Re: advice needed regarding c++ name mangling

2005-03-13 Thread Mark Mitchell
Razya Ladelsky wrote: Mark Mitchell <[EMAIL PROTECTED]> wrote on 11/03/2005 04:55:38: Daniel Berlin wrote: As for why the new name doesn't work, it's not clear from the above. I'd need to see the assembly and the error. :) Likewise. I assume these functions have only internal linkage? Or that

Re: advice needed regarding c++ name mangling

2005-03-13 Thread Steven Bosscher
On Sunday 13 March 2005 16:53, Daniel Berlin wrote: > Sure you could. > You can come up with your own way to mangle the addition you are making > to the name based on how you versioned (IE what parameters you have > removed/made constant, and their values), so that it would merge > properly when pu

Re: advice needed regarding c++ name mangling

2005-03-13 Thread Steven Bosscher
On Sunday 13 March 2005 16:31, Daniel Berlin wrote: > > bl operator<<.585 > > ^^^ > > You are using the demangled name instead of the mangled one, which is > where your problem comes from. Right. But the mangled name is not available at this point. > Mark should be able t

Re: advice needed regarding c++ name mangling

2005-03-13 Thread Daniel Berlin
On Sun, 2005-03-13 at 16:42 +0100, Steven Bosscher wrote: > On Sunday 13 March 2005 16:31, Daniel Berlin wrote: > > > bl operator<<.585 > > > > ^^^ > > > > You are using the demangled name instead of the mangled one, which is > > where your problem comes from. > > Right. B

Re: advice needed regarding c++ name mangling

2005-03-13 Thread Daniel Berlin
On Sun, 2005-03-13 at 10:54 +0200, Razya Ladelsky wrote: > Mark Mitchell <[EMAIL PROTECTED]> wrote on 11/03/2005 04:55:38: > > > Daniel Berlin wrote: > > > > > As for why the new name doesn't work, it's not clear from the above. > > > I'd need to see the assembly and the error. > > > :) > > > >

Re: advice needed regarding c++ name mangling

2005-03-13 Thread Razya Ladelsky
Mark Mitchell <[EMAIL PROTECTED]> wrote on 11/03/2005 04:55:38: > Daniel Berlin wrote: > > > As for why the new name doesn't work, it's not clear from the above. > > I'd need to see the assembly and the error. > > :) > > Likewise. I assume these functions have only internal linkage? Or that

Re: advice needed regarding c++ name mangling

2005-03-10 Thread Mark Mitchell
Daniel Berlin wrote: As for why the new name doesn't work, it's not clear from the above. I'd need to see the assembly and the error. :) Likewise. I assume these functions have only internal linkage? Or that the original function is still provided with external linkage? (I'm just checking that

Re: advice needed regarding c++ name mangling

2005-03-10 Thread Daniel Berlin
On Thu, 2005-03-10 at 18:02 +, Nathan Sidwell wrote: > Razya Ladelsky wrote: > > Hi, > > > > My case is this: > > I version the operator<< function and name it operator<<.number (creating > > an identifier which is not valid in the source code). > > The assembly name created for the versioned

Re: advice needed regarding c++ name mangling

2005-03-10 Thread Nathan Sidwell
Razya Ladelsky wrote: Hi, My case is this: I version the operator<< function and name it operator<<.number (creating an identifier which is not valid in the source code). The assembly name created for the versioned function is the same as the tree identifier, which is not valid for the assembler.

advice needed regarding c++ name mangling

2005-03-10 Thread Razya Ladelsky
Hi, My case is this: I version the operator<< function and name it operator<<.number (creating an identifier which is not valid in the source code). The assembly name created for the versioned function is the same as the tree identifier, which is not valid for the assembler. I tried creating an