From 5d765ed3e065e9df3394ae844cd1adf99720cf58 Mon Sep 17 00:00:00 2001 From: lucasdelimanogueira Date: Mon, 29 Apr 2024 13:39:00 -0300 Subject: [PATCH] Fix makefile cuda --- build/Makefile | 4 ++-- build/cuda.cu.o | Bin 9464 -> 9464 bytes build/libtensor.so | Bin 27776 -> 31872 bytes build/tensor.o | Bin 16496 -> 16568 bytes csrc/tensor.cpp | 3 ++- norch/__pycache__/tensor.cpython-38.pyc | Bin 6405 -> 6420 bytes norch/tensor.py | 1 + 7 files changed, 5 insertions(+), 3 deletions(-) diff --git a/build/Makefile b/build/Makefile index cf3c755..81c04e7 100644 --- a/build/Makefile +++ b/build/Makefile @@ -3,7 +3,7 @@ CC = g++ NVCC = nvcc # Compiler flags -CFLAGS = -Wall -Wextra -std=c++11 +CFLAGS = -Wall -Wextra -std=c++11 -I/usr/local/cuda/include NVCCFLAGS = -std=c++11 # Directories @@ -19,7 +19,7 @@ CU_OBJS = $(patsubst $(SRCDIR)/%.cu, $(BUILDDIR)/%.cu.o, $(CU_SRCS)) # CUDA flags and libraries CUDAFLAGS = -arch=sm_75 -CUDALIBS = -L/usr/local/cuda-10.2/lib64/ -lcudart -lcuda +CUDALIBS = -L/usr/local/cuda -lcudart -lcuda # Rule to build the target $(TARGET): $(OBJS) $(CU_OBJS) diff --git a/build/cuda.cu.o b/build/cuda.cu.o index 63691b1dc1d348085053119d96a2e57d83cc172f..c04ec686d6ffc530cc3de51a1d201490a22b9d82 100644 GIT binary patch delta 17 Ycmez2`NMNVi5QD{N}|c;3b6)007kY3zyJUM delta 17 Ycmez2`NMNVi5QDnim~P93b6)007Z!gmH+?% diff --git a/build/libtensor.so b/build/libtensor.so index 80dc3bc60cfa0c890da8f7febd8165635c3fbebf..69817f2cf97f26421c827b3d22480a140324df1e 100755 GIT binary patch delta 6671 zcmZ`-4OCRs7QT025D;~M8DT(>A5nfpkmNzVP(Vf}dm)Jgk|dV@CSqd$J~UP)O%%Fo zX};D=TeV)<&oj~DStgV}RIb7@MRL99e-bP5FUst`@0@e*;WG8^S~L6XZ}0u>ea_i) z@13*ml)CO0wK+wt+mu>U9~!6d%aDpq=qiLi8|JewT>1Q4%D@Pr8`2yX08?v~u(z%l zEc5+@zEMbuP{TrX*G}EFa|K7xOrWQySyNG z!PL@q$5Z@XZaHh&V^K3{hc!Z-ME|jlPMISZm7S1b%5df6NC8~bYdWaoM@bEcP{cKOxog$D{Jr8Zm)+I;ZB41ss};F-FAYpt*w zJci!sg9jD~zb+lGHv$4z34_-splzDKJ68(HHb|c{2f9pnuyEnuJtjQNgnLbRtfJIQ zQ=iG8FD|$Bn{d5xINqig%DQcYWKonz9UlUW*^DdBggXt8G{J8sQ2 z;Z~D>z6ocmWL$0&&WbXw5{9$+Gtk%uW~?v;1TjFrs!jM%1B6^+!b41WtqC7y!tXZW