Mail Archives: djgpp/1997/06/30/07:44:13
On Thu, 26 Jun 1997, Nate Eldredge wrote:
> I've found a problem with GCC that I don't know whether to describe as a bug
> or just compiler stupidity. The problem is that when calling a function
> returning a struct several times and ignoring the return value, space is
> allocated on the stack for *each* of the ignored structs.
[snip]
> Turning on optimizations has no effect on this problem (except that -O3 or
> -finline-functions make it a lot harder to see). Unless someone can think of
> a justification for this inefficiency, I'm going to report it to GNU.
Please do report this to GNU. Thanks.
- Raw text -