search for: uge

...is the output. Thanks for the help. > Sys.getenv() BASH_FUNC_module() () { eval `/cm/local/apps/environment-modules/3.2.10/Modules/$MODULE_VERSION/bin/modulecmd bash $*` } COLUMNS 152 CPATH /cm/shared/apps/uge/8.2.1/include CVS_RSH ssh DISPLAY localhost:10.0 EDITOR vi G_BROKEN_FILENAMES 1 HISTCONTROL ignoredups HISTSIZE 1000 HOME /root HOSTNAME test LANG en_US.UTF-8 LD_LIB...

...ened URL ================================================== downloaded 54 Kb * installing *source* package ?gee? ... ** libs gfortran -fpic -O2 -g -c dgedi.f -o dgedi.o gfortran -fpic -O2 -g -c dgefa.f -o dgefa.o gcc -I/usr/lib64/R/include -I/usr/local/include -fpic -O3 -g -std=gnu99 -c ugee.c -o ugee.o gcc -shared -Bdirect,--hash-stype=both,-Wl,-O1 -o gee.so dgedi.o dgefa.o ugee.o -lblas -lgfortran -lm -lgfortran -lm -L/usr/lib64/R/lib -lR /usr/bin/ld: cannot find -lblas collect2: ld returned 1 exit status make: *** [gee.so] Error 1 ERROR: compilation failed for package ?gee? * remov...

...urce packages from CRAN using install.packages(). Below is a typical installation output: * installing *source* package ?gee? ... ** libs gfortran -fpic -O3 -pipe -g -c dgedi.f -o dgedi.o gfortran -fpic -O3 -pipe -g -c dgefa.f -o dgefa.o gcc -std=gnu99 -I/usr/share/R/include -fpic -O3 -pipe -g -c ugee.c -o ugee.o gcc -std=gnu99 -shared -o gee.so dgedi.o dgefa.o ugee.o -lblas -lgfortran -lm -lquadmath -lgfortran -lm -lquadmath -L/usr/lib64/R/lib -lR /usr/bin/ld: cannot find -lquadmath collect2: ld returned 1 exit status make: *** [gee.so] Error 1 ERROR: compilation failed for package ?gee? * re...

Hello Everyone, After trying several ways to compile R 3.2.2 without luck, I?m reaching out for help. The ?make check? does not hanges for some reason and when trying to install a package it cannot list the download sites (see below). What could be the problem? ./configure --enable-R-shlib --enable-BLAS-shlib hostname = test uname -m = x86_64 uname -r = 2.6.32-573.7.1.el6.x86_64 uname -s =

> there is no such information. You can still consider every type to have values > in, say, T = [-2^31; 2^31-1]. Probably you are trying to deduce an interval of > possible values for each register. You will need to allow intervals to wrap > around the end of T since (eg) the basic "add" operator in LLVM uses modulo > arithmetic, i.e. if you add 1 to 2^31-1 you get

...100 87 110 191 71 > [20] 3 48 144 101 115 161 52 164 105 154 266 93 1 53 167 161 130 195 > > $uext > [1] 17 12 13 17 14 32 34 5 50 132 93 66 86 66 65 63 54 97 34 > [20] 41 81 123 88 85 213 159 171 158 140 153 75 65 126 252 225 127 192 > > $uges > [1] 101 54 84 99 94 155 47 6 65 209 154 122 148 93 165 150 164 288 105 > [20] 44 129 267 189 200 374 211 335 263 294 419 168 66 179 419 386 257 387 > > $month > [1] "01" "02" "03" "04" "05" "06" "07&quo...

...e code that does address comparisons to check whether a given pointer is within a certain memory range. For example: if (0xff00 <= &a[x] && &a[x] < 0xffff) This results in IR like: %2 = getelementptr inbounds [100 x i32], [100 x i32]* @a, i32 0, i32 %0, !dbg !9 %3 = icmp uge i32* %2, inttoptr (i32 65280 to i32*), !dbg !10 %4 = icmp ult i32* %2, inttoptr (i32 65535 to i32*), !dbg !11 Suppose that 'a' is a global and we know the range of addresses where it can be allocated. Is it a safe transformation to use that range and constant fold the icmps? Is so, is it...

