search for: themap

...erstanding how the " auxvar " variable is supposed to be used in the plot.Map function. I am using R Version 1.8.1 (2003-11-21) on Linux Looking at the plot.Map function itself, I see that it calls a color.ramp function (I am reporting only the relevant part here): ... if (attr(theMap$Shapes, "shp.type") == "poly") { if (!is.null(auxvar) && nclass > 1) { col.rmp <- color.ramp(nclass, nvec = auxvar) for (i in recs) { ... I don't seem to be able to get any information on this color.ramp function. In fact t...

...would never be entered into map, to keep the size of the map low. The first time a query is made within a large block, it would compute the indices of all loads and stores within the block. The nice thing about using a single map is that it makes updating the map really trivial. Just use TheMap.erase(I) when deleting load and store instructions. The numbering can be purely local to a block and the relative ordering of memops don't get invalidated when a load or store is nuked. Does this seem like a reasonable approach? If so, please split the various parts of mem2reg out in...

...the constructor of ECValue before the insert/find calls. If I'm understanding the concern, this looks easy to fix to me by simply making the private construction explicit within the scope of the code which has access to this constructor: iterator insert(const ElemTy &Data) { return TheMapping.insert(ECValue(Data)).first; } You can make this change today, and it will work with your C++03 compiler with no loss of efficiency, and arguably improved clarity. > 2. There are numerous places where there is a std::pair of pointer types and passing 0 results in the rvalue reference con...

...library(maptools) > district.shp <- read.shape("~/Documents/SAHSU/MD/data/Carthography/ districts_ok.shp") Shapefile type: Polygon, (5), # of Shapes: 354 > plot(district.shp) *** caught segfault *** address 0xc00006d5, cause 'memory not mapped' Traceback: 1: polygon(theMap$Shapes[[ii]]$verts, col = fg[i], border = ol, ...) 2: plot.Map(district.shp) 3: plot(district.shp) 4: plot(district.shp) Possible actions: 1: abort (with core dump) 2: normal R exit 3: exit R without saving workspace 4: exit R saving workspace Selection: I saw that plot.Map is deprecated, so...

Sorry, always end up not replying to the list: The main issue with dealing with next this way is that people adding new uses of next will probably not be using c++0x and therefore won't know it's ambiguous and that it needs to be qualified. There are also two issues with rvalue references and the STL: 1. EquivalenceClasses, in the insert and findLeader functions, it uses map functions

...an Index: include/llvm/ADT/DenseSet.h =================================================================== --- include/llvm/ADT/DenseSet.h (revision 68214) +++ include/llvm/ADT/DenseSet.h (working copy) @@ -90,6 +90,13 @@ std::pair<iterator, bool> insert(const ValueT &V) { return TheMap.insert(std::make_pair(V, 0)); } + + // Range insertion of values. + template<typename InputIt> + void insert(InputIt I, InputIt E) { + for (; I != E; ++I) + insert(*I); + } }; } // end namespace llvm Index: include/llvm/ADT/DenseMap.h =======================================...

On Sep 26, 2008, at 8:41 AM, David Greene wrote: > On Thursday 25 September 2008 13:15, David Greene wrote: > >> My patch builds a map from BasicBlock to lists of loads and stores >> in that >> block in an initialization phase along with ordering information >> about the >> loads and stores. RewriteSingleStoreAlloca then queries that >> information

...would never be entered into map, to keep the size of the map low. The first time a query is made within a large block, it would compute the indices of all loads and stores within the block. The nice thing about using a single map is that it makes updating the map really trivial. Just use TheMap.erase(I) when deleting load and store instructions. The numbering can be purely local to a block and the relative ordering of memops don't get invalidated when a load or store is nuked. Does this seem like a reasonable approach? If so, please split the various parts of mem2reg out in...

...de within a > large block, it would compute the indices of all loads and stores > within the block. This is a simpler approach and should have the same effect. I'll code it up. > The nice thing about using a single map is that it makes updating the > map really trivial. Just use TheMap.erase(I) when deleting load and > store instructions. The numbering can be purely local to a block and > the relative ordering of memops don't get invalidated when a load or > store is nuked. Good points. > Does this seem like a reasonable approach? If so, please split the >...

On Thursday 25 September 2008 13:15, David Greene wrote: > My patch builds a map from BasicBlock to lists of loads and stores in that > block in an initialization phase along with ordering information about the > loads and stores. RewriteSingleStoreAlloca then queries that information > to determine if a load appears before the single store. > > I like your approach of using