Date: Wed, 6 Mar 2019 16:48:45 +0200
Ben Boeckel via Modules <modules_at_[hidden]> writes:
> On Wed, Mar 06, 2019 at 16:26:25 +0200, Boris Kolpackov wrote:
>
> > The compiler may also find a BMI that is out-of-date.
>
> That's a problem in the build graph, not the compiler. An error would be
> nice though if the compiler can divine "out-of-date" somehow (I'd expect
> it to result in a linker error later today).
I was referring to the case where BMIs for header units are found via
some sort of search path. Which is, it seems, where things are leaning
towards with the idea of letting the compiler locate the BMIs and fall
back to real headers if not found.
But if direct, one-to-one mapping is supplied by the build system
(as I believe it should be done), then, yes, an out-of-date BMI
would indicate a build system problem.
> On Wed, Mar 06, 2019 at 16:26:25 +0200, Boris Kolpackov wrote:
>
> > The compiler may also find a BMI that is out-of-date.
>
> That's a problem in the build graph, not the compiler. An error would be
> nice though if the compiler can divine "out-of-date" somehow (I'd expect
> it to result in a linker error later today).
I was referring to the case where BMIs for header units are found via
some sort of search path. Which is, it seems, where things are leaning
towards with the idea of letting the compiler locate the BMIs and fall
back to real headers if not found.
But if direct, one-to-one mapping is supplied by the build system
(as I believe it should be done), then, yes, an out-of-date BMI
would indicate a build system problem.
Received on 2019-03-06 15:48:51