llvm-mirror/docs/DSGraphStatus.html

653 lines
87 KiB
HTML
Raw Normal View History

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
2002-11-07 02:07:57 +00:00
<html><head><title>Data Structure Graph Analysis Status Page</title></head>
2002-11-07 02:07:57 +00:00
<body><h1>Data Structure Graph Analysis Status Page</h1>
This page records information and the progress the data structure analysis
passes have been making. Times are in seconds, sizes are #bytes allocated for a
2002-11-08 06:50:22 +00:00
particular pass. Runs that are all '*'d out were not completed because they
were taking too long (time listed show how long I waited).<p>
Fields that appear in bold have changes from the last entry on the page.
Floating point entries are only highlighted if they double or half their
previous value.<p>
2002-11-12 06:53:43 +00:00
<hr>
<a name="20021111-2"><h2>Nov 11, 2002 #2:</h2>
This is the result after fixing some major bugs. The globals list was supposed
to be kept sorted, but there was one minor, tiny, little, bug that caused it to
get screwed up sometimes. This caused vortex to explode, fixing it gets things
back in control where you'd expect. Viola.<p>
This also includes a merging bug that caused some things to be merged and some
other things to happen when they shouldn't, which accounts for folding
reductions.<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: BCTime: LocSize: BUSize: TDSize: TotSize: BCSize: NumFold NumNodes main __main GlobGr
254.gap.lib 17.65 0.8186 4.5919 2.3718 9.2483 2.8978 | <b>2024984</b> <b>2984936</b> <b>2025480</b> <b>7035400</b> <b>16554904</b> | <b>6853</b> <b>24032+3950</b> 44+9 0+0 0+0
255.vortex.lib 15.02 0.5766 2.4586 <b>3.5717</b> <b>7.6692</b> 2.2728 | <b>1381056</b> <b>1340216</b> <b>2461232</b> <b>5182504</b> <b>11689312</b> | <b>7127</b> <b>7302+2815</b> 166+11 0+0 0+0
300.twolf.lib 3.23 0.2747 0.2856 0.5057 1.5129 0.9846 | <b>466528</b> <b>391512</b> <b>396552</b> <b>1254592</b> <b>6936728</b> | <b>749</b> <b>4128+784</b> <b>90+53</b> 0+0 0+0
197.parser.lib 1.84 0.1528 0.3458 0.2180 0.8983 0.6469 | <b>408392</b> <b>225312</b> <b>212688</b> <b>965648</b> <b>4447944</b> | <b>1357</b> <b>2257+707</b> 14+6 0+0 0+0
burg.llvm.lib 1.20 0.1017 0.1867 0.1765 0.5525 0.4912 | <b>530600</b> <b>473296</b> <b>185792</b> <b>1189688</b> <b>3040208</b> | <b>756</b> <b>2273+576</b> <b>45+7</b> 0+0 0+0
164.gzip.llvm.lib 0.71 0.0392 0.0649 0.0330 0.2008 0.4263 | <b>145952</b> <b>184776</b> <b>70872</b> <b>499912</b> <b>1876248</b> | 24 971+87 8+2 0+0 0+0
optimizer-eval.llvm.lib 0.42 0.0163 0.0446 0.2063 0.2953 <b>0.0798</b> | <b>101184</b> <b>79456</b> <b>61760</b> <b>242400</b> <b>624688</b> | * 309+41 72+27 0+0 0+0
256.bzip2.lib 0.34 0.0238 0.0491 0.0112 0.1311 0.1543 | <b>139504</b> <b>83400</b> <b>39712</b> <b>335144</b> <b>1373216</b> | * 563+37 8+2 0+0 0+0
181.mcf.lib 0.20 0.0174 0.0267 0.0272 0.0906 0.0744 | <b>115448</b> <b>146520</b> <b>43680</b> <b>322864</b> <b>589440</b> | <b>166</b> <b>388+51</b> <b>22+10</b> 0+0 0+0
sim.llvm.lib 0.19 0.0135 0.0136 0.0130 0.0776 0.0794 | <b>81632</b> <b>51496</b> <b>21872</b> <b>226584</b> <b>734152</b> | * <b>260+35</b> 39+22 0+0 0+0
voronoi.llvm.lib 0.17 0.0137 0.0267 0.0286 0.0829 0.0561 | <b>79824</b> <b>84152</b> <b>27112</b> <b>204904</b> <b>420568</b> | 18 299+77 15+8 0+0 0+0
bh.llvm.lib 0.16 0.0116 0.0161 0.0111 0.0591 0.0675 | <b>75000</b> <b>51080</b> <b>27760</b> <b>217632</b> <b>470720</b> | 61 194+35 6+2 0+0 0+0
sgefa.llvm.lib 0.14 0.0086 0.0201 0.0208 0.0656 0.0517 | <b>85688</b> <b>100456</b> <b>67024</b> <b>277864</b> <b>368680</b> | * 130+24 24+14 0+0 0+0
sumarray2d.llvm.lib <b>0.11</b> 0.0005 0.0009 0.0002 0.0033 <b>0.0922</b> | <b>1568</b> <b>3584</b> <b>848</b> <b>8264</b> <b>23288</b> | * 10+1 7+1 0+0 0+0
health.llvm.lib 0.09 0.0050 0.0082 0.0073 0.0280 0.0419 | <b>30632</b> <b>31432</b> <b>12704</b> <b>89592</b> <b>199888</b> | <b>18</b> <b>97+31</b> <b>12+4</b> 0+0 0+0
em3d.llvm.lib 0.09 0.0058 0.0110 0.0059 0.0304 0.0395 | <b>33040</b> <b>36256</b> <b>13008</b> <b>90632</b> <b>227520</b> | <b>1</b> <b>178+28</b> 12+4 0+0 0+0
power.llvm.lib 0.08 0.0042 0.0067 0.0040 0.0222 0.0365 | <b>27776</b> <b>17688</b> <b>18144</b> <b>71424</b> <b>213760</b> | * 81+18 11+2 0+0 0+0
mst.llvm.lib 0.08 0.0039 0.0072 0.0037 0.0218 0.0306 | <b>18624</b> <b>25576</b> 7040 <b>65808</b> <b>178832</b> | 10 97+15 12+3 0+0 0+0
lists.llvm.lib 0.08 0.0043 0.0137 0.0072 0.0306 0.0283 | <b>24552</b> <b>16920</b> 14440 <b>68024</b> 141480 | 30 67+26 3+1 0+0 0+0
tsp.llvm.lib 0.07 0.0048 0.0066 0.0027 0.0217 0.0327 | <b>18232</b> <b>14336</b> <b>4920</b> <b>51888</b> <b>180744</b> | 21 <b>43+15</b> <b>8+4</b> 0+0 0+0
perimeter.llvm.lib 0.06 0.0031 0.0066 0.0024 0.0194 0.0260 | <b>15016</b> <b>14192</b> <b>3936</b> <b>45912</b> <b>156352</b> | * 35+15 7+4 0+0 0+0
bisort.llvm.lib 0.06 0.0029 0.0056 0.0018 0.0230 0.0229 | <b>13288</b> <b>11408</b> 3712 <b>39448</b> <b>127360</b> | * 35+13 7+4 0+0 0+0
objinst.llvm.lib 0.05 0.0019 0.0048 0.0031 0.0153 0.0174 | <b>11280</b> <b>15784</b> <b>4112</b> <b>45296</b> <b>89856</b> | 34 45+14 16+10 0+0 0+0
methcall.llvm.lib 0.05 0.0017 0.0043 0.0028 0.0135 0.0187 | <b>11008</b> <b>9880</b> <b>3928</b> <b>38624</b> <b>87976</b> | 30 43+12 14+8 0+0 0+0
heapsort.llvm.lib 0.05 0.0009 0.0020 0.0007 0.0074 <b>0.0306</b> | <b>2856</b> <b>6968</b> <b>2008</b> <b>26976</b> <b>63160</b> | * 26+3 7+2 0+0 0+0
hash.llvm.lib 0.05 0.0030 0.0052 0.0040 0.0184 0.0173 | <b>25728</b> <b>16608</b> <b>8208</b> <b>63576</b> <b>151680</b> | * 117+17 14+6 0+0 0+0
sumarraymalloc.llvm.lib 0.04 0.0009 0.0015 0.0009 0.0067 0.0156 | 7040 2672 2048 25872 54352 | * 23+5 7+4 0+0 0+0
random.llvm.lib 0.04 0.0009 0.0018 0.0004 0.0060 0.0223 | 2056 5720 1320 22080 45264 | * 15+2 3+1 0+0 0+0
llubenchmark.llvm.lib 0.04 0.0022 0.0028 0.0015 0.0130 0.0118 | <b>11072</b> <b>11760</b> <b>2624</b> <b>36480</b> <b>124344</b> | * 31+7 15+5 0+0 0+0
ackermann.llvm.lib 0.04 0.0007 0.0015 0.0004 0.0055 0.0230 | 1744 5728 1200 22056 46448 | * 13+2 3+1 0+0 0+0
treeadd.llvm.lib 0.03 0.0015 0.0029 0.0009 0.0090 0.0076 | 3856 7384 2200 29784 72064 | * 25+7 6+3 0+0 0+0
sieve.llvm.lib 0.03 0.0007 0.0013 0.0005 0.0060 0.0140 | 2176 6176 1632 23432 216840 | * 22+2 12+1 0+0 0+0
printargs.llvm.lib 0.03 0.0005 0.0015 0.0002 0.0034 0.0077 | 1240 1048 696 5128 19160 | * 6+1 4+1 0+0 0+0
matrixTranspose.llvm.lib 0.03 0.0005 0.0010 0.0003 0.0032 0.0167 | <b>4272</b> 1144 736 <b>8560</b> <b>68448</b> | * 8+1 5+1 0+0 0+0
matrix.llvm.lib 0.03 0.0013 0.0022 0.0014 0.0092 0.0100 | <b>8808</b> <b>8872</b> <b>3072</b> <b>33144</b> <b>69112</b> | * 38+9 12+8 0+0 0+0
hello.llvm.lib 0.03 0.0004 0.0014 0.0001 0.0067 0.0123 | 752 2952 472 6512 <b>14160</b> | * 2+0 0+0 0+0 0+0
fib2.llvm.lib 0.03 0.0008 0.0015 0.0007 0.0061 <b>0.0135</b> | <b>7960</b> <b>6288</b> <b>1640</b> <b>28408</b> <b>42256</b> | * 21+5 8+3 0+0 0+0
ary3.llvm.lib 0.03 0.0012 0.0020 0.0011 0.0086 0.0084 | <b>10240</b> <b>13104</b> <b>10888</b> <b>45832</b> 73064 | 6 44+5 12+3 0+0 0+0
sumarray.llvm.lib 0.02 0.0003 0.0009 0.0002 0.0028 <b>0.0114</b> | 904 3192 624 4720 15152 | * 6+1 4+1 0+0 0+0
pi.llvm.lib 0.02 0.0008 0.0020 0.0004 0.0050 0.0037 | 5392 1560 1392 8344 36032 | * 17+2 13+2 0+0 0+0
indvars.llvm.lib 0.02 <b>0.0005</b> 0.0009 0.0003 0.0035 0.0035 | <b>4744</b> <b>1072</b> <b>4208</b> <b>12544</b> 31336 | * 13+1 8+1 0+0 0+0
