Gpc

gpc@gnu.de

5400 discussions

Re: --
by Ken Linder (KC7RAD) 28 Jul '97

28 Jul '97

At 09:14 PM 7/28/97 +0200, Herbert Voss <Herbert.Voss(a)ck.b.shuttle.de> wrote: >Hello, > >I'm a novice in gpc and I hav a lot of problems. Here comes the first >one: > >unit unit1; { reine testunit fuer gpc } >Interface > Procedure ClrScr; > >Implementation > >Procedure ClrScr; >Var i : Real; >Begin > For i := 0 to 25 Do > WriteLn; >End; > >Begin >End. > >I run gpc under Linux 2.0.30 and had the following output: > >voss@crazy_horse:/home/voss/pascal > gpc unit1.pas >/usr/lib/crt1.o: In function start': >/usr/lib/crt1.o(.text+0x5a): undefined reference to main' > To compile a unit do this: gpc unit1.pas -c The -c option tells gpc to just compile and not to link. >Another problem comes, when I choosed the Type Byte for the variable i. > If you want to define a byte variable do it this way: var JustAByte : __byte__ integer; or even define a 'byte' type (This is what I do) type byte = __byte__ integer; There are many other type 'modifiers' available. I don't have my Linux PC with me here at work but I think the type modifier list is in the 'news' file somewhere in the distribution. Hope this helps! Ken L.

