[CSC 435] Optimization Flags
Andrew J. Pounds
pounds_aj at mercer.edu
Wed Jan 31 11:53:26 EST 2018
So I wrote a little perl program last night to run through and test ALL
of the optimization options for the gcc 6.4.0 compiler suite. In the
following output the first line is the result (dimension, time, and the
two values that we check for consistency) using optimization flag -O0.
Then I list the option I turned on in the makefile, and the results from
running the code with that optimization option only. Some things help a
little, some things do not. At least this way you can see what options
might help you. I did not try combinations of options.
I ran these on a "quiet" machine so there was little competition for
processor time with others. There are only a few options that, by
themselves, help. You may want to look for combinations of options to
get better speedup.
dock:/tmp/pounds/condusty % ./testoptions.pl | tee dump | pace
rm: cannot remove `*.o': No such file or directory
make: *** [clean] Error 1
100 61.750000000000000 386743.81287443032
0.81989322403772058
-fauto-inc-dec
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fbranch-count-reg
100 61.930000000000000 386743.81287443032
0.81989322403772058
-fcombine-stack-adjustments
100 61.729999999999997 386743.81287443032
0.81989322403772058
-fcompare-elim
100 61.840000000000003 386743.81287443032
0.81989322403772058
-fcprop-registers
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fdce
100 61.859999999999999 386743.81287443032
0.81989322403772058
-fdefer-pop
f951: Warning: this target machine does not have delayed branches
100 61.82000f951: Warning: this target machine does not have delayed
branches
0000000000 cc1: warning: this target machine does not have delayed
branches
386743.81287443032 0.81989322403772058
-fdelayed-branch
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fdse
100 61.890000000000001 386743.81287443032
0.81989322403772058
-fforward-propagate
100 61.899999999999999 386743.81287443032
0.81989322403772058
-fguess-branch-probability
100 61.850000000000001 386743.81287443032
0.81989322403772058
-fif-conversion2
100 61.840000000000003 386743.81287443032
0.81989322403772058
-fif-conversion
100 61.920000000000002 386743.81287443032
0.81989322403772058
-finline-functions-called-once
100 61.789999999999999 386743.81287443032
0.81989322403772058
-fipa-pure-const
100 61.920000000000002 386743.81287443032
0.81989322403772058
-fipa-profile
100 61.890000000000001 386743.81287443032
0.81989322403772058
-fipa-reference
100 61.600000000000001 386743.81287443032
0.81989322403772058
-fmerge-constants
100 61.770000000000003 386743.81287443032
0.81989322403772058
-fmove-loop-invariants
100 61.869999999999997 386743.81287443032
0.81989322403772058
-freorder-blocks
100 61.899999999999999 386743.81287443032
0.81989322403772058
-fshrink-wrap
100 61.579999999999998 386743.81287443032
0.81989322403772058
-fsplit-wide-types
100 61.820000000000000 386743.81287443032
0.81989322403772058
-fssa-backprop
100 62.020000000000003 386743.81287443032
0.81989322403772058
-fssa-phiopt
100 61.899999999999999 386743.81287443032
0.81989322403772058
-ftree-bit-ccp
100 61.880000000000003 386743.81287443032
0.81989322403772058
-ftree-ccp
100 61.869999999999997 386743.81287443032
0.81989322403772058
-ftree-ch
100 61.820000000000000 386743.81287443032
0.81989322403772058
-ftree-coalesce-vars
100 61.469999999999999 386743.81287443032
0.81989322403772058
-ftree-copy-prop
100 61.740000000000002 386743.81287443032
0.81989322403772058
-ftree-dce
100 61.880000000000003 386743.81287443032
0.81989322403772058
-ftree-dominator-opts
100 61.850000000000001 386743.81287443032
0.81989322403772058
-ftree-dse
100 61.890000000000001 386743.81287443032
0.81989322403772058
-ftree-forwprop
100 61.759999999999998 386743.81287443032
0.81989322403772058
-ftree-fre
100 61.930000000000000 386743.81287443032
0.81989322403772058
-ftree-phiprop
100 61.820000000000000 386743.81287443032
0.81989322403772058
-ftree-sink
100 61.899999999999999 386743.81287443032
0.81989322403772058
-ftree-slsr
100 61.869999999999997 386743.81287443032
0.81989322403772058
-ftree-sra
100 61.770000000000003 386743.81287443032
0.81989322403772058
-ftree-pta
100 61.890000000000001 386743.81287443032
0.81989322403772058
-ftree-ter
100 62.880000000000003 386743.81287443032
0.81989322403772058
-funit-at-a-time
100 61.840000000000003 386743.81287443032
0.81989322403772058
-fthread-jumps
100 61.770000000000003 386743.81287443032
0.81989322403772058
-falign-functions
100 61.920000000000002 386743.81287443032
0.81989322403772058
-falign-jumps
