From 9f9f8190f65209cd3f304f93246740b3983f4ce8 Mon Sep 17 00:00:00 2001
From: zxm5010 <zxm5010@gmail.com>
Date: Sun, 21 Sep 2014 08:30:52 +0800
Subject: [PATCH 1/4] Part 2 - Nbody simulation

Implemented accelerate, updateS, updateF, cudaNBodyUpdateWrapper.

Change N = 1000 from 5000
---
 Part1/PROJ_WIN/CIS565_PROJ_1.suo              | Bin 14336 -> 21504 bytes
 .../CIS565_PROJ_1/CIS565_PROJ_1.vcxproj       |   4 +-
 .../CIS565_PROJ_1/CIS565_PROJ_1.vcxproj.user  |   5 +
 Part1/PROJ_WIN/CIS565_PROJ_1/vc100.pdb        | Bin 446464 -> 487424 bytes
 Part1/PROJ_WIN/src/kernel.cu.deps             | 740 +++++++++---------
 Part1/src/kernel.cu                           |  38 +-
 Part1/src/main.cpp                            |   2 +-
 7 files changed, 412 insertions(+), 377 deletions(-)

diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1.suo b/Part1/PROJ_WIN/CIS565_PROJ_1.suo
index ad46c84a9a00137f71e370ba57e24d2f64f37088..13c9b75a172028367cca1938e8e28cd0762500fd 100644
GIT binary patch
literal 21504
zcmeHPeRv$jd0zo0*iIk>2NH+E1)E~PcfK#(N%$k&r)_Lu;ge)&Hki|CB`><WRc`l;
zY)eU?2}v9vE+rvNN)r;&5867ImW1X5jA@(Lr7<PsNuT^xo~KFsN7Ll#U+vQd_4nJ^
z(VgydPnIp2C!^<gvpX}p?>jT^yz|aGvwHRP<^S^JvDI%VHDQZ#q4MU;a^>Qx^e$Wz
zc6g_vT!0JW&6$}Qk*FxR{7pEITHx2<^&)iM)xa`<{rxszIdBQE5?BGe3wS&54&YMY
zG5{i=Tn=0TtODK*TnStSycc*6uo`$D@LPZfpsZYj_*&ro&i!?W*8ppQ>w(_}ZUEK+
zEdQHu_F7=KV&Kn0-=vfx;*xRzt?oWt38|6XQI1-Ld&;l<DEQ;kT^FLzt=*zzKPVc?
zy~=h_AH`KlnQZlny1Xl)fh2v#bp@IvP<9yxl&oSZ<H}A&M{FsB_|?FZOgBZbroU^I
z3%V$;95Zg|Q-^YFF${s;fZYdqWf<|KlCkOU2R}JnwF}pw{v2-{dv57dXUo3g_>y3n
z)B!S$dVu3!-ZvWm)D6`-fMv;(e1<V~0Cf!Yz+r&n-vdwwv>N|6BL61fW?((w1$@8;
zzz_TZ5ClR%7@$~1fGE%d^a8g48-Yy#dD(*a1I~2}aU4hhNgxHJfvo_`Zg=7xPTYt1
zc3>CK58MIl1~R|^FbE6*dw@HEyMW&TJ_xA5hX4bRZIi|IFpvXA0QUHUz!<O((1Ck_
zaUc&AfFi)>KZf|jK*_nc5SN|n`w(*^vD`t#hkz;P{v(J#>RexS?8TFZX1>(-?AGUg
z`rTE(Ap0GKUnM9Zix|2<8+Dvv%<y(f_uWHgQX4Lh^_MhLD_U7gH;Vn)+<10OTkjdt
zN+#|%2mC(!Z@ni`&RgY@wz;U4tx`6>-qT+m&g;1y+JOOMTq|xK_J>BhN28;Gz(}`0
z8_K@TCMGFfRPNJrCBrmEE#HuCmb3YcRUXj|pUuLUR+_ySFpT^jy*OgrXZn&^D?6Z1
zn7)3!Sj-ORHTxztX=z1MAJ+4_b->q8PEDlbw3~O@7g~wWWh|qlv7Rd{_899vk1zlU
zn}8%!=b5;s4%8rF)ICXHzY@SW<LGM~Zi4=MKe9Of*$s_L%}u5Lmjy{<7*aciKk~2#
z`q%>XeYi8{KGyKg1Nzq^=RJI<OZwE+v{x8v>Bqr?1|G)IDhAqvb7Ct#)`9+Ycn|fW
zTl&=fM*xOe`kepR_7kA1gM(sA>a>+E=vs^Sa1^?wzY=BT{6iuBIFNvrRA*;D6W1~5
zqk`Q6LMlpj&lqSQ6<izOvW$Gn3R_5p#?*1O4XMMplk>WUau&39z)s8JY0J6Z4IO%$
zvKKwR9U7IokwM1duKn*t?Tff7!S{R&5bYqi9|N~p)X{?(L5KFGM0#MAPGV+>C^v%F
zKBVVR%0P|*`-OJY5YkMvLKaVWFfUlR9zmLcJB}qEprTX)zkJq0Y$DDfR+LB1F%Px!
z$C?(wZwCC9ZLT;YH4l_O%AE|&OIC8Yj<}ub3!ypv188|#M4a`S2hg`=neQYI%}Z8t
zI48=xTKawH<pQYE4&LLmg`wPs9%>xc&!hfaBhgl1sHNXn|5S4#o|C^}(B~S9_N?3b
zQ&-CIuiB$Jk2;Uq5-256-k*DkY{n7ih5Vw7ky?&8LUQ1R<65&hlD#p%*cl3Q9zK(_
zIWKB+)72)wUG}m7jHp1nQqA)hf8I%8{>fn75CQgL-jeeYXRPLVK3GOM%mfqA6kNY<
z!taDV8*YIuu)_5`vO@XnrCdF}(3QN!kH167XgrlpM`PY_uRrJw$D=84Z!j45#{J<y
zG!}{UM8b*bsoeHVccgo7|L$G4?G31j>O0^5+~h>bxOXaXi#lk+f>dK81-%IKv6O|~
zs3whExd5ZsRJZEZw#qtJ=$M49aXGUWXV>J-dr#OO)H|qzYyDHLe{1P)wcE3dwyfr)
zIR~)6Y`Wxup?S$lPAmGH_c$lJqi)y#jOfQfk8>Pt&LZSc18rJ<McBj5=A24RxQ6b$
z{_n*LxTa^Iy-AC|1d6m+DwfV%3y$qUOHbN+;%Tm}JEzb2>dx8uf3782J=67rCf6jb
zEMdPh|I(&)+yC@GevSf*P;1Fn)0e)Wq&-Jw#RE0JK8HU2pl<0`%OB;9Gl{IDXy;MX
zdd#-JIO~aZKn-E%7WT)if7UI1s_TDs+P^%k_Cr}{Im;VR`?Do0pqizeI<A$9THco{
zPfZ=_iHCZE`13}B&`5zmGT`m$iTb^Mgmh1|JKCEJPM=sL{ndp3vc(m&u9?2v#02V)
z^rr%`?nJ~J4|Vr?!>Mq<+Z&Jgy`fN~8|pO?j0J<!-&<OBaQa}MTlT-RwAe3X^`g%;
z^GmGlZRR@=o#RWZ?JMH%x&xk<HHZb%sc1MA@W!Lz5Qn?p8^hqiaF6*Dz5aAMod``&
zZR*p_Tw2dl@AW}b<!wF1)vj3Ma|V|)h@x!9ZX&}*+edo{S~m)>YYMYX8dov+2g2~w
z`r*mtRUBz4T=n8Uh&wOfM|!|68-b<9^>-L^^fXq}BUqiyrMnBX=?kq;<!;s7GLkpp
zpJ3Tb?0u(&<ZYwvWnRcP`e2=Ej<vvjiYaG}8bUkyu!4)*GZnpOag-XF(*~Pyx3Iia
z)~=oK5w`h=f#WoKIf{Nvp=a60ybj?ifWIVq+zZKI9_7Q2zZj6P<%DvR#&yWG1km1z
z!D8oZx*si<Kx@&*&VHxwu6es|Y$$cId`4Nj^4GDOAAV`uv7bJD`iigbxooF$>B-gk
z|69Ih*O8e|{`ct<PjO2D`%C|}>iXBdv1ZNIZ~p9!=^KU%YzBmk(lfg1#G5Cs+4<BH
zBS#;<=9j&lJ<r{cGwFvrJc@D(t+CkNzubbX!o>N6w^Ds5Bdg(6<@$i$h7v6PijB@(
z=2v^`>hoEK9@$3TSFHb69GrRKCl_r0$c{&EekOKs^ue7h&+vxz=2sKndiC<7tCg>u
zylm!Wky`)T?D_hV^(mll?_qF{YazAM<`yHGJA-om;QaN_IhTLwWgEA9w20q4#xXtW
z!ZUYd(Kr$-k-W3&qQ~23??ckR_(O0=JyAxg*e_C%N2#L6l+UP;(Yfn6?hNFRV>%ud
z6?3!jn24E|`krgNf^9=oRy!<LkyrW5Ud;7e8`EPhC~KgoLYH&bMuSEdYbn((Rqdl=
zjyjt!>O`Oa;YDldxqZ*QUM@)=X~0iy8`rK`*~ZnLs@mNbwsEFm8~0-mW*FY9KD0~;
z@0h#BAF_AQ!BmTY^9NnakiQOayZ)pl@__Ts1p0+1Fid;jL_`19a^Vzg67l>6>0XU|
zxAd#m-_j1GA7lc&(ECH#7S<yD9j$|;PoIKY`nRI&zf<uc3__YoRcQB-S({mmj`kiP
zwm<30Al};gx4Qn7F(!3V0laXGi7`Spm%@BnE#LF^(u#9BUuZDK&hBofY+Gs}6BxAZ
zP-@JzuzTqN9ES#EUyLK4dx*4m1Ly<(whDKi7M{HM&Y#=B!-#EJ(EFH2zo)?s*A7C%
zx1!BFo<EYZ$GBa2=2y!fS1dWyn)-)3j@*0Pi7SqWR(>~0hx>n>(%*)Mxi(^u<6n%o
z0dT?D&6@Rnb8f55O5&O7U-%z8re9ru?gM*jxTt=AyK9_(>*5&Yxhl8qe;dl~cV5hA
zIS-~>?G`lgw#|AA`O=Fw70dZN&sZUh%=fkS&c>eYx6j?%n0qI8I!H)T4ehh6iCZ>o
zFmwB~Le}&ZW{>9NjKYQyZNHW`CbZIqf;n8)^CKIAet(G14}noQLdQzkf(LF5&Gf9B
z?1|i{_GRKMr&f;mBqk=*{b8S9wKEcB(=rPBLHmf2sU`-KF;B$T?X$9_G0n2|cT3i6
z<Ch)Bj$k;)=a+Ut-np+zjW2c`Dl=aaDLh-&Dq#NWSpK=dxLv5NI_+1X-im4^@9KHb
zzZ;L!USW{?Pdiapj{6MkO0fe(YrOT@A3i^y`tQYam+)bPR_#xlq+1kg`YY`ZsRNr9
ztjd_c^*1Z`i++CXT#NrQj<sTCM4#!Qy{cn+^2nZZpf6$M^VnL!X+vAj@`&FuZoDj~
zJM^8xC(2WN^Pm4MoKxY-aRPthg#Ke4SvN<z>U}?FbG^~Z-W0Vk{(0z#_iOrdJED(Y
zb>!L?qpMIaDcv$gk`or6|Jrrj!=HJ*tMvSjzyJKtPwmJ|KK(TNTk<Vq$vu1zi`@S0
z^bgNgw+d4KtU)2xu2or?$5~P8pLX<HTfesc?V^NS47jCF+viz;p&fn4O)CmF_+|Y)
zNSD3bph#WP0i^twliH5HZJgA-p^`A4`DYrB|F!c9)wXZtdhdVkxa*77{o2~=2QMA^
zpuYDKSt}{G^)38L*IUUa^y%yF{cvRbvJd_8=PU2{@-rJw?m#P5*S)^}k2v)g9;(WT
zQSRV3Ix)%<wY4Wk`K?Fg#3-$hMkhvj-l*QM#nL0T{Wff`*ta}%_mi=^5C8Aw=0$z$
zpD4e;Zf;eBJJJ7n+c(TrSQhIr4P(OX>0>ee_<o)#sc#)puMhVisgI(CFo3!@JAW=p
zwEn#Y<L`Ya!W}-h^hZ(Fo`w;(8-Ed$X%#sBuSJ0y0k`xYM%iCEFKtJk-?6j`QvZ@R
zL#whf@Aq)=pU%zwh#vr60zM9W0^s`elZZbBeA>DH4C04?&jXJDp9PKpEc+<p-v=IZ
z?wR&DuD{^ie-ZJQfIkGD0R9O0GVm4PN#Lu%*MO&huLFM!{0Z=<z@Gt61Ah+u1@M=^
zQQ#ZEGr%!`y8N#I>a^zo&Mn^po(KL0_%`qz;Jd*001k>5f#U#qJ%RXT;BSGy1O6WP
zKJWwJBybA&2jCUpAAuhNr-4_2e*#_uegwP@@O|e~SctRy(!14Y??76nE73NOHn2?3
z)Yd%Qz<xS6zP!11A6JX3mmfX#*he3D`ToO`&%Ja!QG7S*EM?(UEC76fyhj#4TIsUC
z8fxvD(9YP0;^MTZhy1kkRJh=j?_i~%+xMWZB6yD9GkabC(Ubg+A_GrA*`71GpHu%`
zg%7gBHI53;RoM26s#Q-U@GNIS1GAZ~9IW!#iM^3Q_!47^2fKbgtf+k`$#n?lI<D|*
z&+ZC~MxOrT8=6bhsdFxWl@rSH%bA7QJNGD`|3&IQ-uT2_7wWHvroJ867~(liDX%h?
zy(;4$EP?tj%-+C!YA^P#Dc;<FYR-pjt@?N&^)Gi93Xs+@NHuLm;khnhL`>pjJHPGb
z8DkT>rSxzWY<p4ozj<<<r=AaBjP%*1ge6$2XSgOs?my9&K^dvZZg~4Sq^@&v+Sxxn
z4c|~n<of3Z++GQ|rT@DqOWR<ff*c!e=`R#r`JOYOPn}w;|Kxk!#G1K<9~ztm*Ex>8
zremQ*G8DiU5IxbjHyjSdaB8<F;0^Rd0{(ESCmBlwrl-0Tdk2Kv>bEY|cNFb*M%<pG
zUZKQ)B2Qv=2P5HVum|Ur<H3|S90{boJ+W}mn@sw<6aL;vDj1ATPh|@DMmeJwRCyt{
zMcRJ=A{?I{@wu`m=nweiVQ^DT<BMj?`IT>0Ge)_T(-Ovrrsj;IX=U*liTg9Y@nf=P
z?mlvH4yK1BQ=$5r?aP)5MqaP_s6~7Wqo(y@Ht)Q^KG8mwH}iw`sqh1xoC%MAO7h&<
zxmaghh!1dfYX!s7)TCyPTgHT{7cG32BmBZ0G+**OKO<eMPK2Sja@IH$A7#G_#gga$
zaq7mEtxn~xE3Jm)d3|obx<<Sfow}(i`6gUn0W3A~?ez1Csw&x+qNdfU6kUeoTj66E
zQI*QCsBUj1!T#4!G`*7Z{B&Rmm=0C7u`fj>KY;bC2pf*~<9a!6lIOc&A*&=~QPrQx
zTf_x`OTLamtFwje_xi;D8#?Ovg~VE<;zO^TbFTfb3ybM35Z_Lo4PRDgZ8_Kfv=6;T
zu|0y7AJ6D6mHi0C{_^L+2A;p>K9z-B6LSqLcelhY!Lv>ax%2NVBro$>pNoBcYM@un
dc^jTRPm|cMUA(i`6{(Q^>34Lb_1-##{~u^w9V-9;

