From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <9front-bounces@9front.inri.net>
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on inbox.vuxu.org
X-Spam-Level: 
X-Spam-Status: No, score=-0.6 required=5.0 tests=DKIM_INVALID,DKIM_SIGNED,
	HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI autolearn=ham
	autolearn_force=no version=3.4.4
Received: from 9front.inri.net (9front.inri.net [168.235.81.73])
	by inbox.vuxu.org (Postfix) with ESMTP id 6586224E2C
	for <ml@inbox.vuxu.org>; Wed,  8 May 2024 16:24:38 +0200 (CEST)
Received: from wopr.sciops.net ([216.126.196.60]) by 9front; Wed May  8 10:22:43 -0400 2024
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sciops.net; s=20210706;
	t=1715178126;
	h=from:from:reply-to:subject:subject:date:date:message-id:message-id:
	 to:to:cc:mime-version:mime-version:content-type:content-type:
	 content-transfer-encoding:content-transfer-encoding:  in-reply-to:in-reply-to;
	bh=zApdWJODQC5H7hkL24cUwoLBYWdI/IJskdW1WNtBzO8=;
	b=Nj1hLrIUZ4KBPucJPPNdg1PIn0Tzs+cRYNDtug8JN0rgUdj55BMmJrEC1FtLbK1u/m000a
	LWw5f0s87Duov7myJDQKJmQFJzCVvfxB9xsz0UOTw2IYIDrLEy47eEc1WeorlEj4XjpnMg
	kJBERsaPIXHKyFGjMee9iI+F2an9sww=
Received: 
	by wopr.sciops.net (OpenSMTPD) with ESMTPSA id ae5fc260 (TLSv1.2:ECDHE-RSA-CHACHA20-POLY1305:256:NO)
	for <9front@9front.org>;
	Wed, 8 May 2024 07:22:05 -0700 (PDT)
Message-ID: <94909FDAB9454C00B8D7ECE0602C3978@wopr.sciops.net>
Date: Wed, 08 May 2024 16:22:30 +0200
From: qwx@sciops.net
To: 9front@9front.org
In-Reply-To: <d173af93-db09-4c20-ae3d-548ea39ae9eb@sirjofri.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
List-ID: <9front.9front.org>
List-Help: <http://lists.9front.org>
X-Glyph: ➈
X-Bullshit: app hardware-scale API-based callback generator 
Subject: Re: [9front] Linking error while porting
Reply-To: 9front@9front.org
Precedence: bulk

On Wed May  8 13:23:38 +0200 2024, sirjofri+ml-9front@sirjofri.de wrote:
> Hey all,
> 
> I'm trying to port the simon tatham portable puzzle collection using npe and I get incompatible type signatures while linking the files.
> 
> For npe, I had to #undef PI after #including <libc.h> (in npe.h), because puzzles #defines its own PI. Other than that, I didn't change the source repo or npe.
> 
> My sources are here: http://sirjofri.de/oat/tmp/puzzles/
> 
> repo.url contains the url to the source repo for puzzles. To get started, clone that url to your local filesystem and copy the remaining files into the same directory.
> 
> The other files are an mkfile and plan9.c (for the plan 9 frontend, which is mostly a stub I still have to fill).
> 
> b.rc is a short build script for building the blackbox program. Literally `mk install 6.blackbox`. It's not about that program specifically, it's just the first game in the list.
> 
> The linking errors are all roughly the same, like:
> 
> new_window: incompatible type signatures 71a60b79(blackbox.6) and 846c216a(plan9.6) for thegame
> 
> In this specific case, I double checked the type and the only difference is that one is declared as the struct itself, the other as the typedef'd struct (struct game vs game). I also changed that temporarily, without a difference. The compiler handles typedef's like it should.
> 
> I can't imagine any other reason why it shouldn't work. Thanks to npe, it's the same compiler (6c) for both source files. The compilers/linkers should be up to date (sysupdated yesterday).
> 
> If anyone has any idea what could be wrong, that would be very helpful. Also if I should give you more data, just let me know.
> 
> sirjofri

 From what I saw, some of these structs are actually not defined in the
header but in individual c files.  Putting `#pragma incomplete' for a
few of them in puzzles.h fixes most linking errors:

#pragma incomplete midend
#pragma incomplete drawing
#pragma incomplete frontend
#pragma incomplete random_state

One error still remains for a big and complex struct, you might have
to go through all type declarations and make sure whatever isn't
defined in puzzles.h is marked incomplete.

Cheers,
qwx