} option. @item -fmerge-constants +@opindex fmerge-constants Attempt to merge identical constants (string constants and floating point constants) across compilation units. @@ -5285,6 +5286,7 @@ Enabled at levels @option{-O}, @option{-O2}, @option{-O3}, @option{-Os}. @item -fmerge-all-constants +@opindex fmerge-all-constants Attempt to merge identical constants and identical variables. This option implies @option{-fmerge-constants}. In addition to @@ -5477,14 +5479,14 @@ if it finds this kind of loop. @item -fcrossjumping -@opindex crossjumping +@opindex fcrossjumping Perform cross-jumping transformation. This transformation unifies equivalent code and save code size. The resulting code may or may not perform better than without cross-jumping. Enabled at levels @option{-O2}, @option{-O3}, @option{-Os}. @item -fif-conversion -@opindex if-conversion +@opindex fif-conversion Attempt to transform conditional jumps into branch-less equivalents. This include use of conditional moves, min, max, set flags and abs instructions, and some tricks doable by standard arithmetics. The use of conditional execution @@ -5493,7 +5495,7 @@ Enabled at levels @option{-O}, @option{-O2}, @option{-O3}, @option{-Os}. @item -fif-conversion2 -@opindex if-conversion2 +@opindex fif-conversion2 Use conditional execution (where available) to transform conditional jumps into branch-less equivalents. @@ -5584,10 +5586,15 @@ sense when scheduling before register allocation, i.e.@: with @option{-fschedule-insns} or at @option{-O2} or higher. -@item -fsched-stalled-insns=@var{n} +@item -fsched-stalled-insns +@itemx -fsched-stalled-insns=@var{n} @opindex fsched-stalled-insns -Define how many insns (if any) can be moved prematurely from the queue +Define how many insns (if any) can be removed prematurely from the queue of stalled insns into the ready list, during the second scheduling pass. +@option{-fno-fsched-stalled-insns} and @option{-fsched-stalled-insns=0} +are equivalent and mean that no insns will be removed prematurely. +If @var{n} is unspecified then there is no limit on how many queued +insns can be removed prematurely. @item -fsched-stalled-insns-dep=@var{n} @opindex fsched-stalled-insns-dep @@ -5626,7 +5633,7 @@ ones to optimal placement using LCM. @item -freschedule-modulo-scheduled-loops -@opindex fscheduling-in-modulo-scheduled-loops +@opindex freschedule-modulo-scheduled-loops The modulo scheduling comes before the traditional scheduling, if a loop was modulo scheduled we may want to prevent the later scheduling passes from changing its schedule, we use this option to control that. @@ -5644,14 +5651,17 @@ Enabled at levels @option{-O2}, @option{-O3}, @option{-Os}. @item -ftree-reassoc +@opindex ftree-reassoc Perform Reassociation on trees This flag is enabled by default at @option{-O} and higher. @item -ftree-pre +@opindex ftree-pre Perform Partial Redundancy Elimination (PRE) on trees. This flag is enabled by default at @option{-O2} and @option{-O3}. @item -ftree-fre +@opindex ftree-fre Perform Full Redundancy Elimination (FRE) on trees. The difference between FRE and PRE is that FRE only considers expressions that are computed on all paths leading to the redundant computation. @@ -5659,43 +5669,52 @@ This flag is enabled by default at @option{-O} and higher. @item -ftree-copy-prop +@opindex ftree-copy-prop Perform copy propagation on trees. This pass eliminates unnecessary copy operations. This flag is enabled by default at @option{-O} and higher. @item -ftree-store-copy-prop +@opindex ftree-store-copy-prop Perform copy propagation of memory loads and stores. This pass eliminates unnecessary copy operations in memory references (structures, global variables, arrays, etc). This flag is enabled by default at @option{-O2} and higher. @item -ftree-salias +@opindex ftree-salias Perform structural alias analysis on trees. This flag is enabled by default at @option{-O} and higher. @item -fipa-pta +@opindex fipa-pta Perform