</pre><p>
2002-11-11 20:48:45 +00:00
<hr>
<a name="20021111"><h2>Nov 11, 2002:</h2>
This is the first successful run of all of the benchmarks! This is still
compiled in debug mode, still has lots of tweaks and minor improvements that
could be made, lots of optimizations even, but they will have to wait for a
bit.<p>
Also note that we now include time taken the parse the bytecode file into LLVM
IR, and the space used by the LLVM IR.<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: BCTime: LocSize: BUSize: TDSize: TotSize: BCSize: NumFold NumNodes main __main GlobGr
255.vortex.lib <b>24.44</b> <b>0.5896</b> <b>2.7741</b> <b>11.5906</b> <b>16.5760</b> <b>2.3257</b> | <b>1381816</b> <b>1338688</b> <b>14822760</b> <b>17543264</b> <b>11691568</b> | <b>7288</b> <b>7321+3135</b> <b>166+11</b> <b>0+0</b> <b>0+0</b>
254.gap.lib 19.51 0.8708 4.6071 3.5835 10.6354 2.9075 | 2024600 <b>3155592</b> <b>2323136</b> <b>7503328</b> 16555920 | <b>6934</b> <b>24027+3950</b> <b>44+9</b> 0+0 0+0
300.twolf.lib 3.38 0.2753 0.2930 0.5233 1.5807 0.9966 | 466248 <b>392352</b> <b>398296</b> <b>1256896</b> 6936800 | <b>765</b> <b>4129+784</b> <b>91+53</b> 0+0 0+0
197.parser.lib 1.89 0.1478 0.3243 0.2178 0.8755 0.6667 | 408560 <b>224928</b> <b>213736</b> <b>966696</b> 4451256 | <b>1350</b> <b>2259+711</b> <b>14+6</b> 0+0 0+0
burg.llvm.lib 1.23 0.1039 0.1877 0.1779 0.5589 0.4972 | 530224 <b>471200</b> <b>183504</b> <b>1184928</b> 3039960 | <b>790</b> <b>2239+575</b> <b>42+7</b> 0+0 0+0
164.gzip.llvm.lib 0.71 0.0408 0.0578 0.0332 0.1965 0.4323 | 201664 <b>128768</b> <b>70864</b> <b>498912</b> 1870880 | 24 <b>971+87</b> <b>8+2</b> 0+0 0+0
optimizer-eval.llvm.lib 0.51 0.0166 0.0480 0.2086 0.3041 <b>0.1716</b> | 81584 <b>78352</b> <b>60304</b> <b>238616</b> 620296 | * <b>309+41</b> <b>72+27</b> 0+0 0+0
256.bzip2.lib 0.33 0.0240 0.0383 0.0114 0.1214 0.1565 | 139536 <b>84224</b> <b>39704</b> <b>335680</b> 1370784 | * <b>563+37</b> <b>8+2</b> 0+0 0+0
181.mcf.lib 0.22 0.0184 0.0286 0.0287 0.0959 0.0874 | 111384 <b>145792</b> <b>43360</b> <b>317776</b> 589808 | <b>156</b> <b>393+51</b> <b>23+10</b> 0+0 0+0
sim.llvm.lib 0.20 0.0136 0.0137 0.0130 0.0792 0.0848 | 55376 <b>81000</b> <b>21960</b> <b>253344</b> 721520 | * <b>261+36</b> <b>39+22</b> 0+0 0+0
voronoi.llvm.lib 0.17 0.0135 0.0262 0.0289 0.0833 0.0553 | 79696 <b>87968</b> <b>27128</b> <b>208568</b> 419920 | 18 <b>299+77</b> <b>15+8</b> 0+0 0+0
bh.llvm.lib 0.16 0.0116 0.0164 0.0118 0.0590 0.0665 | 55864 71416 27432 219376 473840 | 61 194+35 6+2 0+0 0+0
sgefa.llvm.lib 0.15 0.0089 0.0222 0.0246 0.0720 0.0530 | 85904 82240 67064 260872 364520 | * 130+24 24+14 0+0 0+0
health.llvm.lib 0.10 0.0050 0.0096 0.0055 0.0275 0.0450 | 29880 27920 7608 73008 199064 | 71 79+29 10+4 0+0 0+0
tsp.llvm.lib 0.08 0.0049 0.0072 0.0027 0.0224 0.0350 | 25704 14472 4872 52216 176600 | 21 42+15 7+4 0+0 0+0
power.llvm.lib 0.08 0.0041 0.0067 0.0037 0.0222 0.0361 | 28512 17760 7848 61960 214128 | * 81+18 11+2 0+0 0+0
em3d.llvm.lib 0.08 0.0060 0.0113 0.0055 0.0312 0.0315 | 24280 32960 12536 86576 224616 | 13 172+25 12+4 0+0 0+0
perimeter.llvm.lib 0.07 0.0031 0.0069 0.0020 0.0196 0.0261 | 14936 14112 3928 45592 155208 | * 35+15 7+4 0+0 0+0
mst.llvm.lib 0.07 0.0039 0.0067 0.0037 0.0216 0.0285 | 27408 16752 7040 66048 178328 | 10 97+15 12+3 0+0 0+0
lists.llvm.lib 0.07 0.0043 0.0128 0.0072 0.0299 0.0276 | 24544 16912 14440 68008 141480 | 30 67+26 3+1 0+0 0+0
hash.llvm.lib 0.06 0.0030 0.0051 0.0054 0.0202 0.0186 | 17936 24808 8216 64232 153552 | * 117+17 14+6 0+0 0+0
methcall.llvm.lib 0.05 0.0017 0.0047 0.0029 0.0142 0.0174 | 10984 9856 9144 44000 87328 | 30 43+12 14+8 0+0 0+0
objinst.llvm.lib 0.04 0.0019 0.0042 0.0033 0.0147 0.0114 | 11072 15328 10536 50496 87064 | 34 45+14 16+10 0+0 0+0
matrix.llvm.lib 0.04 0.0012 0.0021 0.0016 0.0094 <b>0.0164</b> | 8720 8784 3048 36600 68048 | * 38+9 12+8 0+0 0+0
llubenchmark.llvm.lib 0.04 0.0021 0.0027 0.0014 0.0130 0.0117 | 4728 <b>9368</b> <b>2608</b> <b>33440</b> 122712 | * 31+7 15+5 0+0 0+0
bisort.llvm.lib 0.04 0.0029 0.0056 0.0019 0.0160 0.0125 | 13608 11744 3712 45984 126584 | * 35+13 7+4 0+0 0+0
ary3.llvm.lib 0.04 0.0012 0.0020 0.0011 0.0088 0.0138 | 8648 19856 2976 43080 73064 | 6 44+5 12+3 0+0 0+0
ackermann.llvm.lib 0.04 0.0007 0.0017 0.0004 0.0058 <b>0.0226</b> | 1744 5728 1200 22056 46448 | * 13+2 3+1 0+0 0+0
treeadd.llvm.lib 0.03 0.0014 0.0029 0.0009 0.0090 0.0077 | 3856 7384 2200 29784 72064 | * 25+7 6+3 0+0 0+0
sumarraymalloc.llvm.lib 0.03 0.0009 <b>0.0017</b> 0.0009 0.0069 0.0100 | 7040 2672 2048 25872 54352 | * 23+5 7+4 0+0 0+0
sieve.llvm.lib 0.03 0.0007 0.0012 0.0005 0.0060 0.0138 | 2176 6176 1632 23432 216840 | * 22+2 12+1 0+0 0+0
random.llvm.lib 0.03 0.0008 0.0018 0.0004 0.0060 <b>0.0137</b> | 2056 5720 1320 22080 45264 | * 15+2 3+1 0+0 0+0
matrixTranspose.llvm.lib 0.03 0.0005 0.0010 0.0003 0.0033 <b>0.0151</b> | 4248 1144 736 8520 68104 | * 8+1 5+1 0+0 0+0
hello.llvm.lib <b>0.03</b> 0.0004 0.0018 0.0001 <b>0.0064</b> <b>0.0105</b> | 752 2952 472 6512 12264 | * 2+0 0+0 0+0 0+0
heapsort.llvm.lib 0.03 0.0009 0.0020 0.0007 0.0074 0.0080 | 7448 7224 1992 28032 62704 | * 26+3 7+2 0+0 0+0
sumarray2d.llvm.lib 0.02 0.0005 0.0009 0.0002 0.0033 0.0030 | 4208 856 3656 10904 21968 | * 10+1 7+1 0+0 0+0
printargs.llvm.lib 0.02 0.0004 0.0014 0.0002 0.0033 <b>0.0061</b> | 1240 1048 696 5128 19160 | * 6+1 4+1 0+0 0+0
pi.llvm.lib 0.02 0.0007 0.0019 0.0004 0.0050 0.0037 | 5392 1560 1392 8344 36032 | * 17+2 13+2 0+0 0+0
indvars.llvm.lib 0.02 <b>0.0011</b> 0.0010 0.0003 0.0043 0.0035 | 1744 4088 1024 9696 31336 | * 13+1 8+1 0+0 0+0
fib2.llvm.lib 0.02 0.0008 0.0015 0.0007 0.0061 0.0050 | 2416 10376 1648 30088 43792 | * 21+5 8+3 0+0 0+0
sumarray.llvm.lib 0.01 <b>0.0003</b> 0.0009 0.0002 0.0027 0.0022 | 904 3192 624 4720 15152 | * 6+1 4+1 0+0 0+0
</pre><p>
<hr>
<a name="20021110"><h2>Nov 10, 2002:</h2>
This data set has the prototype Tarjan SCC code in to help with cases that were
causing exponential behavior. This gets Gap working, but Vortex is still either
infinite looping or really slow. I'm almost certain there are a couple of bugs
still remaining in the implementation, so these should be regarded as
preliminary numbers.<p>
Here is some trivia for you: Vortex contains a SCC with 38 nodes, and another
with 19. Gap contains SCCs of size 2, 2, 20, 6, 2.<p>
Also note that the bytecode file for GAP is 3x the size of the bytecode file for
twolf.<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: BCTime: LocSize: BUSize: TDSize: TotSize: BCSize: NumFold NumNodes main __main GlobGr
255.vortex.lib <b>7.56</b> * * * * * | * * * * * | * * * * *
254.gap.lib <b>22.53</b> <b>0.8908</b> <b>7.0430</b> <b>3.2424</b> <b>12.8506</b> <b>3.1879</b> | <b>2024600</b> <b>5229888</b> <b>3882656</b> <b>11137144</b> <b>16555920</b> | <b>6959</b> <b>42275+3949</b> <b>59+9</b> <b>0+0</b> <b>0+0</b>
300.twolf.lib 3.92 0.3259 0.3470 0.8163 2.0253 1.0287 | <b>466248</b> <b>405936</b> <b>456456</b> <b>1328640</b> <b>6936800</b> | <b>821</b> <b>4119+784</b> <b>90+53</b> 0+0 0+0
197.parser.lib 2.23 0.1939 <b>0.4020</b> 0.2499 1.0864 0.7768 | 408560 <b>229568</b> <b>220672</b> <b>978272</b> 4451256 | <b>1362</b> <b>2272+704</b> <b>16+6</b> 0+0 0+0
burg.llvm.lib 1.28 0.1085 0.1942 0.1954 0.5911 0.5072 | 530224 <b>477440</b> <b>290376</b> <b>1298040</b> 3039960 | <b>796</b> <b>2248+575</b> <b>39+7</b> 0+0 0+0