1 0

Some bugs fixed :-)
by Peter Gerwinski 28 Jul '97

28 Jul '97

Hello, everybody! According to myself (Peter Gerwinski): > > bo5-6u.pas: In function `Writeit': > bo5-6u.pas:12: prior parameter's size depends on `What' > > (Read: A bug I already fixed twice in the foregoing days > is back again and again ... )-: It's fixed now {again}. According to Bernhard Tschirren: > > [Subject: __Volatile__ BUG ...] > > OK, the following code does not seem to actually define the variable - just > take a look at the asm produced by gpc! It's fixed now. AND the volatility of the variable now survives transport through the GPI file - which wasn't the case before, either. (-: UUencoded gzipped diff below.) There is one remaining bug I want to fix before I release the next beta, the GPI problem reported by Bill Currie. (See our recent mail on this list.) Unfortunately, I cannot reproduce this bug on any platform I have access to. If somebody else can reproduce it and help to find out under what circum- stances the error occurs and when it doesn't, it would be great help. Yours, Peter Dipl.-Phys. Peter Gerwinski, Essen, Germany, free physicist and programmer peter.gerwinski(a)uni-essen.de - http://home.pages.de/~peter.gerwinski/ [970201] maintainer GNU Pascal [970714] - http://home.pages.de/~gnu-pascal/ [970125] 8< ---- gpc-970721-970727.diff.gz.uu ----------------------------------------- begin 644 gpc-970721-970727.diff.gz M'XL(`,P2VS,"`^Q=>W?:2++_VYS[(7JR/A,PB$'XC3<[(1@[S&+P!3R3N7/G M:(5HC-8@L9*(X\WDNV]5=4MJ/7"PYV;.Y!Z<!Z*[NKNZNOI7U0^56S/3N>4^ MF[H>NUU:;.JY"WS03H]KQW6]6BBT)(7ML*D]AX<'=\4LTV$>MUS'#[R5%;![ M.YBQ-[;O.A7&`_/6K["%><=M9^I6"J8S82/^CLW,]YR-.7>8N["#@$^JC`W= M!6?NE`4S[G/9@.EQMEI.3*!@_USY`1L_%,8K>SZQG5LD9):[6`*E!^5_ELP( M>LKU.!1)50DU\`_<6@6B"MO'.A;`6*'`&+'*1LW+(4.&V>2]S;0INX)4+%VU MG4(!>H]U3VSH=>!Z#]C`Y76+^>[*LSAT=^FY[^T)<(QD#H<TWP2RP'7G?J&8 ME@G)`^42\`]V'1HL4:]MD*<YGX-D"H61R\SE<OX@^S&QIU.HX=:%*E.LQ,-% M5:Y\'DHIZB%;FH$U8]I2+Q2FG),D[8#(INY\[MYC`C7!3!SJY2JH%D8SCJ/L MF`N>:A%8@$:!R0*689K'M)M]IO68]H'M55WQ,7=OY??Y!!]">>(S:,[4OJU: MIC53OT/O@Y6/"=BG]]SS;=>I6E3-TO1\GGB>A82W2^Y-Q5<:19%<_?