search for: pathtest

...b.cpp // does the same thing as a.cpp, but with its own external function, b() if you run "clang++ -emit-llvm -S -Iinclude -c lib/a/a.cpp lib/b/b.cpp -g" you get two .ll files both with the obvious: !9 = metadata !{metadata !"include/foo.h", metadata !"/tmp/dbginfo/pathtest"} But if you do this instead: "cd lib/a; clang++ -emit-llvm -S -I../../include -c a.cpp -g; cd ../../lib/b; clang++ -emit-llvm -S -I../../include -c b.cpp -g" you get two different nodes: !9 = metadata !{metadata !"../../include/foo.h", metadata !"/tmp/dbginfo/pathte...

All software development progresses so quickly that anyone's script may not be current. Even a script I may have written. And that is the overall challenge with scripts, in general. We users become dependent on scripts. And we users need the learn independence. I have been learning what scripts do and how scripts do what they do. I was wanting to install 4.3.1 on Debian8 and see if I

Just working on some of the gmlt+fission debug info stuff and I came across a comment that might be relevant to reducing the number of distinct debugloc metadata nodes: "or some sub-optimal metadata that // isn't structurally identical (see: file path/name info from clang, which // includes the directory of the cpp file being built, even when the file name // is absolute (such as

...atest samba tarball HERE=$(pwd) cd ${SOURCELOCATION} # NOTE: it downloads the latest tarball! wget -cq https://www.samba.org/samba/ftp/samba-latest.tar.gz tar zxf samba-latest.tar.gz cd samba* ./configure && make && make install cd ${HERE} # Set the path in /etc/profile.d/samba4.sh PATHTEST=$(echo $PATH | grep [s]amba) if [ -x "$PATHTEST" ]; then echo "PATH=/usr/local/samba/bin:/usr/local/samba/sbin:\$PATH" > /etc/profile.d/samba4.sh export PATH=/usr/local/samba/bin:/usr/local/samba/sbin:$PATH fi echo "Samba compiled." exit 0 This will i...