43627 – [4.5 Regression] slow compilation (tree canonical iv takes 75%)

Bug 43627 - [4.5 Regression] slow compilation (tree canonical iv takes 75%)

Summary: [4.5 Regression] slow compilation (tree canonical iv takes 75%)

Status:	RESOLVED FIXED

Alias:	None

Product:	gcc
Classification:	Unclassified
Component:	tree-optimization (show other bugs)
Version:	4.5.0

Importance:	P2 normal
Target Milestone:	4.5.1
Assignee:	Richard Biener

URL:	http://gcc.gnu.org/ml/gcc-patches/201...
Keywords:	compile-time-hog, missed-optimization

Depends on:
Blocks:	41043
	Show dependency tree / graph

Reported:	2010-04-02 08:14 UTC by Joost VandeVondele
Modified:	2010-04-15 13:47 UTC (History)
CC List:	2 users (show)

See Also:
Host:
Target:	x86_64--
Build:
Known to work:	4.4.3 4.5.1 4.6.0
Known to fail:	4.5.0
Last reconfirmed:	2010-04-02 14:23:22

Attachments
testcase (37.57 KB, application/octet-stream) 2010-04-02 08:16 UTC, Joost VandeVondele	Details
smaller testcase (needs 3s, 80% in tree canonical iv) (2.11 KB, text/plain) 2010-04-02 14:07 UTC, Joost VandeVondele	Details
reduced testcase (1.92 KB, text/plain) 2010-04-02 14:08 UTC, Richard Biener	Details
minimal patch (896 bytes, patch) 2010-04-02 15:10 UTC, Richard Biener	Details \| Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Joost VandeVondele 2010-04-02 08:14:09 UTC

The to-be-attached file compiles very slowly with 4.5:

4.3 ([gcc-4_3-branch revision 135036]): 37s
4.4 ([gcc-4_4-branch revision 150482]): 30s
4.5 ([trunk revision 157940]):        6m35s

 gfortran -fbounds-check -g -O3 -ffast-math -funroll-loops -ftree-vectorize -march=native -c hog.f90

Comment 1 Joost VandeVondele 2010-04-02 08:16:17 UTC

Created attachment 20287 [details]
testcase

reproduce with 

gfortran -fbounds-check -g -O3 -ffast-math -funroll-loops -ftree-vectorize -march=native -c hog.f90

Comment 2 Joost VandeVondele 2010-04-02 08:27:40 UTC

And a timing report as well (notice the machine is not fully idle). The major consumer is tree canonical.

