[LLVMdev] Win32 COFF Support - Patch 4

[Michael Spencer]

Would COFFSymbol be better? I'll fix the others and commit, thanks.

Sent from my iPhone

On Jul 24, 2010, at 12:24 AM, Daniel Dunbar <daniel at zuster.org> wrote:

> Hi Michael,
> 
> Looks great to me, with the caveat that I know nothing about COFF.
> 
> A few comments:
> - The term symbol is already very overloaded, I would avoid adding a
> new 'class Symbol'.
> - I find it annoying to have more versions of write little endian
> 32-bit data, but I'm not sure what the right solution is. We can clean
> this up later, perhaps.
> - There are a couple of instances of: "}\nelse" instead of "} else".