interprocedural pointer analysis. @item -ftree-sink +@opindex ftree-sink Perform forward store motion on trees. This flag is enabled by default at @option{-O} and higher. @item -ftree-ccp +@opindex ftree-ccp Perform sparse conditional constant propagation (CCP) on trees. This pass only operates on local scalar variables and is enabled by default at @option{-O} and higher. @item -ftree-store-ccp +@opindex ftree-store-ccp Perform sparse conditional constant propagation (CCP) on trees. This pass operates on both local scalar variables and memory stores and loads (global variables, structures, arrays, etc). This flag is enabled by default at @option{-O2} and higher. @item -ftree-dce +@opindex ftree-dce Perform dead code elimination (DCE) on trees. This flag is enabled by default at @option{-O} and higher. @item -ftree-dominator-opts +@opindex ftree-dominator-opts Perform a variety of simple scalar cleanups (constant/copy propagation, redundancy elimination, range propagation and expression simplification) based on a dominator tree traversal. This also @@ -5710,24 +5729,29 @@ flag is enabled by default at @option{-O} and higher. @item -ftree-ch +@opindex ftree-ch Perform loop header copying on trees. This is beneficial since it increases effectiveness of code motion optimizations. It also saves one jump. This flag is enabled by default at @option{-O} and higher. It is not enabled for @option{-Os}, since it usually increases code size. @item -ftree-loop-optimize +@opindex ftree-loop-optimize Perform loop optimizations on trees. This flag is enabled by default at @option{-O} and higher. @item -ftree-loop-linear +@opindex ftree-loop-linear Perform linear loop transformations on tree. This flag can improve cache performance and allow further loop optimizations to take place. @item -fcheck-data-deps +@opindex fcheck-data-deps Compare the results of several data dependence analyzers. This option is used for debugging the data dependence analyzers. @item -ftree-loop-im +@opindex ftree-loop-im Perform loop invariant motion on trees. This pass moves only invariants that would be hard to handle at RTL level (function calls, operations that expand to nontrivial sequences of insns). With @option{-funswitch-loops} it also moves @@ -5736,12 +5760,14 @@ store motion. @item -ftree-loop-ivcanon +@opindex ftree-loop-ivcanon Create a canonical counter for number of iterations in the loop for that determining number of iterations requires complicated analysis. Later optimizations then may determine the number easily. Useful especially in connection with unrolling. @item -fivopts +@opindex fivopts Perform induction variable optimizations (strength reduction, induction variable merging and induction variable elimination) on trees. @@ -5754,17 +5780,20 @@ rather than constrained e.g. by memory bandwidth. @item -ftree-sra +@opindex ftree-sra Perform scalar replacement of aggregates. This pass replaces structure references with scalars to prevent committing structures to memory too early. This flag is enabled by default at @option{-O} and higher. @item -ftree-copyrename +@opindex ftree-copyrename Perform copy renaming on trees. This pass attempts to rename compiler temporaries to other variables at copy locations, usually resulting in variable names which more closely resemble the original variables. This flag is enabled by default at @option{-O} and higher. @item -ftree-ter +@opindex ftree-ter Perform temporary expression replacement during the SSA->normal phase. Single use/single def temporaries are replaced at their use location with their defining expression. This results in non-GIMPLE code, but gives the expanders @@ -5772,6 +5801,7 @@ enabled by default at @option{-O} and higher. @item -ftree-vectorize +@opindex ftree-vectorize Perform loop vectorization on trees. @item -ftree-vect-loop-version @@ -5784,9 +5814,11 @@ except at level @option{-Os} where it is disabled. @item -fvect-cost-model +@opindex fvect-cost-model Enable cost