Execution times (seconds)
 garbage collection    :   7.71 ( 2%) usr   0.07 ( 4%) sys  14.12 ( 2%) wall       0 kB ( 0%) ggc
 callgraph construction:   0.18 ( 0%) usr   0.01 ( 1%) sys   0.24 ( 0%) wall    6675 kB ( 1%) ggc
 callgraph optimization:   0.61 ( 0%) usr   0.03 ( 2%) sys   0.61 ( 0%) wall    1655 kB ( 0%) ggc
 ipa cp                :   0.19 ( 0%) usr   0.00 ( 0%) sys   0.19 ( 0%) wall     539 kB ( 0%) ggc
 ipa reference         :   0.15 ( 0%) usr   0.00 ( 0%) sys   0.15 ( 0%) wall       0 kB ( 0%) ggc
 ipa pure const        :   0.17 ( 0%) usr   0.00 ( 0%) sys   0.17 ( 0%) wall       0 kB ( 0%) ggc
 ipa SRA               :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 cfg cleanup           :   0.78 ( 0%) usr   0.01 ( 1%) sys   1.27 ( 0%) wall    3661 kB ( 0%) ggc
 CFG verifier          :   2.10 ( 1%) usr   0.00 ( 0%) sys   3.40 ( 1%) wall       0 kB ( 0%) ggc
 trivially dead code   :   0.38 ( 0%) usr   0.00 ( 0%) sys   0.40 ( 0%) wall       0 kB ( 0%) ggc
 df multiple defs      :   0.59 ( 0%) usr   0.00 ( 0%) sys   0.92 ( 0%) wall       0 kB ( 0%) ggc
 df reaching defs      :   0.86 ( 0%) usr   0.00 ( 0%) sys   1.83 ( 0%) wall       0 kB ( 0%) ggc
 df live regs          :   4.92 ( 1%) usr   0.01 ( 1%) sys   8.23 ( 1%) wall       0 kB ( 0%) ggc
 df live&initialized regs:   1.48 ( 0%) usr   0.01 ( 1%) sys   3.37 ( 1%) wall       0 kB ( 0%) ggc
 df use-def / def-use chains:   0.71 ( 0%) usr   0.00 ( 0%) sys   1.39 ( 0%) wall       0 kB ( 0%) ggc
 df reg dead/unused notes:   4.15 ( 1%) usr   0.01 ( 1%) sys   7.47 ( 1%) wall    9314 kB ( 1%) ggc
 register information  :   1.29 ( 0%) usr   0.01 ( 1%) sys   3.00 ( 0%) wall       0 kB ( 0%) ggc
 alias analysis        :   0.64 ( 0%) usr   0.00 ( 0%) sys   0.74 ( 0%) wall   21770 kB ( 3%) ggc
 alias stmt walking    :   1.94 ( 1%) usr   0.06 ( 4%) sys   3.50 ( 1%) wall       0 kB ( 0%) ggc
 register scan         :   0.18 ( 0%) usr   0.00 ( 0%) sys   0.11 ( 0%) wall       0 kB ( 0%) ggc
 rebuild jump labels   :   0.23 ( 0%) usr   0.00 ( 0%) sys   0.26 ( 0%) wall       0 kB ( 0%) ggc
 parser                :   1.27 ( 0%) usr   0.12 ( 7%) sys   1.50 ( 0%) wall   42200 kB ( 5%) ggc
 inline heuristics     :   0.43 ( 0%) usr   0.02 ( 1%) sys   0.34 ( 0%) wall       0 kB ( 0%) ggc
 tree gimplify         :   0.69 ( 0%) usr   0.03 ( 2%) sys   0.79 ( 0%) wall   52375 kB ( 6%) ggc
 tree eh               :   0.03 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall       0 kB ( 0%) ggc
 tree CFG construction :   0.08 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall    9418 kB ( 1%) ggc
 tree CFG cleanup      :   0.49 ( 0%) usr   0.00 ( 0%) sys   0.80 ( 0%) wall     418 kB ( 0%) ggc
 tree VRP              :   2.08 ( 1%) usr   0.05 ( 3%) sys   3.67 ( 1%) wall   54923 kB ( 7%) ggc
 tree copy propagation :   0.37 ( 0%) usr   0.00 ( 0%) sys   0.59 ( 0%) wall     237 kB ( 0%) ggc
 tree find ref. vars   :   0.07 ( 0%) usr   0.02 ( 1%) sys   0.09 ( 0%) wall    3774 kB ( 0%) ggc
 tree PTA              :   0.19 ( 0%) usr   0.00 ( 0%) sys   0.19 ( 0%) wall     425 kB ( 0%) ggc
 tree PHI insertion    :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall     315 kB ( 0%) ggc
 tree SSA rewrite      :   0.44 ( 0%) usr   0.03 ( 2%) sys   0.80 ( 0%) wall   20682 kB ( 3%) ggc
 tree SSA other        :   0.22 ( 0%) usr   0.02 ( 1%) sys   0.23 ( 0%) wall     434 kB ( 0%) ggc
 tree SSA incremental  :   0.62 ( 0%) usr   0.04 ( 2%) sys   0.91 ( 0%) wall     438 kB ( 0%) ggc
 tree operand scan     :   0.27 ( 0%) usr   0.14 ( 8%) sys   0.53 ( 0%) wall   21791 kB ( 3%) ggc
 dominator optimization:   0.42 ( 0%) usr   0.00 ( 0%) sys   0.72 ( 0%) wall    4190 kB ( 1%) ggc
 tree CCP              :   0.56 ( 0%) usr   0.01 ( 1%) sys   0.70 ( 0%) wall    3081 kB ( 0%) ggc
 tree PHI const/copy prop:   0.05 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall      22 kB ( 0%) ggc
 tree split crit edges :   0.03 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall    3268 kB ( 0%) ggc
 tree reassociation    :   0.17 ( 0%) usr   0.00 ( 0%) sys   0.36 ( 0%) wall     161 kB ( 0%) ggc
 tree PRE              :   6.54 ( 2%) usr   0.02 ( 1%) sys  11.71 ( 2%) wall   25200 kB ( 3%) ggc
 tree FRE              :   0.76 ( 0%) usr   0.03 ( 2%) sys   1.15 ( 0%) wall    8100 kB ( 1%) ggc
 tree code sinking     :   0.23 ( 0%) usr   0.04 ( 2%) sys   0.44 ( 0%) wall   12275 kB ( 2%) ggc
 tree linearize phis   :   0.03 ( 0%) usr   0.00 ( 0%) sys   0.05 ( 0%) wall       0 kB ( 0%) ggc
 tree forward propagate:   0.19 ( 0%) usr   0.01 ( 1%) sys   0.25 ( 0%) wall    9572 kB ( 1%) ggc
 tree phiprop          :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 tree conservative DCE :   0.19 ( 0%) usr   0.02 ( 1%) sys   0.51 ( 0%) wall      17 kB ( 0%) ggc
 tree aggressive DCE   :   0.49 ( 0%) usr   0.01 ( 1%) sys   0.74 ( 0%) wall    2998 kB ( 0%) ggc
 tree buildin call DCE :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 tree DSE              :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall      27 kB ( 0%) ggc
 tree loop bounds      :   0.21 ( 0%) usr   0.00 ( 0%) sys   0.47 ( 0%) wall    6310 kB ( 1%) ggc
 tree loop invariant motion:   0.29 ( 0%) usr   0.01 ( 1%) sys   0.45 ( 0%) wall     498 kB ( 0%) ggc
 tree canonical iv     : 230.79 (62%) usr   0.10 ( 6%) sys 393.03 (61%) wall  146373 kB (18%) ggc
 scev constant prop    :   0.11 ( 0%) usr   0.00 ( 0%) sys   0.35 ( 0%) wall    5809 kB ( 1%) ggc
 tree loop unswitching :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall       0 kB ( 0%) ggc