z^}|hubtZ$&gl{zAp(ec9gzJ6Cty@jF-u)ckZo=(?N0^^C862hryJ}l1hwZkEo~J18 z2ixi_TBp0=VBkC4L(BdG@Hs8%%Xuwnwj0#?ThIri(cb4r#QqAn$wz!AE!b=y^VRwvV3fqbK?3F@yD6lN^JAq+vd~)knXAo2PljE*iSK zDNlv3yUF@1Aotb%bq=@38c;=DVWR>W(vJ8tm-prLbX|&hAEFld6L01salo!dySAN6Tr zQ)Y(8CBqC_6EW64mDS$p>chhJiG>ZSU~5^Hnv5)mq<8Nxd<{E1`Q!N-J^)ft(<)(U zGb~G_rJPxs^5#om8aX4giqpX?LW3k}a)VdB+I4(wDEw{G{#`#?_2X*qvw(jahk_mw zl%GqN!}Mfi*5sFA6O@ezVXJ|?=9FEWvRT@fa>_12c|W5Z@(XDRkreN&PRqWDuexW= z>4L^SH1YSa%eAjYq*(ir7*VWqB<+tKL*F=u+t2*W7tz$6yUN%i+Kwe5!R<0|Go?q} zXL;>s+CIh^h&~((YZ1L4H8%0kY-7LQ2Pku1x@6>h=@N~IbE@q$Av!9dMtXmR1k(Fy z>0L(4qQ}~cp%30$`)pR$wb^5kxKnyxy2#h8)!4x%kc!p3pIMsnl2MZYmS3d4xJe;> z%!Bt~+C|EYx4W~zC_IKDq``Gm?cIUESeEDoLBGqO?~}ftUEuT{nePMIG%9;&sbqhY zj*W6;JdtRS*C3$5)hWo=1NMf93KvPv>o1TkVdBs%88e)2jB#W*r2fNsQ&Xvo;}KED zH`3+(^VAHE-3dls)Dh!|eM0IroI~%T^?d%U|Xz_Y&Gn%k2V1IhKhQZ%N!*% zA|b0dM}~w8N=(&!uZoiP^FNr{Du7XL5>@s}mri;tAuI3o7^4b8h^W$hui7TuC(g3|N@&nynzDZZ zyR2!?S;Ku9l_h1_%b^c$ub3IUrpZwtKPA1jUTRP3wP*GEPNCn;en_f3%Cy%u^e8=N^@Qmvb&$#QR^KM>-{|jF086!T8`I)-;?MP#% zUiHG#lvjm(qFcAqf#mF=NlfS6TuoZCW5Qza341;Q8e9j(jjEKT`LCeJm7FnDn36Tr zUxxHheTpOITl}>V6m8!!iVpFl`WUeH7J4c*SgaP-CN!5 zJr-=AcO#Xj#aarAc!FE;PVT(|7vY>pPo-5^zW$mnr^Q(QIhDH{z-zv@Qo5|8DdTG_ z*A>yf$48FKDN>Z?s(0ke5AF^?ydUA+&+W0sW6Uycq94Y`s#!EFeSJ`FA;YZqOrYbfCy)d-t71 z&Afb%V?PR+cyUdO zUpk^`$07f9RMRd%K66ad>=s3_f1+uVAm2QRc*w1%5D&QNhTIj>7&idbAZBy|dz4MT!@$(z$aL(Uw$VA;QzW4v_ zC9rS!P}A(R
    AgK((-U;Yaekd2+4>Kfi5!ekc-6{y3cGVv|lo4v>7T8bz?p zjXcQ~D}e29m>9Ax4mn9@fU;c%8ayo};tAY@qoWbo3^U%V_^0m#&6t)@e?Qy`yEH8u zu*`Ei%OY0mDcH+lzs+p7RB{Klns36s4fcvb_LMSP7+VTFaS|4^-8kj?Hq^qndm@&?NknP#JH66+avb z+Dgrc82c&Vig5~Qw)!b{+?UOW;qOrT@h^p+3z^M+%1F%Cj2HnCig6}rw)&y*jYHAF zxKlBX@&@DR3e$p3HjI8sw9;xa4B(@bjz7g?l>S-@p8c3px3aZm*ED=og_TRYXJ=ra zzQFe?3MP7h#|(F-oKl+atg&II?ZU zkpsIhQGV^y{o{Q6dvyPgeEfgs{sa4*Vw^fY13H#C*@q|`-Nq81rTI5S)Q6>uw0OVb zmoZ7>fK`VTW0F1k^06BjyQ;v(vSj*X*`jCJmLdH3PYc77x__?lPnAcELvSNvWqrn8 z`*88!GGvC17gBtAhULOdG^;$L-ls=b!bkm+j<{dP+w`K1o!qK(_UXkLC&5p1e1u~B zpm`zqdXD44c)$o`b4BPxpRV7-Q~j~|+BnX-m7)J{+xL@_@6%~N1a1Y&iPP%o0eAIrS5F`ys}62&si{kQp=&ap9ZUi zv}18burUe$OzK{o6O_+0;r2mMOLBtn+QY8EBC3R@FQ=`Pwhd@;rPRG7$GCkMw*?!; zFU<(%Hv`1?