model for vectorization. @item -ftree-vrp +@opindex ftree-vrp Perform Value Range Propagation on trees. This is similar to the constant propagation pass, but instead of values, ranges of values are propagated. This allows the optimizers to remove unnecessary range @@ -6222,7 +6254,7 @@ The default is @option{-fno-unsafe-math-optimizations}. @item -fassociative-math -@opindex -fassociative-math +@opindex fassociative-math Allow re-association of operands in series of floating-point operations. This violates the ISO C and C++ language standard by possibly changing @@ -6237,7 +6269,7 @@ The default is @option{-fno-associative-math}. @item -freciprocal-math -@opindex -freciprocal-math +@opindex freciprocal-math Allow the reciprocal of a value to be used instead of dividing by the value if this enables optimizations. For example @code{x / y} ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [doc] cleanups for optimization options in invoke.texi 2007-10-18 19:13 [doc] cleanups for optimization options in invoke.texi Janis Johnson @ 2007-10-20 20:11 ` Gerald Pfeifer 2007-10-23 3:15 ` Janis Johnson 0 siblings, 1 reply; 5+ messages in thread From: Gerald Pfeifer @ 2007-10-20 20:11 UTC (permalink / raw) To: Janis Johnson; +Cc: gcc-patches Hi Janis, On Thu, 18 Oct 2007, Janis Johnson wrote: > This patch cleans up several little things for optimization options in > the GCC Manual. Tested with make dvi/info/make. OK for mainline? > Should I backport it to 4.2, 4.1? thanks for working on this. Yes, as far as applicable it definitely would be good to also backport this set of fixes to the 4.2 branch and also 4.1 would be desirable. > 2007-10-18 Janis Johnson <janis187@us.ibm.com> > > * doc/invoke.texi (Optimization Options): In summary, show that > falign-* and -fsched-stalled-insns options take an optional value > and that -ftree-parallelize-loops takes a value; add opindex for > several optimization options; fix types in opindex for several > others; in description for -fsched-stalled-insns, show it with > and without a value. When it comes to the name of functions in ChangeLog entries, the GNU Coding Conventions advise not to use wild cards or regular expressions; I don't think command-line options are explicitly covered, but following the spirit we probably have to spell those out as well. Using full stops instead of semicolons will make the changelog easier to parse, and I believe "type" is a typo for "typo"? ;-) > --fsched-stalled-insns=@var{n} -fsched-stalled-insns-dep=@var{n} @gol > +-fsched-stalled-insns[=@var{n}] -fsched-stalled-insns-dep=@var{n} @gol As far as I can see the parameter is optional also in the case of -fsched-stalled-insns-dep? > -Define how many insns (if any) can be moved prematurely from the queue > +Define how many insns (if any) can be removed prematurely from the queue > of stalled insns into the ready list, during the second scheduling pass. Would't "moved" be better here than "removed" since the second half of the sentence says "into the ready list")? If so, we may want to adjust this in the following two new sentences added by your patch. The patch is fine with these changes. Thanks, Gerald ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [doc] cleanups for optimization options in invoke.texi 2007-10-20 20:11 ` Gerald Pfeifer @ 2007-10-23 3:15 ` Janis Johnson 2007-11-02 12:21 ` Jakub Jelinek 0 siblings, 1 reply; 5+ messages in thread From: Janis Johnson @ 2007-10-23 3:15 UTC (permalink / raw) To: Gerald Pfeifer; +Cc: gcc-patches On Sat, 2007-10-20 at 17:35 +0200, Gerald Pfeifer wrote: > Hi Janis, > > On Thu, 18 Oct 2007, Janis Johnson wrote: > > This patch cleans up several little things for optimization options in > > the GCC Manual. Tested with make dvi/info/make. OK for mainline? > > Should I backport it to 4.2, 4.1? > > thanks for working on this. Yes, as far as applicable it definitely > would be good to also backport this set of fixes to the 4.2 branch and > also 4.1 