Re: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Martin Buchholz
Christian, I care, and agree with you. I recall many years ago, debugging a mysterious failure building Emacs. Building on Solaris-Sparc defined the symbol "sparc", and this caused the build to fail, only if a component of the current working directory was "sparc". Martin Christian Thalinger wr

Re: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Kelly O'Hair
I very much suspect that this -D$(ARCH) could be removed with no consequences. The difficult part would be verifying it. C macros are a very powerful tool, but some of the global names we have chosen over the years have come back to haunt us. :^( -kto Martin Buchholz wrote: Christian, I care,

RE: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Ted Neward
This may be an area where you have to implement the fix and submit it as a patch, because I can see the developers at Sun not having the bandwidth to take time away from the other things they're working on to take care of this themselves--they're horribly underresourced (if you'll pardon the verbiz

RE: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Christian Thalinger
On Wed, 2008-01-23 at 11:14 -0800, Ted Neward wrote: > This may be an area where you have to implement the fix and submit it as a > patch, because I can see the developers at Sun not having the bandwidth to > take time away from the other things they're working on to take care of this > themselves-

Re: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Christian Thalinger
On Wed, 2008-01-23 at 11:04 -0800, Kelly O'Hair wrote: > I very much suspect that this -D$(ARCH) could be removed with no consequences. > The difficult part would be verifying it. > > C macros are a very powerful tool, but some of the global names we have chosen > over the years have come back to

RE: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Ted Neward
I won't speak for them, but based on Kelly's comments, you may be the guy on the spot to do that verification. :-) Isn't open source fun? :-) I would suggest this: do a grep looking for where that macro is used, and see if removing it entirely has any detrimental effects. That is, pull it, do a b

Re: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Kelly O'Hair
Ignore the java code of course, but in all the *.c, *.h, *.cpp, and *.hpp files I'd look for uses of the symbols: i386, i586, sparc, sparcv9, amd64, x86, x64, and ia64. I suspect you won't find many, but you may get lots of false hits. I'd be happy to help, and in fact someone on the Sun side w

Re: RFC: architecture define probably masks variable names in header files

2008-01-23 Thread Kelly O'Hair
the native code is usually in the src/solaris, src/windows, src/linux, and src/share directories. But to be safe you might want to create a complete list from the root of the repository forest with: find . -type f -name \*.c -o -name \*.h -o -name \*.cpp -o -name \*.hpp Maybe with a little di