 complete unrolling    :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.15 ( 0%) wall    1123 kB ( 0%) ggc
 tree vectorization    :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall      40 kB ( 0%) ggc
 tree slp vectorization:   0.48 ( 0%) usr   0.00 ( 0%) sys   0.83 ( 0%) wall   19329 kB ( 2%) ggc
 tree iv optimization  :   0.59 ( 0%) usr   0.00 ( 0%) sys   0.77 ( 0%) wall   13315 kB ( 2%) ggc
 predictive commoning  :   1.44 ( 0%) usr   0.00 ( 0%) sys   2.29 ( 0%) wall   40577 kB ( 5%) ggc
 tree loop init        :   0.17 ( 0%) usr   0.01 ( 1%) sys   0.31 ( 0%) wall    5246 kB ( 1%) ggc
 tree loop fini        :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 tree copy headers     :   0.02 ( 0%) usr   0.01 ( 1%) sys   0.07 ( 0%) wall     758 kB ( 0%) ggc
 tree SSA uncprop      :   0.04 ( 0%) usr   0.00 ( 0%) sys   0.09 ( 0%) wall       0 kB ( 0%) ggc
 tree rename SSA copies:   0.06 ( 0%) usr   0.00 ( 0%) sys   0.13 ( 0%) wall       0 kB ( 0%) ggc
 tree SSA verifier     :   9.57 ( 3%) usr   0.01 ( 1%) sys  15.09 ( 2%) wall       0 kB ( 0%) ggc
 tree STMT verifier    :  18.08 ( 5%) usr   0.10 ( 6%) sys  30.59 ( 5%) wall       0 kB ( 0%) ggc
 tree switch initialization conversion:   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 callgraph verifier    :   1.64 ( 0%) usr   0.00 ( 0%) sys   1.83 ( 0%) wall       0 kB ( 0%) ggc
 dominance frontiers   :   0.08 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall       0 kB ( 0%) ggc
 dominance computation :   0.58 ( 0%) usr   0.00 ( 0%) sys   0.84 ( 0%) wall       0 kB ( 0%) ggc
 control dependences   :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 expand                :   8.51 ( 2%) usr   0.05 ( 3%) sys  15.28 ( 2%) wall   76554 kB ( 9%) ggc
 jump                  :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall       0 kB ( 0%) ggc
 forward prop          :   1.18 ( 0%) usr   0.00 ( 0%) sys   2.75 ( 0%) wall    6749 kB ( 1%) ggc
 CSE                   :   1.51 ( 0%) usr   0.01 ( 1%) sys   2.73 ( 0%) wall    1375 kB ( 0%) ggc
 dead code elimination :   0.73 ( 0%) usr   0.00 ( 0%) sys   1.60 ( 0%) wall       0 kB ( 0%) ggc
 dead store elim1      :   0.75 ( 0%) usr   0.01 ( 1%) sys   1.18 ( 0%) wall    5337 kB ( 1%) ggc
 dead store elim2      :   1.39 ( 0%) usr   0.00 ( 0%) sys   2.67 ( 0%) wall    6079 kB ( 1%) ggc
 loop analysis         :   0.08 ( 0%) usr   0.01 ( 1%) sys   0.06 ( 0%) wall      61 kB ( 0%) ggc
 loop invariant motion :   0.10 ( 0%) usr   0.01 ( 1%) sys   0.16 ( 0%) wall       1 kB ( 0%) ggc
 loop unswitching      :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall       0 kB ( 0%) ggc
 loop unrolling        :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall     190 kB ( 0%) ggc
 CPROP                 :   1.05 ( 0%) usr   0.00 ( 0%) sys   1.94 ( 0%) wall    7896 kB ( 1%) ggc
 PRE                   :   0.29 ( 0%) usr   0.00 ( 0%) sys   0.52 ( 0%) wall     882 kB ( 0%) ggc
 web                   :   1.08 ( 0%) usr   0.00 ( 0%) sys   1.81 ( 0%) wall      23 kB ( 0%) ggc
 CSE 2                 :   1.53 ( 0%) usr   0.00 ( 0%) sys   2.51 ( 0%) wall     793 kB ( 0%) ggc
 branch prediction     :   0.14 ( 0%) usr   0.01 ( 1%) sys   0.25 ( 0%) wall    4053 kB ( 0%) ggc
 combiner              :   2.39 ( 1%) usr   0.02 ( 1%) sys   4.13 ( 1%) wall   26323 kB ( 3%) ggc
 if-conversion         :   0.08 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall     130 kB ( 0%) ggc
 regmove               :   0.36 ( 0%) usr   0.00 ( 0%) sys   0.47 ( 0%) wall       4 kB ( 0%) ggc
 integrated RA         :   8.51 ( 2%) usr   0.01 ( 1%) sys  14.18 ( 2%) wall    8933 kB ( 1%) ggc
 reload                :   1.93 ( 1%) usr   0.04 ( 2%) sys   3.31 ( 1%) wall    1774 kB ( 0%) ggc
 reload CSE regs       :   0.80 ( 0%) usr   0.01 ( 1%) sys   1.54 ( 0%) wall    9904 kB ( 1%) ggc
 load CSE after reload :   0.15 ( 0%) usr   0.00 ( 0%) sys   0.19 ( 0%) wall       0 kB ( 0%) ggc
 thread pro- & epilogue:   0.14 ( 0%) usr   0.00 ( 0%) sys   0.24 ( 0%) wall     572 kB ( 0%) ggc
 if-conversion 2       :   0.04 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall      59 kB ( 0%) ggc
 combine stack adjustments:   0.04 ( 0%) usr   0.00 ( 0%) sys   0.05 ( 0%) wall       0 kB ( 0%) ggc
 peephole 2            :   0.44 ( 0%) usr   0.00 ( 0%) sys   0.56 ( 0%) wall    2057 kB ( 0%) ggc
 rename registers      :   0.44 ( 0%) usr   0.00 ( 0%) sys   0.85 ( 0%) wall     701 kB ( 0%) ggc
 hard reg cprop        :   0.64 ( 0%) usr   0.00 ( 0%) sys   1.03 ( 0%) wall      35 kB ( 0%) ggc
 scheduling 2          :   1.70 ( 0%) usr   0.03 ( 2%) sys   3.15 ( 0%) wall     257 kB ( 0%) ggc
 machine dep reorg     :   0.18 ( 0%) usr   0.00 ( 0%) sys   0.41 ( 0%) wall       0 kB ( 0%) ggc
 reorder blocks        :   0.13 ( 0%) usr   0.00 ( 0%) sys   0.26 ( 0%) wall    2145 kB ( 0%) ggc
 final                 :   0.91 ( 0%) usr   0.03 ( 2%) sys   1.67 ( 0%) wall    5904 kB ( 1%) ggc
 symout                :   0.47 ( 0%) usr   0.07 ( 4%) sys   1.15 ( 0%) wall   50781 kB ( 6%) ggc
 variable tracking     :  26.64 ( 7%) usr   0.32 (19%) sys  48.05 ( 7%) wall   38563 kB ( 5%) ggc
 plugin execution      :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall       0 kB ( 0%) ggc
 TOTAL                 : 374.92             1.71           641.15             817719 kB
Extra diagnostic checks enabled; compiler may run slowly.
Configure with --enable-checking=release to disable checks.