delta 1688
zcmcgt&2JM|5Z||1uU%*DI3Y1Lwc@oK$cI#8Ly{(?4NF-C64N+tBpS*VYV38`%Jy2`
zbxJK%a;i8WHF&7tfK>Iyg+s8!6(cH9p^7S1B{;Sce?SgKswx$C-eyhR9>}R3>E}1|
zn~(S2?CyB&7FjE;ON4A3c}TbytyT*JcG*Fd7aOVTq{f{fP6%@$@77`R?a1a&q>Dij
zp~^iH22l$jFt`Pq+)%BLgA0y?hwMgjpu+);UJibDT(mMm=&&|g<0TS@`}}_2dq~Y7
z4kIK)%^B=Jj_VwPh9+=5fq<BE=m;e~LSxx>evIpBgksxg?DbQ7J%?)skwvJ89DL&p
z3Dym<l68`XQ}S)y&hgW!!tCwG$IQzg{3$`o6{|}wjukwZgproeYl7)s6sYyO?Xbuu
zlCL4r*Fid5XxxA=x)bpIcnq4|zOE~Hi>rHeC>?OaR$m{;z0rC@=wdyD#tI$D(DWuf
z42R6FPS)%}wig*+2U~CUTwx(P^kdJvjLwwso<(e^hjGJZ@@xOUILxDRr~d)*kT+0s
zAn1Lz?=g^squw~b2gxr9RQ(Gss<!>Sh(^{8fB5GGH>%woJ}`G%xdZb^K4b9*#m$_-
zI0thNHs#ZBZ#J-Nv8;clX3T?cEop^HK5c4x5t4hR$^BPG#Eu^#x4<2ooKuaNi<x(n
zQdwUqrwd|E%d3Vsa$#z6R4HhgvTo=(Q(V=IN;)qZW+kiXVmua&DXSSpD`xVQtg0Ah
zHm{w7?*e1wA+34du30hEvZ0($mkWAcQ?h!dQc#Pgp=2~;Vse6RxqWiu<;j&&x@^WR
z&)LtlqGDjOnpJXTRb9zf%$GZ^<P~aLTfX|D>w!aF!Vt3EK1P#v>>TH0-rvCHMe8?b
zePqWzru=W(wedr4Ttl6{0P_Fe2$1Ie{>vR_xoYe3fAFB`X0SZd&&oTt-wKryZb5AD
zc%mNKogE*7og_HuAPeKC6s}9bXDmdxAV{IL+CEvJg@TCow?s}tW#1C~Fg#TomO-9~
z@u$%HD+ATxui%tRa->9t`S~B?BKqMhJQjKQ1*dJKQ5$QLPvH4jtNSb_t|Cw%JQ#qd
zkwN-p+QP%=&=XkSTf(sugy4va)UfJOEh-Ciocj^&Gx2FO#=D>zt&$=<k6t1=RQ4~x
z11|0;q3bS8FC?sV9byL~@F;nc&d0w6E{)E?N!Qe!14mg86Gle_(Jl|>_#((&oQ30y
W0(}kXzd_qsHz)Ee$--}MhyDVVm^k48

diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj b/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj
index 4c88226..6b18a53 100644
--- a/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj
+++ b/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj
@@ -30,7 +30,7 @@
   </PropertyGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
   <ImportGroup Label="ExtensionSettings">
-    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 5.5.props" />
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.props" />
   </ImportGroup>
   <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
     <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
@@ -114,6 +114,6 @@
   </ItemGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
   <ImportGroup Label="ExtensionTargets">
-    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 5.5.targets" />
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.targets" />
   </ImportGroup>
 </Project>
\ No newline at end of file
diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj.user b/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj.user
index ace9a86..8a448df 100644
--- a/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj.user
+++ b/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj.user
@@ -1,3 +1,8 @@
 <?xml version="1.0" encoding="utf-8"?>
 <Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <LocalDebuggerEnvironment>
+    </LocalDebuggerEnvironment>
+    <DebuggerFlavor>WindowsLocalDebugger</DebuggerFlavor>
+  </PropertyGroup>
 </Project>
\ No newline at end of file
diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1/vc100.pdb b/Part1/PROJ_WIN/CIS565_PROJ_1/vc100.pdb
index b662f859b27e4fb8b4753032a64cb6eb2e33341b..f833075565be29486991e26896cb67d9cd498469 100644
GIT binary patch
delta 37442
zcmeI5dtg+>xxl~8OAL<y8nhxHDvF}G8}a}uNEDUuj64Jt4IzmT2_b}lu~j!(wc4Tu
z!ID<2)LO;b_KH-|mbNHXtwKv}duz3iSZk%%THxMmeO0)>**)iEvQExS@9o|Fql|v+
znVE0roB6)qnK|cVcc*RBz_yM-Ylio}T156l_=mJ|YkB31+3}0!NUyewcNVCk^JJ{d
zmpQUR>ZF-}=S#WN%2KJ9Mr4JwP+HBu%XpnH6M3zXmDJZ!-;&V35WN<eMCn2_D$!de
zS5aR>>6sEslvPvO;vLVNATwmLRLNYakm*#<M7aT}PIO{EU8;~-QX-{D6%Eex^kXto
zn&{lZE#((S;%Z1)BI;$*9phyE$XMi|*N>lFLcadwF{^KFoiyWy{I_4b`>~(qw2iEg
zJj#`9(vJI9O-a1p^A5Gil+DW-1r@TJ_IGId9oq4UeXF~FvufryS!>G&^o-E<gk3)>
zXzf+8w!Gxr=aU9x?-{#;9zCxeH~npDw>?sC9hJSd{LZ5aXmmtGP9Hf(q{j)nW%!-H
zT3cS26^UT6E)rQ>=UI^@l1+1J*qy1eX}b*1k%Yt<;NY!^x-LFgv!HWq`>*Icj{0L>
zTYgOM@#-zV*jaFJ-@$!Y9$3N{6P<tgxC&9C(ggn<99L?`3;bhR8um*noBRTf+qB~~
z{xOP?KVl~S78=*_|7(n%kxTpV-LR%8UYP(WNpW=u#kB^C7bPfO>`~lgptuP@p?G!z
zV_%_4u1=gP^XY;*U3_Xg>(vR+oI?02v||$Gzkuk9B+)V~iS+12Nu)P}G{KUyHuhfi
z#N2t4AAIue^{?N3{LU!;Rpm<F8`ycgrW>RQv}`=OLZ0Jb?<YF-dvw0{B<avXjed~S
zn^k%?Pe*!hzXZ}9i7Mt8t<}Yyt;&Rjo&0#Ji5I`oRb?Pu__bV-o)XEwevb6uFv08*
z>rPv6VUHdXIkKJ!YhBZV;`mck<wzvA9|fmyd@3>=ISn}-QRzzlFNBSk-=8YSMww#Z
zbOut1T!hR-W+AhYImpGxTx1?{334fN88RQa9JvCiLKZ|MU#dA+h}5jh?J*)gnWL#l
zBeEPB%*=5jG6)%hoPeB!oQjM<&O%C&X^2Ww`TqsFx}a}GqN_b#qsy;k%Z~NW%3A{@
zA!|BbkT(3O#82%Ux9)$oAffySnavw_%e-w~bb6hGYkKWqswj-B`DX}kdrNuCOB5x}
zQ~y;JtK*(8#3G|rrf*VT^(0pg9a4;Cx2Xz|Z(f#^_c*?PexgVc?@1oYBN~5gRa2sY
zp-ozar<-``&@*jMZSB{0U%|SAL;7UPgoBC9)`Mr|pAZ?p^~lN(3x@8ZpBM6u?S~X3
zBA-ASZWm5Gp{vXy+lD=oR}~pl$1xnR4v8#A>X9W#1G03(4O{X~jVxm=EbDV)(=APr
zTbi<NY0AE(Dd&Nv+>UMA*Jt%OGCoOaR@PRZ-`u>cc~W(A?bX%Ib+cNkYnCKR<UbE&
z6=t(c1v(WC4RwpE8!8%Cv^1}*X{le<=;=)E)R|VlqPn4eQDa?gd2MZT-HH|RU&(f+
zbm~+zu0p4_Jf75;)l^+mCkbOyKc-Y!<2Nu(t4nsow~#b2U8_$TsBCHVIymEFbr&{Q
zH}zRuU$f%8WsNO$tt}{S*l|zR#QaKD+!j2zEViM~SF(;9+DCfyZHkE;!9x7%Ad${|
zw)~d-Pc|&SFRL(Kb>Y+*=ax^cs+wL^RUE5oS=!WkWlL35rLt&o@_^&2lBy+j&5d;p
zRW(gbv8wv|s^Vx}&6v8fvaw^Us+6Z^)YPnOs&1^g=G>x*lj<6p>YC-^?rJ6kR^$I$
zvlAw8!e{ke!3gtj5;;ze?5oaOm2W|MW<{iTKMF{yZY8fl9o(Pd)x|H6?EbwABP=Z{
zHA+9wSxE!kbzMbO>8wI$8H22DS=K!KG&%4S$)Pka>uSpVlq117yBlUa;pzW3_jYGC
zuo~|ss<`)KF2!xWOO@+ehvND+9V{+*n4R(^(qGM33O)th7Rrv`z1MnqAIjVj)Gr7q
z&5?Xnpz^#~y<woA=Z!~jvDBdD)ugyw%Pal(3F`gXh3EETqNk1aD+r5?@>t2JOi=D<
z)}#DpJfKsq^0y!+%7Z#Q5=!}JX~0JLt>~C2zm0M~g;DMeGu@P@dbi9@x$ZWfayvY+
zvYHG`kuf*39z3gwl@5=}-;S8z3F>U{+(832c)oy+37$2S`zaKj<S^3?Pl|Uhw8Nvj
zt#`lV*z}obhto_{l$n{x&8!F0dSa);r1EzmCYXXc8%$rM0UJyk&@sW(PPv~#VM-1&
z?J%WycZwY*-EBTh{k3H+tF8TW>MF!KJ4MS%GqJQen)QI%Nc?o5RQ?{s1XNIG1L`X@
zU<2x2bWA{PqTEkmKzYMVH&ChGby(VTxA{O>;fakZ&ICM;W<7YmMyzyrRQ`U%1W!<B
zgXilsV1wrYbWHF(NV%WF;PHl;Zg^6?>o6PXZu8;kIvX`T<Te{E95uQsdEb%N+etUG
z9#9VxKOHEQ|0ZGrDyXvo^(`8(0rhQkOhEk`<$ekSDw-T-x`Fb{2Dg(AvytvLA1M25
z6f4clg|?Gf51vPfl@5=}Z$(V-1a&rezC!~xcpgK?1kdA?`zZ{bSkk-chQ~7-z~e9*
z>2CAk={g&&ah#1tkIuyM7Iic00rg$trvs((-$P751$8!{w$p$OsPCg=0_q2p`zZ`4
zZ<y%@D%HD#!!NcUZ_1lg(BFUJq@F0LCr%&aZy2yX>w#(AKJX+s{-d$C@yA~g{x}jm
z8taN<+kH-pW303;BlDY^Sr5l|7^#k<%KuG`-teq7sI%etE)Cdld=DKHj_*_Mr!X8n
z@1`5a)Sx;nj=J0Y`OUsKmRy;Mxh&>p)`REo#7c)p<v&79@C0=>cn;El4W2{jnBb8j
zA072q%kCf4d&5jSJSpCFSR8e?`S5g|jrQE{I2)B^1gKIcvmQ`6#7_rG<#Q1eP(hsy
zs5~050d)jACZKvy?x#?olEX|pP$}MZn2mI|`9RrcqoRylS;gGUdhqllRysT?-wQFp
z6V%z@=}iMRc>17Yf~SCTKZU}R9A?_#N%5}3Y^1x*ho|dobgSE?I*;QsG8?&>^?*8x
z_~}5Y{LzRBsG!aUR3Q!6fEs{~38-Tz_fsfP$zi4)s1)xy0HwRl3{<P@#dc9<F1DS_
zdO!^%emYPpe>`FWDyXvoHHZdmKn+I61k@19{S*pRX=0dZ2P$dyA9<LdyUh&Lq}!dY
zm!c)HOaSF*)&uG!;->?p@+TuEpn^IZP{U}z2Gl9&n1DK!azBLu<qb34K&5)u;d)7T
zn-7%zda1N7Bg>ncSr4AmiIonI%8x)y@C0=>c+Q{!8$4&CV}fTS<$el-hkJ14-E_m_
znf*s*Bi(I2JYBDs7Pu{M(J>jhUUD?+0aZl&bf8pz6k-A@sIvi8OanHcO3*O@RZ6*^
z!hrIInQov`z3Xtjq`S=w)Fd~cVx_g2nW3D_dO(dOemYPp|4GCIR8VIF>TDXY0W}UC
z6HwzR_fr^9o_Es?RBBKimN(sPK2Y}MEi)fNiMg5e;3+3oIy@?WE@FZwsI$Ry9u3&w
znTU=Fp7Sa9Qz$&iVWu6P6z@9BM!MU4c)Bicm9Dc<O(rh3`4g#&Sr4cR;->?p@{<u0
zP(hsys3|mH18OQdCZMKK?x#?olEX|pP$}MZ07`e88K~&XPIqsySVr#N9L;(_T}1qJ
zpj3V)Vgf3tvjH`W25dmhM#luy9LoI^3RKd&X$LALs188sZZiY5|BG%w6^+Tr^^%)e
z52#CspAMADUy7K33hFeVE;oH~<}w<v0W}{TA1L*W9sPp~zT2%=717Zd`K+{~S<l5!
z@s~&4#VY?aVsdd%r@7ep-K3o~U~};^==d(yznk<U%Ka2}&U(X4_c@#DU59g4cbgAS
z*Q<&N54)YSv5b7#E#_v{1M0`bPX|inUqDPi1$8!{UZepVP`l7E0kxZQKZOD14Kv+9
zrFz%l9!GbZ50rg2DyqxKM_b*@dhqNaRysT?|1-n{Pf%xr=Or4j!Si!;Oz`}IazBN^
z6HN{?-SBv3gGYD{@aS%Hh9_2%k;iq8W<7X*MXYpqRQ_ee1W!<BgXh;YV1wrsbWHHP
zO1YoH;PHl;Zg^6?>j01LHg|Z+V$rH3qUroleay|Q2hZ<_l@5=}{~j^H6V%z@>7W4{
zJg=c+g69vE`zaKj<S^3?Pl|UP;L+XY4o~sIjNGTVnf2g#gIMYCsQjN06Ffnk4W2h?
zzy{A=bWHI4nQ}jc!jl|k+Tls@uERyN?lvEut{2hs93RaUk8*iN_Rs#sh?tvM4=DAS
zZ5^oKGu(zxF$Z-vp!PWcwZ9vn4mbmq@NW8n(t|quw0P%lU0lwsyq`i(1-B;?Zd}B>
z{>Vt@Pl<f&w=QCCX5qsBJiq>+qnk6~!Cbfd$+FCRebUXWcPdm9KmAlt`GtsSZVl>e
zr$P-4*yh$+bWC$=9p!!s1u8krw4Vwo-gTH;b+?&`+Pu~YsA#k<6IXhUW<8+liJuOX
z%3p<;fC}ntKrNvG8&D1Cn1EVJxu3#-@`jmipi;f-0F>@FGf=Htoq&qPMrQ&jN3$MK
z%ZZ;3l*%_FCZK{k8&E4~zy?$cIwqi2QtqcPpuAzG8>m$8Ism1+%?#8De{ln<v^FC^
zxtaBVx`z1aK&kwv5ED>Aoeii@(|`@AYtb<Qbsgn?3I!@T%(MfQ;@$3k1rGQ67g4&~
z%s|b1RRPuY7Gv{oogj-Y%*Zb;Ihyq#yOEgckg5D<5EEoUoei>^Xut;9&FGjQyM=N;
zg+b;GGu@D-de`BBhwe5XGV62tSXoA{p&ZS6@Z3hMba+($bBGC^pw0%*=V`zO&uVl`
z@Z3(hpTgkrhM8`7QoZW{kM1^ic#56Ao#=KG+uc3v$A1np=4RG|XAQB^;ZgauhzXvc
z&IZq&G+={g9Xcj>)>H1MP<WEVOglU&-gUTM(%t66({(93cC*t`7AwlkcmCbXdO)=k
zKOHEQ{}N&XDyXvo^<^5c0d+SzCZIM_?x#?olEX|pP$}MZ07`e88K})WoPdfJIelRI
zpFKl4n)QI%MErE1RQ{`o38<jX2Go5tU<2xF=$L@IpK?Ei0p$%d-9V*!*8wQqZDyeM
zeA^AEg(VsJT!Wig52($=PX|inA3{t(1$8!{zCi;vpdLoY1k^Vv_fsfP$zi4)s1)xy
z0HwRl3{?BAPC)UY%#8ej$jz(=)FZ@C2TJ9)ASR%KIvY@r(tr)9t>~D5`VQrO3IodX
zZn}X=4XOiBy4%b^?RER+Z7eh2ymd6|0rdp&(}7a?ClM1+L7fe#@6vz`sPCa;0%|+u
zehLLD>D{ygl@e43pmevHf!g5)RJ5!(6Av*R&3ZuXAbvVfD*qH>0xGDp0rfNu*nrxJ
zjtQt|DECtsP~I@p4OFUk9e~o^W(KNbwcC<cR+NcrsF<5s52)vfpAMAD{}?d=71Y^)
zdVvOPK)r~L38-C^`zZ`4Z<y%@D%HCVK<REX1J%CH4XDzE8M%gXGwT7hhxqA0sr=6n
z6Hq~&4XBrBzy{RM(J=w_3(EZz3RH5KX$LCByABUAb+`FI*`I)ymStjji@BNg;Q1A?
z(&170mk|>@L7ferU(<jMo>$N@!SgERehP;t;obDZ(>16L@aS%Hh9_E_k(tNQtOw8U
zh?NeH%KsiQ!4uTk;OU?N8$7R}V}j=ol=~?R9&ecGh9}j#4)Ex1bA~5YmXSZ)I-2$1
zd4pK#@TmNs5EDE>oeiEhX}|{0UUW?G{F!n;g~8(uGu`l{de`APNOzkLPuD-6MBRR*
zG+L9HFI~Br^?-Vt_~}5Y{9h3hP(hsysC_hG18P4yCZG;b?x!%IJnyC(sMMf3%tpG~
ze4y;tK_wa48Ykvv)`RE2iIonI%D;n{;0fw%@cfMiZ1B8`jtQRkDECtsJl-(V4Nt0f
z9pKU3_K(1mk&Eb(QWvuxJRcG(9UhhcJ7R(-sI$TI5e?YjIf#x4o<o%T$pViWW(MHt
z;@#r#y#v%Q7wB$F!m};w#=OU~;|H#1<Kty)Vmy}1he_;&*@I1u&lfo`jLp){5jm!n
z8ya>;{tKJDpFfbFVLVn6yAk(2k-dF+?}z=^Ns}Fh=UyQ47`x<tFixZh_q7jZLlJf)
ze&AwuG^T7STNHow43XzA5xJT6Umt_(`$=3Ln?d7eqf94uPh{5M4J9+lA0_f@3?Dd}
zZEfbV3vaH-i~Z^3k^EJ^PUJ^Xkz<QR7NK_qI}@*9fBOHRlOr%T{UUburL2riW^ZGg
z-!IVdXV@>{=}(CivhnaPb{a0ALyg688aPYjB}Q`&4MZv2doH_rHKIfV?d%DCDPGw{
z#~xfFa`4l9>);A>Fw#N(hm0s^l*rp`ZTt`$iN3<|Gx=->O{b%D`s=gMr7n;6y~L>h
z+91-Ky@uas$Kyrw_(6$yIWA#8*LL3U<SIr;#chOuK|(KEXyl2!O6UHaf?{-+(&$wa
zFwSc`t~~Q3I)s7u(4D@3&CI9qUIy?iUYS5hE<m>zVW=M|vWmJX3q>|i_Kk^n_)L-b
z%_q~dBiI0$SpHxb8(h=7pA?GRis75-*nPNc;&PEsWBmLqI*IN^Tsr!bBDK>+j>2n=
z7#M@je%`x|P92&|2QlzNI?2|$@(v#BpzV80MV@1jkCwpAaUws>C;kyCxAy@MV!H>~
zPltwHM{iFNc`X+vPRGUcdg*LNOne70%7<`qKCVBGQ9WEO62tg4=<TLsgL|@N<duB%
zx)Cqbh@3r>P7I`M1s#Ezt8x4Dwb(18qy340`yxh0gLl!hdx`&&p(2}!@0+;rxfyhB
zG9Cb=CkBb^I|g?Th6RjHX0O<<RzUo<Z21i<>v{k82}40KXAxf6$Y?*pi@&={WG5Z}
z;CPYWFrYr;IaAO%1@|wBqx2a@{S|uNx)@hpn&e>}uZwwOAue4)uYZm^r;ifpPe|UU
zBgaGl?L;0_jpY=(hF9A7GE^^FDzci<zBdJ9aC8IVc$ZE;MF+O!5XJ!_e~gK|aSq`=
zo%lb;D6g-FMBK8blGsD_>-2PPKRm-5BXsN~T)qbvAEfTsV@1AI&d51leZI&q&t@K&
z#ppjJ@&~-Nh)$e`N6+ZR_GTE_!TWcgfHBx9CS)I>yYT`B%V5TgBL4G+;{h7Fla9n_
ztdQfKjIInwX5ylk8O<{g8>bWdhO*20e33h-{|@opkHNc|3EO4N2gGzRGLBI_O&cR{
z^LSq8q5nO+vt%w|#d8gXoLTTTf>708A7`s}+&qqWeez^}l>c&wW^`Y=maW@yZxO~X
zS;1&0vi~}+>_sPTx|mMV>7^HOZsVoljBXDhikyunm*Gjmp|*xU6OUdwB+12I2DcvL
zrF3*2gSedzy?eY!bsRT+K8Kk1#67E^l@L6C9l(ua7Hkz6i;=tN&^@^EWX|<%z;`db
zf04%Dq4PgwUU&}IzePvxU=VMSe}cMxN8=%kKEr79M?yR8Zh)0HW^>-xiQJ7xU*N1d
zaVN$;V1;;-LbVg|-=_f+klcg2^7~@wf^DU9@)kyhG{yK)=g;LPeii1CO}TPf5i`t=
zT)B(avpaIPlyAv;ye;pE+`}u%;T7fZigI{GIlQ9$Ypy8c?JElFE9>I+5v;8hEXXWp
zAC6|xXMJrQ=DOP4mG#91dA-h5_cb>i%9UI>bckhCvfKEF^q0>krHzHS;r3+NiR;!@
z4A?p&`e?znNL${v!z<O{d!NJiK8Npp4&VFyi{ATO`hQ!gcs#`o)Mr_}#<6m(<2Lgf
ztTQ_==PPhSlB?LMmV2heL(bWfx21e}ay82u#iOQ3-p5w69;<oChTO_(wnksgw#`_R
z_hdn2{I(-E<y{@w(6%KnXOOy)<N1ehTPr+Kka~ohb!kABZn$Af-l>sgTgxlU`rNr7
zXXq|k<knJhvUpNK`UIjZs2umb(!6chBY9Pkp`neU*O+X0ISr5M6KUBRX*yJ}wQtk@
Jg0+=>{|7DMG_L>v