would be desirable. Thanks for the review, here's what I checked into mainline: 2007-10-22 Janis Johnson <janis187@us.ibm.com> * doc/invoke.texi (Optimization Options): In the summary, show that the value for options -falign-jumps, -falign-labels, -falign-loops, -falign-functions, -fsched-stalled-insns, and -fsched-stalled-insns-dep is optional. In the summary, show that -fsched-stalled-insns takes an optional argument and that -fparallelize-loops takes an argument. Add opindex for several optimization options, and fix typos in the opindex commands for several others. In the descriptions for -fsched-stalled-insns and -fsched-stalled-insns-dep, show the options with and without values. Index: gcc/doc/invoke.texi =================================================================== --- gcc/doc/invoke.texi (revision 129552) +++ gcc/doc/invoke.texi (working copy) @@ -315,8 +315,8 @@ @item Optimization Options @xref{Optimize Options,,Options that Control Optimization}. -@gccoptlist{-falign-functions=@var{n} -falign-jumps=@var{n} @gol --falign-labels=@var{n} -falign-loops=@var{n} @gol +@gccoptlist{-falign-functions[=@var{n}] -falign-jumps[=@var{n}] @gol +-falign-labels[=@var{n}] -falign-loops[=@var{n}] @gol -fbounds-check -fmudflap -fmudflapth -fmudflapir @gol -fbranch-probabilities -fprofile-values -fvpt -fbranch-target-load-optimize @gol -fbranch-target-load-optimize2 -fbtr-bb-exclusive @gol @@ -347,7 +347,7 @@ -fschedule-insns -fschedule-insns2 @gol -fno-sched-interblock -fno-sched-spec -fsched-spec-load @gol -fsched-spec-load-dangerous @gol --fsched-stalled-insns=@var{n} -fsched-stalled-insns-dep=@var{n} @gol +-fsched-stalled-insns[=@var{n}] -fsched-stalled-insns-dep[=@var{n}] @gol -fsched2-use-superblocks @gol -fsched2-use-traces -fsee -freschedule-modulo-scheduled-loops @gol -fsection-anchors -fsignaling-nans -fsingle-precision-constant @gol @@ -358,7 +358,7 @@ -fvariable-expansion-in-unroller -ftree-reassoc @gol -ftree-pre -ftree-ccp -ftree-dce -ftree-loop-optimize @gol -ftree-loop-linear -ftree-loop-im -ftree-loop-ivcanon -fivopts @gol --fcheck-data-deps -ftree-parallelize-loops @gol +-fcheck-data-deps -ftree-parallelize-loops=@var{n} @gol -ftree-dominator-opts -ftree-dse -ftree-copyrename -ftree-sink @gol -ftree-ch -ftree-sra -ftree-ter -ftree-fre -ftree-vectorize @gol -ftree-vect-loop-version -fvect-cost-model -ftree-salias -fipa-pta -fweb @gol @@ -5277,6 +5277,7 @@ optimization is turned on, use the @option{-fno-keep-static-consts} option. @item -fmerge-constants +@opindex fmerge-constants Attempt to merge identical constants (string constants and floating point constants) across compilation units. @@ -5287,6 +5288,7 @@ Enabled at levels @option{-O}, @option{-O2}, @option{-O3}, @option{-Os}. @item -fmerge-all-constants +@opindex fmerge-all-constants Attempt to merge identical constants and identical variables. This option implies @option{-fmerge-constants}. In addition to @@ -5479,14 +5481,14 @@ if it finds this kind of loop. @item -fcrossjumping -@opindex crossjumping +@opindex fcrossjumping Perform cross-jumping transformation. This transformation unifies equivalent code and save code size. The resulting code may or may not perform better than without cross-jumping. Enabled at levels @option{-O2}, @option{-O3}, @option{-Os}. @item -fif-conversion -@opindex if-conversion +@opindex fif-conversion Attempt to transform conditional jumps into branch-less equivalents. This include use of conditional moves, min, max, set flags and abs instructions, and some tricks doable by standard arithmetics. The use of conditional execution @@ -5495,7 +5497,7 @@ Enabled at levels @option{-O}, @option{-O2}, @option{-O3}, @option{-Os}. @item -fif-conversion2 -@opindex if-conversion2 +@opindex fif-conversion2 Use conditional execution (where available) to transform conditional jumps into branch-less equivalents. @@ -5586,17 +5588,27 @@ sense when scheduling before register allocation, i.e.