real    10m46.566s
user    6m17.140s

Comment 3 Steven Bosscher 2010-04-02 09:18:29 UTC

This tells me you are comparing apples and cows: "Extra diagnostic checks enabled; compiler may run slowly."

Could you try again with a compiler configured with --enable=checking=release?

Comment 4 Joost VandeVondele 2010-04-02 09:26:30 UTC

(In reply to comment #3)
> This tells me you are comparing apples and cows: "Extra diagnostic checks
> enabled; compiler may run slowly."
> 
> Could you try again with a compiler configured with --enable=checking=release?
> 

I'll do now...

for reference, 4.4 has:

> gfortran -ftime-report -fbounds-check -g -O3 -ffast-math -funroll-loops -ftree-vectorize -march=native hog.f90

Execution times (seconds)
 garbage collection    :   0.15 ( 1%) usr   0.00 ( 0%) sys   0.14 ( 1%) wall       0 kB ( 0%) ggc
 callgraph construction:   0.33 ( 1%) usr   0.03 ( 4%) sys   0.33 ( 1%) wall    9447 kB ( 2%) ggc
 callgraph optimization:   0.46 ( 2%) usr   0.01 ( 1%) sys   0.50 ( 2%) wall     239 kB ( 0%) ggc
 ipa cp                :   0.22 ( 1%) usr   0.00 ( 0%) sys   0.24 ( 1%) wall       0 kB ( 0%) ggc
 ipa reference         :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 cfg cleanup           :   0.08 ( 0%) usr   0.00 ( 0%) sys   0.12 ( 0%) wall     914 kB ( 0%) ggc
 trivially dead code   :   0.10 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall       0 kB ( 0%) ggc
 df reaching defs      :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.12 ( 0%) wall       0 kB ( 0%) ggc
 df live regs          :   0.22 ( 1%) usr   0.00 ( 0%) sys   0.18 ( 1%) wall       0 kB ( 0%) ggc
 df live&initialized regs:   0.10 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall       0 kB ( 0%) ggc
 df use-def / def-use chains:   0.11 ( 0%) usr   0.00 ( 0%) sys   0.05 ( 0%) wall       0 kB ( 0%) ggc
 df reg dead/unused notes:   0.17 ( 1%) usr   0.00 ( 0%) sys   0.09 ( 0%) wall    3443 kB ( 1%) ggc
 register information  :   0.07 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall       0 kB ( 0%) ggc
 alias analysis        :   0.14 ( 1%) usr   0.00 ( 0%) sys   0.12 ( 0%) wall    6273 kB ( 1%) ggc
 register scan         :   0.04 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall       0 kB ( 0%) ggc
 rebuild jump labels   :   0.08 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 parser                :   1.27 ( 5%) usr   0.12 (16%) sys   1.31 ( 5%) wall   50936 kB ( 9%) ggc
 inline heuristics     :   0.13 ( 1%) usr   0.05 ( 6%) sys   0.25 ( 1%) wall       0 kB ( 0%) ggc
 tree gimplify         :   0.44 ( 2%) usr   0.04 ( 5%) sys   0.54 ( 2%) wall   61550 kB (11%) ggc
 tree eh               :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall       0 kB ( 0%) ggc
 tree CFG construction :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall    9734 kB ( 2%) ggc
 tree CFG cleanup      :   0.28 ( 1%) usr   0.00 ( 0%) sys   0.18 ( 1%) wall     668 kB ( 0%) ggc
 tree VRP              :   1.21 ( 5%) usr   0.03 ( 4%) sys   1.26 ( 5%) wall   42193 kB ( 8%) ggc
 tree copy propagation :   0.21 ( 1%) usr   0.00 ( 0%) sys   0.24 ( 1%) wall     315 kB ( 0%) ggc
 tree find ref. vars   :   0.07 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall    8937 kB ( 2%) ggc
 tree PTA              :   0.10 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall     758 kB ( 0%) ggc
 tree alias analysis   :   0.12 ( 0%) usr   0.05 ( 6%) sys   0.12 ( 0%) wall      77 kB ( 0%) ggc
 tree call clobbering  :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall      18 kB ( 0%) ggc
 tree flow sensitive alias:   0.02 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall     121 kB ( 0%) ggc
 tree flow insensitive alias:   0.02 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 tree memory partitioning:   0.00 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall      21 kB ( 0%) ggc
 tree PHI insertion    :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall     201 kB ( 0%) ggc
 tree SSA rewrite      :   0.17 ( 1%) usr   0.01 ( 1%) sys   0.13 ( 1%) wall   19668 kB ( 4%) ggc
 tree SSA other        :   0.11 ( 0%) usr   0.03 ( 4%) sys   0.18 ( 1%) wall     360 kB ( 0%) ggc
 tree SSA incremental  :   0.24 ( 1%) usr   0.02 ( 3%) sys   0.25 ( 1%) wall      40 kB ( 0%) ggc
 tree operand scan     :   0.36 ( 1%) usr   0.15 (19%) sys   0.58 ( 2%) wall   27070 kB ( 5%) ggc
 dominator optimization:   0.26 ( 1%) usr   0.00 ( 0%) sys   0.14 ( 1%) wall    2270 kB ( 0%) ggc
 tree SRA              :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 tree CCP              :   0.33 ( 1%) usr   0.01 ( 1%) sys   0.24 ( 1%) wall    4060 kB ( 1%) ggc
 tree reassociation    :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall     124 kB ( 0%) ggc
 tree PRE              :   5.18 (21%) usr   0.05 ( 6%) sys   5.07 (20%) wall   87699 kB (16%) ggc
 tree FRE              :   0.51 ( 2%) usr   0.00 ( 0%) sys   0.55 ( 2%) wall    7664 kB ( 1%) ggc
 tree code sinking     :   0.04 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall      75 kB ( 0%) ggc
 tree linearize phis   :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 tree forward propagate:   0.11 ( 0%) usr   0.02 ( 3%) sys   0.11 ( 0%) wall   11274 kB ( 2%) ggc