delta 3434
zcmcK7dz{T>8UXNf=4PhMg^)tVU8~F3ZChOziP3IMx=@snh+H~Ksi-rQAu2QyiX;g|
zF|8SrP;RNV-R&MV-EG><Zf$L=R^NALkIv3!{k7-w`Tm~geV^xjf9H3Z^P!}wNy&mn
z)7xf`h(scfMIC&Gmv<`Kks3c;n$R&R4|8E0B<jgAQA^Yl<C^Od^NGIXJ>>BNCrC7e
z*nt|22gZrMgVn=)=t=Y?@|v-_FaE=cYA#4@lsM6$>WRKY4E-$+#rY52bda7dN%P71
z!A|;PJ@JDz|JeL<UY(lrCC#cISK^U{_`ig^lT?o<5B{%tjT8U-T<(9;xA~7edBKJw
zs;@lEg*y<|T#^!rT*^QOGmJFz>142)3MyH{TGsJ2>v@I^;rVZ56VLJ-o7uuvwy~Y(
z*})6E$V>c|m-!vP=MTKXtGveRyuq8i#oO#;7w_;c*;Nkj@jiE!jml{kAL%JyO^i`o
z!?ld&I<997H*h02F_v-M%q_$zVLTJKmD`xeEAG>)yvFOi!JE9r+bMn#c1m{fl&^G-
z1$Qxx>C9jjPqUt9*uX|M@hssr-OLuYvW@LL&kkPTMPA|#`tYl8F1kB^q)^h0I^>eg
z;T%S7a>&C=5vfZ9>Qj$~97$KYa5iUfCP#A=CzDST8q<hlXv)vS8*viN_$fc5J*^3^
zUJIJjiUQivhM#aOZ8?q;If3K(1*dT;;nh8b@G5trGadOQXAytgVHUsUA}-_t&ZigW
zQAAIAa4r|qhu&O5xQKrA<x(!AnEni600X(4E4Y%u3=K<%NQQA0BN@&JVqDENjAAs`
zay{2^17o;}8yUw~ZsBIeQ^KuGU?R6MiBfK7a`@^~B~!SQJ1FCBW^fnNn9e=SWDc{r
zm$}@}eavSbD_G70Ji>!4U?C5&h=*CsvT*B`vV=!@jFmh=IjiU#^{nW^*>t5Fg`7t(
zA}Nu`K5>vZil*|@#c#!X#7zFA?PH?m{w97&yn(?G#jE&AUWX4AvW#oX&lFR|W#T$k
za1Nj9`6rd~T=kyfXzH+*)9At;J-x)b{6pVf+R5`-AO8OPOBOJTOvR5ZXd<3q{C9bQ
zcoI{X&j7O2cdLEH9OKLQi#*4C8yfMAaXqo4S~g=DLmGMJ7mDdA?*A2%dbH$^7EKk$
zvyF(xV>nlS7TeiMAO5U-m|ANtHr{KTE#D|_A?`E2QvR*j*tiyJjK2`Kh-FNRJAAES
zsMt#3Sn&(vw(`Gnit!S<v6SI_pk6LMNh3OOD)Wf3oFlBwXPo><@pEg^#QL=4aGI;d
zzf+hL7Esq@D{;TM&uA?FM9dW@(_DR~*iby0@a5CR9%36dv6;CvAw#V<bC}J&%;i4r
zXI|9vzew^iOL&C&Jir1TWFZf+h=*CsQo_G%m+>f1u!7}0PB|-C#cz0$r>I~xYp7&x
z)boE|vWuOp;~n0kig)>l5BQKz*v%e3<ue9wIRhzW5|g=|DNN-{zNQ)9kmBy7hC4LE
zp*HntKtobAv=Cc!94FGAe&$aW&!Dfot5`^Pia4JOxRCzp{lqGdFYzg^DzB6Gd2ZQv
zO>?6q)oFfS+Ny%g{Agx*R?)uPX5krh(>`u!`o9((<v81qW%z9lGlR=p759ynEgH}=
zI>>)0`sPfCmByl_v6RwSYH2KOaV))R_1=sQ=@l>U$taGNcP<)|HM3Uj<9oFCTVTud
z$o~Clwn;5<@aZtITCvlH(}Yj+vKieQl=m!HnY%icQJ6cUQ>7g%f7o$%X4rA09rNP<
zy<?>vgRo=6Nl80a^vtidCO4W{kyTJ@O!5`nU7U8{vJ34PKH;*%>I^Yl`cRkMzk1OS
Pm)$!jI;=bz+n@VAOgo}h