@: with @option{-fschedule-insns} or at @option{-O2} or higher. -@item -fsched-stalled-insns=@var{n} +@item -fsched-stalled-insns +@itemx -fsched-stalled-insns=@var{n} @opindex fsched-stalled-insns Define how many insns (if any) can be moved prematurely from the queue of stalled insns into the ready list, during the second scheduling pass. +@option{-fno-fsched-stalled-insns} and @option{-fsched-stalled-insns=0} +are equivalent and mean that no insns will be moved prematurely. +If @var{n} is unspecified then there is no limit on how many queued +insns can be moved prematurely. -@item -fsched-stalled-insns-dep=@var{n} +@item -fsched-stalled-insns-dep +@itemx -fsched-stalled-insns-dep=@var{n} @opindex fsched-stalled-insns-dep Define how many insn groups (cycles) will be examined for a dependency on a stalled insn that is candidate for premature removal from the queue -of stalled insns. Has an effect only during the second scheduling pass, +of stalled insns. This has an effect only during the second scheduling pass, and only if @option{-fsched-stalled-insns} is used and its value is not zero. +@option{-fno-sched-stalled-insns-dep} is equivalent to +@option{-fsched-stalled-insns-dep=0}. +@option{-fsched-stalled-insns-dep} without a value is equivalent to +@option{-fsched-stalled-insns-dep=1}. @item -fsched2-use-superblocks @opindex fsched2-use-superblocks @@ -5628,7 +5640,7 @@ ones to optimal placement using LCM. @item -freschedule-modulo-scheduled-loops -@opindex fscheduling-in-modulo-scheduled-loops +@opindex freschedule-modulo-scheduled-loops The modulo scheduling comes before the traditional scheduling, if a loop was modulo scheduled we may want to prevent the later scheduling passes from changing its schedule, we use this option to control that. @@ -5646,14 +5658,17 @@ Enabled at levels @option{-O2}, @option{-O3}, @option{-Os}. @item -ftree-reassoc +@opindex ftree-reassoc Perform Reassociation on trees This flag is enabled by default at @option{-O} and higher. @item -ftree-pre +@opindex ftree-pre Perform Partial Redundancy Elimination (PRE) on trees. This flag is enabled by default at @option{-O2} and @option{-O3}. @item -ftree-fre +@opindex ftree-fre Perform Full Redundancy Elimination (FRE) on trees. The difference between FRE and PRE is that FRE only considers expressions that are computed on all paths leading to the redundant computation. @@ -5661,43 +5676,52 @@ This flag is enabled by default at @option{-O} and higher. @item -ftree-copy-prop +@opindex ftree-copy-prop Perform copy propagation on trees. This pass eliminates unnecessary copy operations. This flag is enabled by default at @option{-O} and higher. @item -ftree-store-copy-prop +@opindex ftree-store-copy-prop Perform copy propagation of memory loads and stores. This pass eliminates unnecessary copy operations in memory references (structures, global variables, arrays, etc). This flag is enabled by default at @option{-O2} and higher. @item -ftree-salias +@opindex ftree-salias Perform structural alias analysis on trees. This flag is enabled by default at @option{-O} and higher. @item -fipa-pta +@opindex fipa-pta Perform interprocedural pointer analysis. @item -ftree-sink +@opindex ftree-sink Perform forward store motion on trees. This flag is enabled by default at @option{-O} and higher. @item -ftree-ccp +@opindex ftree-ccp Perform sparse conditional constant propagation (CCP) on trees. This pass only operates on local scalar variables and is enabled by default at @option{-O} and higher. @item -ftree-store-ccp +@opindex ftree-store-ccp Perform sparse conditional constant propagation (CCP) on trees. This pass operates on both local scalar variables and memory stores and loads (global variables, structures, arrays, etc). This flag is enabled by default at @option{-O2} and higher. @item -ftree-dce +@opindex ftree-dce Perform dead code elimination (DCE) on trees. This flag is enabled by default at @option{-O} and higher. @item -ftree-dominator-opts +@opindex ftree-dominator-opts