[[^"FL MM&JN/M#C]U851V)/IN]59\%B+OBE\<>'P!7-17F=WG#4[';Q\>+Z[^_]R]X- M/OM+;OFB&QYU1W*.CU[@)SN"";*[80?"?EC+I7S4\7-NCXEC14!48OR+7;%^ MU8CUF9H1J%]`EQQMK_J+59G]BM_=90`\^"*3.):/@;E84EW!0G+$G3V9<\M_ MT;6#7T5]UGPUH>Y]7RX3@27XM"R+^);L<^=?*UO(83IWS:`Z$T)6ORQ6\\"& M_HD&X6%/='/E3$!6*$8AK`\P3O0YMP$R0H[M@%?YAZ48F[%Y)Q[\>R$]R]+J MU>-J'?X(K8_GQW?+[P2<=4$W(5G]7M`T;0WISA6,Y0^K.8,9IA\UZJ>-PQ.F MGYX>%\KE<KJ>G>%*$A\S76_HIXW:@2!^_9II>F6?E?5*_82]?ETH(RF0(352 M('WMD+%K'G"/77(/IJ5_9[._+C&A>ALFO%XYM@8`PYWJA/^M4"Z4=_:(]>`! MQO6N:C78K>?>&1/;MSP0G&,Z0;'4``!V)C#_0'2`D3YI`Q1E,,>-N0FP'AA8 M`5).7.=E`%KBVG.:^J.?K]O&>?^JV>DA\IE4#(D!P!;+.4*OZ7GF0S5BQM9F MIC]#5O#3",SQG/_R:P64"@;7F'$3!KI8JC!0BHFQ<">K>9S:H.H!"%T+:XXK M8),'0"/;@BP"(9]S=C\S`R!9+CETA[$1@CO\#5RJ9&H#(H)]X)X'%L[C2]=# MLS)^`&LUG[/6RO-LSCIH0QPW0`(/>+&@40XZR%A#*\HN"1ZQ0\1RX'&2E&4" MW/[8'!CG[58W$O$_WKL@%`"[EX@)")OSAS-B2!&^$`W*=P%R<=D])U-&9A** MK``TWB=&BRHH^H"9"S,PP<H(<[,HA6(7N/C08/YJ[*$^TH`VV+WI.<P<PP@# MB74'Q4("J0$F-'#K++@3&*YGH'P-Q&*.*0U(a)1G,]&!D^.6.."U_-.;-(WT7I M]Z9GX_!@V>G*L1!H#--"&]A@-NB';=D(a)`(953.&?8PF#(GK"BKN[I0I5A$-@ M,M`Z!R<`###8*!(a)V?P6SN+BK1_T$.S['D0#)Q#H><$-4F%9@V4RHJ8+G(BD* M*-`"=44('?K&IU,8+`;S':S[2W`<..388.0K;`S2`W[N33$.T"_0)9A9Y#N! M'P0FU*^6*&_"K3G8<P,$XR,W.(+<@Q;^M3+G]M0&C*,15]0DE*$O6*=J;GG` M_L$_@"P<<_X2.\\0B0"%8L0X:NCZDQ&#%1A(L7DS>ML?#!N@2/?D)U6WYGQK MSK\N<ZY:/6F)U:2L45=S-['K"7K%M-?T1OVDH=<4TWZB[U>.P;KC)V2`@4?/ MFWT4'V`OP60P[[TYKY#I/(-IJ/W%GH*,:"U1*,.7Z%D4^6Z/79L^X#$#_\(3 M2R-S?F\^^"!;A&Q<*DA:MO<=5,'G8(^@&$(I+E`@D45U#5=C1,LE+8-HS48V M'+#<0YLYM^^X,*QSU^>`$8B#)JQ:K)GMH*4%["9BP4C8/M`-76%CD-YY"$L( MS!2:BB:[*N6T?UC1T0TZV3_!!Q046D8RT<47Y$B@@1*L(F-D%Z!V#Q8N(`MD M^D7IC$IY/%AYTKX;"].[,QQW@K+=^03216D\*A1["I9(a)T&X;K?YY6SZBL\.* MX#CP#Z42^^85X,2H?=D>4$8):OZX(;]HQ@""-V35`?!%)B6#Y3A%95L9S:8C MG"ZPI28U2RP+[P:\`]?1:($.)@J;EQ[.F`L7;((+>A-XNK5],!5GZ'TL:+$- M&H:N%FC!Q!4#=EH[KIS">)WJ-?C$X5(T-59M4*)7($)T%=%1-,`KZC1[HX14 MB6&0ZJM7S!(VDUP4DD0I[%OT\[WL8`,!$V`W,.B[$;A&Z"3(&L\^/]@QC\"Z M"8AEB$H)VM5J-AN(J$KT,//K%"IT%@X:3N:56'1+S<`QZ/3.R94E/VGZ(-4* M#*EU!]*ITK06HU`_(7@YW=?A,T07ML?T$ON)B\%'K1,BPY5_2%`O"141+2,9 M*`NX\=(UI?E4B#!