[patch] Excessive alignment in ix86_data_alignment
Petr Murzin
petrmurzin1@gmail.com
Wed Nov 26 23:41:00 GMT 2014
On 09 Oct 08:25, H.J. Lu wrote:
> On Thu, Oct 9, 2014 at 1:37 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
> > On Thu, Oct 9, 2014 at 10:25 AM, Kirill Yukhin <kirill.yukhin@gmail.com> wrote:
> >> On 08 Oct 23:02, Petr Murzin wrote:
> >>> Hi,
> >>> I have measured performance impact on Haswell platform according to this
> >>> input:
> >>> https://gcc.gnu.org/ml/gcc-patches/2014-06/msg00978.html
>
> Kirill, please mention:
>
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61296
>
> in your ChangeLog.
>
> > What about older processors?
>
> Kirill, please collect data on Nehelam/Westmere, Sandybrigde/Ivybride
> and Silvermont.
>
> > The optimization was introduced well before Haswell for then current
> > processors, and it was based on the recommendation from Intel
> > optimization guide. If this optimization doesn't apply for new
> > processors, then tune option should be introduced and set accordingly.
> >
>
> I believe the original excessive alignment was introduced by cut/paste
> from
>
> https://gcc.gnu.org/git/?p=gcc.git;a=patch;h=ed45e834f305d1f2709bf200a13d5beebc2fcfee
>
> to improve x86 FP performance, which might be partially copied from
> CONSTANT_ALIGNMENT:
>
> https://gcc.gnu.org/git/?p=gcc.git;a=patch;h=f7d6703c5d83fc9fb06246d6eb49e9b61098045c
>
>
> --
> H.J.
Hi,
Here is the info about code and data size.
CODE SEGMENTS
SLM
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 1970381 | 1969805 | -0.0292329250028
hmmer | 1244935 | 1244775 | -0.0128520766144
Xalan | 6464124 | 6448572 | -0.240589444138
gcc | 5613433 | 5611001 | -0.0433246464329
gamess | 16775109 | 16744869 | -0.180267085001
h264ref | 2004787 | 2003155 | -0.0814051567573
libquantum | 915693 | 915693 | 0.0
astar | 956881 | 956849 | -0.0033441984949
gobmk | 3000807 | 3000263 | -0.0181284567785
namd | 1525077 | 1525045 | -0.00209825471107
lbm | 777141 | 777141 | 0.0
milc | 1151431 | 1151303 | -0.0111166018632
sjeng | 994163 | 993555 | -0.0611569732529
GemsFDTD | 2361733 | 2360773 | -0.0406481172935
leslie3d | 1474453 | 1473909 | -0.0368950383634
bwaves | 1076385 | 1075777 | -0.0564853653665
mcf | 778835 | 778835 | 0.0
gromacs | 2348746 | 2348106 | -0.0272485828608
wrf | 7742741 | 7739637 | -0.040089162223
cactusADM | 2820689 | 2820049 | -0.0226894918227
zeusmp | 1676605 | 1675997 | -0.0362637592039
sphinx | 1379854 | 1379822 | -0.00231908593228
omnetpp | 2872183 | 2871511 | -0.0233968378756
dealII | 2514404 | 2512324 | -0.0827233809682
povray | 3212579 | 3211395 | -0.0368551248078
perlbench | 3123490 | 3122178 | -0.0420042964761
calculix | 3914026 | 3909866 | -0.106284424273
bzip2 | 887974 | 887910 | -0.00720741823522
tonto | 7213573 | 7206981 | -0.0913832853705
SNB and IVB
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 1969317 | 1968741 | -0.029248719226
