From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1Lyq7Z-00065U-5v
	for qemu-devel@nongnu.org; Tue, 28 Apr 2009 12:25:25 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1Lyq7U-00063c-CM
	for qemu-devel@nongnu.org; Tue, 28 Apr 2009 12:25:24 -0400
Received: from [199.232.76.173] (port=59168 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1Lyq7U-00063T-0Z
	for qemu-devel@nongnu.org; Tue, 28 Apr 2009 12:25:20 -0400
Received: from mail-fx0-f219.google.com ([209.85.220.219]:53516)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <blauwirbel@gmail.com>) id 1Lyq0X-00059U-6p
	for qemu-devel@nongnu.org; Tue, 28 Apr 2009 12:18:09 -0400
Received: by fxm19 with SMTP id 19so584182fxm.34
	for <qemu-devel@nongnu.org>; Tue, 28 Apr 2009 09:18:08 -0700 (PDT)
MIME-Version: 1.0
In-Reply-To: <200904281432.42291.paul@codesourcery.com>
References: <f43fc5580904221234l22c4e5f8h7362e8e9adee0b49@mail.gmail.com>
	<200904281432.42291.paul@codesourcery.com>
Date: Tue, 28 Apr 2009 19:18:07 +0300
Message-ID: <f43fc5580904280918v7bbfc480v9124a9896880a63d@mail.gmail.com>
Subject: Re: [Qemu-devel] [PATCH, RFC] Smarter compilation for target devices
From: Blue Swirl <blauwirbel@gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Paul Brook <paul@codesourcery.com>
Cc: qemu-devel@nongnu.org

On 4/28/09, Paul Brook <paul@codesourcery.com> wrote:
> > Compile target devices only once for each endian and word size combination,
>  > saving a few compiles if large number of targets are enabled.
>
>
> > +CPPFLAGS=-I. -I.. -I$(SRC_PATH) -MMD -MT $@ -MP -DNEED_CPU_H
>
>
> > +# Warning: Do not add new files here if they have conditional code
>  > +# with #ifdef TARGET_xxx etc, use TARGET_PAGE_SIZE or reference
>  > +# CPUState
>
>
> This is just asking for trouble. Anything that includes cpu.h must be rebuilt
>  for every cpu.  IMO a necessary prerequisite for this change is reworking
>  header files so that you can't accidentally use the wrong symbols.

Yes, this was a hack. I think with your header change it will be much cleaner.

>  Also, which bitwidth are you intending to distinguish here? target_ulong or
>  target_phys_addr_t? With a few exceptions, devices don't care about the

target_phys_addr_t, of course. The devices that know about
target_ulong (or CPUState, TARGET_PAGE_SIZE) can't use this system.

>  former. Duplicating this logic is both cpu.h and configure is also asking for
>  trouble. At minimum the build should fail if they disagree.

Which logic?