 tree phiprop          :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 tree conservative DCE :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall       2 kB ( 0%) ggc
 tree aggressive DCE   :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall       5 kB ( 0%) ggc
 tree DSE              :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.05 ( 0%) wall      65 kB ( 0%) ggc
 tree loop bounds      :   0.23 ( 1%) usr   0.00 ( 0%) sys   0.11 ( 0%) wall    5701 kB ( 1%) ggc
 loop invariant motion :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.15 ( 1%) wall       0 kB ( 0%) ggc
 tree canonical iv     :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall      23 kB ( 0%) ggc
 scev constant prop    :   0.08 ( 0%) usr   0.00 ( 0%) sys   0.12 ( 0%) wall     520 kB ( 0%) ggc
 tree loop unswitching :   0.03 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall       0 kB ( 0%) ggc
 complete unrolling    :   0.09 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall    1121 kB ( 0%) ggc
 tree iv optimization  :   2.30 (10%) usr   0.01 ( 1%) sys   2.33 ( 9%) wall   34677 kB ( 6%) ggc
 predictive commoning  :   0.94 ( 4%) usr   0.02 ( 3%) sys   1.00 ( 4%) wall   42843 kB ( 8%) ggc
 tree loop init        :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall     232 kB ( 0%) ggc
 tree SSA uncprop      :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 tree SSA to normal    :   0.11 ( 0%) usr   0.02 ( 3%) sys   0.22 ( 1%) wall   17790 kB ( 3%) ggc
 tree rename SSA copies:   0.06 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall       0 kB ( 0%) ggc
 dominance frontiers   :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 dominance computation :   0.11 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall       0 kB ( 0%) ggc
 control dependences   :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 expand                :   0.87 ( 4%) usr   0.01 ( 1%) sys   0.89 ( 4%) wall   27910 kB ( 5%) ggc
 forward prop          :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall    3922 kB ( 1%) ggc
 CSE                   :   0.53 ( 2%) usr   0.01 ( 1%) sys   0.65 ( 3%) wall     639 kB ( 0%) ggc
 dead code elimination :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall       0 kB ( 0%) ggc
 dead store elim1      :   0.11 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall    2892 kB ( 1%) ggc
 dead store elim2      :   0.07 ( 0%) usr   0.00 ( 0%) sys   0.11 ( 0%) wall    2730 kB ( 1%) ggc
 loop analysis         :   0.03 ( 0%) usr   0.01 ( 1%) sys   0.04 ( 0%) wall     389 kB ( 0%) ggc
 CPROP 1               :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall     430 kB ( 0%) ggc
 PRE                   :   0.08 ( 0%) usr   0.01 ( 1%) sys   0.11 ( 0%) wall      15 kB ( 0%) ggc
 CPROP 2               :   0.07 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall    1014 kB ( 0%) ggc
 bypass jumps          :   0.07 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall    1039 kB ( 0%) ggc
 web                   :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall      14 kB ( 0%) ggc
 CSE 2                 :   0.20 ( 1%) usr   0.00 ( 0%) sys   0.25 ( 1%) wall     228 kB ( 0%) ggc
 branch prediction     :   0.14 ( 1%) usr   0.00 ( 0%) sys   0.12 ( 0%) wall    4586 kB ( 1%) ggc
 combiner              :   0.79 ( 3%) usr   0.01 ( 1%) sys   0.78 ( 3%) wall   15629 kB ( 3%) ggc
 if-conversion         :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall     140 kB ( 0%) ggc
 regmove               :   0.11 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall       0 kB ( 0%) ggc
 integrated RA         :   0.81 ( 3%) usr   0.00 ( 0%) sys   1.02 ( 4%) wall    2360 kB ( 0%) ggc
 reload                :   0.57 ( 2%) usr   0.00 ( 0%) sys   0.43 ( 2%) wall    2090 kB ( 0%) ggc
 reload CSE regs       :   0.39 ( 2%) usr   0.01 ( 1%) sys   0.43 ( 2%) wall    4804 kB ( 1%) ggc
 load CSE after reload :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.11 ( 0%) wall      31 kB ( 0%) ggc
 thread pro- & epilogue:   0.02 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall     451 kB ( 0%) ggc
 if-conversion 2       :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall      59 kB ( 0%) ggc
 peephole 2            :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall      81 kB ( 0%) ggc
 rename registers      :   0.35 ( 1%) usr   0.00 ( 0%) sys   0.36 ( 1%) wall     379 kB ( 0%) ggc
 scheduling 2          :   0.51 ( 2%) usr   0.00 ( 0%) sys   0.56 ( 2%) wall     342 kB ( 0%) ggc
 machine dep reorg     :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.09 ( 0%) wall       2 kB ( 0%) ggc
 reorder blocks        :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall     129 kB ( 0%) ggc
 final                 :   0.16 ( 1%) usr   0.03 ( 4%) sys   0.23 ( 1%) wall     745 kB ( 0%) ggc
 symout                :   0.04 ( 0%) usr   0.01 ( 1%) sys   0.04 ( 0%) wall    3436 kB ( 1%) ggc
 variable tracking     :   0.09 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall    1278 kB ( 0%) ggc
 TOTAL                 :  24.14             0.77            24.91             537367 kB
/data03/vondele/gcc_4_4_branch/build/lib/gcc/x86_64-unknown-linux-gnu/4.4.2/libgfortranbegin.a(fmain.o): In function `main':
/data03/vondele/gcc_4_4_branch/gcc/libgfortran/fmain.c:21: undefined reference to `MAIN__'
collect2: ld returned 1 exit status