hmmer | 1252815 | 1252655 | -0.0127712391694
Xalan | 6517380 | 6501828 | -0.238623495945
gcc | 5634249 | 5631817 | -0.0431645814731
gamess | 16555117 | 16525197 | -0.18072961973
h264ref | 2062507 | 2060779 | -0.0837815338324
libquantum | 915085 | 915085 | 0.0
astar | 957025 | 956993 | -0.00334369530577
gobmk | 3010535 | 3009991 | -0.0180698779453
namd | 1505805 | 1505773 | -0.00212510916088
lbm | 775733 | 775733 | 0.0
milc | 1140567 | 1140471 | -0.00841686634805
sjeng | 997235 | 996627 | -0.0609685781185
GemsFDTD | 2451645 | 2450653 | -0.0404626281537
leslie3d | 1479125 | 1478581 | -0.0367785008028
bwaves | 1072497 | 1071953 | -0.0507227526044
mcf | 778803 | 778803 | 0.0
gromacs | 2322770 | 2322130 | -0.0275533091955
wrf | 8092645 | 8090693 | -0.0241206675938
cactusADM | 2808633 | 2807993 | -0.0227868860047
zeusmp | 1733717 | 1733077 | -0.0369149059506
sphinx | 1388206 | 1388174 | -0.00230513338798
omnetpp | 2875179 | 2874475 | -0.0244854320374
dealII | 2527560 | 2525576 | -0.078494674706
povray | 3164931 | 3163971 | -0.0303324148299
perlbench | 3123690 | 3122346 | -0.0430260365145
calculix | 3905210 | 3901114 | -0.104885524722
bzip2 | 888262 | 888230 | -0.00360254069183
tonto | 7210917 | 7204101 | -0.0945233456438
WSM
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 1972141 | 1971597 | -0.0275842345958
hmmer | 1243935 | 1243775 | -0.0128624084056
Xalan | 6480260 | 6464676 | -0.240484178104
gcc | 5614945 | 5612481 | -0.0438828875439
gamess | 16432005 | 16401829 | -0.183641618902
h264ref | 1998315 | 1996715 | -0.0800674568324
libquantum | 915597 | 915597 | 0.0
astar | 957065 | 957033 | -0.00334355555788
gobmk | 3004111 | 3003599 | -0.0170433116486
namd | 1523745 | 1523713 | -0.00210008892564
lbm | 776589 | 776589 | 0.0
milc | 1148391 | 1148295 | -0.00835952214882
sjeng | 994883 | 994275 | -0.0611127137563
GemsFDTD | 2353397 | 2352405 | -0.0421518341359
leslie3d | 1472469 | 1471925 | -0.0369447506195
bwaves | 1075393 | 1074817 | -0.0535618141461
mcf | 778771 | 778771 | 0.0
gromacs | 2342426 | 2341722 | -0.0300543112141
wrf | 7549709 | 7546605 | -0.0411141674467
cactusADM | 2798905 | 2798233 | -0.0240093893862
zeusmp | 1672901 | 1672293 | -0.0363440514412
sphinx | 1377542 | 1377446 | -0.0069689345225
omnetpp | 2867139 | 2866467 | -0.023437998646
dealII | 2512444 | 2510396 | -0.0815142546461
povray | 3208771 | 3207747 | -0.0319125297505
perlbench | 3120402 | 3119122 | -0.0410203557106
calculix | 3878210 | 3874082 | -0.106440858025
bzip2 | 885710 | 885678 | -0.00361292070769
tonto | 7185589 | 7178965 | -0.0921845098572
HSW
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 1963401 | 1962761 | -0.0325964996453
hmmer | 1252327 | 1252199 | -0.0102209726373
Xalan | 6526735 | 6511023 | -0.240732923889
gcc | 5688321 | 5685825 | -0.0438793802249
gamess | 17467069 | 17437245 | -0.170744158622