164.gzip.llvm.lib 0.67 0.0392 0.0506 0.0326 0.1873 0.4044 | 201664 <b>130312</b> <b>72000</b> <b>501592</b> 1870880 | 24 <b>988+87</b> <b>12+2</b> 0+0 0+0
optimizer-eval.llvm.lib 0.43 0.0172 0.0570 0.2122 0.3200 0.0725 | 81584 <b>78640</b> <b>60384</b> <b>238984</b> 620296 | * <b>299+41</b> <b>62+27</b> 0+0 0+0
256.bzip2.lib 0.34 0.0250 <b>0.0374</b> 0.0117 0.1242 0.1508 | 139536 <b>84856</b> <b>40208</b> <b>336816</b> 1370784 | * <b>569+37</b> <b>12+2</b> 0+0 0+0
181.mcf.lib 0.23 0.0304 0.0296 0.0438 0.1255 0.0685 | 111384 <b>145928</b> <b>38568</b> <b>313120</b> 589808 | <b>161</b> <b>384+51</b> <b>27+10</b> 0+0 0+0
sim.llvm.lib 0.22 0.0138 0.0162 0.0146 0.0850 0.0891 | 55376 <b>81680</b> <b>21784</b> <b>253848</b> 721520 | * <b>259+36</b> <b>37+22</b> 0+0 0+0
voronoi.llvm.lib 0.20 0.0140 0.0311 0.0293 0.0962 0.0547 | 79696 <b>85176</b> <b>27272</b> <b>205920</b> 419920 | 18 <b>294+76</b> <b>17+8</b> 0+0 0+0
sgefa.llvm.lib 0.17 0.0088 0.0250 0.0223 0.0730 0.0615 | 85904 <b>82240</b> <b>67064</b> <b>260872</b> 364520 | * 130+24 24+14 0+0 0+0
bh.llvm.lib 0.15 0.0120 0.0173 0.0126 0.0614 0.0548 | 55864 <b>71416</b> <b>27432</b> <b>219376</b> 473840 | 61 194+35 6+2 0+0 0+0
em3d.llvm.lib 0.09 0.0066 0.0111 0.0088 0.0350 0.0284 | 24280 <b>32960</b> <b>12536</b> <b>86576</b> 224616 | <b>13</b> <b>172+25</b> 12+4 0+0 0+0
health.llvm.lib 0.08 0.0052 0.0098 0.0062 0.0287 0.0330 | 29880 <b>27920</b> <b>7608</b> <b>73008</b> 199064 | 71 79+29 10+4 0+0 0+0
mst.llvm.lib 0.07 0.0048 <b>0.0068</b> 0.0056 0.0266 0.0218 | 27408 <b>16752</b> <b>7040</b> <b>66048</b> 178328 | 10 97+15 12+3 0+0 0+0
tsp.llvm.lib 0.06 0.0049 <b>0.0069</b> 0.0026 0.0222 0.0212 | 25704 <b>14472</b> <b>4872</b> <b>52216</b> 176600 | 21 42+15 7+4 0+0 0+0
power.llvm.lib 0.06 0.0042 <b>0.0065</b> 0.0036 0.0217 0.0214 | 28512 <b>17760</b> <b>7848</b> <b>61960</b> 214128 | * 81+18 11+2 0+0 0+0
lists.llvm.lib 0.06 0.0042 0.0113 0.0071 0.0293 0.0146 | 24544 <b>16912</b> <b>14440</b> <b>68008</b> 141480 | 30 67+26 3+1 0+0 0+0
perimeter.llvm.lib 0.05 0.0030 0.0058 0.0020 0.0186 0.0161 | 14936 <b>14112</b> <b>3928</b> <b>45592</b> 155208 | * 35+15 7+4 0+0 0+0
methcall.llvm.lib 0.05 0.0017 <b>0.0042</b> 0.0028 0.0137 0.0099 | 10984 <b>9856</b> <b>9144</b> <b>44000</b> 87328 | 30 43+12 14+8 0+0 0+0
hash.llvm.lib 0.05 0.0031 0.0050 0.0039 0.0187 0.0163 | 17936 <b>24808</b> <b>8216</b> <b>64232</b> 153552 | * 117+17 14+6 0+0 0+0
bisort.llvm.lib 0.05 0.0048 <b>0.0053</b> 0.0018 0.0188 0.0125 | 13608 <b>11744</b> <b>3712</b> <b>45984</b> 126584 | * 35+13 7+4 0+0 0+0
objinst.llvm.lib 0.04 0.0018 <b>0.0044</b> 0.0032 0.0163 0.0103 | 11072 <b>15328</b> <b>10536</b> <b>50496</b> 87064 | <b>34</b> 45+14 16+10 0+0 0+0
llubenchmark.llvm.lib 0.04 0.0021 <b>0.0022</b> 0.0014 0.0125 0.0124 | 4728 <b>9520</b> <b>2624</b> <b>33608</b> 122712 | * 31+7 15+5 0+0 0+0
treeadd.llvm.lib 0.03 0.0015 <b>0.0024</b> 0.0009 0.0085 0.0077 | 3856 <b>7384</b> <b>2200</b> <b>29784</b> 72064 | * 25+7 6+3 0+0 0+0
sumarraymalloc.llvm.lib 0.03 0.0009 <b>0.0035</b> 0.0009 0.0088 0.0058 | 7040 <b>2672</b> <b>2048</b> <b>25872</b> 54352 | * 23+5 7+4 0+0 0+0
sieve.llvm.lib 0.03 0.0007 <b>0.0011</b> 0.0005 0.0060 0.0144 | 2176 <b>6176</b> <b>1632</b> <b>23432</b> 216840 | * 22+2 12+1 0+0 0+0
random.llvm.lib 0.03 0.0008 <b>0.0016</b> 0.0004 0.0063 0.0050 | 2056 <b>5720</b> <b>1320</b> <b>22080</b> 45264 | * 15+2 3+1 0+0 0+0
matrix.llvm.lib 0.03 0.0012 0.0021 0.0014 0.0093 0.0081 | 8720 <b>8784</b> <b>3048</b> <b>36600</b> 68048 | * 38+9 12+8 0+0 0+0
heapsort.llvm.lib 0.03 0.0010 <b>0.0018</b> 0.0006 0.0084 0.0064 | 7448 <b>7224</b> <b>1992</b> <b>28032</b> 62704 | * 26+3 7+2 0+0 0+0
fib2.llvm.lib 0.03 0.0008 <b>0.0016</b> 0.0007 0.0063 0.0048 | 2416 <b>10376</b> <b>1648</b> <b>30088</b> 43792 | * 21+5 8+3 0+0 0+0
ary3.llvm.lib 0.03 0.0012 0.0019 0.0011 0.0086 0.0073 | 8648 <b>19856</b> <b>2976</b> <b>43080</b> 73064 | 6 44+5 12+3 0+0 0+0
sumarray2d.llvm.lib 0.02 0.0004 <b>0.0008</b> 0.0003 0.0031 0.0029 | 4208 856 3656 10904 21968 | * 10+1 7+1 0+0 0+0
printargs.llvm.lib 0.02 0.0005 <b>0.0012</b> 0.0002 0.0031 0.0024 | 1240 <b>1048</b> <b>696</b> <b>5128</b> 19160 | * 6+1 4+1 0+0 0+0
pi.llvm.lib 0.02 0.0007 <b>0.0015</b> 0.0004 0.0050 0.0035 | 5392 <b>1560</b> <b>1392</b> <b>8344</b> 36032 | * 17+2 13+2 0+0 0+0
matrixTranspose.llvm.lib 0.02 0.0006 <b>0.0011</b> 0.0003 0.0036 0.0051 | 4248 <b>1144</b> <b>736</b> 8520 68104 | * 8+1 5+1 0+0 0+0
indvars.llvm.lib 0.02 0.0005 <b>0.0009</b> 0.0003 0.0035 0.0037 | 1744 4088 <b>1024</b> <b>9696</b> 31336 | * 13+1 8+1 0+0 0+0
ackermann.llvm.lib 0.02 0.0008 <b>0.0014</b> 0.0004 0.0056 0.0060 | 1744 <b>5728</b> <b>1200</b> <b>22056</b> 46448 | * 13+2 3+1 0+0 0+0
sumarray.llvm.lib 0.01 <b>0.0014</b> <b>0.0008</b> 0.0002 0.0036 0.0021 | 904 <b>3192</b> 624 <b>4720</b> 15152 | * 6+1 4+1 0+0 0+0
hello.llvm.lib 0.01 0.0004 <b>0.0015</b> 0.0001 0.0028 0.0019 | 752 2952 472 6512 12264 | * 2+0 0+0 0+0 0+0
</pre><p>
2002-11-10 07:31:38 +00:00
<hr>
<a name="20021109-2"><h2>Nov 9, 2002 #2:</h2>
This has the first part of the globals graph implemented, this simply deletes
the nodes that should be moved to the globals graph, without saving them and
then later merging them back in (The GlobGr size indicates the size of the
globals graph: right now it is empty, this will be fixed in the future). From
this we can get some <b>best case</b> times that the final algorithm will not be
able to match, but it is interesting to see badly how the N<sup>2</sup> case
hurts the TD pass.<p>
Note that I now know that gap and vortex are not working because of an infinite
loop in the algorithm, now that this is identified, hopefully it will be fixed
shortly.<p>
Note that these numbers are still with a debug build. Compiling with
optimizations and without assertions improves run time 3-4x.<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main GlobGr
254.gap.lib 4.92 * * * * | * * * * | * * * * *
255.vortex.lib 2.78 * * * * | * * * * | * * * * *
300.twolf.lib <b>3.50</b> 0.2890 <b>0.3094</b> <b>0.5272</b> <b>1.6337</b> | <b>487976</b> <b>434248</b> <b>418472</b> <b>1340696</b> | <b>794</b> <b>4223+808</b> <b>88+45</b> <b>0+0</b> 0+0
197.parser.lib <b>1.70</b> 0.1517 <b>0.1978</b> <b>0.2152</b> <b>0.7588</b> | <b>408560</b> <b>243688</b> <b>213776</b> <b>985496</b> | <b>1327</b> <b>2257+706</b> <b>14+6</b> <b>0+0</b> 0+0
burg.llvm.lib <b>1.14</b> 0.1056 <b>0.1296</b> <b>0.1720</b> <b>0.5084</b> | <b>530224</b> <b>478880</b> <b>179080</b> <b>1188184</b> | <b>722</b> <b>2167+577</b> <b>30+7</b> <b>0+0</b> 0+0
164.gzip.llvm.lib 0.65 0.0406 <b>0.0256</b> <b>0.0328</b> <b>0.1651</b> | <b>201664</b> <b>129304</b> <b>70864</b> <b>499448</b> | <b>24</b> <b>971+87</b> <b>8+2</b> <b>0+0</b> 0+0
optimizer-eval.llvm.lib 0.40 0.0170 <b>0.0366</b> 0.2099 0.3047 | <b>81584</b> <b>79352</b> <b>62232</b> <b>241544</b> | * <b>309+41</b> <b>72+27</b> <b>0+0</b> 0+0
256.bzip2.lib <b>0.29</b> 0.0249 <b>0.0107</b> <b>0.0112</b> <b>0.0943</b> | <b>139536</b> <b>84752</b> <b>39720</b> <b>336224</b> | <b>*</b> <b>563+37</b> <b>8+2</b> <b>0+0</b> 0+0
181.mcf.lib 0.20 0.0185 0.0247 <b>0.0285</b> 0.1042 | <b>111384</b> <b>145912</b> <b>43544</b> <b>318080</b> | <b>156</b> <b>393+51</b> <b>23+10</b> <b>0+0</b> 0+0
sim.llvm.lib 0.18 0.0143 0.0106 <b>0.0129</b> 0.0765 | <b>55376</b> <b>83072</b> <b>21968</b> <b>255424</b> | * <b>261+36</b> <b>39+22</b> <b>0+0</b> 0+0
