public inbox for gcc-prs@sourceware.org help / color / mirror / Atom feed
From: Darren Tucker <dtucker@dodgynet.dyndns.org> To: gcc-gnats@gcc.gnu.org Subject: optimization/7625: gcc pessimized 64-bit % operator on hppa2.0 Date: Sun, 18 Aug 2002 14:26:00 -0000 [thread overview] Message-ID: <200208181129.g7IBTUO03308@dodgynet.dyndns.org.> (raw) >Number: 7625 >Category: optimization >Synopsis: gcc pessimized 64-bit % operator on hppa2.0 >Confidential: no >Severity: non-critical >Priority: low >Responsible: unassigned >State: open >Class: pessimizes-code >Submitter-Id: net >Arrival-Date: Sun Aug 18 04:36:00 PDT 2002 >Closed-Date: >Last-Modified: >Originator: Darren Tucker >Release: 3.2 >Organization: none >Environment: System: HP-UX c240 B.11.00 A 9000/782 2007058445 two-user license host: hppa2.0w-hp-hpux11.00 build: hppa2.0w-hp-hpux11.00 target: hppa2.0w-hp-hpux11.00 configured with: ../gcc-3.2/configure --with-as=/usr/local/hppa2.0w-hp-hpux11.00/bin/as --with-gnu-as --with-ld=/usr/ccs/bin/ld --enable-languages=c,c++ >Description: GCC seems to compile code for the 64-bit "%" operator that is about 6 times slower that the HP native compiler on HPPA2.0 machines, even with -march=2.0. This was noticed affecting OpenSSL DSA operations and identified by Deron Meranda . For background, please see http://marc.theaimsgroup.com/?l=openssh-unix-dev&m=102646106016694&w=2 $ cat logmodtest.c #include <stdio.h> int main() { unsigned long long i, a=0; for(i=2000000; i; --i) a += (i+10) % i; printf("Result=%llu\n", a); exit(0); } $ cc +O3 longmodtest.c $ time ./a.out Result=19999913 real 0m0.649s user 0m0.650s sys 0m0.000s $ gcc -O3 -march=2.0 longmodtest.c $ time ./a.out Result=19999913 real 0m3.712s user 0m3.700s sys 0m0.020s >How-To-Repeat: $ gcc -O3 -march=2.0 -v -save-temps longmodtest.c Reading specs from /usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2/specs Configured with: ../gcc-3.2/configure --with-as=/usr/local/hppa2.0w-hp-hpux11.00/bin/as --with-gnu-as --with-ld=/usr/ccs/bin/ld --enable-languages=c,c++ Thread model: single gcc version 3.2 /usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2/cpp0 -lang-c -v -D__GNUC__=3 -D__GNUC_MINOR__=2 -D__GNUC_PATCHLEVEL__=0 -D__GXX_ABI_VERSION=102 -Dhppa -Dhp9000s800 -D__hp9000s800 -Dhp9k8 -DPWB -Dhpux -Dunix -D__hppa__ -D__hp9000s800__ -D__hp9000s800 -D__hp9k8__ -D__PWB__ -D__hpux__ -D__unix__ -D__hppa -D__hp9000s800 -D__hp9k8 -D__PWB -D__hpux -D__unix -Asystem=unix -Asystem=hpux -Acpu=hppa -Amachine=hppa -D__OPTIMIZE__ -D__STDC_HOSTED__=1 -D_PA_RISC1_1 -D__hp9000s700 -D_HPUX_SOURCE -D_HIUX_SOURCE -D__STDC_EXT__ -D_INCLUDE_LONGLONG longmodtest.c longmodtest.i GNU CPP version 3.2 (cpplib) (hppa) ignoring nonexistent directory "NONE/include" ignoring nonexistent directory "/usr/local/hppa2.0w-hp-hpux11.00/include" #include "..." search starts here: #include <...