diff --git a/Part1/PROJ_WIN/src/kernel.cu.deps b/Part1/PROJ_WIN/src/kernel.cu.deps
index 926dd55..79a2ca8 100644
--- a/Part1/PROJ_WIN/src/kernel.cu.deps
+++ b/Part1/PROJ_WIN/src/kernel.cu.deps
@@ -1,87 +1,87 @@
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\cuda_runtime.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_config.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\cuda_runtime.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_config.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\sal.h
 c:\program files (x86)\microsoft visual studio 10.0\vc\include\codeanalysis\sourceannotations.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\vadefs.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\limits.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\stddef.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\channel_descriptor.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_runtime_api.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_device_runtime_api.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_runtime_api.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\common_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\channel_descriptor.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_runtime_api.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_device_runtime_api.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_runtime_api.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\common_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\string.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\time.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\wtime.inl
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\time.inl
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\math_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\math_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\math.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\stdlib.h
@@ -92,159 +92,159 @@ c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\use_ansi.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\math.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\cstdlib
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\channel_descriptor.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\channel_descriptor.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_11_atomic_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_12_atomic_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_13_double_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_20_atomic_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_32_atomic_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_35_atomic_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_32_atomic_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_20_intrinsics.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_30_intrinsics.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_32_intrinsics.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_35_intrinsics.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\sm_32_intrinsics.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_fetch_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\cuda_texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_indirect_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_indirect_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\builtin_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\driver_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\surface_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\texture_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\host_defines.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_functions.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\device_launch_parameters.h
-c:\program files\nvidia gpu computing toolkit\cuda\v5.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\channel_descriptor.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\channel_descriptor.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_11_atomic_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_12_atomic_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_13_double_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_20_atomic_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_32_atomic_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_35_atomic_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_32_atomic_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_20_intrinsics.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_30_intrinsics.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_32_intrinsics.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_35_intrinsics.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\sm_32_intrinsics.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_fetch_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\cuda_texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_indirect_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_indirect_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\builtin_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\driver_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\surface_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\texture_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\host_defines.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_functions.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\device_launch_parameters.h
+c:\program files\nvidia gpu computing toolkit\cuda\v6.5\include\vector_types.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\stdio.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\swprintf.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\cuda.h
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm/glm.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core/_fixes.hpp
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\cuda.h
+c:\users\administrator\documents\github\project-1\part1\src\glm/glm.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core/_fixes.hpp
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\climits
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\cfloat
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\float.h
@@ -259,107 +259,107 @@ c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\xstddef
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\cstddef
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\cstdio
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core/setup.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/_detail.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\setup.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core/setup.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/_detail.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\setup.hpp
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\cassert
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\assert.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/_vectorize.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/type.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_half.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_half.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_detail.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_float.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_half.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\setup.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_int.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\setup.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_detail.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_gentype.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_size.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec1.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_gentype.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_float.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_int.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_size.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_swizzle.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_swizzle_func.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec1.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec2.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_float.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_int.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_size.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_swizzle.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec2.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec3.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_float.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_int.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_size.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_swizzle.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec3.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec4.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_float.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_int.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_size.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_swizzle.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_vec4.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat2x2.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_gentype.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat2x2.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat2x3.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat2x3.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat2x4.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat2x4.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat3x2.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat3x2.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat3x3.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat3x3.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat3x4.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat3x4.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat4x2.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat4x2.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat4x3.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat4x3.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat4x4.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\type_mat4x4.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_trigonometric.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_trigonometric.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_exponential.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_exponential.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_common.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_fixes.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_common.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_packing.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_packing.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_geometric.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_geometric.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_matrix.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_matrix.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_vector_relational.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\_detail.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_vector_relational.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_integer.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_integer.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/func_noise.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core\func_noise.inl
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\./core/_swizzle.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\utilities.h
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm/glm.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core/_fixes.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/_vectorize.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/type.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_half.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_half.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_detail.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_float.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_half.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\setup.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_int.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\setup.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_detail.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_gentype.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_size.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec1.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_gentype.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_float.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_int.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_size.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_swizzle.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_swizzle_func.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec1.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec2.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_float.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_int.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_size.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_swizzle.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec2.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec3.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_float.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_int.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_size.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_swizzle.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec3.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec4.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_float.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_int.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_size.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_swizzle.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_vec4.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat2x2.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_gentype.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat2x2.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat2x3.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat2x3.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat2x4.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat2x4.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat3x2.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat3x2.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat3x3.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat3x3.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat3x4.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat3x4.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat4x2.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat4x2.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat4x3.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat4x3.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat4x4.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\type_mat4x4.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_trigonometric.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_trigonometric.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_exponential.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_exponential.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_common.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_fixes.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_common.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_packing.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_packing.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_geometric.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_geometric.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_matrix.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_matrix.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_vector_relational.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\_detail.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_vector_relational.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_integer.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_integer.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/func_noise.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core\func_noise.inl
+c:\users\administrator\documents\github\project-1\part1\src\glm\./core/_swizzle.hpp
+c:\users\administrator\documents\github\project-1\part1\src\utilities.h
+c:\users\administrator\documents\github\project-1\part1\src\glm/glm.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core/_fixes.hpp
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\algorithm
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\memory
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\xmemory
@@ -415,7 +415,6 @@ c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\emmintrin.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\xmmintrin.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\mmintrin.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
-c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\ammintrin.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\mm3dnow.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\crtdefs.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\mmintrin.h
@@ -514,55 +513,56 @@ c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\sstream
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\string
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\vector
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\xfunctional
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\cudaMat4.h
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm/glm.hpp
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\glm\core/_fixes.hpp
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\cuda_runtime.h
-c:\users\harmoli\documents\cis565\project3-simulation\part1\src\kernel.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/config.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/simple_defines.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/compiler.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\cuda.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/host_system.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/device_system.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/host_device.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\host_defines.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/debug.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/compiler_fence.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/forceinline.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/config/hd_warning_disable.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/cstdint.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/discard_block_engine.h
+c:\users\administrator\documents\github\project-1\part1\src\cudaMat4.h
+c:\users\administrator\documents\github\project-1\part1\src\glm/glm.hpp
+c:\users\administrator\documents\github\project-1\part1\src\glm\core/_fixes.hpp
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\cuda_runtime.h
+c:\users\administrator\documents\github\project-1\part1\src\kernel.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/config.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/simple_defines.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/compiler.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\cuda.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/host_system.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/device_system.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/host_device.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\host_defines.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/debug.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/compiler_fence.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/forceinline.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/hd_warning_disable.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/config/global_workarounds.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/cstdint.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/discard_block_engine.h
 c:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\include\iostream
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/random_core_access.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/discard_block_engine.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/linear_congruential_engine.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/linear_congruential_engine_discard.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/mod.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/linear_congruential_engine.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/linear_feedback_shift_engine.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/linear_feedback_shift_engine_wordmask.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/linear_feedback_shift_engine.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/subtract_with_carry_engine.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/subtract_with_carry_engine.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/xor_combine_engine.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/type_traits.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/type_traits/has_trivial_assign.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/xor_combine_engine_max.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/mpl/math.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/xor_combine_engine.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/uniform_int_distribution.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/pair.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/pair.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/swap.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/detail/integer_traits.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/uniform_int_distribution.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/uniform_real_distribution.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/uniform_real_distribution.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/normal_distribution.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/normal_distribution_base.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\thrust/random/detail/normal_distribution.inl
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\math_constants.h
-C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include\cuda.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/random_core_access.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/discard_block_engine.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/linear_congruential_engine.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/linear_congruential_engine_discard.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/mod.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/linear_congruential_engine.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/linear_feedback_shift_engine.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/linear_feedback_shift_engine_wordmask.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/linear_feedback_shift_engine.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/subtract_with_carry_engine.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/subtract_with_carry_engine.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/xor_combine_engine.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/type_traits.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/type_traits/has_trivial_assign.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/xor_combine_engine_max.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/mpl/math.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/xor_combine_engine.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/uniform_int_distribution.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/pair.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/pair.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/swap.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/detail/integer_traits.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/uniform_int_distribution.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/uniform_real_distribution.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/uniform_real_distribution.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/normal_distribution.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/normal_distribution_base.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\thrust/random/detail/normal_distribution.inl
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\math_constants.h
+C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v6.5\include\cuda.h
diff --git a/Part1/src/kernel.cu b/Part1/src/kernel.cu
index ee9b9e5..5af5cc3 100644
--- a/Part1/src/kernel.cu
+++ b/Part1/src/kernel.cu
@@ -89,19 +89,46 @@ __global__ void generateCircularVelArray(int time, int N, glm::vec3 * arr, glm::
 //		 REMEMBER : F = (G * m_a * m_b) / (r_ab ^ 2)
 __device__  glm::vec3 accelerate(int N, glm::vec4 my_pos, glm::vec4 * their_pos)
 {
-    return glm::vec3(0.0f);
+	int index = (blockIdx.x * blockDim.x) + threadIdx.x;
+	glm::vec3 r;
+	//float mass = my_pos.z; 
+	glm::vec3 acc = glm::vec3(0.0f,0.0f,0.0f);
+	glm::vec3 pos(my_pos.x, my_pos.y, my_pos.z);
+	float l = glm::length(pos);
+	for(int i = 0;i < N; i++)
+	{
+		if(i != index)
+		{
+			r = glm::vec3(their_pos[i].x-my_pos.x, their_pos[i].y-my_pos.y, their_pos[i].z-my_pos.z);
+			float l = glm::length(r);
+			acc +=  (float)( (float)their_pos[i].w / powf(l * l + EPSILON * EPSILON,3/2)) * r;
+			//acc +=  r * (float) (their_pos[i].w / (l* l * l + EPSILON)) ;
+		}
+	}
+	acc += (float)(  (float)starMass / powf(l * l + EPSILON * EPSILON,3/2)) * (-pos); 
+	//acc +=  (-pos) * (float)(starMass / ( s * s * s + EPSILON )) ; 
+    return (float)G * acc;
 }
 