voronoi.llvm.lib 0.15 0.0152 0.0203 0.0309 0.0807 | <b>79696</b> <b>87280</b> <b>27336</b> <b>208088</b> | <b>18</b> <b>299+77</b> <b>15+8</b> <b>0+0</b> 0+0
bh.llvm.lib 0.13 0.0120 0.0100 <b>0.0110</b> 0.0506 | <b>55864</b> <b>72480</b> <b>27632</b> <b>220640</b> | <b>61</b> <b>194+35</b> <b>6+2</b> <b>0+0</b> 0+0
sgefa.llvm.lib 0.12 0.0088 <b>0.0165</b> <b>0.0199</b> 0.0616 | <b>85904</b> <b>82032</b> <b>67032</b> <b>260632</b> | * <b>130+24</b> <b>24+14</b> <b>0+0</b> 0+0
health.llvm.lib 0.07 0.0051 0.0065 0.0054 0.0243 | <b>29880</b> <b>31328</b> <b>7816</b> <b>76624</b> | 71 <b>79+29</b> <b>10+4</b> <b>0+0</b> 0+0
em3d.llvm.lib 0.07 0.0059 0.0058 <b>0.0058</b> 0.0254 | <b>24280</b> <b>33360</b> <b>13344</b> <b>87784</b> | <b>10</b> <b>179+27</b> <b>12+4</b> <b>0+0</b> 0+0
power.llvm.lib 0.06 0.0043 0.0029 0.0051 0.0197 | <b>28512</b> <b>17808</b> <b>8008</b> <b>62168</b> | * <b>81+18</b> <b>11+2</b> <b>0+0</b> 0+0
tsp.llvm.lib 0.05 0.0051 0.0031 <b>0.0027</b> 0.0185 | <b>25704</b> <b>14576</b> <b>5040</b> <b>52488</b> | <b>21</b> <b>42+15</b> <b>7+4</b> <b>0+0</b> 0+0
perimeter.llvm.lib 0.05 0.0030 0.0036 0.0020 0.0163 | <b>14936</b> <b>22008</b> <b>4296</b> <b>53856</b> | * <b>35+15</b> <b>7+4</b> <b>0+0</b> 0+0
mst.llvm.lib 0.05 0.0051 0.0033 <b>0.0038</b> 0.0193 | <b>27408</b> <b>16880</b> <b>7216</b> <b>66352</b> | 10 <b>97+15</b> <b>12+3</b> <b>0+0</b> 0+0
lists.llvm.lib 0.05 0.0042 0.0058 0.0071 0.0227 | <b>24544</b> <b>17384</b> <b>13696</b> <b>67736</b> | <b>30</b> <b>67+26</b> <b>3+1</b> <b>0+0</b> 0+0
hash.llvm.lib 0.05 0.0030 0.0033 0.0041 0.0168 | <b>17936</b> <b>25000</b> <b>8544</b> <b>64752</b> | * <b>117+17</b> <b>14+6</b> <b>0+0</b> 0+0
objinst.llvm.lib 0.04 0.0018 0.0018 0.0033 0.0122 | <b>11072</b> <b>16344</b> <b>12040</b> <b>53016</b> | <b>35</b> <b>45+14</b> <b>16+10</b> <b>0+0</b> 0+0
methcall.llvm.lib 0.04 0.0017 0.0015 0.0029 0.0111 | <b>10984</b> <b>9816</b> <b>11880</b> <b>46696</b> | <b>30</b> <b>43+12</b> <b>14+8</b> <b>0+0</b> 0+0
llubenchmark.llvm.lib 0.04 0.0021 <b>0.0010</b> <b>0.0015</b> 0.0115 | <b>4728</b> <b>9384</b> <b>2808</b> <b>33656</b> | * <b>31+7</b> <b>15+5</b> <b>0+0</b> 0+0
bisort.llvm.lib 0.04 0.0029 0.0022 <b>0.0019</b> 0.0128 | <b>13608</b> <b>11840</b> <b>3888</b> <b>46256</b> | * <b>35+13</b> <b>7+4</b> <b>0+0</b> 0+0
treeadd.llvm.lib 0.03 0.0014 0.0009 0.0010 0.0071 | <b>3856</b> <b>12496</b> <b>2392</b> <b>35088</b> | * <b>25+7</b> <b>6+3</b> <b>0+0</b> 0+0
sumarraymalloc.llvm.lib 0.03 0.0009 0.0008 0.0010 0.0063 | <b>7040</b> <b>2192</b> <b>2656</b> <b>26000</b> | * <b>23+5</b> <b>7+4</b> <b>0+0</b> 0+0
sieve.llvm.lib 0.03 0.0007 0.0005 0.0006 0.0063 | <b>2176</b> <b>5832</b> <b>1968</b> <b>23424</b> | * <b>22+2</b> <b>12+1</b> <b>0+0</b> 0+0
matrixTranspose.llvm.lib <b>0.03</b> 0.0005 0.0002 0.0003 0.0026 | <b>4248</b> <b>792</b> <b>1088</b> <b>8520</b> | * <b>8+1</b> <b>5+1</b> <b>0+0</b> 0+0
matrix.llvm.lib 0.03 0.0013 0.0013 0.0015 0.0085 | <b>8720</b> <b>8704</b> <b>3416</b> <b>36888</b> | * <b>38+9</b> <b>12+8</b> <b>0+0</b> 0+0
ary3.llvm.lib 0.03 0.0012 0.0012 0.0012 0.0078 | <b>8648</b> <b>14136</b> <b>3328</b> <b>37712</b> | 6 <b>44+5</b> <b>12+3</b> <b>0+0</b> 0+0
random.llvm.lib 0.02 0.0008 0.0004 0.0005 0.0047 | <b>2056</b> <b>4720</b> <b>1656</b> <b>21416</b> | * <b>15+2</b> <b>3+1</b> <b>0+0</b> 0+0
pi.llvm.lib 0.02 0.0007 0.0003 <b>0.0005</b> 0.0036 | <b>5392</b> <b>1384</b> <b>4768</b> <b>11544</b> | * <b>17+2</b> <b>13+2</b> <b>0+0</b> 0+0
hello.llvm.lib 0.02 0.0004 <b>0.0001</b> <b>0.0001</b> 0.0015 | <b>752</b> <b>2952</b> <b>472</b> <b>6512</b> | * <b>2+0</b> <b>0+0</b> <b>0+0</b> 0+0
heapsort.llvm.lib 0.02 0.0009 0.0006 0.0007 0.0062 | <b>7448</b> <b>6864</b> <b>2360</b> <b>28040</b> | * <b>26+3</b> <b>7+2</b> <b>0+0</b> 0+0
fib2.llvm.lib 0.02 0.0008 0.0007 0.0007 0.0055 | <b>2416</b> <b>5680</b> <b>2000</b> <b>25744</b> | * <b>21+5</b> <b>8+3</b> <b>0+0</b> 0+0
ackermann.llvm.lib <b>0.02</b> <b>0.0007</b> <b>0.0004</b> <b>0.0005</b> <b>0.0046</b> | <b>1744</b> <b>5368</b> <b>1552</b> <b>22048</b> | * <b>13+2</b> <b>3+1</b> <b>0+0</b> 0+0
sumarray.llvm.lib 0.01 0.0003 0.0002 0.0002 0.0021 | <b>904</b> <b>2232</b> <b>624</b> <b>3760</b> | * <b>6+1</b> <b>4+1</b> 0+0 0+0
sumarray2d.llvm.lib 0.01 0.0004 0.0002 0.0003 0.0026 | <b>4208</b> <b>856</b> <b>3656</b> <b>10904</b> | * <b>10+1</b> <b>7+1</b> <b>0+0</b> 0+0
printargs.llvm.lib 0.01 0.0004 0.0002 0.0003 0.0023 | <b>1240</b> <b>704</b> <b>1064</b> <b>5152</b> | * <b>6+1</b> <b>4+1</b> <b>0+0</b> 0+0
indvars.llvm.lib 0.01 0.0005 0.0003 0.0003 0.0030 | <b>1744</b> <b>4088</b> <b>1032</b> <b>9704</b> | * <b>13+1</b> <b>8+1</b> <b>0+0</b> 0+0
</pre><p>
2002-11-10 06:55:29 +00:00
<hr>
<a name="20021109-1"><h2>Nov 9, 2002:</h2>
Same as the <a href="#20021108-3">previous run</a>, but now with a small tweak:
constants are never put into the scalar map for a function. This is in
preparation for the global graph work, but it does speed analysis up
marginally:<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
254.gap.lib 2.16 * * * * | * * * * | * * * *
255.vortex.lib 1.11 * * * * | * * * * | * * * *
300.twolf.lib 11.63 0.2468 0.7464 7.5333 9.2461 | <b>595576</b> <b>795896</b> <b>10397040</b> <b>11788512</b> | <b>3577</b> <b>161901+1222</b> 854+0 854+0
197.parser.lib 10.97 0.1258 0.4903 8.6508 9.5864 | <b>464920</b> <b>621904</b> <b>12987664</b> <b>14309904</b> | <b>2643</b> <b>197546+1185</b> 601+0 601+0
burg.llvm.lib 5.18 0.0904 0.2948 3.7480 4.3338 | <b>576064</b> <b>893240</b> <b>4821880</b> <b>6356592</b> | <b>1557</b> <b>72659+919</b> 436+0 436+0
164.gzip.llvm.lib 1.28 0.0355 0.1037 0.5037 0.7448 | <b>228456</b> <b>511192</b> <b>1971048</b> <b>2809016</b> | <b>70</b> <b>16642+233</b> 171+0 171+0
256.bzip2.lib 0.69 0.0201 0.0535 0.3301 0.4626 | <b>163016</b> <b>284360</b> <b>1450312</b> <b>1970944</b> | 20 <b>11464+202</b> 148+0 148+0
optimizer-eval.llvm.lib 0.57 0.0195 0.0846 0.3000 0.4423 | <b>86832</b> <b>180296</b> <b>702840</b> <b>988152</b> | * 3693+101 63+0 63+0
sim.llvm.lib 0.26 0.0137 0.0195 0.0619 0.1439 | <b>92280</b> <b>106632</b> <b>330888</b> <b>602464</b> | * <b>2738+67</b> 99+0 99+0
181.mcf.lib 0.23 0.0164 0.0294 0.0572 0.1265 | <b>133872</b> <b>128088</b> <b>229576</b> <b>508384</b> | <b>180</b> <b>1923+73</b> 44+0 44+0
voronoi.llvm.lib 0.19 0.0119 0.0245 0.0606 0.1138 | <b>83592</b> <b>93192</b> <b>211920</b> <b>402648</b> | 17 <b>1734+106</b> 30+0 30+0
sgefa.llvm.lib 0.17 0.0071 0.0333 0.0406 0.0994 | <b>88880</b> <b>102320</b> <b>133176</b> <b>349144</b> | * 634+55 31+0 31+0
bh.llvm.lib 0.15 0.0102 0.0154 0.0239 0.0697 | <b>77528</b> <b>80672</b> <b>110160</b> <b>316400</b> | 66 <b>716+64</b> 13+0 13+0
em3d.llvm.lib 0.09 0.0045 0.0105 0.0229 0.0477 | <b>37816</b> <b>55256</b> <b>123168</b> <b>232944</b> | <b>8</b> <b>1032+59</b> 28+0 28+0
mst.llvm.lib 0.07 0.0033 0.0056 0.0115 0.0310 | <b>30744</b> <b>21600</b> <b>62072</b> <b>129088</b> | 10 <b>471+36</b> 21+0 21+0
health.llvm.lib 0.07 0.0042 0.0080 0.0105 0.0313 | <b>31520</b> <b>31784</b> <b>45624</b> <b>123896</b> | 71 <b>362+40</b> 14+0 14+0
tsp.llvm.lib 0.06 0.0046 0.0047 0.0060 0.0240 | <b>19552</b> <b>26040</b> <b>25800</b> <b>85696</b> | 19 <b>248+31</b> 10+0 10+0
power.llvm.lib 0.06 0.0034 0.0050 0.0086 0.0254 | <b>30904</b> <b>31232</b> <b>29704</b> <b>99512</b> | * 277+35 9+0 9+0