-$a#5bAt4>Xsx1c16oHFbq{ErRTQ->Cuk*4t8~tzN@)7FDAlxWKx?h0 p?g6c%nxd9V&;DwvTs}pGKfdQvH@=rs)a_FO?-0B*XyWA2{{!R+0?Ysa delta 6556 zcmaJ_3se->8J^jd6%>`nBJ3{6Lr{>%3WA_Wgpk#Zijpd7jOb!9K92Q?@fCwPB&jQs zX#TVsPfwH{(?)BRh_=KN4Mvj?Ym}&Ih)EQm9hLY*MWb%NJ9lOnHfiUao&W#8|2g;m zbLZ||^*~(powzPREM4DkZuzZ9LB3Qf1aMgd>BQLnT=u2`p&VDK-p>S8T>;nE6?de$ zMlP@BltRUzK*h9CF>NfS!Q%12d~O(}QL-AQsyD^6OsmN-ue_m;5^~dAgTe>?M~Mc>x7pvcpTx@scyVj3Wi~ zC@5SUMV8=R34nyfiABzeoHe7sI^CDb{GF9KF9Me=Pdpx})Z_7d?(>HNas|hU1@uO6 zQcO5>FG2T77CfXO;GAwpo3mnNo3J<;*$ks7Ix@u2JDjcs?-obO!t*xTYn~lrgM;5= zI8^0?G5+x*@Rg3X%G`MO=Kw=(low(VZS+EjEeHm6tY4(ZKg%ASZL~ThOSIC#5L=j4 zRc?T?!qF;T+X4UWR%#8|Hb_(c`=RXF1Aw#~hU>mO#Cog<&*`ja$#uF{HaOjj>*1Nm zOeRok&pt6tP7FjP8!8h(r~3O-2RTly41|4(rtb`Wh2xZX zZ7X0aVCj7x(xaJNq<@B`J@qdv?L6dP+tg}!$R;`DPEG%N=qnseJme<8QrrWI=p9Bw zEw1>n!{kw~T>L|x`zFS|E7E8q_X<4u$feU4mXT&(mZ?M29dode3%G$bNLr&=*U(7o z3d5=UbRgE^Ta_~qUnglAU9t9wxq#({Px2LjS|am(HPI1DwOPfpG|Fa;S*sZv^v3(| zsm6($ie~lUFsg4 zKC?Rm)l|nRDavZNc#8^bmfq_%>&}RwS$}#ju??r~ALe06E)-H|S=#Ii;?<8h-r^T_bG;EZKPww^qSi zN{&hMFVb96Z_xY*b4r#51j;d#Mv7%Ib^$@x`)M(*)87%}n{H~_53keB2y;Z82CToX zYox7!B|C-pUKU?GM4ne|F)Bt#ZN65+i8sUAa}eZ?ebkgK_t67v^`v(+>wT0L zyFxre`(v}>9}kujINb;LrQnltfDIE$AI%=)qo36HXIm&a%3^43VeJ!Bt@u=BgkrC# zm2;v+HD67O;?m5EA$RPa?7^&@z{%L;OZPSFh!#2C;w4H@=sNR9MVVApMk!@(ZD}dxA31pdYO(VOfa`HKet1|(@Okl_hrbV z^*#PiNG~BRAFV&p)P7^c9ki=ojNg~cr{b{l1zI0$p*#IX81|;q(8Q3uVJJ^p04?{~ zaM|OjbR2bVbRKLrV0xv$0C2iL#iKVl-Gu=%t;`L;ep=pVO-+|GufjhZ$)-igqkW0R z2IFP2Bt;lL9w3|iXoSg9MB|CxqvE8wh8gMfRZ>{rqI5x6SKPiWs>&G&L^^?epR>|u zD0o)NF7i+AB^FX*@~eL1(-`KnVF2w-&J{1wz2s*Nhf*moCBwEhm79lQ-tfo)*ZkRb ziv@T{>r;ZoVf22=0Mm){Y$ry6$)QG{$!XMxIXOa%-PCaUAM zEdb534T+?27ufjtTw+P9e3r0^*`A;>Jn7g)TioJrV$N)zzL zAuitv(iZzOmOeaN0N*1b%#SD|_1=|Z z&s{WcVXc2&bWr(M9#1o(79R6>d@%qn&?wO78$F&Z&_$pVLC;}qmx4|?