$2D#M=U+YUL\9TQ,3YKJ/4R::,$SBH.C&45UTX^A0Z0:- MG224\X2V?`S@E[8=#'`=65&DC0$%O`?#7;+B=?=F:+3?70\JT.\*]J[":N!- MLL_^R'E+\@!!O]A$A1Y%BURL"+O.=IX$%T)2Q[J0U/&!E%3A_TX^.SL[CPIB MTTGPJ8!__XLXKI_L$U#43PYJE2/B^!,R+2R/Y^+V&^WV)1><#<8N8;D*:.1# M^_-H7:`N7ZL%C>WA/U:$ELAW7]BW,X0K"[<0X_72/>Y:W7-:/MZKJR?(L:T9 MU0&3-#!Q0]/U%F!/U(9HHQ,G2G)U6ZU6<32QS^C5VA;-_4)VG<V*V&1)RN/@ M!.10IL]]72I[2HF('/&PU[^FD0J5YMMOV75SV&IV#:(^[PQ;@\Y5IT>02J6D M?E%V_[H]:`*44`Z.,6#2%%WU8GXVC')93KNR4%'I1OC@%^!R*V+1:'6;PV&6 M9V@^]#SBTOA#2^*7^R\;:F(^F_52DH8!USDBS2\JNA#_@)I=@#L_>COHWUR^ M):W-LE5/LB42_[H)K_KS>=6?QZN>QRO?A-?:\WFM/957:?D:F4(3%[%BAKXF ME@ES/HG'3X48RK#E,P$68MX<DZFKG];V55.G.M.1^9VX"YS*KQ)[5$6<O6AY MXY]HV@G`%L5`AS65ZZBR)$)%Y"G)K"6WW.4#^31Q0PEF5'.7JBI;-%%2N%?D M775OXGZL&=]\8JAQN]#?+O2_LH5^N*,<+<S#A+Q%?IBWV1(_HK[P;$%]B-2' M]<;A:;S`/T(WK'P4.F-_`>!#?Z;7''5^;!LWG?,B_"NQ(GW\K0:K%Y'08!H] ME-`7"DM=-=\9;YO#M\:H^:;;9GJM=EK0I&LA#U57]L18!A[X*\H6>K+<KV<[ M`+3V1.PDCU?6'0</!M$V78=:R;I"P)X[G1H!M`B?OEAL`?K,78#P.,7P[7]S MX:/6CVL$U/!Q(G':]ZMW_`'PJ`8EY5("DJ?W'@P_*W[KX]8G5`!N8-'W2V`< MX3OZG(R-_\T]EQ5C1B/*.$E`L'(D`$O.3%\SUH]9=)K`BDGQ*9RDJI!X;3MV M8`"LCV&]7'0<5H:!JA^4SI2!O![T6^WAT.B!9U1T*K209>Q_";.I.N#0YX$! MSU`%Y%880#I2H`#W:V3I]FNG4H!DX,0H6,NS^(NW_+.*"+*^L1<@(0.MA.&. M08NM.P-%QR>EU!X8S/($!2A$MNS6/FWMT]=FG\+C06E3PJ]9VQ3F;&*9(MKD MF7*M#G]CNU0_/@3P*(L/VDP5:WOR=J]O!M?]8;A9(?U`WS+!"XQW`6IRAY+V M/2)'.2I*Y+1$O>F*U2BXS3O"]X\/NI)-Y#3'=I3%`=XJ(:.C$0^?Y54M2SXV M&KEG\)#\41DI;\9(NGS(#%M7O/665B39JDH5-,,(@LGB*O>/%*;6\4`Q+EX+ MMXX.#FH576?E\$%=0H$1\;AYIZPJ/D4;NH3[[:LW[7.Q+H(6]F&&<\]T)F*T MHH7HF\[(N.BTN^?&H'W14#):_=XY;62HB5?MT=O^N=$"E%/SE)V#3%-E9=6; M:DS-4II3D[,-EG/7D.(^`JTD4\OA@';*&+@K$[X,9FC[DTNQ1XOJSR]:SQ85 M@WJX7SFLX:"*A\\.:JRC]^Z:46SUKZ[[O79OE![%YF#0_#D[M$!]DS>\_?/. MQ<^9Y$X/QBV=F-8!2HSV*1.U=GIYJ3?=;*6CP4VO99QW?LPVU^YT<S,NNOW^ M(#=G0'W,RQ'-0&_SF\G+$)7EYN2UT'[7;(URV\9U?"8-?+5T6G?XMG.1E=`@ M/[D[Z(^:HW:6?$TZSL0.R"TO_=V:]&:.QLCT44ODL*201V\QLW.1-P"0UQ^L MS<IK*RRT)B./[1P5ZV:%<9FENLQ2M?\[D]1KY\R53-*P/3)R6@6@R&D%YV:W M_6X-NN9-=C5+F>Z9$C?Y`*M.>35#F?2)JC)(G)[XB=J5J9](CR>_FIR>_HF& MDP"@9J4A0,U+@T"VN7A.9YK+STI#02(OOZ4T'*1DE)<:08*:F@"%1+OK,I+` MD"BR-B<)#NF<=VMSFKD:EH&(U!"D0"*;J\!