Re: [musl] [PATCH v2] riscv64: correct struct __ucontext name

[Rich Felker <dalias@libc.org>]

On Sun, Dec 06, 2020 at 04:55:39PM +0000, Ariadne Conill wrote:
> 
> Hello,
> 
> On Sunday, December 6, 2020 5:49:25 AM MST Drew DeVault wrote:
> > On Sun Dec 6, 2020 at 3:51 AM EST, Szabolcs Nagy wrote:
> > > * Drew DeVault <sir@cmpwn.com> [2020-12-05 18:10:06 +0000]:
> > > > This makes it consistent with other architectures and fixes some issues
> > > > with downstream software.
> > > 
> > > which software?
> > > 
> > > glibc uses struct ucontext_t too and user code should use ucontext_t
> > > without struct.
> 
> Some glibc architecture ports use the struct __ucontext and even struct 
> ucontext names, or at least did in the past.
> 
> > libucontext, which does use ucontext_t.
> > 
> > In fact, the issue was more related to the type conflict with
> > ucontext.h, which declared struct __ucontext in the scope of its
> > function declarations due to the naming mismatch.
> 
> glibc uses the POSIX 2004 standardized ucontext_t type in its public 
> definitions.  I believe musl should do the same.

This produces a compile-time error is ucontext.h is included without
the right feature test macros, since signal.h will not have defined
ucontext_t in that case. That's why the public declarations must use
the struct tag.

> As far as libucontext goes, this is increasingly moot because 0.13 will 
> introduce freestanding mode which avoids the musl definitions entirely, instead 
> using simplified (though ABI compatible) definitions, allowing it to not only be 
> used on musl but on other libc and other OS entirely (for example, it is known 
> to now build on AmigaOS and Darwin).
> 
> libucontext using its own definitions is an important step toward eventually 
> taking ucontext.h out of musl entirely, and providing it in libucontext 
> instead, too, which I think musl should do since the ucontext API was dropped 
> from POSIX.

It's still an open question whether musl will eventually add these,
but the contents of ucontext.h are independent of the implementation;
they're just function declarations. The *type* declarations are in
signal.h and are *not* dropped from POSIX, so they can't be removed.

> But right now, I think the best way forward is to leave the architecture 
> headers alone and just fix the ucontext.h definitions instead.  I can send a 
> patch doing that if you want to focus on other things.

As described above this does not work.

Rich