[patch] prevent tree sinking of trapping stmts
Thu Sep 4 10:28:00 GMT 2014
On Thu, Sep 04, 2014 at 12:22:33PM +0200, Richard Biener wrote:
> But then the proposed patch pessimizes code too much IMHO.
Yeah. IMHO if Ada wants to __builtin_longjmp from synchronous signal handlers
(SIGFPE, SIGSEGV, ...) then the middle-end needs to be aware of that and
have it represented in the IL by edges from those possibly signalling
instructions to a setjmp dispatch pad or similar.
More information about the Gcc-patches