h264ref | 2080675 | 2078915 | -0.084587934204
libquantum | 915205 | 915205 | 0.0
astar | 956325 | 956293 | -0.00334614278619
gobmk | 3023167 | 3022591 | -0.0190528674069
namd | 1516953 | 1516889 | -0.00421898371274
lbm | 775733 | 775733 | 0.0
milc | 1150343 | 1150247 | -0.00834533699948
sjeng | 999275 | 999627 | 0.0352255385154
GemsFDTD | 2430581 | 2429589 | -0.0408132870289
leslie3d | 1474005 | 1473493 | -0.0347352960132
bwaves | 1071921 | 1071313 | -0.0567205978799
mcf | 778867 | 778867 | 0.0
gromacs | 2336282 | 2335578 | -0.0301333486283
wrf | 8055589 | 8053637 | -0.0242316235349
cactusADM | 2868025 | 2867353 | -0.0234307581001
zeusmp | 1728405 | 1727797 | -0.0351769405897
sphinx | 1390126 | 1390094 | -0.00230194960744
omnetpp | 2848923 | 2848091 | -0.029204018501
dealII | 2488573 | 2486429 | -0.0861537917513
povray | 3146543 | 3147855 | 0.0416965539642
perlbench | 3120914 | 3119666 | -0.039988285483
calculix | 4007986 | 4004050 | -0.0982039358421
bzip2 | 890390 | 890326 | -0.00718786149889
tonto | 7299565 | 7293037 | -0.0894299865814
DATA SEGMENTS
SLM
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 24360 | 24168 | -0.788177339901
hmmer | 9992 | 9672 | -3.20256204964
Xalan | 160712 | 160360 | -0.219025337249
gcc | 12696 | 12312 | -3.02457466919
gamess | 74668 | 70092 | -6.12846199175
h264ref | 7960 | 7800 | -2.01005025126
libquantum | 6280 | 6280 | 0.0
astar | 10120 | 10024 | -0.948616600791
gobmk | 2115816 | 2004488 | -5.26170517663
namd | 10280 | 10152 | -1.24513618677
lbm | 6264 | 6264 | 0.0
milc | 6728 | 6632 | -1.42687277051
sjeng | 6488 | 6456 | -0.493218249075
GemsFDTD | 7724 | 7532 | -2.48575867426
leslie3d | 7320 | 7256 | -0.874316939891
bwaves | 7336 | 7208 | -1.74482006543
mcf | 6296 | 6296 | 0.0
gromacs | 13960 | 13192 | -5.50143266476
wrf | 21244 | 20796 | -2.10883072868
cactusADM | 8348 | 8316 | -0.383325347389
zeusmp | 7416 | 7288 | -1.7259978425
sphinx | 8840 | 8808 | -0.361990950226
omnetpp | 31976 | 31368 | -1.90142606955
dealII | 26744 | 26584 | -0.598265031409
povray | 24152 | 22712 | -5.96223915204
perlbench | 21384 | 20520 | -4.0404040404
calculix | 8428 | 8236 | -2.27812055055
bzip2 | 6248 | 6248 | 0.0
tonto | 143496 | 143272 | -0.156101912248
SNB and IVB
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 24344 | 24184 | -0.657246138679
hmmer | 9992 | 9672 | -3.20256204964
Xalan | 160696 | 160376 | -0.199133768109
gcc | 12680 | 12296 | -3.02839116719
gamess | 74668 | 70124 | -6.08560561419
h264ref | 7944 | 7784 | -2.01409869084
libquantum | 6280 | 6280 | 0.0
astar | 10168 | 10040 | -1.25885129819
gobmk | 2115816 | 2004488 | -5.26170517663
namd | 10280 | 10152 | -1.24513618677
lbm | 6264 | 6264 | 0.0
milc | 6744 | 6648 | -1.42348754448
sjeng | 6488 | 6456 | -0.493218249075
GemsFDTD | 7740 | 7548 | -2.48062015504
leslie3d | 7320 | 7256 | -0.874316939891