Comment 5 Joost VandeVondele 2010-04-02 09:47:22 UTC

(In reply to comment #3)

cows with cows now (i.e. --enable-checking=release), on an idle machine.

Execution times (seconds)
 garbage collection    :   0.29 ( 0%) usr   0.00 ( 0%) sys   0.31 ( 0%) wall       0 kB ( 0%) ggc
 callgraph construction:   0.11 ( 0%) usr   0.01 ( 1%) sys   0.12 ( 0%) wall    5939 kB ( 1%) ggc
 callgraph optimization:   0.29 ( 0%) usr   0.00 ( 0%) sys   0.25 ( 0%) wall     184 kB ( 0%) ggc
 ipa cp                :   0.10 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall     539 kB ( 0%) ggc
 ipa reference         :   0.07 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall       0 kB ( 0%) ggc
 ipa pure const        :   0.09 ( 0%) usr   0.00 ( 0%) sys   0.14 ( 0%) wall       0 kB ( 0%) ggc
 cfg cleanup           :   0.67 ( 0%) usr   0.00 ( 0%) sys   0.83 ( 0%) wall    3661 kB ( 1%) ggc
 trivially dead code   :   0.21 ( 0%) usr   0.00 ( 0%) sys   0.17 ( 0%) wall       0 kB ( 0%) ggc
 df multiple defs      :   0.35 ( 0%) usr   0.00 ( 0%) sys   0.36 ( 0%) wall       0 kB ( 0%) ggc
 df reaching defs      :   0.69 ( 0%) usr   0.00 ( 0%) sys   0.65 ( 0%) wall       0 kB ( 0%) ggc
 df live regs          :   3.08 ( 1%) usr   0.00 ( 0%) sys   3.07 ( 1%) wall       0 kB ( 0%) ggc
 df live&initialized regs:   1.17 ( 0%) usr   0.00 ( 0%) sys   1.07 ( 0%) wall       0 kB ( 0%) ggc
 df use-def / def-use chains:   0.53 ( 0%) usr   0.00 ( 0%) sys   0.35 ( 0%) wall       0 kB ( 0%) ggc
 df reg dead/unused notes:   2.50 ( 1%) usr   0.00 ( 0%) sys   2.73 ( 1%) wall    9314 kB ( 1%) ggc
 register information  :   1.05 ( 0%) usr   0.00 ( 0%) sys   0.84 ( 0%) wall       0 kB ( 0%) ggc
 alias analysis        :   0.58 ( 0%) usr   0.00 ( 0%) sys   0.61 ( 0%) wall   21770 kB ( 3%) ggc
 alias stmt walking    :   1.29 ( 0%) usr   0.04 ( 4%) sys   1.36 ( 0%) wall       0 kB ( 0%) ggc
 register scan         :   0.09 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall       0 kB ( 0%) ggc
 rebuild jump labels   :   0.21 ( 0%) usr   0.00 ( 0%) sys   0.25 ( 0%) wall       0 kB ( 0%) ggc
 parser                :   1.15 ( 0%) usr   0.12 (11%) sys   1.26 ( 0%) wall   42200 kB ( 6%) ggc
 inline heuristics     :   0.24 ( 0%) usr   0.01 ( 1%) sys   0.24 ( 0%) wall       0 kB ( 0%) ggc
 tree gimplify         :   0.43 ( 0%) usr   0.05 ( 4%) sys   0.47 ( 0%) wall   52375 kB ( 8%) ggc
 tree eh               :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall       0 kB ( 0%) ggc
 tree CFG construction :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall    9418 kB ( 1%) ggc
 tree CFG cleanup      :   0.27 ( 0%) usr   0.00 ( 0%) sys   0.46 ( 0%) wall     418 kB ( 0%) ggc
 tree VRP              :   1.57 ( 1%) usr   0.06 ( 5%) sys   1.60 ( 1%) wall   54731 kB ( 8%) ggc
 tree copy propagation :   0.20 ( 0%) usr   0.00 ( 0%) sys   0.29 ( 0%) wall     237 kB ( 0%) ggc
 tree find ref. vars   :   0.03 ( 0%) usr   0.01 ( 1%) sys   0.10 ( 0%) wall    3774 kB ( 1%) ggc
 tree PTA              :   0.16 ( 0%) usr   0.00 ( 0%) sys   0.08 ( 0%) wall     423 kB ( 0%) ggc
 tree PHI insertion    :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall     315 kB ( 0%) ggc
 tree SSA rewrite      :   0.24 ( 0%) usr   0.02 ( 2%) sys   0.19 ( 0%) wall   20682 kB ( 3%) ggc
 tree SSA other        :   0.10 ( 0%) usr   0.04 ( 4%) sys   0.19 ( 0%) wall     434 kB ( 0%) ggc
 tree SSA incremental  :   0.56 ( 0%) usr   0.02 ( 2%) sys   0.66 ( 0%) wall     438 kB ( 0%) ggc
 tree operand scan     :   0.21 ( 0%) usr   0.20 (18%) sys   0.42 ( 0%) wall   21791 kB ( 3%) ggc
 dominator optimization:   0.35 ( 0%) usr   0.01 ( 1%) sys   0.36 ( 0%) wall    4189 kB ( 1%) ggc
 tree SRA              :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 tree CCP              :   0.49 ( 0%) usr   0.00 ( 0%) sys   0.34 ( 0%) wall    3081 kB ( 0%) ggc
 tree PHI const/copy prop:   0.02 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 0%) wall      22 kB ( 0%) ggc
 tree split crit edges :   0.04 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall    3265 kB ( 0%) ggc
 tree reassociation    :   0.12 ( 0%) usr   0.01 ( 1%) sys   0.11 ( 0%) wall     161 kB ( 0%) ggc
 tree PRE              :   4.88 ( 2%) usr   0.00 ( 0%) sys   4.89 ( 2%) wall   25200 kB ( 4%) ggc
 tree FRE              :   0.65 ( 0%) usr   0.02 ( 2%) sys   0.67 ( 0%) wall    8099 kB ( 1%) ggc
 tree code sinking     :   0.16 ( 0%) usr   0.05 ( 4%) sys   0.17 ( 0%) wall   12275 kB ( 2%) ggc
 tree linearize phis   :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall       0 kB ( 0%) ggc
 tree forward propagate:   0.14 ( 0%) usr   0.00 ( 0%) sys   0.17 ( 0%) wall    9572 kB ( 1%) ggc
 tree phiprop          :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 tree conservative DCE :   0.21 ( 0%) usr   0.03 ( 3%) sys   0.15 ( 0%) wall      17 kB ( 0%) ggc
 tree aggressive DCE   :   0.16 ( 0%) usr   0.00 ( 0%) sys   0.16 ( 0%) wall    2998 kB ( 0%) ggc
 tree DSE              :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall      26 kB ( 0%) ggc
 PHI merge             :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       5 kB ( 0%) ggc
 tree loop bounds      :   0.11 ( 0%) usr   0.00 ( 0%) sys   0.19 ( 0%) wall    6263 kB ( 1%) ggc
 tree loop invariant motion:   0.19 ( 0%) usr   0.01 ( 1%) sys   0.19 ( 0%) wall     497 kB ( 0%) ggc
 tree canonical iv     : 223.30 (75%) usr   0.01 ( 1%) sys 223.28 (75%) wall   21873 kB ( 3%) ggc
 scev constant prop    :   0.05 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall    5809 kB ( 1%) ggc
 tree loop unswitching :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 complete unrolling    :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.06 ( 0%) wall    1123 kB ( 0%) ggc