>G5m?eHXM2 z^v!QQo+eQD4<1h&=xgUZo-q8ay>}7*pzmRhCxb4>o3sk_9ndYHJ3;F~GfshN26G!U z3R6CZz1pDFpff=`;oWB^Fq1()0IdOSpmSs6#iP_^Twhi{mTc|9_gI*WUl#CO)T3V5 z{|jaKS&n-=*dPmnqkF&`hWu`(=lnaPqXnja2ftwG=b{`);d%d-gkWnVudWV|oo{;S$XrZ#8WNn<~*7u!q5x zNwgj8MX)<0IsrnFhAy*mDLlb?DR6-x(;mjqq3vohKbPwI+k#FR*}m}XCzC%S4+ zPw6W{$TRe$FjsY|OSaW+ALdURh9rV|{-ho^F&qU8^#g0Kx{itEY7)&6jFBBi;4SjL zj6$FmPp>oLgf1Xok8s{71Pkhspk8Yf%!0)2%k?-CV-q{P10)MZp{JmpBI>n9AyiNg z=sMzIf_g&P5w{5Hxf;jEabgrOn=ZYnira*`U&X_PhF`@agfySS-KBQS+j-neC}_GpwL~`1PNeS0;x9+~i5!9DrG?Xnmn&Xs{1?H9LWB_J6>TkW z1I|x3CGjlss{wYOVjtyY->lg0@v?s++qchi0`^w&^=I5!;&`t>!RS7g_!yPE7+T({ zgu@A>e5ntqo^~p*pgv@?iz8u*q8QOwmi}H@HYizMR^An5K80@;`z&tXPaECNMqbuI zwt>U-TTTSKg2z(y47=gcF#635dwIU1RLDHoh{7*B;$;P|cX30uoU&^pa5e;uN^$tJ zAzzziyoccZ?!7PL*zYOp`P(xCQa(We6ux9urkG9d&N?n`rCGC=h-+)l&2APu z4J;insP?lt-**vXsCtRLi~0~3?R0ucrk}cO^e5}mOuuxV*f*UfLCWV+Kt641m!k6N z^wLaq`B0VxA&jDz*}G_K!El2|51kgO{-%eF$Av|=bWPMdIlj!xtL0$E=k>N|}`8bPU}BmWB+f8Bln diff --git a/build/tensor.o b/build/tensor.o index 377f3e86741c61ddbc0d9a1671950bca0beac2a4..3b6337d819d6599c6c495a600c406e6fa36f9b07 100644 GIT binary patch delta 4071 zcmZ{n4Qy0J5XWy@D_m`9uR@olrR^!jwwBhjrB~?3QH8)E5e|jKU}d35rj}K zAhf}%WekCcXhV!(B<8HqiUc@Kj6|YdRDw~VHHbt^p0k(v0}y3tDmY^*R}fDHEVrPTk^o*o(XfagX1ygaSFuo7)`Ur6z<4+ zF#mSEEyZgCP|Ozs-H!i7!MOArO?RKHyr{X1lNNX*vKM!FmxLj0bH#pe*f z`5gI+OVEM$|2Xs}Q8igFjb8Fu{$?0JI$u+xiWJaz6w*kSve-hj8f~jV+#}TnMyXSV z_BeF6h3O7#ef<@@sqh#`H3 zQT}AtK?3O(3;ht1NRM0S_b`I=oP}P7K&|Y)!YJ>U&-p0y`OI3j>0ae(yLUeaJpT0* zh@-^fA9L;SPHjLfvG~LxT`Scl3&oQ=MXEkVx&GH6g!BNTysK&GMEaA3W}Ld$Ep#*_ z>SQ#Zwj&Pe1Ft()+fCoh$t`34TuBUqby6*{&^W}9ZnDrGM@b)}T*;_A9D%o-T&e9w zVXJ;plVGGyiaLnbOYs5(>ZN+glHnk9AU&ihb%3AT4=6*lj-=oqN&?O2#zWd}s+*I$ zfcf)PJP&ECmGDGu*P34jai7F*)qb1@A%yeSfL$>J*I`(2sO3i=}0m%wb3&jo#_F-qOiskU`p|skaYnAQ+bF z^|5ZOw+}}lhV%rZ{MlWEL|8_zY3hA~1+;h^-nER?cCV5$7{OX~o+$I3+YEtb{EYHY zb8@2&?e@^H-ZQ-pebYntx$=C69r}@n?$`8FNMqg4Jk*?*(@yM758b0<{m?m0s`6s@ zt?7fQTkX&$4>fgLpdVvFQ|c9c)R`y#Hl&aq^Te9nl7h@M8U2dI^4(G*T|Jbugyn2! zc77J#h1hhdQkt4oY}RMt3rHaSmc{Z~Baoaf@gHPIi&T>S9^x<=jM`F&w@5J=0_d|{ z`*_!j-?m7PS~8q)GMu6e@K(#5?