perimeter.llvm.lib 0.05 0.0025 0.0044 0.0037 0.0195 | <b>15760</b> <b>24296</b> <b>8552</b> <b>61448</b> | * <b>106+24</b> 4+0 4+0
lists.llvm.lib 0.05 0.0037 0.0075 0.0105 0.0283 | <b>25912</b> <b>29360</b> <b>28352</b> <b>101472</b> | 50 <b>252+34</b> 10+0 10+0
hash.llvm.lib 0.05 0.0025 0.0040 0.0059 0.0194 | <b>18464</b> <b>25408</b> <b>24192</b> <b>81464</b> | * <b>242+23</b> 7+0 7+0
llubenchmark.llvm.lib 0.04 0.0017 0.0022 0.0038 0.0198 | <b>12040</b> <b>13368</b> <b>17600</b> <b>54016</b> | * <b>145+20</b> 15+0 15+0
bisort.llvm.lib 0.04 0.0024 0.0035 0.0051 0.0176 | <b>14648</b> <b>28160</b> <b>22256</b> <b>76136</b> | * <b>211+27</b> 10+0 10+0
treeadd.llvm.lib 0.03 0.0011 0.0015 0.0020 0.0090 | <b>9320</b> <b>8552</b> <b>10120</b> <b>44152</b> | * <b>77+14</b> 5+0 5+0
sieve.llvm.lib 0.03 0.0006 0.0006 0.0008 0.0063 | <b>8032</b> <b>1720</b> <b>2552</b> <b>22672</b> | * <b>31+4</b> 2+0 2+0
objinst.llvm.lib 0.03 0.0015 0.0021 <b>0.0036</b> 0.0129 | <b>11272</b> <b>19896</b> <b>6096</b> <b>50920</b> | 33 <b>74+17</b> 3+0 3+0
methcall.llvm.lib 0.03 0.0014 0.0018 0.0028 0.0111 | <b>10936</b> <b>15720</b> <b>5152</b> <b>45064</b> | 26 <b>59+14</b> 2+0 2+0
matrix.llvm.lib 0.03 0.0014 0.0023 0.0025 0.0118 | <b>8984</b> <b>8096</b> <b>4352</b> <b>37776</b> | * <b>47+11</b> 1+0 1+0
ary3.llvm.lib 0.03 0.0010 0.0014 0.0014 0.0084 | <b>8992</b> <b>8624</b> <b>4344</b> <b>37624</b> | 6 <b>52+7</b> 1+0 1+0
sumarraymalloc.llvm.lib 0.02 0.0007 0.0009 0.0012 0.0069 | <b>2936</b> <b>6984</b> <b>7216</b> <b>31496</b> | * <b>39+8</b> 2+0 2+0
random.llvm.lib 0.02 0.0006 0.0006 0.0008 0.0057 | <b>2336</b> <b>6000</b> <b>2856</b> <b>26944</b> | * <b>32+5</b> 3+0 3+0
pi.llvm.lib 0.02 0.0006 0.0006 0.0013 0.0047 | <b>5816</b> <b>7240</b> <b>6688</b> <b>19744</b> | * 34+6 3+0 3+0
heapsort.llvm.lib 0.02 0.0007 0.0009 0.0012 0.0070 | <b>3128</b> <b>7336</b> <b>4008</b> <b>29640</b> | * <b>46+6</b> 3+0 3+0
fib2.llvm.lib 0.02 0.0007 0.0008 0.0008 0.0056 | <b>5704</b> <b>1928</b> <b>2520</b> <b>22440</b> | * <b>26+6</b> 1+0 1+0
sumarray.llvm.lib 0.01 0.0003 0.0002 0.0003 0.0022 | <b>960</b> <b>3176</b> <b>608</b> <b>6856</b> | * 7+2 0+0 0+0
sumarray2d.llvm.lib 0.01 0.0003 0.0003 0.0004 0.0028 | <b>1736</b> <b>6056</b> <b>1064</b> <b>11176</b> | * 15+3 1+0 1+0
printargs.llvm.lib 0.01 0.0003 0.0004 0.0005 0.0026 | <b>1464</b> <b>848</b> <b>3512</b> <b>5824</b> | * 13+4 1+0 1+0
matrixTranspose.llvm.lib 0.01 0.0004 0.0004 0.0005 0.0030 | <b>4384</b> <b>1248</b> <b>3824</b> <b>9456</b> | * 17+4 3+0 3+0
indvars.llvm.lib 0.01 0.0004 0.0005 0.0005 0.0033 | <b>4800</b> <b>7440</b> <b>1224</b> <b>13464</b> | * 18+3 1+0 1+0
hello.llvm.lib 0.01 0.0003 0.0003 0.0003 0.0021 | <b>976</b> <b>3024</b> <b>2968</b> <b>9224</b> | * 8+3 1+0 1+0
ackermann.llvm.lib 0.00 0.0000 0.0000 0.0001 0.0004 | <b>72</b> 0 832 <b>904</b> | * 0+0 * *
</pre><p>
<hr>
<a name="20021108-3"><h2>Nov 8, 2002 #3:</h2>
Same as the <a href="#20021108-2">previous run</a>, but now we automatically
collapse arrays down to a single element in the representation. The graphs
produced should be exactly the same, it's just memory consumption that goes
dramatically down, for example: 197.parser (64M-&gt;14M), 164.gzip
(14M-&gt;2.7M), 256.bzip (9M -&gt; 1.9M), etc.<p>
With reduced memory consumption comes higher performance of course (better cache
behavior n stuff).<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
254.gap.lib 71.65 * * * * | * * * * | * * * *
255.vortex.lib <b>4.62</b> * * * * | * * * * | * * * *
300.twolf.lib 12.22 0.2369 0.8840 8.0785 9.9177 | <b>573880</b> <b>770752</b> <b>10364040</b> <b>11708672</b> | <b>3514</b> 161528+1101 854+0 854+0
197.parser.lib <b>11.73</b> 0.1216 <b>0.5217</b> <b>9.3368</b> <b>10.3099</b> | <b>461720</b> <b>619960</b> <b>12948496</b> <b>14266336</b> | <b>2641</b> 196932+1184 601+0 601+0
burg.llvm.lib 5.35 0.0857 0.2873 3.9878 4.5484 | <b>650608</b> <b>906016</b> <b>4691952</b> <b>6248576</b> | <b>1555</b> 72533+902 436+0 436+0
164.gzip.llvm.lib <b>1.33</b> <b>0.0309</b> <b>0.1085</b> <b>0.5445</b> <b>0.7708</b> | <b>284312</b> <b>521048</b> <b>1877416</b> <b>2779368</b> | <b>67</b> 16632+232 171+0 171+0
256.bzip2.lib <b>0.73</b> 0.0191 <b>0.0588</b> <b>0.3611</b> <b>0.4995</b> | <b>156760</b> <b>336272</b> <b>1390240</b> <b>1956528</b> | <b>20</b> 11456+201 148+0 148+0
optimizer-eval.llvm.lib 0.56 0.0129 0.0957 0.2881 0.4359 | <b>85720</b> <b>178016</b> <b>750424</b> <b>1032352</b> | <b>*</b> 3693+101 63+0 63+0
sim.llvm.lib <b>0.29</b> 0.0126 <b>0.0237</b> <b>0.0731</b> <b>0.1630</b> | <b>93240</b> <b>106144</b> <b>331688</b> <b>626088</b> | <b>*</b> 2739+67 99+0 99+0
181.mcf.lib 0.28 0.0153 0.0379 <b>0.0649</b> 0.1457 | <b>136744</b> <b>155416</b> <b>202496</b> <b>511544</b> | <b>178</b> 1914+72 44+0 44+0
voronoi.llvm.lib 0.22 0.0122 0.0296 0.0728 0.1324 | <b>83568</b> <b>94008</b> <b>212560</b> <b>404128</b> | <b>17</b> 1733+106 30+0 30+0
sgefa.llvm.lib 0.19 0.0072 0.0375 0.0491 0.1131 | <b>102144</b> <b>109744</b> <b>117064</b> <b>353720</b> | <b>*</b> 634+55 31+0 31+0
bh.llvm.lib 0.17 0.0098 0.0164 0.0260 0.0728 | <b>78048</b> <b>82616</b> <b>86688</b> <b>311168</b> | 66 715+64 13+0 13+0
em3d.llvm.lib 0.11 0.0046 0.0114 0.0327 0.0590 | <b>36344</b> <b>64760</b> <b>125360</b> <b>243088</b> | <b>9</b> 1031+59 28+0 28+0
power.llvm.lib 0.09 0.0043 0.0053 0.0100 0.0296 | <b>22840</b> <b>33912</b> <b>30440</b> <b>103016</b> | <b>*</b> 277+35 9+0 9+0
tsp.llvm.lib 0.08 0.0043 0.0057 0.0066 0.0276 | <b>20640</b> <b>25416</b> <b>25784</b> <b>86464</b> | <b>19</b> 247+31 10+0 10+0
mst.llvm.lib 0.08 0.0033 0.0060 0.0123 0.0308 | <b>29088</b> 31400 <b>51632</b> <b>126792</b> | <b>10</b> 470+36 21+0 21+0
lists.llvm.lib 0.08 <b>0.0035</b> 0.0080 0.0111 0.0294 | <b>25048</b> <b>33872</b> <b>25520</b> <b>96760</b> | <b>50</b> 246+34 10+0 10+0
health.llvm.lib 0.08 <b>0.0043</b> 0.0083 0.0120 0.0344 | 31232 <b>31256</b> <b>45008</b> <b>122464</b> | <b>71</b> 361+40 14+0 14+0
perimeter.llvm.lib 0.07 0.0036 0.0047 0.0049 0.0251 | 15336 22904 <b>15968</b> <b>66864</b> | * 105+24 4+0 4+0
objinst.llvm.lib 0.06 0.0016 0.0024 <b>0.0090</b> 0.0191 | 11032 <b>15560</b> <b>12416</b> <b>52664</b> | <b>33</b> 73+17 3+0 3+0
bisort.llvm.lib 0.06 0.0023 0.0036 0.0065 0.0202 | 14376 <b>20968</b> <b>22152</b> <b>68568</b> | <b>*</b> 210+27 10+0 10+0
llubenchmark.llvm.lib 0.05 0.0018 0.0024 0.0042 0.0166 | 11824 <b>15464</b> <b>23480</b> <b>61776</b> | <b>*</b> 144+20 15+0 15+0
hash.llvm.lib 0.05 0.0026 0.0044 0.0062 0.0206 | <b>18192</b> <b>24928</b> <b>23976</b> <b>80496</b> | <b>*</b> 241+23 7+0 7+0
methcall.llvm.lib 0.04 0.0014 0.0019 0.0031 0.0120 | 10824 9704 11128 45312 | <b>26</b> 58+14 2+0 2+0
ary3.llvm.lib 0.04 0.0010 0.0015 0.0014 0.0102 | 8752 17136 4168 45720 | 6 50+7 1+0 1+0
treeadd.llvm.lib 0.03 0.0011 0.0016 0.0023 0.0098 | 9080 8408 <b>9944</b> <b>39424</b> | <b>*</b> 76+14 5+0 5+0
sieve.llvm.lib 0.03 0.0006 <b>0.0006</b> <b>0.0009</b> 0.0063 | <b>7856</b> <b>6056</b> <b>2480</b> <b>26760</b> | * 30+4 2+0 2+0
pi.llvm.lib 0.03 0.0006 0.0007 0.0012 0.0050 | 5696 5272 <b>5584</b> <b>16552</b> | * 34+6 3+0 3+0
matrix.llvm.lib 0.03 0.0010 0.0017 0.0018 0.0096 | 8584 17680 4208 46232 | * 46+11 1+0 1+0
heapsort.llvm.lib 0.03 0.0010 0.0009 0.0013 0.0076 | 2936 7224 3912 29240 | * 45+6 3+0 3+0