> search starts here: /usr/local/include /usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2/include /usr/include End of search list. /usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2/cc1 -fpreprocessed longmodtest.i -quiet -dumpbase longmodtest.c -march=2.0 -O3 -version -o longmodtest.s GNU CPP version 3.2 (cpplib) (hppa) GNU C version 3.2 (hppa2.0w-hp-hpux11.00) compiled by GNU C version 3.2. /usr/local/hppa2.0w-hp-hpux11.00/bin/as --traditional-format -o longmodtest.o longmodtest.s /usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2/collect2 -L/lib/pa1.1 -L/usr/lib/pa1.1 -z -u main /usr/ccs/lib/crt0.o -L/usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2 -L/usr/ccs/bin -L/usr/ccs/lib -L/opt/langtools/lib -L/usr/local/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.2/../../.. longmodtest.o -lgcc -lgcc_eh -lc -lgcc -lgcc_eh begin 644 longmodtest.i.bz2 M0EIH.3%!62936=43M8T`!)W?@'X46G__[P-D$$Z_[__Z8`J_>`Z:6VN[?>JC ME=ZUMPY`.SW<V`'">W<"22:)C0B83:GJ::,33(TT```&@$IJ9`IE3\BGFJ,F MGJ8F0``-``]0!S3$R9-&$P3$TP"8!#!&!&`2:2131/0F1,$#1ZC:GJ!H&@:# M0`(DE3U/5/4/U31LD9/4WI1H,:C33$--,$9&!$D0"",(:AZJ>4_5,8H`'I`! MIID8AF06"(2$B$0.E%$BQ6$4!`('O3R,(I`4BS_;20Q;1:J-M9Y:%8"C#&5J M`+`J5A60UACF.#,B7"RX%D,3,ZV88JQE:VE!2L*@T.E9_UT(JL$8:$E%8JBR M_QDS(52VI14$4=Z<H!(3CJ`I).*2'0G0:Z6B.4K:6NBR8*6R:88FJ62;,V8: M2%0#$,2;@S')K21S<Z[CADI<L/S173GO65Y24J7O9*JM-Z1>W(;[[9#DZ^Q/ M8'H/%;%6+*A41@HQ2"(*(BHBGW"$]26V8HSRVKS_;YOJ^"/="(R0"#P"A7B% M0DD)2V5.P9L0_4G'I?PEBJ<X,VD>C5U:&AGXI6"A!8/4&F<"L+"Q)J<VT`H: M/;+ZW;2ASCE:AD20M!K[/&S;961-5B.,*D@$!4A1-#`D,^+LR0_3#MJ1HBZA M10@T\O+FM*HC4O#D)I)IFG&3D9I,3E<8X!$HS,UH<<%HK>`F5K1J720&1<S` M%,F%WP,9(9*S",G[)MPX\!_,])ZX#F0D31@):F6G.M#G)"-+('\&UWZYNAH7 M77J[E_4X_[T>?VW5A#<#?^+.H:2)V<<?(HG9]M#$Q6=>0Z1[W)R^S;N]P^0[ M>ZIQ^(AWDGJMQ,2*5JC6HU(.F)DU==$33!&L*+KS6A1&+INM8D?\U02`2,8) MCEU03[D$_ECYH)Q%^6<3*@FL@G$@E[GQK18V85))())!(1(U0\>J"4)]P@GD M:BD4BAH@E()G`;H*Q0]0HT7X<<X\G)MUJ8WB)L()+Y3G\<L4$\C6#C)MEG=' MTS02K_9\?P02XX0.PMZ,Z*U_C2&?J@G!!,K\?&BVU2$WVAL0`;BC;7;E\T$P MM#@B'E#Z_C7Y('LF/!=B_N@X]=)H#6N.<CIN,-:E9LX]G'U#!DFRRS>8S97V M*\9QB2FMM!Z)2L!M*[)SHL^\6."`WID?RP/I%<+%:_+=\[LP*(@_7[+V`<X: M];6^15#ZO<^YJ+L[8([?%=P=!1/$KK\H;!#8ID)>=;"7._]E.X[=^W?T]77( M0\WH%6,51(L@I6//=(0Q\)"*,5G$QAQ:WPYC9G,V6SAT1YHFOSV<@46!$&"N M'4PXI-@/FE@>=@:R%T!O9B-HWDK0KH^$"T*BM!5.(P._D)-NPD=J'>:U:MLH M92:',E@M>;:\6:F]Y=H&VX5,4VXVB8IU45SH_&G:F@DBJ*-/[<CXW3.89T3F M=5"9]4L40^9+H-3%7:4;#8_I5\>M"O:E*;C#V`2IPXO,P)/-2_D7]$Z\Y;7G M5>=BO0S)A6VF"";<`M8?M10W:F_U<6_F221A)(T?,03MW^WJ4<NDDEN24X/+ M`DE]_*^V9]\^CBG-!