Perform a variety of simple scalar cleanups (constant/copy propagation, redundancy elimination, range propagation and expression simplification) based on a dominator tree traversal. This also @@ -5712,24 +5736,29 @@ flag is enabled by default at @option{-O} and higher. @item -ftree-ch +@opindex ftree-ch Perform loop header copying on trees. This is beneficial since it increases effectiveness of code motion optimizations. It also saves one jump. This flag is enabled by default at @option{-O} and higher. It is not enabled for @option{-Os}, since it usually increases code size. @item -ftree-loop-optimize +@opindex ftree-loop-optimize Perform loop optimizations on trees. This flag is enabled by default at @option{-O} and higher. @item -ftree-loop-linear +@opindex ftree-loop-linear Perform linear loop transformations on tree. This flag can improve cache performance and allow further loop optimizations to take place. @item -fcheck-data-deps +@opindex fcheck-data-deps Compare the results of several data dependence analyzers. This option is used for debugging the data dependence analyzers. @item -ftree-loop-im +@opindex ftree-loop-im Perform loop invariant motion on trees. This pass moves only invariants that would be hard to handle at RTL level (function calls, operations that expand to nontrivial sequences of insns). With @option{-funswitch-loops} it also moves @@ -5738,12 +5767,14 @@ store motion. @item -ftree-loop-ivcanon +@opindex ftree-loop-ivcanon Create a canonical counter for number of iterations in the loop for that determining number of iterations requires complicated analysis. Later optimizations then may determine the number easily. Useful especially in connection with unrolling. @item -fivopts +@opindex fivopts Perform induction variable optimizations (strength reduction, induction variable merging and induction variable elimination) on trees. @@ -5756,17 +5787,20 @@ rather than constrained e.g. by memory bandwidth. @item -ftree-sra +@opindex ftree-sra Perform scalar replacement of aggregates. This pass replaces structure references with scalars to prevent committing structures to memory too early. This flag is enabled by default at @option{-O} and higher. @item -ftree-copyrename +@opindex ftree-copyrename Perform copy renaming on trees. This pass attempts to rename compiler temporaries to other variables at copy locations, usually resulting in variable names which more closely resemble the original variables. This flag is enabled by default at @option{-O} and higher. @item -ftree-ter +@opindex ftree-ter Perform temporary expression replacement during the SSA->normal phase. Single use/single def temporaries are replaced at their use location with their defining expression. This results in non-GIMPLE code, but gives the expanders @@ -5774,6 +5808,7 @@ enabled by default at @option{-O} and higher. @item -ftree-vectorize +@opindex ftree-vectorize Perform loop vectorization on trees. @item -ftree-vect-loop-version @@ -5786,9 +5821,11 @@ except at level @option{-Os} where it is disabled. @item -fvect-cost-model +@opindex fvect-cost-model Enable cost model for vectorization. @item -ftree-vrp +@opindex ftree-vrp Perform Value Range Propagation on trees. This is similar to the constant propagation pass, but instead of values, ranges of values are propagated. This allows the optimizers to remove unnecessary range @@ -6224,7 +6261,7 @@ The default is @option{-fno-unsafe-math-optimizations}. @item -fassociative-math -@opindex -fassociative-math +@opindex fassociative-math Allow re-association of operands in series of floating-point operations. This violates the ISO C and C++ language standard by possibly changing @@ -6239,7 +6276,7 @@ The default is @option{-fno-associative-math}. @item -freciprocal-math -@opindex -freciprocal-math +@opindex freciprocal-math Allow the reciprocal of a value to be used instead of dividing by the value if this enables optimizations. For example @code{x / y} ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [doc] cleanups for optimization options in invoke.texi 2007-10-23 3:15 ` Janis Johnson @ 2007-11-02 12:21 ` Jakub Jelinek 2007-11-13 22:13 ` Diego Novillo 0 siblings, 1 reply; 5+ messages in thread From: Jakub Jelinek @ 2007-11-02 12:21 UTC (permalink / raw) To: Janis Johnson; +Cc: Gerald Pfeifer, gcc-patches On Mon, Oct 22, 2007 at 04:24:01PM -0700, Janis Johnson wrote: > 2007-10-22 Janis Johnson <janis187@us.ibm.com> > > * doc/invoke.texi (Optimization Options): In the summary, show > that the value for options -falign-jumps, -falign-labels, > -falign-loops, -falign-functions, -fsched-stalled-insns, and > -fsched-stalled-insns-dep is optional. In the summary, show that > -fsched-stalled-insns takes an optional argument and that > -fparallelize-loops takes an argument. Add opindex for several > optimization options, and fix typos in the opindex commands for > several others. In the descriptions for -fsched-stalled-insns > and -fsched-stalled-insns-dep, show the options with and without > values. The description of -fsched-stalled-insns doesn't match what the compiler actually does. -fno-sched-stalled-insns (invoke.texi has there an extra f) means what is written, but -fsched-stalled-insns=0 means infinity, not zero and -fsched-stalled-insns without argument is the same as -fsched-stalled-insns=1. Ok for trunk? 2007-11-02 Jakub Jelinek <jakub@redhat.com> * doc/invoke.texi: Fix description of -fsched-stalled-insns=0 and -fsched-stalled-insns. --- gcc/doc/invoke.texi.jj 2007-11-01 20:48:56.000000000 +0100 +++ gcc/doc/invoke.texi 2007-11-02 13:14:10.000000000 +0100 @@ -5601,10 +5601,11 @@ sense when scheduling before register al @opindex fsched-stalled-insns Define how many insns (if any) can be moved prematurely from the queue of stalled insns into the ready list, during the second scheduling pass. -@option{-fno-fsched-stalled-insns} and @option{-fsched-stalled-insns=0} -are equivalent and mean that no insns will be moved prematurely. -If @var{n} is unspecified then there is no limit on how many queued -insns can be moved prematurely. +@option{-fno-sched-stalled-insns} means that no insns will be moved +prematurely, @option{-fsched-stalled-insns=0} means there is no limit +on how many queued insns can be moved prematurely. +@option{-fsched-stalled-insns} without a value is equivalent to +@option{-fsched-stalled-insns=1}. @item -fsched-stalled-insns-dep @itemx -fsched-stalled-insns-dep=@var{n} @@ -5612,7 +5613,7 @@ insns can be moved prematurely. Define how many insn groups (cycles) will be examined for a dependency on a stalled insn that is candidate for premature removal from the queue of stalled insns. This has an effect only during the second scheduling pass, -and only if @option{-fsched-stalled-insns} is used and its value is not zero. +and only if @option{-fsched-stalled-insns} is used. @option{-fno-sched-stalled-insns-dep} is equivalent to @option{-fsched-stalled-insns-dep=0}. @option{-fsched-stalled-insns-dep} without a value is equivalent to Jakub ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [doc] cleanups for optimization options in invoke.texi 2007-11-02 12:21 ` Jakub Jelinek @ 2007-11-13 22:13 ` Diego Novillo 0 siblings, 0 replies; 5+ messages in thread From: Diego Novillo @ 2007-11-13 22:13 UTC (permalink / raw) To: Jakub Jelinek; +Cc: Janis Johnson, Gerald Pfeifer, gcc-patches Jakub Jelinek wrote: > 2007-11-02 Jakub Jelinek <jakub@redhat.com> > > * doc/invoke.texi: Fix description of -fsched-stalled-insns=0 > and -fsched-stalled-insns. OK. ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2007-11-13 20:28 UTC | newest] Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2007-10-18 19:13 [doc] cleanups for optimization options in invoke.texi Janis Johnson 2007-10-20 20:11 ` Gerald Pfeifer 2007-10-23 3:15 ` Janis Johnson 2007-11-02 12:21 ` Jakub Jelinek 2007-11-13 22:13 ` Diego Novillo
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).