bwaves | 7352 | 7224 | -1.74102285092
mcf | 6296 | 6296 | 0.0
gromacs | 13976 | 13208 | -5.49513451631
wrf | 21228 | 20876 | -1.65818729979
cactusADM | 8348 | 8316 | -0.383325347389
zeusmp | 7416 | 7288 | -1.7259978425
sphinx | 8840 | 8808 | -0.361990950226
omnetpp | 31992 | 31384 | -1.90047511878
dealII | 26776 | 26584 | -0.71706005378
povray | 24184 | 22712 | -6.08666887198
perlbench | 21384 | 20520 | -4.0404040404
calculix | 8444 | 8252 | -2.27380388441
bzip2 | 6248 | 6248 | 0.0
tonto | 143496 | 143272 | -0.156101912248
WSM
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 24328 | 24168 | -0.657678395265
hmmer | 10008 | 9688 | -3.19744204636
Xalan | 160696 | 160376 | -0.199133768109
gcc | 12680 | 12296 | -3.02839116719
gamess | 74668 | 70092 | -6.12846199175
h264ref | 7944 | 7784 | -2.01409869084
libquantum | 6280 | 6280 | 0.0
astar | 10120 | 10024 | -0.948616600791
gobmk | 2115816 | 2004488 | -5.26170517663
namd | 10264 | 10168 | -0.935307872175
lbm | 6248 | 6248 | 0.0
milc | 6728 | 6632 | -1.42687277051
sjeng | 6472 | 6440 | -0.494437577256
GemsFDTD | 7740 | 7548 | -2.48062015504
leslie3d | 7320 | 7256 | -0.874316939891
bwaves | 7336 | 7208 | -1.74482006543
mcf | 6296 | 6296 | 0.0
gromacs | 13976 | 13208 | -5.49513451631
wrf | 21228 | 20780 | -2.11042019974
cactusADM | 8348 | 8316 | -0.383325347389
zeusmp | 7400 | 7272 | -1.72972972973
sphinx | 8840 | 8808 | -0.361990950226
omnetpp | 32024 | 31384 | -1.99850112416
dealII | 26776 | 26584 | -0.71706005378
povray | 24184 | 22712 | -6.08666887198
perlbench | 21400 | 20536 | -4.03738317757
calculix | 8428 | 8236 | -2.27812055055
bzip2 | 6248 | 6248 | 0.0
tonto | 143512 | 143288 | -0.156084508613
HSW
Benchmark | Trunk | Trunk w/ patch | Diff, %
soplex | 24360 | 24168 | -0.788177339901
hmmer | 9992 | 9672 | -3.20256204964
Xalan | 160728 | 160376 | -0.219003533921
gcc | 12680 | 12296 | -3.02839116719
gamess | 74684 | 70140 | -6.0843018585
h264ref | 7944 | 7784 | -2.01409869084
libquantum | 6280 | 6280 | 0.0
astar | 10168 | 10040 | -1.25885129819
gobmk | 2115832 | 2004504 | -5.26166538742
namd | 10264 | 10168 | -0.935307872175
lbm | 6264 | 6264 | 0.0
milc | 6728 | 6632 | -1.42687277051
sjeng | 6472 | 6440 | -0.494437577256
GemsFDTD | 7740 | 7548 | -2.48062015504
leslie3d | 7320 | 7256 | -0.874316939891
bwaves | 7352 | 7224 | -1.74102285092
mcf | 6296 | 6296 | 0.0
gromacs | 13976 | 13208 | -5.49513451631
wrf | 21228 | 20876 | -1.65818729979
cactusADM | 8348 | 8316 | -0.383325347389
zeusmp | 7416 | 7288 | -1.7259978425
sphinx | 8840 | 8808 | -0.361990950226
omnetpp | 32040 | 31368 | -2.09737827715
dealII | 26776 | 26584 | -0.71706005378
povray | 24184 | 22712 | -6.08666887198
perlbench | 21400 | 20536 | -4.03738317757
calculix | 8444 | 8252 | -2.27380388441
bzip2 | 6264 | 6264 | 0.0
tonto | 143432 | 143240 | -0.13386134196
Thanks,
Petr
More information about the Gcc-patches
mailing list