$;M%'"J[-RN]$KE)V\T1TF4=Y MF4<9P8::V&OGSKB<Q`1X)(9/@8_T7(\!Y$_M+8DM_&"-GW18.ZCL'X&?)!\V M]Y-</)X4?E+23>J`G@W:K8R7].;FXJ(]2*=>=-X9`B(R/@?D$$#D90@TRLL9 MY#I=0-\<Y9B6RSR[W7R3]:0N+H:Y=KG7'ZTQE7DYK>8@QQWJ]WYL#W*XDX>< MR;2>T:7%39;K\_.L1$#:[4&[U\JQM+W1X&<CORK@Z(><JIK=ZV8.GYVKYF4Z M(VU.DTJ10"Y%+1+V)J48Z3Q%-=)9JG*D\P9K3+2J($D0N<S'[UA)$M5<#->@ M<ZP.6:#,SU.4);6(4]0EP6ND,,G4E,HD>A$K30+W4FJ3P-J,XJ2X^R&WPH3R M)%0CJ3Y?'DS7(F*FT'1NWOK4A)_,S@*CECC>E7/`]'WN!:Q8BP]#LR?`*I$\ M!(ANKOUE13<>AZ/^('E1K"";I3VMFD[7A.1G$L+QJKW1:UY1C]FK^$(^*Z[O M?4),FQ:BEH:=__E$=P/D]<UJ!#,RYOK;F<X^FQ)91_^>14\;MRD./?W*_4: MRA,?3I3;1"SU"H-ZUJXE=KH&7*/[1>)Z#')"Q]2>"=\]O&'F$)-,;&AIJ7-Z MW&>K@&(LQ(.X`5M.-/`:?I[0BKC(G[A;L+ZMBG+WGBD[F>*MC)6#V^I85AR] M![6$6.D5N`V'0+:Z*3EM'SZJ6,DK#%'UKXBK]/V%Q%ZHN`U&O9K;/DS'J-<5 M4;B2+)W\$5)3BBB/*<;H.L,?QLACM)F6(B40[U$8]J0DV"_3P(/6W8#VFVRR M6BP>%'VC2]"FT$(!F8Y[7Y$:A[?.3'S/\IY';^&P\%T.(L_531IDB;_?!C6! MP/0N270WY6UG:/S8[S9'G6Z;B$HEN<V=4%]K9MH.7C%57BB1=!O*11F!D'&4 M31*/5(7[)E2X+$.SC`E67FR0I4WUFOUI9ST[*2IZ=ZME@8)H4B5)W1?'51I M5)6FQWB=AQ6+I%A[):BPS.H5%GV'$18I!V0I,[OP*5[H4!=;Y!-Y1PB:7(&P MXK-V`?>'!Q6]5D>\/ZK#TV'"@GYZI*4_8L=?4[N$[T3+OEAX=JX(-\<IV@PJ M<]:7SRQ8?TK!C;P+[1'%T5)V7$O<F/LR9R'EI(9EAZ.\L0J6UUU*_/S0/594 M?W[1^M.*;C2`Y8WGOAC$\O9$9GLBLSV1V9[(?!W6[\L9,6V[R[K=9=UHE_6/ MF2)?6-,WV#C;GLIN3V6WI[+;4]GGG,K^OUNW?>G%5WG=,>_V3&][IO>[SO3^ MG%/Q#YI/&QW^I<[^0(?/<X_^MJ]];%_[^*I>^Y`1U^3[&O);]J4/F9'SSL=I M^IV/D'1H!H+TB-7W&_!7/U%B#1V?UC&R`'V&+WU\@O]^8]?-UM_;YRP,_<)& M/_7/^Z.A$@D!:3_BZP[B@&?))T!G6^RWW^CHRL"(>L:2P@Z5!!EC0$0!?HHO M.L-^&)-HXG(1O$*^"I>*+4<G+;Z(?@.5[-8)&Q_&89RVU,O*,AP,4!A`0C&) MBKOTEK\H?O",X@=1<>SI=?L<L*;38D4`5D,<MLM(#;6C0WD^?G0<GH]#*1EQ MR*#@&(:(DP+"8[N[P,RN?L:$S,5KD8*(SJ->5E[FE=6BLM&)V:Y>8;O[^$:A MAFP^/&"<D#NLN!R-TKH29;4$?OL4G_'\)(a)+)9`.A?(P"%V&$H%@K*!Y0JA\P M=BQ19Z:CHJYL5T4!T>&PO87M8X`/$9CV!1U]ACU<2Y)S?J=*(Q)&WIFP*LOU MN1FYY0ZF$*:V7IC\0^"9,=_:&KZ%SF3828DA55EY?66?YY[JW&$LCWGPA<,F M[<720Y\8XW]Q3\13G.0J1/)NQQ-Z]]S6/M_]'8JQ1#/YX$"'F:RS\L%!_3"^ MZ1)&NT$8"$P/(RM^\RH=BJM$KU%]E*^DA1=2KA!S1-!H/.OWV73ET5V4,$:3 M?',,Z$<NP\I$M*5[+J)*BMB:>$G`I5HD.