linux-kernel - Re: [PATCH] loongarch: Only select HAVE_OBJTOOL and allow ORC unwinder if the inline assembler supports R_LARCH_{32,64}

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: <CAAhV-H6j0WgN6QpHBYcdprdWBpcQq4ObYK5YcE=TVsDNxGFEtA@mail.gmail.com>
Date: Mon, 17 Jun 2024 21:11:33 +0800
From: Huacai Chen <chenhuacai@...nel.org>
To: Xi Ruoyao <xry111@...111.site>
Cc: Jinyang He <hejinyang@...ngson.cn>, Nathan Chancellor <nathan@...nel.org>, 
	Peter Zijlstra <peterz@...radead.org>, WANG Xuerui <kernel@...0n.name>, 
	Tiezhu Yang <yangtiezhu@...ngson.cn>, Nick Desaulniers <ndesaulniers@...gle.com>, 
	Bill Wendling <morbo@...gle.com>, Justin Stitt <justinstitt@...gle.com>, 
	Youling Tang <tangyouling@...inos.cn>, loongarch@...ts.linux.dev, 
	linux-kernel@...r.kernel.org, llvm@...ts.linux.dev, mengqinggang@...ngson.cn, 
	cailulu@...ngson.cn, wanglei@...ngson.cn, luweining@...ngson.cn, 
	Yujie Liu <yujie.liu@...el.com>, Heng Qi <hengqi@...ux.alibaba.com>, 
	Tejun Heo <tj@...nel.org>
Subject: Re: [PATCH] loongarch: Only select HAVE_OBJTOOL and allow ORC
 unwinder if the inline assembler supports R_LARCH_{32,64}_PCREL

On Sat, Jun 15, 2024 at 6:22 PM Xi Ruoyao <xry111@...111.site> wrote:
>
> On Sat, 2024-06-15 at 17:33 +0800, Huacai Chen wrote:
> > > To me for now we should just make OBJTOOL and ORC depend on BROKEN and
> > > backport to stable...
> > But this patch allows clang to build objtool, which seems broken, too.
>
> Yes, so I mean make objtool depend on CONFIG_BROKEN because it is indeed
> broken as at now.
I don't like CONFIG_BROKEN here, that means telling everyone not to
enable OBJTOOL, but that is not the case.

>
> Or we'll end up at least:
>
> select HAVE_OBJTOOL if AS_HAS_EXPLICIT_RELOCS && AS_HAS_THIN_ADD_SUB && !CC_IS_CLANG && !RUST
Maybe we needn't consider RUST here? And can we think
AS_HAS_THIN_ADD_SUB always imply AS_HAS_EXPLICIT_RELOCS here?

If so, we can simplify the condition. And please submit an updated version.

Huacai

>
> this is already nasty and I still don't know if it covers all broken
> cases (I've no idea if GCC will generate frame pointer in some cases as
> well...)
>
> > > Even if we can fix both the -mno-thin-add-sub problem and the frame
> > > pointer problem in these weeks, they'll be some nontrivial large change
> > > and improper to backport.  Thus we have to admit objtool doesn't really
> > > work for old releases and mark it broken.
> > I don't like to disable objtool, unless there is no better solution.
> > And it seems there has already been some "large fix" in objtool's
> > history.
>
> Then we can still backport the large fix to the stable branches when we
> finish it up.
>
> --
> Xi Ruoyao <xry111@...111.site>
> School of Aerospace Science and Technology, Xidian University