100 61.859999999999999 386743.81287443032
0.81989322403772058
-falign-loops
100 61.880000000000003 386743.81287443032
0.81989322403772058
-falign-labels
100 61.850000000000001 386743.81287443032
0.81989322403772058
-fcaller-saves
100 61.960000000000001 386743.81287443032
0.81989322403772058
-fcrossjumping
100 61.829999999999998 386743.81287443032
0.81989322403772058
-fcse-follow-jumps
100 61.740000000000002 386743.81287443032
0.81989322403772058
-fcse-skip-blocks
100 61.899999999999999 386743.81287443032
0.81989322403772058
-fdelete-null-pointer-checks
100 62.140000000000001 386743.81287443032
0.81989322403772058
-fdevirtualize
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fdevirtualize-speculatively
100 61.710000000000001 386743.81287443032
0.81989322403772058
-fexpensive-optimizations
100 61.469999999999999 386743.81287443032
0.81989322403772058
-fgcse
100 61.759999999999998 386743.81287443032
0.81989322403772058
-fgcse-lm
100 61.850000000000001 386743.81287443032
0.81989322403772058
-fhoist-adjacent-loads
100 61.930000000000000 386743.81287443032
0.81989322403772058
-finline-small-functions
100 61.770000000000003 386743.81287443032
0.81989322403772058
-findirect-inlining
100 61.759999999999998 386743.81287443032
0.81989322403772058
-fipa-cp
100 61.969999999999999 386743.81287443032
0.81989322403772058
-fipa-cp-alignment
100 61.810000000000002 386743.81287443032
0.81989322403772058
-fipa-sra
100 61.899999999999999 386743.81287443032
0.81989322403772058
-fipa-icf
100 61.920000000000002 386743.81287443032
0.81989322403772058
-fisolate-erroneous-paths-dereference
100 61.840000000000003 386743.81287443032
0.81989322403772058
-flra-remat
100 61.780000000000001 386743.81287443032
0.81989322403772058
-foptimize-sibling-calls
100 61.880000000000003 386743.81287443032
0.81989322403772058
-foptimize-strlen
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fpartial-inlining
100 61.590000000000003 386743.81287443032
0.81989322403772058
-fpeephole2
100 61.759999999999998 386743.81287443032
0.81989322403772058
-freorder-blocks-algorithm=stc
100 61.950000000000003 386743.81287443032
0.81989322403772058
-freorder-blocks-and-partition
100 61.799999999999997 386743.81287443032
0.81989322403772058
-freorder-functions
100 61.890000000000001 386743.81287443032
0.81989322403772058
-frerun-cse-after-loop
100 61.950000000000003 386743.81287443032
0.81989322403772058
-fsched-interblock
100 61.770000000000003 386743.81287443032
0.81989322403772058
-fsched-spec
100 61.899999999999999 386743.81287443032
0.81989322403772058
-fschedule-insns
100 61.829999999999998 386743.81287443032
0.81989322403772058
-fschedule-insns2
100 61.909999999999997 386743.81287443032
0.81989322403772058
-fstrict-aliasing
100 61.850000000000001 386743.81287443032
0.81989322403772058
-fstrict-overflow
100 61.850000000000001 386743.81287443032
0.81989322403772058
-ftree-builtin-call-dce
100 61.810000000000002 386743.81287443032
0.81989322403772058
-ftree-switch-conversion
100 61.770000000000003 386743.81287443032
0.81989322403772058
-ftree-tail-merge
100 61.750000000000000 386743.81287443032
0.81989322403772058
-ftree-pre
100 61.869999999999997 386743.81287443032
0.81989322403772058
-ftree-vrp
100 61.820000000000000 386743.81287443032
0.81989322403772058
-fipa-ra
100 61.930000000000000 386743.81287443032
0.81989322403772058
-finline-functions
100 61.920000000000002 386743.81287443032
0.81989322403772058
-funswitch-loops
100 61.789999999999999 386743.81287443032
0.81989322403772058
-fpredictive-commoning
100 61.909999999999997 386743.81287443032
0.81989322403772058
-fgcse-after-reload
100 61.850000000000001 386743.81287443032
0.81989322403772058
-ftree-loop-vectorize
100 61.859999999999999 386743.81287443032
0.81989322403772058
-ftree-loop-distribute-patterns
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fsplit-paths
100 61.750000000000000 386743.81287443032
0.81989322403772058
-ftree-slp-vectorize
100 61.869999999999997 386743.81287443032
0.81989322403772058
-fvect-cost-model
100 61.780000000000001 386743.81287443032
0.81989322403772058
-ftree-partial-pre
100 61.789999999999999 386743.81287443032
0.81989322403772058
-fipa-cp-clone
100 61.920000000000002 386743.81287443032
0.81989322403772058
dock:/tmp/pounds/condusty %
--
Andrew J. Pounds, Ph.D. (pounds_aj at mercer.edu)
Professor of Chemistry and Computer Science
Mercer University, Macon, GA 31207 (478) 301-5627
http://faculty.mercer.edu/pounds_aj
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://theochem.mercer.edu/pipermail/csc435/attachments/20180131/aea4cb2f/attachment-0001.html>
More information about the csc435
mailing list