sumarraymalloc.llvm.lib 0.02 0.0007 0.0009 0.0015 0.0072 | 2720 <b>6872</b> <b>7128</b> <b>31152</b> | * 38+8 2+0 2+0
sumarray.llvm.lib 0.02 0.0003 0.0002 0.0003 0.0025 | <b>864</b> <b>3144</b> <b>592</b> <b>6712</b> | <b>*</b> 7+2 0+0 0+0
random.llvm.lib 0.02 0.0006 0.0006 0.0009 0.0057 | 6184 1808 2752 23600 | * 31+5 3+0 3+0
printargs.llvm.lib 0.02 0.0003 0.0004 0.0006 0.0028 | 1360 816 <b>3480</b> <b>5656</b> | * 13+4 1+0 1+0
matrixTranspose.llvm.lib 0.02 0.0004 <b>0.0004</b> <b>0.0006</b> <b>0.0034</b> | <b>5888</b> <b>1224</b> <b>4864</b> <b>11976</b> | * 17+4 3+0 3+0
indvars.llvm.lib <b>0.02</b> <b>0.0005</b> <b>0.0005</b> <b>0.0005</b> <b>0.0037</b> | <b>4680</b> <b>3552</b> <b>1200</b> <b>9432</b> | * 18+3 1+0 1+0
fib2.llvm.lib 0.02 0.0007 0.0008 0.0009 0.0059 | 6176 1808 2400 23008 | * 25+6 1+0 1+0
sumarray2d.llvm.lib <b>0.01</b> <b>0.0004</b> <b>0.0004</b> <b>0.0004</b> <b>0.0030</b> | <b>1648</b> <b>6024</b> <b>4144</b> <b>14136</b> | * 15+3 1+0 1+0
hello.llvm.lib 0.01 0.0003 0.0003 0.0003 0.0022 | 880 592 2944 6672 | * 8+3 1+0 1+0
ackermann.llvm.lib <b>0.00</b> 0.0000 0.0000 0.0002 0.0004 | 0 0 832 832 | * 0+0 * *
</pre><p>
<hr>
<a name="20021108-2"><h2>Nov 8, 2002 #2:</h2>
Same as the <a href="#20021108">previous run</a>, but with a huge bug fix:
before, all of the call nodes (from the local pass) were being passed up in the
BU pass, causing graph exposions. Now these call sites are not propogated. The
times and sized of graphs are all much smaller now.<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
254.gap.lib <b>130.47</b> * * * * | * * * * | * * * *
255.vortex.lib <b>86.17</b> * * * * | * * * * | * * * *
197.parser.lib <b>26.50</b> 0.1601 <b>1.4931</b> <b>22.7485</b> <b>24.7647</b> | <b>2187656</b> <b>4037056</b> <b>57704672</b> <b>64165544</b> | <b>4550</b> <b>196932+1184</b> 601+0 601+0
300.twolf.lib <b>14.33</b> 0.2311 1.0000 <b>10.1733</b> <b>12.1591</b> | 629008 <b>1085616</b> <b>15564712</b> <b>17279336</b> | <b>4655</b> <b>161528+1101</b> 854+0 854+0
burg.llvm.lib <b>6.71</b> 0.0904 <b>0.3799</b> <b>5.1871</b> <b>5.8809</b> | <b>745152</b> <b>1254280</b> <b>6822216</b> <b>8821648</b> | <b>3253</b> <b>72533+902</b> <b>436+0</b> <b>436+0</b>
164.gzip.llvm.lib <b>6.07</b> 0.1158 0.9992 <b>4.1925</b> <b>5.3966</b> | <b>2113120</b> <b>2772872</b> <b>9043760</b> <b>14026344</b> | <b>233</b> <b>16632+232</b> 171+0 171+0
256.bzip2.lib <b>2.16</b> 0.0239 0.1987 <b>1.6047</b> <b>1.8899</b> | 372776 <b>1059888</b> <b>7503640</b> <b>9009560</b> | <b>132</b> <b>11456+201</b> 148+0 148+0
sim.llvm.lib 1.54 0.0163 0.2356 1.0723 1.3738 | <b>106096</b> <b>124280</b> <b>400760</b> <b>726152</b> | <b>41</b> <b>2739+67</b> 99+0 99+0
optimizer-eval.llvm.lib <b>0.58</b> 0.0126 <b>0.0890</b> <b>0.3282</b> <b>0.4696</b> | <b>110464</b> <b>188584</b> <b>1055832</b> <b>1373072</b> | <b>287</b> <b>3693+101</b> 63+0 63+0
181.mcf.lib 0.33 0.0155 0.0465 <b>0.1437</b> <b>0.2308</b> | <b>156864</b> <b>225184</b> <b>805584</b> <b>1204520</b> | <b>209</b> <b>1914+72</b> 44+0 44+0
voronoi.llvm.lib <b>0.19</b> 0.0111 0.0253 <b>0.0644</b> <b>0.1178</b> | <b>83248</b> <b>93616</b> <b>238648</b> <b>429504</b> | <b>31</b> <b>1733+106</b> 30+0 30+0
sgefa.llvm.lib 0.18 0.0069 0.0398 0.0492 0.1156 | 104712 <b>108472</b> <b>178920</b> <b>416872</b> | <b>67</b> <b>634+55</b> 31+0 31+0
bh.llvm.lib <b>0.16</b> 0.0100 0.0177 <b>0.0355</b> <b>0.0838</b> | <b>81376</b> <b>93656</b> <b>190104</b> <b>428952</b> | 66 <b>715+64</b> 13+0 13+0
health.llvm.lib 0.10 <b>0.0230</b> 0.0084 0.0138 0.0551 | 31232 <b>31552</b> <b>56984</b> <b>134736</b> | <b>75</b> <b>361+40</b> 14+0 14+0
em3d.llvm.lib 0.09 0.0045 0.0117 <b>0.0255</b> 0.0519 | <b>37336</b> <b>67808</b> <b>126960</b> <b>248728</b> | <b>50</b> <b>1031+59</b> <b>28+0</b> <b>28+0</b>
power.llvm.lib 0.07 0.0035 0.0053 <b>0.0105</b> <b>0.0287</b> | <b>23432</b> <b>31680</b> <b>42680</b> <b>113616</b> | <b>3</b> <b>277+35</b> 9+0 9+0
mst.llvm.lib 0.07 0.0033 0.0060 <b>0.0145</b> 0.0333 | <b>29080</b> <b>31400</b> <b>56720</b> <b>131872</b> | <b>35</b> <b>470+36</b> 21+0 21+0
lists.llvm.lib 0.07 <b>0.0079</b> 0.0129 0.0170 0.0453 | <b>27616</b> <b>39648</b> <b>75280</b> <b>154864</b> | <b>54</b> <b>246+34</b> 10+0 10+0
indvars.llvm.lib 0.07 0.0020 0.0257 0.0269 0.0569 | <b>85496</b> <b>87560</b> 161216 <b>334272</b> | * 18+3 1+0 1+0
tsp.llvm.lib 0.06 0.0041 0.0049 <b>0.0065</b> 0.0250 | <b>19360</b> <b>26216</b> <b>26712</b> <b>86912</b> | <b>23</b> <b>247+31</b> 10+0 10+0
perimeter.llvm.lib 0.05 0.0024 0.0045 <b>0.0039</b> 0.0194 | <b>15336</b> <b>22904</b> <b>16288</b> <b>67184</b> | * <b>105+24</b> 4+0 4+0
hash.llvm.lib 0.05 0.0029 0.0043 <b>0.0066</b> 0.0211 | 25880 <b>18176</b> <b>28168</b> <b>85624</b> | <b>2</b> <b>241+23</b> 7+0 7+0
bisort.llvm.lib <b>0.05</b> 0.0022 0.0036 0.0064 0.0192 | 14376 <b>28616</b> <b>22416</b> <b>76480</b> | <b>3</b> <b>210+27</b> 10+0 10+0
llubenchmark.llvm.lib 0.04 0.0017 0.0026 0.0045 0.0169 | 11824 <b>24304</b> <b>18168</b> <b>65304</b> | <b>8</b> <b>144+20</b> 15+0 15+0
treeadd.llvm.lib 0.03 0.0013 0.0016 0.0023 0.0101 | <b>9080</b> <b>8408</b> <b>11224</b> <b>40704</b> | <b>2</b> <b>76+14</b> 5+0 5+0
sumarray2d.llvm.lib 0.03 0.0014 0.0097 0.0106 0.0238 | 42056 46440 83192 174008 | * 15+3 1+0 1+0
sieve.llvm.lib 0.03 0.0008 0.0021 0.0032 0.0103 | 16064 <b>14264</b> <b>35312</b> <b>76008</b> | * <b>30+4</b> 2+0 2+0
objinst.llvm.lib <b>0.03</b> 0.0015 0.0024 <b>0.0037</b> 0.0134 | 11032 <b>15488</b> <b>11440</b> <b>51616</b> | 39 <b>73+17</b> 3+0 3+0
methcall.llvm.lib 0.03 0.0014 0.0018 0.0030 0.0115 | <b>10824</b> <b>9704</b> <b>11128</b> <b>45312</b> | 31 <b>58+14</b> 2+0 2+0
matrix.llvm.lib 0.03 0.0011 0.0017 0.0018 0.0109 | <b>8584</b> <b>17680</b> <b>4208</b> <b>46232</b> | * <b>46+11</b> 1+0 1+0
ary3.llvm.lib 0.03 0.0010 0.0014 0.0014 0.0085 | 8752 <b>17136</b> <b>4168</b> <b>45720</b> | 6 <b>50+7</b> 1+0 1+0
sumarraymalloc.llvm.lib 0.02 0.0007 0.0009 0.0013 0.0070 | 2720 <b>12824</b> <b>3376</b> <b>33352</b> | * <b>38+8</b> 2+0 2+0
random.llvm.lib 0.02 0.0006 0.0009 0.0009 0.0060 | <b>6184</b> <b>1808</b> <b>2752</b> <b>23600</b> | * <b>31+5</b> 3+0 3+0
pi.llvm.lib 0.02 0.0006 0.0007 0.0011 0.0049 | 5696 <b>5272</b> <b>6544</b> <b>17512</b> | * <b>34+6</b> 3+0 3+0
matrixTranspose.llvm.lib 0.02 0.0007 0.0033 0.0034 0.0095 | 14088 9424 29480 52992 | * 17+4 3+0 3+0
heapsort.llvm.lib <b>0.02</b> 0.0008 0.0009 0.0013 <b>0.0076</b> | 2936 <b>7224</b> <b>3912</b> <b>29240</b> | * <b>45+6</b> 3+0 3+0
fib2.llvm.lib 0.02 0.0014 0.0008 0.0009 0.0067 | <b>6176</b> <b>1808</b> <b>2400</b> <b>23008</b> | * <b>25+6</b> 1+0 1+0
sumarray.llvm.lib <b>0.01</b> 0.0003 0.0003 0.0003 0.0022 | 1280 3552 584 7528 | 3 7+2 0+0 0+0
printargs.llvm.lib 0.01 0.0005 0.0004 0.0006 0.0030 | 1360 <b>816</b> <b>4760</b> <b>6936</b> | * <b>13+4</b> 1+0 1+0
hello.llvm.lib 0.01 0.0003 0.0003 0.0003 0.0022 | 880 <b>592</b> <b>2944</b> <b>6672</b> | * <b>8+3</b> 1+0 1+0
ackermann.llvm.lib 0.01 0.0000 0.0000 0.0001 0.0004 | 0 0 832 832 | * 0+0 * *
</pre><p>
<hr>
<a name="20021108"><h2>Nov 8, 2002:</h2>
Same as the <a href="#20021107-2">previous run</a>, but now the Top-Down pass
has been rewritten to:<p>
<ul>
<li>Actually work on all of the benchmarks
<li>Require less book-keeping
<li>Only clone a caller graph into a particular callee graph once
<li>Not prune nodes as aggressively, this is probably the reason for many
slowdowns below.
