Benjamin Smedberg <[EMAIL PROTECTED]> writes: > Diego Novillo wrote: >> Before plug-ins: put your gimple-to-myIR converter in passes.c >> After plug-ins: dlopen gimple-to-myIR.so >> >> Both represent the same effort. Both require your converter to be kept >> up-to-date with GCC's ever shifting ABI/API. > > They represent the same effort for somebody writing a plugin pass. They do > *not* represent the same effort for a Mozilla hacker who just wants to > compile Mozilla with the extra static-checking pass enabled. [...]
There may be a subtle connection between this and the debugging information correctness thread by Alex Oliva a few weeks ago. It may be that if only gcc's debuginfo for deployed (-O2 -g) code were more complete & more correct, that some such analysis code could work by on-the-fly probing-based instrumentation rather than compiled-in stuff. At least, this applies to the kinds of assertions expressible by something like the old GNU Nana tool. - FChE