 tree slp vectorization:   0.38 ( 0%) usr   0.00 ( 0%) sys   0.31 ( 0%) wall   19328 kB ( 3%) ggc
 tree iv optimization  :   0.39 ( 0%) usr   0.00 ( 0%) sys   0.47 ( 0%) wall   13309 kB ( 2%) ggc
 predictive commoning  :   1.13 ( 0%) usr   0.01 ( 1%) sys   1.17 ( 0%) wall   40528 kB ( 6%) ggc
 tree loop init        :   0.13 ( 0%) usr   0.01 ( 1%) sys   0.07 ( 0%) wall    5208 kB ( 1%) ggc
 tree copy headers     :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall     758 kB ( 0%) ggc
 tree SSA uncprop      :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%) wall       0 kB ( 0%) ggc
 tree rename SSA copies:   0.08 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall       0 kB ( 0%) ggc
 dominance frontiers   :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.07 ( 0%) wall       0 kB ( 0%) ggc
 dominance computation :   0.26 ( 0%) usr   0.00 ( 0%) sys   0.21 ( 0%) wall       0 kB ( 0%) ggc
 control dependences   :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall       0 kB ( 0%) ggc
 expand                :   2.97 ( 1%) usr   0.04 ( 4%) sys   3.11 ( 1%) wall   76883 kB (11%) ggc
 lower subreg          :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall       0 kB ( 0%) ggc
 forward prop          :   0.89 ( 0%) usr   0.00 ( 0%) sys   0.85 ( 0%) wall    6749 kB ( 1%) ggc
 CSE                   :   1.46 ( 0%) usr   0.01 ( 1%) sys   1.51 ( 1%) wall    1369 kB ( 0%) ggc
 dead code elimination :   0.45 ( 0%) usr   0.00 ( 0%) sys   0.43 ( 0%) wall       0 kB ( 0%) ggc
 dead store elim1      :   0.60 ( 0%) usr   0.00 ( 0%) sys   0.44 ( 0%) wall    5337 kB ( 1%) ggc
 dead store elim2      :   0.48 ( 0%) usr   0.00 ( 0%) sys   0.42 ( 0%) wall    6072 kB ( 1%) ggc
 loop invariant motion :   0.06 ( 0%) usr   0.00 ( 0%) sys   0.09 ( 0%) wall       1 kB ( 0%) ggc
 loop unswitching      :   0.03 ( 0%) usr   0.00 ( 0%) sys   0.05 ( 0%) wall       0 kB ( 0%) ggc
 loop unrolling        :   0.02 ( 0%) usr   0.00 ( 0%) sys   0.01 ( 0%) wall     190 kB ( 0%) ggc
 CPROP                 :   0.84 ( 0%) usr   0.02 ( 2%) sys   0.81 ( 0%) wall    7746 kB ( 1%) ggc
 PRE                   :   0.22 ( 0%) usr   0.00 ( 0%) sys   0.35 ( 0%) wall     777 kB ( 0%) ggc
 web                   :   0.16 ( 0%) usr   0.00 ( 0%) sys   0.11 ( 0%) wall      16 kB ( 0%) ggc
 CSE 2                 :   1.42 ( 0%) usr   0.00 ( 0%) sys   1.54 ( 1%) wall     793 kB ( 0%) ggc
 branch prediction     :   0.16 ( 0%) usr   0.00 ( 0%) sys   0.10 ( 0%) wall    4053 kB ( 1%) ggc
 combiner              :   2.05 ( 1%) usr   0.02 ( 2%) sys   2.10 ( 1%) wall   26058 kB ( 4%) ggc
 if-conversion         :   0.04 ( 0%) usr   0.00 ( 0%) sys   0.04 ( 0%) wall     130 kB ( 0%) ggc
 regmove               :   0.26 ( 0%) usr   0.00 ( 0%) sys   0.21 ( 0%) wall       4 kB ( 0%) ggc
 integrated RA         :   4.46 ( 1%) usr   0.00 ( 0%) sys   4.24 ( 1%) wall    8905 kB ( 1%) ggc
 reload                :   1.47 ( 0%) usr   0.00 ( 0%) sys   1.55 ( 1%) wall    1737 kB ( 0%) ggc
 reload CSE regs       :   0.73 ( 0%) usr   0.01 ( 1%) sys   0.76 ( 0%) wall    9904 kB ( 1%) ggc
 load CSE after reload :   0.11 ( 0%) usr   0.00 ( 0%) sys   0.12 ( 0%) wall       0 kB ( 0%) ggc
 thread pro- & epilogue:   0.09 ( 0%) usr   0.00 ( 0%) sys   0.15 ( 0%) wall     572 kB ( 0%) ggc
 if-conversion 2       :   0.01 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall      59 kB ( 0%) ggc
 combine stack adjustments:   0.07 ( 0%) usr   0.00 ( 0%) sys   0.03 ( 0%) wall       0 kB ( 0%) ggc
 peephole 2            :   0.22 ( 0%) usr   0.00 ( 0%) sys   0.30 ( 0%) wall    2057 kB ( 0%) ggc
 rename registers      :   0.48 ( 0%) usr   0.00 ( 0%) sys   0.50 ( 0%) wall     701 kB ( 0%) ggc
 hard reg cprop        :   0.29 ( 0%) usr   0.00 ( 0%) sys   0.35 ( 0%) wall      35 kB ( 0%) ggc
 scheduling 2          :   1.42 ( 0%) usr   0.00 ( 0%) sys   1.42 ( 0%) wall     222 kB ( 0%) ggc
 machine dep reorg     :   0.24 ( 0%) usr   0.00 ( 0%) sys   0.23 ( 0%) wall       0 kB ( 0%) ggc
 reorder blocks        :   0.12 ( 0%) usr   0.00 ( 0%) sys   0.09 ( 0%) wall    2144 kB ( 0%) ggc
 final                 :   0.56 ( 0%) usr   0.06 ( 5%) sys   0.76 ( 0%) wall    5904 kB ( 1%) ggc
 symout                :   0.39 ( 0%) usr   0.06 ( 5%) sys   0.44 ( 0%) wall   50781 kB ( 7%) ggc
 variable tracking     :  23.48 ( 8%) usr   0.17 (15%) sys  23.48 ( 8%) wall   38556 kB ( 6%) ggc
 plugin execution      :   0.02 ( 0%) usr   0.01 ( 1%) sys   0.03 ( 0%) wall       0 kB ( 0%) ggc
 TOTAL                 : 298.36             1.14           299.51             690347 kB
COLLECT_GCC_OPTIONS='-v' '-ftime-report' '-fbounds-check' '-g' '-O3' '-ffast-math' '-funroll-loops' '-ftree-vectorize'  '-c'
 as -V -Qy --64 -o hog.o /tmp/cclB9I15.s
GNU assembler version 2.18.50 (x86_64-suse-linux) using BFD version (GNU Binutils; openSUSE 11.0) 2.18.50.20080409-11.1
COMPILER_PATH=/data03/vondele/gcc_trunk/build/libexec/gcc/x86_64-unknown-linux-gnu/4.5.0/:/data03/vondele/gcc_trunk/build/libexec/gcc/x86_64-unknown-linux-gnu/4.5.0/:/data03/vondele/gcc_trunk/build/libexec/gcc/x86_64-unknown-linux-gnu/:/data03/vondele/gcc_trunk/build/lib/gcc/x86_64-unknown-linux-gnu/4.5.0/:/data03/vondele/gcc_trunk/build/lib/gcc/x86_64-unknown-linux-gnu/
LIBRARY_PATH=/data03/vondele/gcc_trunk/build/lib/gcc/x86_64-unknown-linux-gnu/4.5.0/:/data03/vondele/gcc_trunk/build/lib/gcc/x86_64-unknown-linux-gnu/4.5.0/../../../../lib64/:/lib/../lib64/:/usr/lib/../lib64/:/data03/vondele/gcc_trunk/build/lib/gcc/x86_64-unknown-linux-gnu/4.5.0/../../../:/lib/:/usr/lib/
COLLECT_GCC_OPTIONS='-v' '-ftime-report' '-fbounds-check' '-g' '-O3' '-ffast-math' '-funroll-loops' '-ftree-vectorize'  '-c'