=N+8CY6&04BY0X&R70=^&_E"WK$ MXY@^C*T)!><KK@3M]*/HE"+*5UQD3$&E[8D(N'D!RY11!_QLO.F#)F)BB]AL M<0%S[+ZGEXHQ2">&^:$X.B+J-X8$$TAOQ@7H7!Q#8(K8VQ3/0_`G8G!GV:^( MR-<H(RHL(R`5X?N<F[["O15*I431QKV)"%M&=S)D93&QN"%@8F)5<$[R\9D9 M76ZBJQT*0Y)KY*(:6KHPF%Y,!CIP`92)?3J0.(RO/W-7X(V*NTE*O3!J@DF; MST%AL+MXBX#)&%S@$H-P8GJU/QB_B+0+Y!Z6O^>>PG6H3]58JZ/073L[.YD` M0R)$.;V_&>WV2<I4F"&,%C)H#T>#3FL$+H1ZLU!6@O%29,F/2AVD(P8&7O=L M"V:R(<@Q*,O@IGWV&*6<C%E*Y9%>,=^`25&79#(+0EBA&K>/Q6PD%1=\"#$- MQ_'L)5#?R;G2$E8'DR-1(UE!=_Q/;@5KRGZ*N@A/<GQID8;B$'TYD\GW,USI MJ*,:B`%5=;(4Z@@3\LPL'\/:4@HBJ@KG2%R+E+A803>'P\YE#P5.*TK]+"*3 M/TEM!##\*0XZ_,;UYC@)I-_X`F4S$2[J"P8V'^`N1YL3"<AS,>V>LF_9&^,\ M].2P%S5I!`Y.*_HI&H&CHTJ]+AWDI/A5A<+_SEO]WG"4JU`RFD6R!KQ;J3T- M^;4G(K_V=.37GHK\VE.17]L8^;4G([_V%.37GHS\VEKDWY$W7!/3UZ]&J*M: M@XU\@LVP62I8<OYIGYE_89EX$FZ"^.5G(WXY0OSRQHA?WACQ8TKE\>F(7\Y# M_/+O1?SR[T#\<A[BER/$9_*=?;ZP`_!V'=PRQ``Y],LW*!R*N!")68Y+440% MM!WK%4*V4[VBA\B&%3UQ6$M1](=(/-1Q7'ZMZU84K!+'5$B(J=XR&^'O7L%P MXM&OAL%8IKC%B=&&78I^:BMA'^+"K>AWR9@8+D?^LA$1C9ABQ4-:B)IQM`:, M+,F#-?-T0V[R"C^%F[A\&(9A?9C0V*"\HJ,@3%5B<<(0KIG&:)H&-ZU1?Q"/ M6S3EHNX9R)<L5Y$N1"6*"G&X+R+('NX?AA%DI0P[^3'ZP]\T$-KFT":O[]_N MKNC;H-WJ#\Y%][0UN#-LO6U?-9/%]<14WJ#`KE3D:%$H;@;C;PAR'?QU!A06 M=7>W(CMCB+NRX6;9([^Y`*8C$&]WI+<[TE_7CK3XW1AR,UE\R>Y'B_1-MJ,E MI1KYOMZH'S?J2@`BO0ZVZ(25Z5,&=$_]T$H6)S3`L1%]61_?-B+[3VM7LQ(Q M#(3/]2GF6'?!=1&J5L1W\`5D+X+@8;&U[,5W=_Z2F:1)6T'84_DR369FTR3M M]TWA#%":ZPY'!7($.]=]MMNE>+.?,6@J%"S#HZ$*)O1WW_R3(<CX4:^R9-$B M9YAALOCWM4Y4%GTXQ%E\2,[/=F$O`BVFHJN=@OT)=5+P>7^3-CKDI!#<*7R) MH!_MN0(GQ,NKXI+\.D['+/$N3(@<4R7IS8!HC9RGI1.RCZJ5-=@]WJL>8<H: M].\YZ'FQ_#IC^#Y3;1<F3.(,`;B[&`9'$]%UC[[0'?%)P3+2[HP^7O3L$CO) M76OC-&!=\^"J'Z'NT9(a)XV.\G7"4$.4:NJ(`.H!H4=PG-#^T+N<;"4N3/I-*Y M=!?7!_=IGC/GQR`7R^->;U4?^=4V"XO*[TO#3HA'ML$'WBNV$]X-%PO<[`DF M_"432CN)&O86+#"X4!I$:8->ZVNR0,C)T,A2SC'L\.+\T5M>!75.7NB1,EP7 M%GPS%K+Z5*5'_Y(:4$N1IO$&?9#D8IX<Z_CUM"@-1=X6\N<8>LULJN)=G#D> M;CN1KC\>M<8%SY8<#)Z5G^%,GS)0G42<CL)A!+E(H9GX6V\4LQ0(a)H04$1./\ MK]J5,M:;;'JIU+#?`B;[L:Y#'1TZ0R4VR.]T=!/&B3DLZGC2/1.HXS1N?/_W L!4`Z0E!$09JP9Q&\C]/G&V_&2U#SF84_]:K>$JB:U66LQ^H7CZFPF/%T```` ` end