</ul><p>
Looking deeper into the results, gzip got slower by a factor of 4, optimizer
eval got slower by a factor of 3, sim slowed down just a little bit.<p>
Note that without the globals graph, some benchmarks are still VERY slow (those
with heavy use of globals).<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
254.gap.lib 330.50 * * * * | * * * * | * * * *
255.vortex.lib * * * * * | * * * * | * * * *
197.parser.lib 120.10 0.1649 11.8198 105.3900 118.0119 | 2188536 4525320 60972272 67921544 | 22575 202604+8604 601+0 601+0
300.twolf.lib 72.38 0.2431 1.8500 30.5671 72.3845 | 629008 1375352 16515832 18520192 | 18090 163168+3906 854+0 854+0
burg.llvm.lib 34.18 0.0871 1.0919 15.8467 34.1754 | 749304 1755232 7209416 9779360 | 21707 78464+4387 393+0 393+0
164.gzip.llvm.lib 21.81 0.1149 1.2083 19.6981 21.1201 | 2055024 3145544 8929008 14227896 | 2387 18236+1588 171+0 171+0
256.bzip2.lib 7.44 0.0250 0.3032 6.2987 7.4430 | 372776 1205520 7660752 9312304 | 1765 11947+1011 148+0 148+0
optimizer-eval.llvm.lib 3.24 0.0131 1.0648 1.9117 3.0792 | 110456 343032 1971952 2443624 | 5293 4186+1033 63+0 63+0
sim.llvm.lib 2.27 0.0170 0.2274 1.7720 2.0708 | 104728 130376 451320 759088 | 62 2755+122 99+0 99+0
181.mcf.lib 0.61 0.0162 0.0556 0.3631 0.4646 | 156648 234864 866704 1275064 | 243 2023+163 44+0 44+0
voronoi.llvm.lib 0.45 0.0117 0.0447 0.2557 0.3385 | 83088 150936 337552 585520 | 157 2063+390 30+0 30+0
bh.llvm.lib 0.38 0.0101 0.0253 0.1778 0.2394 | 79736 129328 245408 502512 | 66 858+198 13+0 13+0
sgefa.llvm.lib 0.27 0.0071 0.0396 0.0780 0.1465 | 104712 112576 204600 446656 | 130 640+103 31+0 31+0
power.llvm.lib 0.14 0.0035 0.0080 0.0451 0.0675 | 31272 49768 66968 155680 | 6 359+96 9+0 9+0
em3d.llvm.lib 0.13 0.0046 0.0140 0.0524 0.0825 | 37096 65320 122624 241744 | 141 920+127 21+0 21+0
mst.llvm.lib 0.13 0.0033 0.0084 0.0381 0.0605 | 29104 36104 70496 150376 | 100 506+91 21+0 21+0
lists.llvm.lib 0.12 0.0036 0.0118 0.0328 0.0664 | 27880 33064 98728 177520 | 58 270+63 10+0 10+0
bisort.llvm.lib 0.11 0.0023 0.0047 0.0125 0.0270 | 14376 29520 33808 88776 | 6 226+42 10+0 10+0
indvars.llvm.lib 0.11 0.0020 0.0254 0.0310 0.0606 | 85520 88232 161216 334968 | * 18+3 1+0 1+0
health.llvm.lib 0.10 0.0044 0.0113 0.0255 0.0519 | 31232 44592 53456 144248 | 79 385+69 14+0 14+0
tsp.llvm.lib 0.10 0.0042 0.0062 0.0176 0.0382 | 27840 19008 51376 112528 | 27 271+55 10+0 10+0
hash.llvm.lib 0.10 0.0026 0.0056 0.0168 0.0336 | 25880 29936 35312 104528 | 6 271+44 7+0 7+0
heapsort.llvm.lib 0.09 0.0007 0.0009 0.0015 0.0429 | 2936 7384 4064 29552 | * 47+7 3+0 3+0
perimeter.llvm.lib 0.08 0.0025 0.0064 0.0090 0.0271 | 15432 25616 18552 72440 | * 121+39 4+0 4+0
objinst.llvm.lib 0.08 0.0015 0.0027 0.0076 0.0180 | 11032 16088 19848 60624 | 39 78+25 3+0 3+0
sumarray2d.llvm.lib 0.06 0.0011 0.0098 0.0111 0.0241 | 42056 46440 83192 174008 | * 15+3 1+0 1+0
ary3.llvm.lib 0.06 0.0010 0.0016 0.0021 0.0096 | 8752 17528 4696 46640 | 6 55+10 1+0 1+0
llubenchmark.llvm.lib 0.05 0.0018 0.0034 0.0063 0.0199 | 11824 20384 28056 71272 | 32 141+31 15+0 15+0
treeadd.llvm.lib 0.05 0.0012 0.0023 0.0036 0.0121 | 9120 17640 10960 53880 | 4 82+18 5+0 5+0
sumarray.llvm.lib 0.04 0.0003 0.0003 0.0003 0.0022 | 1280 3552 584 7528 | 3 7+2 0+0 0+0
methcall.llvm.lib 0.03 0.0014 0.0021 0.0045 0.0147 | 10680 16000 11784 51720 | 31 63+19 2+0 2+0
sieve.llvm.lib 0.03 0.0008 0.0022 0.0047 0.0120 | 16064 14528 39672 80632 | * 32+5 2+0 2+0
fib2.llvm.lib 0.03 0.0007 0.0009 0.0011 0.0063 | 5512 6088 2544 26432 | * 27+7 1+0 1+0
matrix.llvm.lib 0.02 0.0010 0.0017 0.0021 0.0100 | 8760 8104 8976 42184 | * 48+12 1+0 1+0
matrixTranspose.llvm.lib 0.02 0.0007 0.0034 0.0037 0.0098 | 14088 9424 29480 52992 | * 17+4 3+0 3+0
sumarraymalloc.llvm.lib 0.02 0.0007 0.0010 0.0016 0.0075 | 2720 12928 3536 33544 | * 40+9 2+0 2+0
pi.llvm.lib 0.02 0.0006 0.0007 0.0014 0.0052 | 5696 5368 5680 16744 | * 35+7 3+0 3+0
random.llvm.lib 0.02 0.0006 0.0007 0.0012 0.0061 | 2136 6040 7248 31176 | * 33+6 3+0 3+0
printargs.llvm.lib 0.02 0.0003 0.0004 0.0007 0.0030 | 1360 1024 1728 4112 | * 14+5 1+0 1+0
hello.llvm.lib 0.02 0.0003 0.0003 0.0005 0.0023 | 880 648 768 4552 | * 9+4 1+0 1+0
ackermann.llvm.lib 0.02 0.0000 0.0000 0.0002 0.0004 | 0 0 832 832 | * 0+0 * *
</pre><p>
<hr>
2002-11-08 06:50:22 +00:00
<a name="20021107-2"><h2>Nov 7, 2002 #2:</h2>
Same as the <a href="#20021107">previous run</a>, but now we actually handle
constants (especially constantexprs) correctly in the local and all subsequent
passes. This seems to dramatically improve many benchmarks, only slowing down
gzip significantly.<p>
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
300.twolf.lib 30.88 * * * * | * * * * | * * * *
255.vortex.lib 15.71 * * * * | * * * * | * * * *
2002-11-08 06:50:22 +00:00
254.gap.lib 6.07 analyze: ../../../include/llvm/Analysis/DSNode.h:77: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
164.gzip.llvm.lib 5.32 0.1157 1.0296 3.4025 4.6326 | 2054352 2631576 7804656 12588904 | 398 13372+217 174+3 174+3
256.bzip2.lib 2.36 0.0250 0.1913 1.8190 2.0966 | 371928 1045320 7533440 9023600 | 407 11415+149 151+2 151+2
sim.llvm.lib 1.97 0.0174 0.2155 1.4731 1.7539 | 104584 120112 449736 747416 | 46 2745+52 101+2 101+2
197.parser.lib 1.27 analyze: ../../../include/llvm/Analysis/DSNode.h:77: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
optimizer-eval.llvm.lib 1.09 0.0132 0.0926 0.8104 0.9528 | 109416 186592 1827184 2141376 | 1709 8189+43 63+0 63+0
burg.llvm.lib 0.81 analyze: ../../../include/llvm/Analysis/DSNode.h:77: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
181.mcf.lib 0.38 0.0163 0.0470 0.1847 0.2714 | 156304 219896 808112 1201160 | 226 1859+0 44+0 44+0
voronoi.llvm.lib 0.22 0.0118 0.0241 0.0776 0.1293 | 82728 87872 214584 399128 | 31 1650+0 30+0 30+0
sgefa.llvm.lib 0.19 0.0074 0.0324 0.0716 0.1303 | 104560 104744 200024 434096 | 89 610+1 32+0 32+0
bh.llvm.lib 0.17 0.0106 0.0158 0.0344 0.0813 | 80112 86984 210008 425144 | 66 678+3 16+0 16+0
em3d.llvm.lib 0.16 0.0067 0.0111 0.0619 0.0899 | 37600 55936 217632 327872 | 90 1584+522 38+3 38+3
health.llvm.lib 0.10 0.0044 0.0071 0.0117 0.0322 | 31120 38712 35248 120048 | 81 328+0 14+0 14+0
lists.llvm.lib 0.09 0.0037 0.0075 0.0296 0.0477 | 27792 32968 83288 162264 | 61 222+0 10+0 10+0
llubenchmark.llvm.lib 0.09 0.0018 0.0021 0.0060 0.0178 | 11104 12176 32552 66840 | 35 152+22 17+1 17+1
power.llvm.lib 0.08 0.0036 0.0045 0.0139 0.0308 | 21672 31000 40512 109008 | 5 246+0 9+0 9+0
mst.llvm.lib 0.07 0.0033 0.0051 0.0125 0.0296 | 28368 27552 49088 119680 | 54 435+1 21+0 21+0
indvars.llvm.lib 0.07 0.0020 0.0199 0.0281 0.0521 | 85496 84840 164280 334616 | * 15+0 1+0 1+0
perimeter.llvm.lib 0.06 0.0027 0.0035 0.0037 0.0180 | 15312 12112 14536 54800 | * 83+0 4+0 4+0
tsp.llvm.lib 0.05 0.0045 0.0041 0.0063 0.0254 | 27736 15016 25296 82352 | 28 228+1 11+0 11+0
hash.llvm.lib 0.04 0.0025 0.0038 0.0054 0.0202 | 18288 24528 26104 82320 | 2 220+0 7+0 7+0
bisort.llvm.lib 0.04 0.0025 0.0030 0.0050 0.0178 | 14296 19056 28384 72808 | 5 186+0 10+0 10+0
sumarray2d.llvm.lib 0.04 0.0011 0.0077 0.0102 0.0210 | 42032 46264 83960 174576 | * 12+0 1+0 1+0
treeadd.llvm.lib 0.04 0.0011 0.0013 0.0017 0.0086 | 9008 7360 10264 42432 | 2 64+0 5+0 5+0
objinst.llvm.lib 0.03 0.0021 0.0019 0.0049 0.0147 | 10928 15344 24592 64520 | 31 151+72 9+4 9+4
methcall.llvm.lib 0.03 0.0014 0.0016 0.0030 0.0111 | 15520 8928 8032 45736 | 27 99+44 8+4 8+4
matrix.llvm.lib 0.03 0.0010 0.0015 0.0017 0.0086 | 8680 8424 8200 41304 | * 39+0 1+0 1+0
sieve.llvm.lib 0.03 0.0007 0.0021 0.0032 0.0101 | 10320 13864 34512 72456 | * 26+0 2+0 2+0
sumarraymalloc.llvm.lib 0.03 0.0007 0.0008 0.0008 0.0060 | 2656 6144 6848 30024 | * 31+0 2+0 2+0
matrixTranspose.llvm.lib 0.03 0.0006 0.0027 0.0034 0.0087 | 14064 9080 28600 51744 | * 14+0 3+0 3+0
ary3.llvm.lib 0.02 0.0011 0.0013 0.0012 0.0081 | 8720 6968 8256 39992 | 6 44+0 1+0 1+0
heapsort.llvm.lib 0.02 0.0008 0.0008 0.0008 0.0077 | 2896 13464 2552 33728 | * 39+0 3+0 3+0
random.llvm.lib 0.02 0.0007 0.0006 0.0006 0.0056 | 6296 4848 6112 29856 | * 26+0 3+0 3+0
fib2.llvm.lib 0.01 0.0007 0.0007 0.0006 0.0052 | 5488 5408 1344 24888 | * 19+0 1+0 1+0
pi.llvm.lib 0.01 0.0006 0.0006 0.0009 0.0043 | 5656 4768 5416 15840 | * 29+0 3+0 3+0
printargs.llvm.lib 0.01 0.0003 0.0003 0.0003 0.0023 | 1336 560 728 2624 | * 9+0 1+0 1+0
sumarray.llvm.lib 0.01 0.0003 0.0003 0.0002 0.0022 | 1256 2456 456 6280 | 3 5+0 0+0 0+0
hello.llvm.lib 0.01 0.0003 0.0003 0.0002 0.0019 | 840 2872 3136 6848 | * 5+0 1+0 1+0
ackermann.llvm.lib 0.01 * * * * | * * * * | * 0+0 * *
</pre><p>
<hr>
<a name="20021107"><h2>Nov 7, 2002:</h2>
Notes: Debug build, no globals graph, no array collapsing.<p>
Things are a big slower than before due to corrected handling of global
variables.