-@+ZS>@F@^2"7X0?T,"0`A(2$4/Z174F>!F&'D\'D)" M;Z?.>!6EIG'J\6'0(L%1!(B(O&=^3L/'-X`Q@<4A8@;9A4<R\X%U_&'#2>)B M!J*/B&HBN.`)-NFUC[]^I[0,+9S5YOD1[=L;9BCD=<,FV`8?^,H!"489<+9X M?#-#+0%)V$5F0BO+N;LYB%<32:EM,00*E\5':S^.+P21DW;].7X)TG>*D@C" MJ'OR$5W=?1HU)($)+%O$X51V`Q-:"2':8$,@C:^3J\[!A&32QJ9<,1](VC4$ MN"^#V*,OD@DL:8X0K.46?:"FP&/O%'A[E4(K_=!-R"1!-P3W[,[B"<T,6TXM M+!NRX;=*$B\A2$%''2ZH2%D=E)-P:"]K!TA/=9`/"JP3G%!(C$1=`RK%51C$ M&"OCI5$541\,"U4599,025613?W!OS-171"X-N9XN2F@](B9)D=L3T8G<,&` M9\`*]$$[()P<M^*CL;3(&*+NR=`BBG5"7*R"MA*LJAU*<GNO;`\2$^7`1E;6 MR=69'+&62EI]>2&15S#18<Q:2`/Z>0)Z"8EO+84<(6PE.`*77#8N%=B-86OL MY<SGXF(*9H!BJ+@P`C&$$(D0=0,<-WLP!(U*\D4KI".YWN?)38R0"(!IX9<= MNAWV>AGZJZ%F)L5QU>.*"<\+Y();PU?(17H%4(Z8%C#?6,_-#$OQ.*;T$]:- M3B=-#89N-%QZ9X,FJ";LK>:"'CRV;.HH[7AA?'I8ND&&UR85'KT0YW;`@:'% MPWW#>KJ!"!1PM0!(DCT%://#'PXX,Q+;GB^4+<Y?(\7S<9J86E4VL#"U=04J MI$)E7N,F]M*K8M1F*-^-[FG6`!%NY@%,/:J8+?(KC<5RN;I1"=26L6ZE)85C M:-[P::LL8B*"&K*UJS2%M*QB*"B"B8T1%B-*-"M:JEIIAG#8YO:WVWL%M@JJ M)UY+L:)@)1A`H(`Q%*4*82!LD#@A@R%*4U&0#."Y6N`2)RJ;9Y!Q9@J:AR[; M_50#/1V4\#?>QD;6A9:$V76YN-V*6+>[W[$@2%_G<;QPTE\H.7A(=$)(Z"D% M)J(FY5YQ(D%$01FJ9O36\VVFH(JPS.&&:J".C:QBM0IDS:@93YE.TZX&==44 M?DMT(HEYYSS\&H!`^$!J#"#WACWTO?B*/!5.UU+^'&C$FVU#1$P/H@F?!FW# AV()[RZ('+',$0@K]0=Q=D\X2/L[VM_\7<D4X4)#5$[6- ` end For comparison, if you split the "%" operation out into a separate source file: unsigned long long longmod(unsigned long long a, unsigned long long b) { return(a % b); } the HP compiler produces the following assembler output: .LEVEL 2.0N .SPACE $TEXT$,SORT=8 .SUBSPA $CODE$,QUAD=0,ALIGN=4,ACCESS=0x2c,CODE_ONLY,SORT=24 longmod .PROC .CALLINFO FRAME=0,ARGS_SAVED,ORDERING_AWARE .ENTRY DEPD %r25,31,32,%r26 ;offset 0x0 DEPD %r23,31,32,%r24 ;offset 0x4 EXTRD,U %r26,31,32,%r25 ;offset 0x8 .CALL ;in=23,24,25,26;out=21,22,28,29; (MILLICALL) B,L $$rem2U,%r31 ;offset 0xc EXTRD,U %r24,31,32,%r23 ;offset 0x10 DEPD %r28,31,32,%r29 ;offset 0x14 $00000002 $L0 BVE (%r2) ;offset 0x18 .EXIT EXTRD,U %r29,31,32,%r28 ;offset 0x1c .PROCEND ;in=23,25;out=28,29;fpin=105,107; .SPACE $TEXT$ .SUBSPA $CODE$ .SPACE $PRIVATE$,SORT=16 .SPACE $TEXT$ .SUBSPA $CODE$ .EXPORT longmod,ENTRY,PRIV_LEV=3,ARGW0=GR,ARGW1=GR,ARGW2=GR,ARGW3=GR,RTNVAL=GR,LONG_RETURN .IMPORT $$rem2U,MILLICODE .END >Fix: Be patient :-) >Release-Note: >Audit-Trail: >Unformatted:
next reply other threads:[~2002-08-18 11:36 UTC|newest] Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top 2002-08-18 14:26 Darren Tucker [this message] 2002-11-21 14:26 danglin
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=200208181129.g7IBTUO03308@dodgynet.dyndns.org. \ --to=dtucker@dodgynet.dyndns.org \ --cc=dtucker@zip.com.au \ --cc=gcc-gnats@gcc.gnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).