2 1

--
by Herbert Voss 28 Jul '97

28 Jul '97

Hello, I'm a novice in gpc and I hav a lot of problems. Here comes the first one: unit unit1; { reine testunit fuer gpc } Interface Procedure ClrScr; Implementation Procedure ClrScr; Var i : Real; Begin For i := 0 to 25 Do WriteLn; End; Begin End. I run gpc under Linux 2.0.30 and had the following output: voss@crazy_horse:/home/voss/pascal > gpc unit1.pas /usr/lib/crt1.o: In function start': /usr/lib/crt1.o(.text+0x5a): undefined reference to main' Another problem comes, when I choosed the Type Byte for the variable i. Herbert

1 0

Re: Allocation/Memory Leak question
by Peter Gerwinski 26 Jul '97

26 Jul '97

According to Ken: > > program allo2; > var > mm: array[1..500000] of real; > indx : __long__ integer; (* JFYI, `__long__ Integer's have 64 bits. For your program, an ordinary 32-bit `Integer' would be sufficient. *) > ma: array[1..500000] of real; > [...] > This program realy does nothing of value but when the second FOR > loop is running, the program starts eating up memory. Really at the second FOR loop, not already at the first one? I don't know the details of the memory management of Linux, but AFAIK, the storage is physically allocated when the array is referenced. > According to TOP, > the program uses 8094 kBytes just before the loop ends. Why should > this program be so memory hungry at that time? [...] Simply because one `Real' has eight bytes, and you are allocating one million of `Real's. 8000000 bytes are 7812.5kB; if you include code, runtime library, etc., 8094kB seems realistic for me. Peter Dipl.-Phys. Peter Gerwinski, Essen, Germany, free physicist and programmer peter.gerwinski(a)uni-essen.de - http://home.pages.de/~peter.gerwinski/ [970201] maintainer GNU Pascal [970714] - http://home.pages.de/~gnu-pascal/ [970125]

1 0

Re: problems with allocation?
by Peter Gerwinski 26 Jul '97

26 Jul '97

According to Giuseppe Petrella: > > The following simple program gave segmentation fault compiled on > digital alpha running osf > > PROGRAM prova(input,output); > var > mm: array[1..500000]of real; > BEGIN > END. Global variables of the program are allocated in the stack segment, not in the data segment. DJGPP limits the stack to 256kB, Linux seems to have no limit here. Obviously, OSF has a limit, too. This is a known problem. I would like to do something against it, but I was told that this would break Extended Pascal's nonlocal `goto' (whatever that is), so I don't dare to touch this part of the compiler as long as I don't have test programs. To work around, either move the array to a Unit (or Module), or make it a `__static__ array [...] of Real'. ^^^^^^^^^^ Hope this helps, Peter Dipl.-Phys. Peter Gerwinski, Essen, Germany, free physicist and programmer peter.gerwinski(a)uni-essen.de - http://home.pages.de/~peter.gerwinski/ [970201] maintainer GNU Pascal [970714] - http://home.pages.de/~gnu-pascal/ [970125]