 // TODO : update the acceleration of each body
 __global__ void updateF(int N, float dt, glm::vec4 * pos, glm::vec3 * vel, glm::vec3 * acc)
 {
-	// FILL IN HERE
+	int index = (blockIdx.x * blockDim.x) + threadIdx.x;
+    if(index < N)
+    {
+		acc[index] =  accelerate(N, pos[index], pos) ;
+    }
 }
 
 // TODO : update velocity and position using a simple Euler integration scheme
 __global__ void updateS(int N, float dt, glm::vec4 * pos, glm::vec3 * vel, glm::vec3 * acc)
 {
-	// FILL IN HERE
+	int index = (blockIdx.x * blockDim.x) + threadIdx.x;
+    if(index < N)
+    {	
+		pos[index] += glm::vec4(vel[index].x, vel[index].y, vel[index].z, 0.0f) * dt; 
+		vel[index] += acc[index] * dt;
+    }
 }
 
 // Update the vertex buffer object
@@ -179,7 +206,10 @@ void initCuda(int N)
 // TODO : Using the functions you wrote above, write a function that calls the CUDA kernels to update a single sim step
 void cudaNBodyUpdateWrapper(float dt)
 {
-	// FILL IN HERE
+	dim3 fullBlocksPerGrid((int)ceil(float(numObjects)/float(blockSize)));
+	updateF<<<fullBlocksPerGrid, blockSize>>>( numObjects, dt, dev_pos, dev_vel, dev_acc);
+	updateS<<<fullBlocksPerGrid, blockSize>>>( numObjects, dt, dev_pos, dev_vel, dev_acc);
+	cudaThreadSynchronize();
 }
 
 void cudaUpdateVBO(float * vbodptr, int width, int height)
diff --git a/Part1/src/main.cpp b/Part1/src/main.cpp
index ea86207..53ba4a3 100644
--- a/Part1/src/main.cpp
+++ b/Part1/src/main.cpp
@@ -4,7 +4,7 @@
 
 #include "main.h"
 
-#define N_FOR_VIS 5000
+#define N_FOR_VIS 1000
 #define DT 0.2
 #define VISUALIZE 1
 //-------------------------------

From 6494d88c32c433d9045c90ca874a55d2508e52b6 Mon Sep 17 00:00:00 2001
From: zxm5010 <zxm5010@gmail.com>
Date: Mon, 22 Sep 2014 02:52:42 +0800
Subject: [PATCH 2/4] Nbody finished & matrix_math finished

---
 Part1/src/kernel.cu                        |   5 +-
 Part1/src/main.cpp                         |   4 +-
 Part2/matrix/matrix.sln                    |  20 ++++
 Part2/matrix/matrix/matrix.vcxproj         |  76 ++++++++++++
 Part2/matrix/matrix/matrix.vcxproj.filters |  22 ++++
 Part2/matrix/matrix/matrix_math.cu         | 129 +++++++++++++++++++++
 6 files changed, 252 insertions(+), 4 deletions(-)
 create mode 100644 Part2/matrix/matrix.sln
 create mode 100644 Part2/matrix/matrix/matrix.vcxproj
 create mode 100644 Part2/matrix/matrix/matrix.vcxproj.filters
 create mode 100644 Part2/matrix/matrix/matrix_math.cu

diff --git a/Part1/src/kernel.cu b/Part1/src/kernel.cu
index 5af5cc3..c738de4 100644
--- a/Part1/src/kernel.cu
+++ b/Part1/src/kernel.cu
@@ -101,11 +101,12 @@ __device__  glm::vec3 accelerate(int N, glm::vec4 my_pos, glm::vec4 * their_pos)
 		{
 			r = glm::vec3(their_pos[i].x-my_pos.x, their_pos[i].y-my_pos.y, their_pos[i].z-my_pos.z);
 			float l = glm::length(r);
-			acc +=  (float)( (float)their_pos[i].w / powf(l * l + EPSILON * EPSILON,3/2)) * r;
+			glm::vec3 mag = r / l;
+			acc +=  (float)( (float)their_pos[i].w / powf(l * l + ZERO_ABSORPTION_EPSILON * ZERO_ABSORPTION_EPSILON,3/2)) * mag;
 			//acc +=  r * (float) (their_pos[i].w / (l* l * l + EPSILON)) ;
 		}
 	}
-	acc += (float)(  (float)starMass / powf(l * l + EPSILON * EPSILON,3/2)) * (-pos); 
+	acc += (float)(  (float)starMass / powf(l * l + ZERO_ABSORPTION_EPSILON * ZERO_ABSORPTION_EPSILON,3/2)) * (-pos)/glm::length(-pos); 
 	//acc +=  (-pos) * (float)(starMass / ( s * s * s + EPSILON )) ; 
     return (float)G * acc;
 }
diff --git a/Part1/src/main.cpp b/Part1/src/main.cpp
index 53ba4a3..7ac825d 100644
--- a/Part1/src/main.cpp
+++ b/Part1/src/main.cpp
@@ -4,8 +4,8 @@
 
 #include "main.h"
 
