linux-kernel - Re: [PATCH] lib/string: shrink lib/string.i via IWYU

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <2023120650-faucet-palpable-f1a3@gregkh>
Date:   Wed, 6 Dec 2023 12:09:17 +0900
From:   Greg KH <greg@...ah.com>
To:     Al Viro <viro@...iv.linux.org.uk>
Cc:     Nick Desaulniers <ndesaulniers@...gle.com>, tanzirh@...gle.com,
        Kees Cook <keescook@...omium.org>,
        Andy Shevchenko <andy@...nel.org>,
        linux-hardening@...r.kernel.org, linux-kernel@...r.kernel.org,
        Nick DeSaulniers <nnn@...gle.com>,
        Andrew Morton <akpm@...ux-foundation.org>, llvm@...ts.linux.dev
Subject: Re: [PATCH] lib/string: shrink lib/string.i via IWYU

On Wed, Dec 06, 2023 at 03:00:47AM +0000, Al Viro wrote:
> On Wed, Dec 06, 2023 at 12:55:42AM +0000, Al Viro wrote:
> > On Wed, Dec 06, 2023 at 08:46:50AM +0900, Greg KH wrote:
> > > > >
> > > > > But of course, it doesn't always hold true, there are a few minor
> > > > > exceptions, but they are rare.
> > > > 
> > > > $ grep -r \\#include lib | grep asm
> > > > 
> > > > shows quite a few exceptions, and just in lib/.
> > > > 
> > > > For example, lib/math/int_log.c includes asm/bug.h.  Is that a case
> > > > where lib/math/int_log.c should be #include 'ing linux/bug.h rather
> > > > than asm/bug.h?
> > > 
> > > Probably yes, but we don't normally go back and take coding style fixes
> > > for old files like this as it doesn't make much sense to do so.
> > > 
> > > But, if you are cleaning up the headers for large portions with the goal
> > > of faster builds, that's a good reason.
> > 
> > FWIW, the most common (by far - about 13% of such includes, over drivers/, fs/,
> > mm/, net/ and sound/) is asm/unaligned.h.
> 
> Why the hell is unaligned.h in asm/*, anyway?
> 
> We have 3 variants: arc, parisc and generic (== everything else).
> Both arc and parisc instances have an explicit include of
> asm-generic/unaligned.h (i.e. the generic variant).
> 
> On arc there's also misaligned_fixup() extern or stub, with exactly
> one user (in arch/arc/kernel/traps.c).  On parisc there are
> externs for handle_unaligned() and check_unaligned() (3 call sites,
> all in arch/parisc/kernel/traps.c).
> 
> How about we take those into arch/{arc,parisc}/kernel/unaligned.h,
> slap #include "unaligned.h" into their traps.c and unaligned.c
> (callers and definitions resp.) and strip those from asm/unaligned.h?
> At that point we can remove arch/{arc,parisc}/asm/unaligned.h - everything
> will pick include/asm-generic/unaligned.h.
> 
> Then the next cycle we ask Linus to run the following:
> for i in `git grep -l -w asm/unaligned.h`; do
> 	sed -i -e "s/asm\/unaligned.h/linux\/unaligned.h/" $i
> done
> git mv include/asm-generic/unaligned.h include/linux/unaligned.h
> sed -i -e "/unaligned.h/d" include/asm-generic/Kbuild
> right before releasing -rc1 and asm/unaligned.h is gone...

Please do, it's really annoying and would be great to fix up.

thanks,

greg k-h