1 0

Allocation/Memory Leak question
by Ken 25 Jul '97

25 Jul '97

Just a quick note/question/observation. Consider this program: ---------------------------------------------------------------------------- program allo2; var mm: array[1..500000] of real; indx : __long__ integer; ma: array[1..500000] of real; begin for indx := 1 to 500000 do begin mm[indx] := indx / 2.2; if indx = 1 then ma[indx] := 1.0 else ma[indx] := (ma[indx-1] + mm[indx]) / indx; end; for indx := 1 to 500000 do writeln(indx, mm[indx], ma[indx]); end. ---------------------------------------------------------------------------- This program realy does nothing of value but when the second FOR loop is running, the program starts eating up memory. According to TOP, the program uses 8094 kBytes just before the loop ends. Why should this program be so memory hungry at that time? FYI, I am using GPC 2.0 with Linux 2.??? later... Ken L.

1 0

Re: problems with allocation?
by Ken 25 Jul '97

25 Jul '97

At 05:28 PM 7/25/97 +0200, Giuseppe Petrella wrote: >Hallo! >The following simple program gave segmentation fault compiled on >digital alpha running osf > >PROGRAM prova(input,output); >var > mm: array[1..500000]of real; >BEGIN >END. > >however array[1..1000] gives no problem. ???? >Thank you for your attention >Fulvio Ciriaco > > Compiles fine on a Linux box. What version of GPC are you using? Later... Ken L.

1 0

problems with allocation?
by Giuseppe Petrella 25 Jul '97

25 Jul '97

Hallo! The following simple program gave segmentation fault compiled on digital alpha running osf PROGRAM prova(input,output); var mm: array[1..500000]of real; BEGIN END. however array[1..1000] gives no problem. ???? Thank you for your attention Fulvio Ciriaco

1 0

Re: SizeOf with Objects BUG
by Frank Heckenbach 25 Jul '97

25 Jul '97

Bernhard Tschirren wrote: > You cannot use SizeOf with objects?!?! It just does not work. It either > aborts during compilation, or it returns rediculous values during execution. Did you initialize the object with a constructor call? BTW: There is a little difference to BP: In BP, objects that have neither constructors, destructors nor virtual methods don't have a VMT link. So it's possible in BP to get their size without initalizing them. I think that's kind of a misfeature, since the following program produces the wrong result for vy in BP, so IMHO, BP's behaviour should not be duplicated by gpc. (Though it's not 100% BP compatible then, a good BP program shouldn't make use of it, anyway.) type x=object a:integer; end; y=object(x) b:integer; end; procedure w(var p:x); begin writeln(sizeof(p)) end; var vx:x; vy:y; begin w(vx); w(vy); end. -- Frank Heckenbach, Erlangen, Germany heckenb(a)mi.uni-erlangen.de http://www.mi.uni-erlangen.de/~heckenb/links.htm

1 0

Re: SizeOf with Objects BUG
by Peter Gerwinski 24 Jul '97

24 Jul '97

According to Bernhard Tschirren: > You cannot use SizeOf with objects?!?! It just does not work. It either > aborts during compilation, or it returns rediculous values during execution. Some of my test programs check that `SizeOf' *does* work with objects. If it doesn't for one of your programs, please post it here. Thanks for your cooperation, Peter Dipl.-Phys. Peter Gerwinski, Essen, Germany, free physicist and programmer peter.gerwinski(a)uni-essen.de - http://home.pages.de/~peter.gerwinski/ [970201] maintainer GNU Pascal [970714] - http://home.pages.de/~gnu-pascal/ [970125]

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

Gpc