#*r?rxI10skNHkl8r9&XY!cGzfnv3*<&|A4 z1NsD1xW5l`|1TVccunNu=j_MsQ>=Qu~2)ozeH7OthC2*v8u_%C` zC)MO}*op1pDF40}Z+c7+qMj5xVWn3o%6gyiPG-W>)AK|T3{01*Hq#%Rfw$3kUG#qj z>NORcAWjupS6{N2cM3n~_a?%Cjv#U%jU%U6 zsiinzf)ghMVun=nEY-J(4y}U_(nd>3`A(bVZipjo8i8&zbUUNZTi-TA4>Gz3bL;k* zp~o2Q^kHC7(}pvQc3FDP_UoJ8Ct5MuESd!=84cb{scQ6=U?WOO>J44NC@EQIXbYpH zWE;e0%7)XH@|C~Bz3j^5X8B6A{(QVhe?BB;NfmeKc!zF<6w-D}X%@Cl1G^wSONve% zLF9udm1=IbpM^$JHHINnD#bm}k9H%((LMx2Xm>yddcADDv^KXI-_zsOrz|E_<5woa zcRE7ovJ=78KT#i4YjDFPG|JM(GsbGY2F0GD@f56GOjcl=2=o^c16%_l(JVo>9shfLMi8Rb&4o+ z=rTqr?;1m!7^S=|;8n3~8OEG#9G#C6GiKW0dNAm7yJs zQk}nIXb+>b?;ja@m{HpIqY#}d)p15i)k))fno&|UY-oip6l^pjQdI*9jC=>9q-uqs z8yF>3O%SS*D#0k7g6+n4KcjRC4jTG3qm=iUp}#Upc~6Zvf+fKyspkMuM>Kyv57qY&QD)11goW5?ohpf^q+APM3J^y zO1oxDJ(;50(}X@O;Ww;P(1_6`x=Ch?surG zY5?L${}_P^=s=pQ3kVyZ(RI*|bgDz$y!iHo^eIbuwd@6NzEt}hUw4Jyh8WVz4lTg^ zG^1ijB7II8QU81L?}HwsZ&<1|yOH08f%!6i$f0gM42}@P*wI z^|jMh^9_Fg*nbn3!3PU}7TM5Q8y?d*&39^qA8Ny5F$8apppBuu4LZAXpH2DI_QjqMK`3e_b>$OBKRQ~L;DDH)Xn$@s7)#+ diff --git a/csrc/tensor.cpp b/csrc/tensor.cpp index 6e2e336..ad3f4f1 100644 --- a/csrc/tensor.cpp +++ b/csrc/tensor.cpp @@ -100,7 +100,8 @@ extern "C" { tensor->device = device; } #else - printf("Warning: CUDA is not available. Cannot perform GPU operations.\n"); + fprintf(stderr, "ERROR: CUDA is not available. Cannot perform GPU operations.\n"); + exit(0); #endif } diff --git a/norch/__pycache__/tensor.cpython-38.pyc b/norch/__pycache__/tensor.cpython-38.pyc index 20683050c21c440d262e5de6367fa86e3b7aa075..84e4a97ccde336fbc20106d81a6e28d91bda31b1 100644 GIT binary patch delta 130 zcmZoQnqtHo%FD~e00c*l=%?M-$oof_F>5lHNVH51Ll#dBb39K9Baq~+VGd@{Wb!MD z0jkvGn_Ma4!pJdsr--en2uLpz7n2yH0FwX{8zUPd4ywQn4hed5@NBO&P8LrGlBdN>8Lv%tlsF{l#lg?P O$iv71g3M}+JWK#5F&KFO diff --git a/norch/tensor.py b/norch/tensor.py index 779c1c3..66a1a59 100644 --- a/norch/tensor.py +++ b/norch/tensor.py @@ -206,6 +206,7 @@ class Tensor: result_data = Tensor() result_data.tensor = result_tensor_ptr result_data.shape = self.shape.copy() + result_data.device = self.device result_data.ndim = 2 return result_data