-#define N_FOR_VIS 1000
-#define DT 0.2
+#define N_FOR_VIS 2000
+#define DT 0.1
 #define VISUALIZE 1
 //-------------------------------
 //-------------MAIN--------------
diff --git a/Part2/matrix/matrix.sln b/Part2/matrix/matrix.sln
new file mode 100644
index 0000000..ba69b9b
--- /dev/null
+++ b/Part2/matrix/matrix.sln
@@ -0,0 +1,20 @@
+
+Microsoft Visual Studio Solution File, Format Version 11.00
+# Visual Studio 2010
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "matrix", "matrix\matrix.vcxproj", "{6D40842A-A7DB-4254-9652-8C9A21F55B4F}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Win32 = Debug|Win32
+		Release|Win32 = Release|Win32
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{6D40842A-A7DB-4254-9652-8C9A21F55B4F}.Debug|Win32.ActiveCfg = Debug|Win32
+		{6D40842A-A7DB-4254-9652-8C9A21F55B4F}.Debug|Win32.Build.0 = Debug|Win32
+		{6D40842A-A7DB-4254-9652-8C9A21F55B4F}.Release|Win32.ActiveCfg = Release|Win32
+		{6D40842A-A7DB-4254-9652-8C9A21F55B4F}.Release|Win32.Build.0 = Release|Win32
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+EndGlobal
diff --git a/Part2/matrix/matrix/matrix.vcxproj b/Part2/matrix/matrix/matrix.vcxproj
new file mode 100644
index 0000000..619a761
--- /dev/null
+++ b/Part2/matrix/matrix/matrix.vcxproj
@@ -0,0 +1,76 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup Label="ProjectConfigurations">
+    <ProjectConfiguration Include="Debug|Win32">
+      <Configuration>Debug</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Release|Win32">
+      <Configuration>Release</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+  </ItemGroup>
+  <PropertyGroup Label="Globals">
+    <ProjectGuid>{6D40842A-A7DB-4254-9652-8C9A21F55B4F}</ProjectGuid>
+    <RootNamespace>matrix</RootNamespace>
+  </PropertyGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>false</UseDebugLibraries>
+    <WholeProgramOptimization>true</WholeProgramOptimization>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
+  <ImportGroup Label="ExtensionSettings">
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.props" />
+  </ImportGroup>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <PropertyGroup Label="UserMacros" />
+  <PropertyGroup />
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+    </Link>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>MaxSpeed</Optimization>
+      <FunctionLevelLinking>true</FunctionLevelLinking>
+      <IntrinsicFunctions>true</IntrinsicFunctions>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <EnableCOMDATFolding>true</EnableCOMDATFolding>
+      <OptimizeReferences>true</OptimizeReferences>
+      <AdditionalDependencies>kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;cudart.lib;%(AdditionalDependencies)</AdditionalDependencies>
+    </Link>
+    <CudaCompile>
+      <Include>$(CudaToolkitIncludeDir)</Include>
+    </CudaCompile>
+  </ItemDefinitionGroup>
+  <ItemGroup>
+    <CudaCompile Include="matrix_math.cu">
+      <FileType>Document</FileType>
+    </CudaCompile>
+  </ItemGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
+  <ImportGroup Label="ExtensionTargets">
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.targets" />
+  </ImportGroup>
+</Project>
\ No newline at end of file
diff --git a/Part2/matrix/matrix/matrix.vcxproj.filters b/Part2/matrix/matrix/matrix.vcxproj.filters
new file mode 100644
index 0000000..f07abd3
--- /dev/null
+++ b/Part2/matrix/matrix/matrix.vcxproj.filters
@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup>
+    <Filter Include="Source Files">
+      <UniqueIdentifier>{4FC737F1-C7A5-4376-A066-2A32D752A2FF}</UniqueIdentifier>
+      <Extensions>cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx</Extensions>
+    </Filter>
+    <Filter Include="Header Files">
+      <UniqueIdentifier>{93995380-89BD-4b04-88EB-625FBE52EBFB}</UniqueIdentifier>
+      <Extensions>h;hpp;hxx;hm;inl;inc;xsd</Extensions>
+    </Filter>
+    <Filter Include="Resource Files">
+      <UniqueIdentifier>{67DA6AB6-F800-4c08-8B7A-83BB121AAD01}</UniqueIdentifier>
+      <Extensions>rc;ico;cur;bmp;dlg;rc2;rct;bin;rgs;gif;jpg;jpeg;jpe;resx;tiff;tif;png;wav;mfcribbon-ms</Extensions>
+    </Filter>
+  </ItemGroup>
+  <ItemGroup>
+    <CudaCompile Include="matrix_math.cu">
+      <Filter>Source Files</Filter>
+    </CudaCompile>
+  </ItemGroup>
+</Project>
\ No newline at end of file
diff --git a/Part2/matrix/matrix/matrix_math.cu b/Part2/matrix/matrix/matrix_math.cu
new file mode 100644
index 0000000..436c6a4
--- /dev/null
+++ b/Part2/matrix/matrix/matrix_math.cu
@@ -0,0 +1,129 @@
+#include <cuda.h>
+#include <cuda_runtime.h>
+#include <iostream>
+
+using namespace std;
+const int N = 3;
+
+//data on device
+float * dev_matA;
+float * dev_matB;
+float * dev_res;
+
+//data on host
+float * matA = (float *)malloc(N*N*sizeof(float));
+float * matB = (float *)malloc(N*N*sizeof(float));
+float * res = (float *)malloc(N*N*sizeof(float));
+
+void initMat(float * mat)
+{
+	for(int i = 0; i < N; i++)
+		for(int j = 0; j < N; j++)
+			mat[i + j*N] = i + j * N;
+}
+
+void printRes()
+{
+	for(int j =0; j <N; j++)
+	{
+		cout << endl;
+		for(int i=0;i<N;i++)
+			cout << res[j*N + i] << " ";
+	}
+}
+
+__global__ void mat_add(float * A, float * B, float * res)
+{
+	int x = threadIdx.x;
+	int y = threadIdx.y;
+	res[x + y*N] = A[x + y*N] + B[x + y*N];
+}
+
+__global__ void mat_sub(float * A, float * B, float * res)
+{
+	int x = threadIdx.x;
+	int y = threadIdx.y;
+	res[x + y*N] = A[x + y*N] - B[x + y*N];
+}
+
+__global__ void mat_mult(float * A, float * B, float * res)
+{
+	int x = threadIdx.x;
+	int y = threadIdx.y;
+
+	float result = 0;
+	for(int i = 0; i < N; i++)
+	{
+		result += A[y * N + i] * B[x + i * N];
+	}
+	res[x + y*N] = result;
+}
+
+void mat_add_serial(float * A, float * B, float * res)
+{
+	for(int i = 0; i < N * N; i++)
+	{
+		res[i] = A[i] + B[i];
+	}
+}
+void mat_sub_serial(float * A, float * B, float * res)
+{
+	for(int i = 0; i < N * N; i++)
+	{
+		res[i] = A[i] - B[i];
+	}
+}
+void mat_mult_serial(float * A, float * B, float * res)
+{
+	for(int i = 0; i < N; i++)
+		for(int j = 0; j < N; j++)
+		{
+			float result = 0;
+			for(int k = 0; k < N; k++)
+			{
+				result += A[i*N + k] * B[k*N + j];
+			}
+			res[i*N + j] = result;
+		}
+}
+
+int main(int argc, char** argv)
+{
+	dim3 threadsPerBlock(N,N);
+	cudaMalloc((void**)&dev_matA, N*N*sizeof(float));
+	cudaMalloc((void**)&dev_matB, N*N*sizeof(float));
+	cudaMalloc((void**)&dev_res, N*N*sizeof(float));
+
+	initMat(matA);
+	initMat(matB);
+	cudaMemcpy(dev_matA, matA, N*N*sizeof(float), cudaMemcpyHostToDevice);
+	cudaMemcpy(dev_matB, matB, N*N*sizeof(float), cudaMemcpyHostToDevice);
+
+	mat_add<<<1, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
+	cudaMemcpy(res, dev_res, N*N*sizeof(float), cudaMemcpyDeviceToHost);
+	cout<< endl << "###########parallel mat_add test######## ";
+	printRes();
+
+	mat_sub<<<1, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
+	cudaMemcpy(res, dev_res, N*N*sizeof(float), cudaMemcpyDeviceToHost);
+	cout<< endl << "###########parallel mat_add test######## ";
+	printRes();
+
+	mat_mult<<<1, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
+	cudaMemcpy(res, dev_res, N*N*sizeof(float), cudaMemcpyDeviceToHost);
+	cout<< endl << "###########parallel mat_mul test######## ";
+	printRes();
+
+	mat_add_serial(matA, matB, res);
+	cout<< endl << "###########serial mat_add test######## ";
+	printRes();
+
+	mat_sub_serial(matA, matB, res);
+	cout<< endl << "###########serial mat_add test######## ";
+	printRes();
+
+	mat_mult_serial(matA, matB, res);
+	cout<< endl << "###########serial mat_mul test######## ";
+	printRes();
+
+}
\ No newline at end of file

From 17431cc3044fe7533d87f6b6b92079e12265624d Mon Sep 17 00:00:00 2001
From: zxm5010 <zxm5010@gmail.com>
Date: Mon, 22 Sep 2014 03:42:15 +0800
Subject: [PATCH 3/4] Performance Analysis