<pre>
Name: Anlyz: LocTime: BUTime: TDTime: TotTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
254.gap.lib 6.23 analyze: ../../../include/llvm/Analysis/DSNode.h:77: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
sim.llvm.lib 3.77 0.0156 0.3313 1.6634 2.0787 | 63080 98008 192424 426656 | * 1825+84 63+2 63+2
300.twolf.lib 374.15 * * * * | * * * * | * * * *
255.vortex.lib 185.26 * * * * | * * * * | * * * *
256.bzip2.lib 1.10 0.0158 0.0679 0.7397 0.8794 | 213592 385312 3368648 4040464 | 18 5238+257 64+1 64+1
164.gzip.llvm.lib 1.08 0.0260 0.0797 0.3326 0.5279 | 235048 421744 1484408 2237792 | 21 9628+634 113+1 113+1
197.parser.lib 1.06 analyze: ../../../include/llvm/Analysis/DSNode.h:77: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
burg.llvm.lib 0.71 analyze: ../../../include/llvm/Analysis/DSNode.h:77: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
optimizer-eval.llvm.lib 0.67 0.0115 0.0793 0.4125 0.5422 | 78080 141896 372592 610760 | 58 666+27 2+0 2+0
181.mcf.lib 0.32 0.0125 0.0418 0.1271 0.2052 | 120504 192920 545696 876008 | 186 753+0 14+0 14+0
voronoi.llvm.lib 0.18 0.0098 0.0204 0.0661 0.1116 | 78560 83344 116304 292200 | 19 1006+0 16+0 16+0
sgefa.llvm.lib 0.16 0.0064 0.0340 0.0413 0.0999 | 82280 96016 67960 270944 | * 125+2 2+0 2+0
sieve.llvm.lib 0.16 0.0007 0.0028 0.0209 0.0295 | 9872 13808 38872 76312 | * 22+0 1+0 1+0
lists.llvm.lib 0.15 0.0027 0.0061 0.0042 0.0217 | 22256 14256 3208 51720 | 52 38+0 0+0 0+0
bh.llvm.lib 0.14 0.0082 0.0140 0.0171 0.0630 | 73320 75448 42920 255160 | 74 402+4 10+0 10+0
em3d.llvm.lib 0.14 0.0037 0.0091 0.0303 0.0522 | 33816 50768 109200 202192 | 10 925+410 19+3 19+3
hash.llvm.lib 0.14 0.0020 0.0033 0.0036 0.0154 | 17488 23928 12584 67400 | * 131+0 2+0 2+0
llubenchmark.llvm.lib 0.13 0.0013 0.0011 0.0019 0.0128 | 9680 2144 10072 32904 | * 51+10 4+1 4+1
indvars.llvm.lib 0.12 0.0022 0.0248 0.0422 0.0713 | 82040 87944 160872 333400 | * 12+0 0+0 0+0
ary3.llvm.lib 0.12 0.0008 0.0011 0.0009 0.0072 | 7992 7808 2416 34264 | 6 37+0 0+0 0+0
health.llvm.lib 0.10 0.0035 0.0067 0.0067 0.0280 | 28432 27984 12640 83984 | 72 151+1 6+0 6+0
hello.llvm.lib 0.10 0.0002 0.0002 0.0002 0.0016 | 504 2816 336 3656 | * 2+0 0+0 0+0
ackermann.llvm.lib 0.09 0.0004 0.0005 0.0004 0.0045 | 6976 992 840 18736 | * 10+0 0+0 0+0
objinst.llvm.lib 0.08 0.0014 0.0019 0.0052 0.0137 | 9480 19064 12216 54416 | 25 148+80 7+4 7+4
sumarray2d.llvm.lib 0.08 0.0012 0.0096 0.0109 0.0235 | 41416 47080 80696 171512 | * 9+0 0+0 0+0
methcall.llvm.lib 0.08 0.0011 0.0016 0.0032 0.0108 | 3952 17528 8584 43720 | 23 104+52 7+4 7+4
heapsort.llvm.lib 0.08 0.0006 0.0008 0.0007 0.0063 | 2264 13720 2216 33016 | * 33+0 2+0 2+0
fib2.llvm.lib 0.08 0.0005 0.0006 0.0004 0.0048 | 1776 5312 1080 23992 | * 14+0 0+0 0+0
tsp.llvm.lib 0.07 0.0036 0.0035 0.0041 0.0197 | 17304 13808 17968 63704 | 18 95+1 4+0 4+0
random.llvm.lib 0.07 0.0005 0.0006 0.0005 0.0066 | 1584 4704 5752 25264 | * 21+0 2+0 2+0
pi.llvm.lib 0.07 0.0004 0.0004 0.0005 0.0031 | 4744 1248 4368 10360 | * 14+0 0+0 0+0
power.llvm.lib 0.06 0.0031 0.0063 0.0070 0.0251 | 19440 28024 23544 86832 | * 146+0 4+0 4+0
bisort.llvm.lib 0.06 0.0019 0.0036 0.0026 0.0168 | 12776 18296 6808 48952 | * 89+0 4+0 4+0
matrixTranspose.llvm.lib 0.06 0.0005 0.0019 0.0025 0.0065 | 9136 14520 25336 48992 | * 10+0 1+0 1+0
printargs.llvm.lib 0.06 0.0003 0.0003 0.0003 0.0021 | 888 504 512 1904 | * 5+0 0+0 0+0
perimeter.llvm.lib 0.05 0.0026 0.0042 0.0044 0.0220 | 13968 12464 3928 43016 | * 35+0 1+0 1+0
mst.llvm.lib 0.05 0.0026 0.0040 0.0052 0.0227 | 16520 25840 14016 71048 | 10 217+1 9+0 9+0
sumarray.llvm.lib 0.05 0.0002 0.0003 0.0002 0.0020 | 976 3424 464 6976 | 3 5+0 0+0 0+0
treeadd.llvm.lib 0.03 0.0010 0.0011 0.0009 0.0085 | 7928 6424 2088 28432 | * 27+0 1+0 1+0
matrix.llvm.lib 0.02 0.0020 0.0013 0.0014 0.0090 | 7856 10416 2200 36232 | * 32+0 0+0 0+0
sumarraymalloc.llvm.lib 0.02 0.0006 0.0007 0.0005 0.0056 | 2032 5984 1368 23472 | * 17+0 0+0 0+0
</pre><p>
<hr>
<a name="20021106"><h2>Nov 6, 2002:</h2>
Notes: Debug build, no globals graph, no array collapsing.<p>
<pre>
2002-11-07 02:18:30 +00:00
Name: LocTime: BUTime: TDTime: TotTime: AnlzTime: LocSize: BUSize: TDSize: TotSize: NumFold NumNodes main __main
ackermann.llvm.lib 0.0004 0.0005 0.0004 0.0045 0m0.040s | 6976 992 840 18736 | * 10+0 0+0 0+0
ary3.llvm.lib 0.0008 0.0012 0.0012 0.0074 0m0.034s | 7992 7808 2416 34264 | 6 37+0 0+0 0+0
fib2.llvm.lib 0.0005 0.0006 0.0004 0.0047 0m0.019s | 1776 5312 1080 23992 | * 14+0 0+0 0+0
hash.llvm.lib 0.0020 0.0031 0.0036 0.0157 0m0.046s | 17488 23928 12584 67400 | * 131+0 2+0 2+0
heapsort.llvm.lib 0.0006 0.0008 0.0007 0.0061 0m0.023s | 2264 13720 2216 33016 | * 33+0 2+0 2+0
hello.llvm.lib 0.0002 0.0002 0.0002 0.0015 0m0.014s | 504 2816 336 3656 | * 2+0 0+0 0+0
lists.llvm.lib 0.0029 0.0057 0.0057 0.0203 0m0.077s | 22272 14264 12128 60688 | 57 38+0 0+0 0+0
llubenchmark.llvm.lib 0.0012 0.0011 0.0017 0.0111 0m0.036s | 9680 2144 9944 32776 | * 49+8 4+1 4+1
matrix.llvm.lib 0.0008 0.0013 0.0014 0.0079 0m0.038s | 7848 10416 2200 36176 | * 32+0 0+0 0+0
matrixTranspose.llvm.lib 0.0004 0.0015 0.0019 0.0054 0m0.031s | 9136 14520 25336 48992 | * 10+0 1+0 1+0
methcall.llvm.lib 0.0010 0.0018 0.0030 0.0108 0m0.080s | 3952 17528 9024 44160 | 30 112+52 7+4 7+4
objinst.llvm.lib 0.0011 0.0018 0.0058 0.0149 0m0.061s | 9480 19064 13192 55392 | 37 163+80 7+4 7+4
2002-11-07 02:18:30 +00:00
pi.llvm.lib 0.0005 0.0004 0.0005 0.0032 0m0.027s | 4744 1248 4368 10360 | * 14+0 0+0 0+0
printargs.llvm.lib 0.0003 0.0005 0.0003 0.0023 0m0.042s | 888 504 512 1904 | * 5+0 0+0 0+0
random.llvm.lib 0.0005 0.0005 0.0005 0.0056 0m0.032s | 1584 4704 5752 25264 | * 21+0 2+0 2+0
sumarray.llvm.lib 0.0002 0.0003 0.0002 0.0020 0m0.034s | 976 3424 464 6976 | 3 5+0 0+0 0+0
sumarray2d.llvm.lib 0.0010 0.0077 0.0111 0.0216 0m0.047s | 41416 47080 80696 171512 | * 9+0 0+0 0+0
sumarraymalloc.llvm.lib 0.0007 0.0007 0.0005 0.0054 0m0.032s | 2032 5984 1368 23472 | * 17+0 0+0 0+0
indvars.llvm.lib 0.0018 0.0189 0.0264 0.0490 0m0.065s | 82040 87944 160872 333400 | * 12+0 0+0 0+0
bh.llvm.lib 0.0113 0.0128 0.0163 0.0597 0m0.148s | 55112 75768 42928 240056 | 74 400+5 11+0 11+0
bisort.llvm.lib 0.0018 0.0024 0.0024 0.0123 0m0.051s | 12776 18296 6808 48952 | * 89+0 4+0 4+0
em3d.llvm.lib 0.0035 0.0098 0.0196 0.0490 0m0.103s | 33816 49744 54144 146112 | 10 518+33 14+1 14+1
health.llvm.lib 0.0033 0.0061 0.0057 0.0226 0m0.088s | 28432 27992 12640 83992 | 72 151+1 6+0 6+0
mst.llvm.lib 0.0025 0.0039 0.0050 0.0192 0m0.074s | 16520 25840 14016 71048 | 10 217+1 9+0 9+0
perimeter.llvm.lib 0.0020 0.0032 0.0025 0.0154 0m0.045s | 13968 12720 3928 43272 | * 35+0 1+0 1+0
power.llvm.lib 0.0027 0.0039 0.0060 0.0232 0m0.062s | 20960 30136 23504 82696 | * 146+0 4+0 4+0
treeadd.llvm.lib 0.0009 0.0010 0.0008 0.0070 0m0.054s | 7928 6424 2088 28432 | * 27+0 1+0 1+0
tsp.llvm.lib 0.0035 0.0034 0.0032 0.0187 0m0.064s | 17304 13808 17968 63704 | 18 95+1 4+0 4+0
voronoi.llvm.lib 0.0093 0.0193 0.0567 0.1033 0m0.199s | 78560 83336 116312 292200 | 19 1006+0 16+0 16+0
optimizer-eval.llvm.lib 0.0099 0.1017 0.4073 0.5492 0m0.674s | 76736 140000 409448 644528 | 58 1259+23 7+0 7+0
sgefa.llvm.lib 0.0057 0.0283 0.0357 0.0863 0m0.153s | 82280 96016 67728 270712 | * 121+2 2+0 2+0
sieve.llvm.lib 0.0006 0.0020 0.0030 0.0096 0m0.033s | 9872 13808 38872 76312 | * 22+0 1+0 1+0
sim.llvm.lib 0.0134 0.2127 1.1230 1.3938 0m1.559s | 62864 98984 200600 457536 | * 1807+206 71+7 71+7
164.gzip.llvm.lib 0.0225 0.0727 0.2910 0.4651 0m1.003s | 235048 421864 1439120 2192624 | 17 9181+106 114+1 114+1
181.mcf.lib 0.0121 0.0354 0.1096 0.1780 0m0.292s | 120504 192920 545720 876032 | 195 753+0 14+0 14+0
197.parser.lib analyze: ../../../include/llvm/Analysis/DSNode.h:74: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
254.gap.lib analyze: ../../../include/llvm/Analysis/DSNode.h:74: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
255.vortex.lib analyze: ../../../include/llvm/Analysis/DSNode.h:74: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
2002-11-07 02:18:30 +00:00
256.bzip2.lib 0.0149 0.0604 0.6503 0.7810 0m1.027s | 210880 387776 3371632 4041800 | 18 5005+76 65+1 65+1
2002-11-08 06:50:22 +00:00
300.twolf.lib 0.1806 0.5248 42.5463 43.8329 0m45.785s | 522056 653816 10076992 11252864 | 1269 78066+4900 410+12 410+12
burg.llvm.lib analyze: ../../../include/llvm/Analysis/DSNode.h:74: DSNode::~DSNode(): Assertion `Referrers.empty() && "Referrers to dead node exist!"' failed.
</pre>
<hr>
<address><a href="mailto:sabre@nondot.org">Chris Lattner</a></address>
<!-- Created: Wed Nov 6 19:57:57 CST 2002 -->
<!-- hhmts start -->
2002-11-12 06:53:43 +00:00
Last modified: Tue Nov 12 00:49:40 CST 2002
<!-- hhmts end -->
</body>
</html>