Comment 6 Richard Biener 2010-04-02 12:19:05 UTC

The issue is for certain the many manually unrolled loops and possibly the
new autoinc code.

What's your native arch?  I can't reproduce this on a core i?86.

Comment 7 Joost VandeVondele 2010-04-02 12:28:51 UTC

(In reply to comment #6)
> What's your native arch?  I can't reproduce this on a core i?86.

-v output:

 /data03/vondele/gcc_trunk/build/libexec/gcc/x86_64-unknown-linux-gnu/4.5.0/f951 hog.f90 -march=k8-sse3 -mcx16 -msahf --param l1-cache-size=64 --param l1-cache-line-size=64 --param l2-cache-size=1024 -mtune=k8 -quiet -dumpbase hog.f90 -auxbase hog -g -O3 -version -fbounds-check -ffast-math -funroll-loops -ftree-vectorize -fintrinsic-modules-path /data03/vondele/gcc_trunk/build/lib/gcc/x86_64-unknown-linux-gnu/4.5.0/finclude -o /tmp/ccA2YvFn.s

Comment 8 Richard Biener 2010-04-02 14:07:05 UTC

Confirmed on x86_64-linux with -O2 -fbounds-check.

find_loop_niter_by_eval takes a lot of time in each of the ints2bits_*
routines because the loops have a lot of exits (due to -fbounds-check).

Comment 9 Joost VandeVondele 2010-04-02 14:07:23 UTC

Created attachment 20290 [details]
smaller testcase (needs 3s, 80% in tree canonical iv)

Comment 10 Richard Biener 2010-04-02 14:08:31 UTC

Created attachment 20291 [details]
reduced testcase

Comment 11 Richard Biener 2010-04-02 14:13:11 UTC

Compared to 4.4 we no longer eliminate most of the bound checks in 4.5.

Comment 12 Joost VandeVondele 2010-04-02 14:17:17 UTC

(In reply to comment #9)
> Created an attachment (id=20290) [edit]
> smaller testcase (needs 3s, 80% in tree canonical iv)

from valgrind, I see some 13000000 cals to get_val_for / fold_binary_loc, for the small testcase

Comment 13 Richard Biener 2010-04-02 14:23:22 UTC

Testcase for that:

MODULE hfx_compression_core_methods

  IMPLICIT NONE

  INTEGER, PARAMETER :: int_8=8

  CONTAINS

  SUBROUTINE ints2bits_3(Ndata,packed_data,full_data)
    INTEGER, INTENT(IN)                      :: Ndata
    INTEGER(KIND=int_8), INTENT(OUT)         :: packed_data(*)
    INTEGER(KIND=int_8), INTENT(IN)          :: full_data(*)

    INTEGER, PARAMETER                       :: Nbits = 3

    INTEGER                                  :: idata, ipack, kdata, Ndata_rep
    INTEGER(KIND=int_8)                      :: data_tmp, pack_tmp

   idata=0
   ipack=0
   Ndata_rep=(Ndata/2)*2
   DO kdata=1,Ndata_rep,2
   pack_tmp=0
     idata=idata+1
        data_tmp = full_data(idata)
        data_tmp = ISHFT(data_tmp,61)
        pack_tmp = IOR(pack_tmp,data_tmp)
        pack_tmp = ISHFT(pack_tmp,-3)
     idata=idata+1
        data_tmp = full_data(idata)
        data_tmp = ISHFT(data_tmp,61)
        pack_tmp = IOR(pack_tmp,data_tmp)
        pack_tmp = ISHFT(pack_tmp,0)
   pack_tmp = ISHFT(pack_tmp,0)
   ipack = ipack + 1
   packed_data(ipack) = pack_tmp
   ENDDO
  END SUBROUTINE ints2bits_3

END MODULE hfx_compression_core_methods


likely caused by

2010-02-16  Richard Guenther  <rguenther@suse.de>

        PR tree-optimization/41043
        * tree-vrp.c  (vrp_var_may_overflow): Only ask SCEV for real loops.
        (vrp_visit_assignment_or_call): Do not ask SCEV for regular
        statements ...
        (vrp_visit_phi_node): ... but only for loop PHI nodes.

Comment 14 Richard Biener 2010-04-02 14:26:58 UTC

Interestingly it works on i?86 ...

Comment 15 Richard Biener 2010-04-02 14:39:17 UTC

C testcase for the missed VRP, fails with long on x86_64 only, with
long long also on i?86:

extern void link_error (void) __attribute__((noreturn));
int n;
float *x;
int main()
{
  if (n > 0)
    {
      int i = 0;
      do
        {
          long index;
          i = i + 1;
          index = i;
          if (index <= 0)
            link_error ();
          x[index] = 0;
          i = i + 1;
          index = i;
          if (index <= 0)
            link_error ();
          x[index] = 0;
        }
      while (i < n);
    }
}

Comment 16 Richard Biener 2010-04-02 14:53:07 UTC

It's the strict-overflow stuff that cripples VRP again here.  I have a kludge.

Comment 17 Richard Biener 2010-04-02 15:10:32 UTC

Created attachment 20292 [details]
minimal patch

I'm testing this minimal patch.

Comment 18 Richard Biener 2010-04-06 11:21:07 UTC

GCC 4.5.0 is being released.  Deferring to 4.5.1.

Comment 19 Richard Biener 2010-04-06 12:32:52 UTC

Subject: Bug 43627

Author: rguenth
Date: Tue Apr  6 12:32:25 2010
New Revision: 157992

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=157992
Log:
2010-04-06  Richard Guenther  <rguenther@suse.de>

	PR tree-optimization/43627
	* tree-vrp.c (extract_range_from_unary_expr): Widenings
	of [1, +INF(OVF)] go to [1, +INF(OVF)] of the wider type,
	not varying.

	* gcc.dg/tree-ssa/vrp49.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/tree-ssa/vrp49.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-vrp.c

Comment 20 Richard Biener 2010-04-06 12:33:00 UTC

Fixed on trunk sofar.  Queued for 4.5.1.

Comment 21 Richard Biener 2010-04-15 13:47:02 UTC

Subject: Bug 43627

Author: rguenth
Date: Thu Apr 15 13:46:42 2010
New Revision: 158377

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=158377
Log:
2010-04-15  Richard Guenther  <rguenther@suse.de>

	PR tree-optimization/43627
	* tree-vrp.c (extract_range_from_unary_expr): Widenings
	of [1, +INF(OVF)] go to [1, +INF(OVF)] of the wider type,
	not varying.

	* gcc.dg/tree-ssa/vrp49.c: New testcase.

Added:
    branches/gcc-4_5-branch/gcc/testsuite/gcc.dg/tree-ssa/vrp49.c
Modified:
    branches/gcc-4_5-branch/gcc/ChangeLog
    branches/gcc-4_5-branch/gcc/testsuite/ChangeLog
    branches/gcc-4_5-branch/gcc/tree-vrp.c

Comment 22 Richard Biener 2010-04-15 13:47:10 UTC

Fixed.