---
 README.md | 119 ++++--------------------------------------------------
 1 file changed, 7 insertions(+), 112 deletions(-)

diff --git a/README.md b/README.md
index 70ae0d3..eafe98a 100644
--- a/README.md
+++ b/README.md
@@ -1,120 +1,15 @@
-Project 1
-=========
-
-# Project 1 : Introduction to CUDA
-
-## NOTE :
-This project (and all other projects in this course) requires a NVIDIA graphics
-card with CUDA capabilityi!  Any card with compute capability 2.0 and up will
-work.  This means any card from the GeForce 400 and 500 series and afterwards
-will work.  If you do not have a machine with these specs, feel free to use
-computers in the SIG Lab.  All computers in SIG lab and Moore 100 C have CUDA 
-capable cards and should already have the CUDA SDK installed. 
-
-## PART 1 : INSTALL NSIGHT
-To help with benchmarking and performance analysis, we will be using NVIDIA's
-profiling and debugging tool named NSight. Download and install it from the
-following link for whichever IDE you will be using:
-http://www.nvidia.com/object/nsight.html. 
-
-NOTE : If you are using Linux / Mac, most of the screenshots and class usage of
-NSight will be in Visual Studio.  You are free to use to the Eclipse version
-NSight during these in class labs, but we will not be able to help you as much.
-
-## PART 2 : NBODY SIMULATION
-To get you used to using CUDA kernels, we will be writing a simple 2D nbody 
-simulator.  The following source files are included in the project:
-
-* main.cpp : sets up graphics stuff for visualization
-* kernel.cu : this contains the CUDA kernel calls
-
-All the code that you will need to modify is in kernel.cu and is marked clearly
-by TODOs.
-
-## PART 3 : MATRIX MATH
-In this portion we will walk you through setting up a project that writes some
-simple matrix math functions. Please put this portion in a folder marked Part2
-in your repository. 
-
-### Step 1 : Create your project.
-Using the instructions on the Google forum, please set up a new Visual Studio project that
-compiles using CUDA. For uniformity, please write your main function and all
-your code in a file named matrix_math.cu.
-
-### Step 2 : Setting up CUDA memory.
-As we discussed in class, there is host memory and device memory.  Host memory
-is the memory that exists on the CPU, whereas device memory is memory on the
-GPU.  
-
-In order to create/reserve memory on the GPU, we need to explicitly do so
-using cudaMalloc.  By calling cudaMalloc, we are calling malloc on the GPU to
-reserve a portion of its memory.  Like malloc, cudaMalloc simply mallocs a
-portion of memory and returns a pointer. This memory is only accessible on the
-device unless we explicitly copy memory from the GPU to the CPU.  The reverse is
-also true.  
-
-We can copy memory to and from the GPU using the function cudaMemcpy. Like the
-POSIX C memcpy, you will need to specify the size of memory you are copying.
-The last argument is used to specify the direction of the copy (from GPU to CPU
-or the other way around).
-
-Please initialize 2 5 x 5 matrices represented as an array of floats on the CPU
-and the GPU where each of the entry is equal to its position (i.e. A_00 = 0,
-A_01 = 1, A_44 = 24). 
-
-### Step 3 : Creating CUDA kernels. 
-In the previous part, we explicitly divided the CUDA kernels from the rest of
-the file for stylistic purposes.  Since there will be far less code in this
-project, we will write the global and device functions in the same file as the
-main function.
-
-Given a matrix A and matrix B (both represented as arrays of floats), please
-write the following functions :
-* mat_add : A + B
-* mat_sub : A - B
-* mat_mult : A * B
-
-You may assume for all matrices that the dimensions of A and B are the same and
-that they are square.
-
-Use the 2 5 x 5 matrices to test your code either by printing directly to the
-console or writing an assert.
-
-THINGS TO REMEMBER :
-* global and device functions only have access to memory that is explicitly on
-  the device, meaning you MUST copy memory from the CPU to the GPU if you would
-  like to use it there
-* The triple triangle braces "<<<" begin and end the global function call.  This
-  provides parameters with which CUDA uses to set up tile size, block size and
-  threads for each warp.
-* Do not fret if Intellisense does not understand CUDA keywords (if they have
-  red squiggly lines underneath CUDA keywords).  There is a way to integrate
-  CUDA syntax highlighting into Visual Studio, but it is not the default.
-
-### Step 4 : Write a serial version.
-For comparison, write a single-threaded CPU version of mat_add, mat_sub and
-mat_mult. We will not introduce timing elements in this project, but please
-keep them in mind as the upcoming lab will introduce more on this topic. 
-
 ## PART 4 : PERFORMANCE ANALYSIS
-Since this is the first project, we will guide you with some example
-questions.  In future projects, please answer at least these questions, as
-they go through basic performance analysis.  Please go above and beyond the
-questions we suggest and explore how different aspects of your code impact
-performance as a whole. 
 
-We have provided a frame counter as a metric, but feel free to add cudaTimers,
-etc. to do more fine-grained benchmarking of various aspects. 
+* How does changing the tile and block sizes change performance? Why?
 
-NOTE : Performance should be measured in comparison to a baseline.  Be sure to
-describe the changes you make between experiments and how you are benchmarking.
+Changing the tile and block sizes may increase the speed if we can utilize the shared memory efficiently. Smaller the block size, smaller the shared memory threads can have. If threads frequently access global memory instead of shared memory, then the performance go down. 
+For Nbody problem, we actually calculated forces between two paired body twice. If we can store them into shared memory, then the performance can be doubled. 
 
-* How does changing the tile and block sizes change performance? Why?
 * How does changing the number of planets change performance? Why?
+
+As we can see from the result, there is only 1.5 FPS for 5000 bodies, and 60FPS for 100 bodies. Performance decreses in O(N2) if the number of planets increases. The main reason is the accelerate() function has a for loop which integrates all forces from other bodies serially.To increase the performance, we can try to make the additions also in parallell.
+
 * Without running experiments, how would you expect the serial and GPU verions
   of matrix_math to compare?  Why?
 
-## SUBMISSION
-Please commit your changes to your forked version of the repository and open a
-pull request.  Please write your performance analysis in your README.md.
-Remember to email Harmony (harmoli+CIS565@seas.upenn.edu) your grade and why.
+GPU version of matrix_math runs faster than serial computation. Since the computation for each element in the matrix is independent, all computations can be done parallely, which means those computations can be done simultaneously rather than one-by-one in CPU. 

From c0ea6336b9b8be207a10ebcf6cbacbe0f4707562 Mon Sep 17 00:00:00 2001
From: zxm5010 <zxm5010@gmail.com>
Date: Mon, 22 Sep 2014 03:59:05 +0800
Subject: [PATCH 4/4] Tile/block for matrix_math

---
 Part2/matrix/matrix/matrix_math.cu | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/Part2/matrix/matrix/matrix_math.cu b/Part2/matrix/matrix/matrix_math.cu
index 436c6a4..7dbf43a 100644
--- a/Part2/matrix/matrix/matrix_math.cu
+++ b/Part2/matrix/matrix/matrix_math.cu
@@ -1,9 +1,10 @@
 #include <cuda.h>
 #include <cuda_runtime.h>
 #include <iostream>
+#include <math.h>
 
 using namespace std;
-const int N = 3;
+const int N = 5;
 
 //data on device
 float * dev_matA;
@@ -34,22 +35,22 @@ void printRes()
 
 __global__ void mat_add(float * A, float * B, float * res)
 {
-	int x = threadIdx.x;
-	int y = threadIdx.y;
+	int x = threadIdx.x + blockIdx.x *blockDim.x;
+	int y = threadIdx.y + blockIdx.y *blockDim.y;
 	res[x + y*N] = A[x + y*N] + B[x + y*N];
 }
 
 __global__ void mat_sub(float * A, float * B, float * res)
 {
-	int x = threadIdx.x;
-	int y = threadIdx.y;
+	int x = threadIdx.x + blockIdx.x *blockDim.x;
+	int y = threadIdx.y + blockIdx.y *blockDim.y;
 	res[x + y*N] = A[x + y*N] - B[x + y*N];
 }
 
 __global__ void mat_mult(float * A, float * B, float * res)
 {
-	int x = threadIdx.x;
-	int y = threadIdx.y;
+	int x = threadIdx.x + blockIdx.x *blockDim.x;
+	int y = threadIdx.y + blockIdx.y *blockDim.y;
 
 	float result = 0;
 	for(int i = 0; i < N; i++)
@@ -89,7 +90,9 @@ void mat_mult_serial(float * A, float * B, float * res)
 
 int main(int argc, char** argv)
 {
-	dim3 threadsPerBlock(N,N);
+	dim3 threadsPerBlock(16,16);
+	dim3 numBlocks((int)ceil((float)N / threadsPerBlock.x), (int)ceil((float)N/threadsPerBlock.y));
+
 	cudaMalloc((void**)&dev_matA, N*N*sizeof(float));
 	cudaMalloc((void**)&dev_matB, N*N*sizeof(float));
 	cudaMalloc((void**)&dev_res, N*N*sizeof(float));
@@ -99,17 +102,17 @@ int main(int argc, char** argv)
 	cudaMemcpy(dev_matA, matA, N*N*sizeof(float), cudaMemcpyHostToDevice);
 	cudaMemcpy(dev_matB, matB, N*N*sizeof(float), cudaMemcpyHostToDevice);
 
-	mat_add<<<1, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
+	mat_add<<<numBlocks, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
 	cudaMemcpy(res, dev_res, N*N*sizeof(float), cudaMemcpyDeviceToHost);
 	cout<< endl << "###########parallel mat_add test######## ";
 	printRes();
 
-	mat_sub<<<1, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
+	mat_sub<<<numBlocks, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
 	cudaMemcpy(res, dev_res, N*N*sizeof(float), cudaMemcpyDeviceToHost);
 	cout<< endl << "###########parallel mat_add test######## ";
 	printRes();
 
-	mat_mult<<<1, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
+	mat_mult<<<numBlocks, threadsPerBlock>>>(dev_matA, dev_matB, dev_res);
 	cudaMemcpy(res, dev_res, N*N*sizeof(float), cudaMemcpyDeviceToHost);
 	cout<< endl << "###########parallel mat_mul test######## ";
 	printRes();
@@ -126,4 +129,6 @@ int main(int argc, char** argv)
 	cout<< endl << "###########serial mat_mul test######## ";
 	printRes();
 
+	int in;
+	cin >> in;
 }
\ No newline at end of file