...check should be removed: %jl_value_t = type {} @jl_domain_exception = external global %jl_value_t* declare void @jl_throw_with_superfluous_argument(%jl_value_t*, i32) declare float @llvm.sqrt.f32(float %Val) define float @julia_f_64805(float) { top: %1 = fmul float %0, %0 %2 = fcmp uge float %1, 0.000000e+00 br i1 %2, label %pass, label %fail fail: ; preds = %top %3 = load %jl_value_t** @jl_domain_exception, align 8 call void @jl_throw_with_superfluous_argument(%jl_value_t* %3, i32 2) unreachable pass:...

...*C = dyn_cast<SCEVConstant>(AddRec->getStart())) if (!C->getValue()->isZero()) ConservativeResult = ConservativeResult.intersectWith( ConstantRange(C->getValue()->getValue(), APInt(BitWidth, 0))); and conclude that %idx.dec is always UGE %begin. This is a real bug waiting to happen. As far as I can tell, in the presence of no-wrap flags, 'sub X Y' is not the same as 'add X -Y', and subtractions can be expressed as additions only if they don't have any no-wrap flags on them. While I have not thought about nsw...

...N set by another instruction to an auxiliary register. # this should translate to one instruction like: r3 = add r1, r2 # n is extended to 257-bit for simplicity define i256 @add(i256 %b, i256 %c, i257 %n) { %1 = zext i256 %b to i257 %2 = zext i256 %c to i257 %3 = add i257 %2, %1 %4 = icmp uge i257 %3, %n %5 = select i1 %4, i257 %n, i257 0 %6 = sub i257 %3, %5 %7 = trunc i257 %6 to i256 ret i256 %7 } The ultimate goal is to use LLVM to optimize and emit machine code for programs full of modular addition, subtraction and multiplication. For example, I might want off-the-shelf LLV...

...x i8> %wide.load.1 to <4 x i32> %11 = getelementptr inbounds i8* %pix2, i64 4 %12 = bitcast i8* %11 to <4 x i8>* %wide.load7.1 = load <4 x i8>* %12, align 1 %13 = zext <4 x i8> %wide.load7.1 to <4 x i32> %14 = sub nsw <4 x i32> %10, %13 %15 = icmp uge <4 x i8> %wide.load.1, %wide.load7.1 %16 = sub nsw <4 x i32> zeroinitializer, %14 %17 = select <4 x i1> %15, <4 x i32> %14, <4 x i32> %16 %18 = add nsw <4 x i32> %17, %7 %19 = getelementptr inbounds i8* %pix1, i64 8 ..... The test case is a perfect examp...

...%4 = mul i64 %0, %0 %5 = fsub float %3, %3 store double* %o0, double** %o0.addr %6 = and i64 %0, %0 %7 = fptoui float %3 to i1 %8 = ashr i1 %7, %7 store double* null, double** %o0.addr %9 = fptosi double %input to i64 %10 = shl i1 %7, %7 %11 = sitofp i1 %8 to float %12 = icmp uge i64 %6, %4 %13 = and i64 %0, %6 %14 = fcmp ole float %11, %11 %15 = ashr i1 %10, %7 %16 = sext i1 %10 to i64 %17 = ashr i1 %7, %12 %18 = alloca [2 x double] %19 = fcmp true float %11, %3 br label %splittedblock7 splittedblock7: ; preds = %entry %...

...i32 1, i32 3, i32 undef> %I40 = insertelement <4 x i16> %Shuff7, i16 %E19, i32 1 %ZE41 = zext i1 true to i16 %Sl42 = select i1 true, i1 true, i1 true br i1 %Sl42, label %CF237, label %CF246 CF246: ; preds = %CF246, %CF271 %Cmp43 = icmp uge i64 %L37, %L18 br i1 %Cmp43, label %CF246, label %CF249 CF249: ; preds = %CF249, %CF263, %CF246 %L44 = load i64, i64* %A store i64 %L5, i64* %Sl17 %E45 = extractelement <4 x i16> %Shuff14, i32 2 %Shuff46 = shufflevector <1 x i32> zer...

