RE: [PATCH v7 12/35] Hexagon (target/hexagon) instruction attributes

From: Taylor Simpson <tsimpson@quicinc.com>
To: "Philippe Mathieu-Daudé" <f4bug@amsat.org>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>
Cc: "ale@rev.ng" <ale@rev.ng>,
	"richard.henderson@linaro.org" <richard.henderson@linaro.org>,
	"alex.bennee@linaro.org" <alex.bennee@linaro.org>,
	"laurent@vivier.eu" <laurent@vivier.eu>,
	Brian Cain <bcain@quicinc.com>
Subject: RE: [PATCH v7 12/35] Hexagon (target/hexagon) instruction attributes
Date: Fri, 29 Jan 2021 23:15:48 +0000	[thread overview]
Message-ID: <BYAPR02MB4886F04D58FB80D87179FC71DEB99@BYAPR02MB4886.namprd02.prod.outlook.com> (raw)
In-Reply-To: <55f89e45-c479-e02b-27c6-f2e6a7fb85ed@amsat.org>

> -----Original Message-----
> From: Philippe Mathieu-Daudé <philippe.mathieu.daude@gmail.com> On
> Behalf Of Philippe Mathieu-Daudé
> Sent: Monday, January 25, 2021 10:21 AM
> To: Taylor Simpson <tsimpson@quicinc.com>; qemu-devel@nongnu.org;
> Paolo Bonzini <pbonzini@redhat.com>
> Cc: ale@rev.ng; alex.bennee@linaro.org; richard.henderson@linaro.org;
> laurent@vivier.eu; Brian Cain <bcain@quicinc.com>
> Subject: Re: [PATCH v7 12/35] Hexagon (target/hexagon) instruction
> attributes
>
> >> On 1/20/21 4:28 AM, Taylor Simpson wrote:
> >>> Signed-off-by: Taylor Simpson <tsimpson@quicinc.com>
> >>> ---
> >>>  target/hexagon/attribs.h     | 30 ++++++++++++++
> >>>  target/hexagon/attribs_def.h | 95
> >> ++++++++++++++++++++++++++++++++++++++++++++
> >>>  2 files changed, 125 insertions(+)
> >>>  create mode 100644 target/hexagon/attribs.h
> >>>  create mode 100644 target/hexagon/attribs_def.h
> >>>
> >>> diff --git a/target/hexagon/attribs.h b/target/hexagon/attribs.h
> >>> new file mode 100644
> >>> index 0000000..e88e5eb
> >>> --- /dev/null
> >>> +++ b/target/hexagon/attribs.h
> >>> @@ -0,0 +1,30 @@
> >>> +
> >>> +enum {
> >>> +#define DEF_ATTRIB(NAME, ...) A_##NAME,
> >>> +#include "attribs_def.h"
> >>
> >> Per QEMU conventions, this file has to be named "attribs_def.h.inc".
> >
> > Didn't know that.  Which files should end in .inc?
>
> Oh you are right, it is not documented in CODING_STYLE.rst.
>
> You can see the rationale in commits:139c1837db7 and 0979ed017f0:
>
>   meson: rename included C source files to .c.inc
>
>   With Makefiles that have automatically generated dependencies, you
>   generated includes are set as dependencies of the Makefile, so that they
>   are built before everything else and they are available when first
>   building the .c files.
>
>   Alternatively you can use a fine-grained dependency, e.g.
>
>           target/arm/translate.o: target/arm/decode-neon-shared.inc.c
>
>   With Meson you have only one choice and it is a third option, namely
>   "build at the beginning of the corresponding target"; the way you
>   express it is to list the includes in the sources of that target.
>
>   The problem is that Meson decides if something is a source vs. a
>   generated include by looking at the extension: '.c', '.cc', '.m', '.C'
>   are sources, while everything else is considered an include---including
>   '.inc.c'.
>
>   Use '.c.inc' to avoid this, as it is consistent with our other convention
>   of using '.rst.inc' for included reStructuredText files.  The editorconfig
>   file is adjusted.

OK, I understand why it's better to have files end .[ch].inc than .inc.[ch].

However, I need some confirmation on which files need .inc instead of simply ending in .h.  From what I can tell these are the guidelines
- If a file is intended to be included in the middle of another file (as opposed to the top), it should end in .inc.
- If a .inc file is intended to be included in a .h file, it should end in .h.inc.
- If a .inc file is intended to be included in a .c file, it should end in .c.inc.
- The above applies to both human-written and generated files.

Thanks,
Taylor