search for: argi

...some more major surgery needs to be done to memdep to make it work both backward (to support GVN) and forward (to support DSE). In the meantime, enhancing memcpyopt is fine, so long as there are good testcases (and your patch has them!). Nit picky stuff about the patch: + for(unsigned argI = 0; argI < CS.arg_size(); ++argI) { + if(CS.getArgument(argI)->stripPointerCasts() == pointer) Please put spaces after if/for. You get this right in some places, but wrong in others. + Value* pointer = L->getPointerOperand(); Please use "Value *pointer" style through...

Hello, This patch fixes PR2218. However, I'm not pretty sure that this optimization should be in MemCpyOpt. I think that GVN is good place as well. Regards -- Jakub Staszak -------------- next part -------------- A non-text attachment was scrubbed... Name: pr2218.patch Type: application/octet-stream Size: 6146 bytes Desc: not available URL:

...to be better in the > future :( > > This patch is looking like a great improvement. Some comments on > formatting: > > Please pull this out to a helper function: > + CallSite CS = CallSite::get(C); > + > + // Pointer must be a parameter (case 1) > + for (argI = 0; argI < CS.arg_size(); ++argI) > + if (CS.getArgument(argI)->stripPointerCasts() == pointer) > + break; > + > + if (argI == CS.arg_size()) > + return false; > > per http://llvm.org/docs/CodingStandards.html#hl_predicateloops > > > + // Store c...

...k for the exact sequence of instructions expected, which also allows the tests to be merged into one file. +/// MemCpyOpt::pointerIsParameter - returns true iff pointer is a parameter of +/// C call instruction. +bool MemCpyOpt::pointerIsParameter(Value *pointer, CallInst *C, unsigned &argI) +{ + CallSite CS = CallSite::get(C); + for (argI = 0; argI < CS.arg_size(); ++argI) Please make this a static function, it doesn't need "this". Also, please do something like this in the for loop: + for (argI = 0, CS.arg_size(); argI != argE; ++argI) pointerIsParam...

...o be done to memdep to make it work both > backward (to support GVN) and forward (to support DSE). In the > meantime, enhancing memcpyopt is fine, so long as there are good > testcases (and your patch has them!). > > Nit picky stuff about the patch: > > + for(unsigned argI = 0; argI < CS.arg_size(); ++argI) { > + if(CS.getArgument(argI)->stripPointerCasts() == pointer) > > Please put spaces after if/for. You get this right in some places, > but wrong in others. > > + Value* pointer = L->getPointerOperand(); > > Please use &qu...

...ructions expected, which also allows the tests to be merged into > one file. > > +/// MemCpyOpt::pointerIsParameter - returns true iff pointer is a > parameter of > +/// C call instruction. > +bool MemCpyOpt::pointerIsParameter(Value *pointer, CallInst *C, > unsigned &argI) > +{ > + CallSite CS = CallSite::get(C); > + for (argI = 0; argI < CS.arg_size(); ++argI) > > Please make this a static function, it doesn't need "this". Also, > please do something like this in the for loop: > > + for (argI = 0, CS.arg_size();...

...akes me a while to get back to it. I'll try to be better in the future :( This patch is looking like a great improvement. Some comments on formatting: Please pull this out to a helper function: + CallSite CS = CallSite::get(C); + + // Pointer must be a parameter (case 1) + for (argI = 0; argI < CS.arg_size(); ++argI) + if (CS.getArgument(argI)->stripPointerCasts() == pointer) + break; + + if (argI == CS.arg_size()) + return false; per http://llvm.org/docs/CodingStandards.html#hl_predicateloops + // Store cannot be volatile (case 2) and must-alias with...

...ning. Neither of these mention the hard-coded limits on the acceptable values of this option in options.c <https://github.com/wch/r-source/blob/a7356bf91b511287aacd3a992abfbcb75b60d93c/src/main/options.c#L546-L552> : if (streql(CHAR(namei), "warning.length")) { int k = asInteger(argi); if (k < 100 || k > 8170) error(_("invalid value for '%s'"), CHAR(namei)); R_WarnLength = k; SET_VECTOR_ELT(value, i, SetOption(tag, argi)); } Further, it appears there's a physical limit on the length of the error message itself which is only slightly larger t...

Hello All, I want to ask about CentOS and money. Please do not start some kind of shitstorm over this, it isn't productive. I have just been looking at archive.org to see when the paypal option went missing from the donate menu at centos.org. I can't pin it down, but it was there on October 16, 2010 and isn't now, so that's the information I have on that point. There was a

Hello, I have an ESRI ArGis geo database directory which I would like to convert to geojson or a shape file (or anything else that I can read into R). Unfortunately that does not work out of the box with rgdal, because it does not come with the fileGDB or openfileGDB driver. I could successfully install gdal and the fileGDB...

Hi, > I believe the LowerCall is doing what it needs to do - passing pointer either on the stack or in register as per ABI. >From very quick test-cases with no understanding of XCore, that looks plausible. > LowerFormalArguments () calls CCInfo.AnalyzeFormalArguments(Ins, CC_XCore), which calls the CC_XCore(). > This is where I placed the CCIfByVal<CCPassByVal<0,4>>

...corrupt the original data viz pass-by-reference!) This should ideally be done early on in the IR in my thinking - to allow optimisation if the data is only ever read. FYI, the clang hack is - notice the "CreateMemCpy": CodeGenFunction::EmitFunctionProlog(){ ... if (ArgI.getIndirectByVal()) { llvm::AllocaInst *ByValue = CreateMemTemp(Ty, V->getName() + "agg.tmp"); llvm::ConstantInt * size = llvm::ConstantInt::get(IntPtrTy, getContext().getTypeSizeInChars(Ty).getQuantity()); Builder.CreateMemCpy(ByVa...

Rather then implementing kafs in MIT Kerberos, I would like to suggest an alternative which has advantages to all parties. The OpenSSH sshd needs to do two things: (1) sets a PAG in the kernel, (2) obtains an AFS token storing it in the kernel. It can use the Kerberos credentials either obtained via GSSAPI delegation, PAM or other kerberos login code in the sshd. The above two

Hi Tim, Thank you for the input. I think I follow you. I believe the LowerCall is doing what it needs to do - passing pointer either on the stack or in register as per ABI. The LowerFormalArguments() is where I am stuck. LowerFormalArguments () calls CCInfo.AnalyzeFormalArguments(Ins, CC_XCore), which calls the CC_XCore(). This is where I placed the CCIfByVal<CCPassByVal<0,4>> which

...gt; return asInteger(GetOption(install("scipen"), rho)); > } 235a240,243 > SET_TAG(v, install("scipen")); > SETCAR(v, ScalarInteger(0)); > v = CDR(v); > 374a383,386 > else if (streql(CHAR(namei), "scipen")) { > k = asInteger(argi); > SET_VECTOR_ELT(value, i, SetOption(tag, ScalarInteger(k))); > } diff -r R-1.7.0/src/main/print.c R-1.7.0.mod/src/main/print.c 88a89 > R_print.scipen = GetOptionSciPen(rho); ***************************** End diff output ********************************

The main limitation of the shapefiles package that I put together is that it does not create shapefiles from R objects - rather it only writes shapefiles that have been read into R and manipulated within the constraints of the existing file structure. By this I mean that for example you can change the coordinates of points and write them back out. Or you can add a bunch of blank columns in the