...liary register. > > # this should translate to one instruction like: r3 = add r1, r2 > # n is extended to 257-bit for simplicity > define i256 @add(i256 %b, i256 %c, i257 %n) { > %1 = zext i256 %b to i257 > %2 = zext i256 %c to i257 > %3 = add i257 %2, %1 > %4 = icmp uge i257 %3, %n > %5 = select i1 %4, i257 %n, i257 0 > %6 = sub i257 %3, %5 > %7 = trunc i257 %6 to i256 > ret i256 %7 > } > > The ultimate goal is to use LLVM to optimize and emit machine code for > programs full of modular addition, subtraction and multiplication. For...

...lling *source* package ‘gee’ ... ** package ‘gee’ successfully unpacked and MD5 sums checked ** libs gfortran -fpic -g -O2 -c dgedi.f -o dgedi.o gfortran -fpic -g -O2 -c dgefa.f -o dgefa.o gcc -std=gnu99 -I/ebio/abt6/jmao/R/R-2.15.0/include -DNDEBUG -I/usr/local/include -fpic -g -O2 -c ugee.c -o ugee.o gcc -std=gnu99 -shared -L/usr/local/lib64 -o gee.so dgedi.o dgefa.o ugee.o -L/ebio/abt6/jmao/R/R-2.15.0/lib -lRblas -lgfortran -lm -lgfortran -lm installing to /ebio/abt6/jmao/R/Rpacks/gee/libs ** R ** preparing package for lazy loading ** help *** installing help indices ** building p...

> We are permitted to turn 'sub nsw i32 %x, 1' into 'add nsw i32 %x, -1' nsw also has the same problem: sub nsw int_min, int_min is 0 but add nsw int_min, (-int_min) is poison -- Sanjoy

...ected to be 0 then. If I run the bitcode with lli (JIT off), no problem. If I use the JIT (lli --force-interpreted=true), then it returns 1. (Converely, fcmp une with the JIT returns 0 instead of the expected 1 ... without JIT it returns the expected 0). The other operators (oeq, oge, ..., ueq, uge, ...) seem to work fine. Is this a bug with the JIT or have I done to muck this up? I've included source below which illustrates the problem (with doubles on fcmp one only ... ). Thanks for your help. ~ David declare void @exit(i32) declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) ; see...

Maybe we can have two versions of the intrinsic function, "ordered" and "unordered", just like fcmp has [1]. Would that work ? [1] - http://llvm.org/docs/LangRef.html#fcmp-instruction On Dec 17, 2012, at 11:14 AM, "Schoedel, Kevin P" <kevin.p.schoedel at intel.com> wrote: > At Monday, December 17, 2012 2:05 PM, Nadav Rotem [mailto:nrotem at apple.com]

...; # this should translate to one instruction like: r3 = add r1, r2 >> # n is extended to 257-bit for simplicity >> define i256 @add(i256 %b, i256 %c, i257 %n) { >> %1 = zext i256 %b to i257 >> %2 = zext i256 %c to i257 >> %3 = add i257 %2, %1 >> %4 = icmp uge i257 %3, %n >> %5 = select i1 %4, i257 %n, i257 0 >> %6 = sub i257 %3, %5 >> %7 = trunc i257 %6 to i256 >> ret i256 %7 >> } >> >> The ultimate goal is to use LLVM to optimize and emit machine code for >> programs full of modular addition, subt...

At Monday, December 17, 2012 2:05 PM, Nadav Rotem [mailto:nrotem at apple.com] wrote: >This part worries me. The new min/max intrinsics will only be useful if we could pattern match cmp/select into them. Yes, that's the obvious alternative. I don't think we have any strong opinion either way, and fcmp/select is certainly easier to implement. -- Kevin Schoedel, Software Developer,