Re: [Qemu-devel] [PATCH 03/15] hw/ssi: Remove SSIBus from "qemu/typedefs.h"

[Thomas Huth <thuth@redhat.com>]

On 2019-01-15 13:28, Markus Armbruster wrote:
> Philippe Mathieu-Daudé <philmd@redhat.com> writes:
> 
>> From: Philippe Mathieu-Daudé <f4bug@amsat.org>
>>
>> There are only three files requiring this typedef, let them
>> include "hw/ssi/ssi.h" directly to simplify "qemu/typedefs.h".
>>
>> To clean "qemu/typedefs.h", move the forward declaration
>> to "hw/ssi/ssi.h".
>>
>> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
>> ---
>>  hw/arm/strongarm.h      | 1 +
>>  include/hw/arm/pxa.h    | 1 +
>>  include/hw/ssi/pl022.h  | 1 +
>>  include/hw/ssi/ssi.h    | 1 +
>>  include/qemu/typedefs.h | 1 -
>>  5 files changed, 4 insertions(+), 1 deletion(-)
> 
> When typedefs.h changes, we recompile the world, but it pretty much only
> ever changes when new typedefs are added.  Thus, *keeping* a typedef
> there is therefore pretty cheap.
> 
> Nevertheless, we shouldn't keep typedefs there without a real reason.
> Being able to move one away without having to add any new #include
> directives is a strong sign for "no real reason".  I like patches doing
> that.
> 
> What I don't like is adding #include directives just so you can move
> typedefs out of typedefs.h: it slows down the build.  Granted, the four
> added by this patch are a drop in the bucket.  The point I'm trying to
> make is typedefs.h's purpose: it's for avoiding #include directives.
> Circular ones in particular, but others, too.

Yes, agreed, removing things from typedefs.h just to add lots of
#includes in other files is not really the best idea. I also dropped
this patch in v2 of my current PULL request because of this reason.
Phil, I suggest to simply drop this patch. What we maybe could do is to
split up typedefs.h per subsystem, so that we additionally have a
hw/arm/typedefs.h, hw/ppc/typedefs.h etc. in the end, then the
target-specific typedefs would not clutter the common qemu/typedefs.h
file anymore.

 Thomas