linux-kernel - Re: [PATCH 2/2] kbuild: userprogs: use lld to link through clang

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: <20250213175437.GA2756218@ax162>
Date: Thu, 13 Feb 2025 10:54:37 -0700
From: Nathan Chancellor <nathan@...nel.org>
To: Thomas Weißschuh <thomas.weissschuh@...utronix.de>
Cc: Masahiro Yamada <masahiroy@...nel.org>,
	Nicolas Schier <nicolas@...sle.eu>,
	Nick Desaulniers <ndesaulniers@...gle.com>,
	Bill Wendling <morbo@...gle.com>,
	Justin Stitt <justinstitt@...gle.com>,
	Sam Ravnborg <sam@...nborg.org>, linux-kbuild@...r.kernel.org,
	linux-kernel@...r.kernel.org, llvm@...ts.linux.dev,
	stable@...r.kernel.org
Subject: Re: [PATCH 2/2] kbuild: userprogs: use lld to link through clang

Hi Thomas,

On Thu, Feb 13, 2025 at 03:55:18PM +0100, Thomas Weißschuh wrote:
> The userprog infrastructure links objects files through $(CC).
> Either explicitly by manually calling $(CC) on multiple object files or
> implicitly by directly compiling a source file to an executable.
> The documentation at Documentation/kbuild/llvm.rst indicates that ld.lld would
> be used for linking if LLVM=1 is specified.
> However clang instead will use either a globally installed cross linker from
> $PATH called ${target}-ld or fall back to the system linker, which probably
> does not support crosslinking.
> For the normal kernel build this is not an issue because the linker is always
> executed directly, without the compiler being involved.
> 
> Fix this by passing -fuse-lld and let clang find its matching lld.
> 
> Fixes: 7f3a59db274c ("kbuild: add infrastructure to build userspace programs")
> Cc: stable@...r.kernel.org
> Signed-off-by: Thomas Weißschuh <thomas.weissschuh@...utronix.de>

First of all, thank you for catching and noticing this!

> ---
>  Makefile | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/Makefile b/Makefile
> index bb5737ce7f9e79f4023c9c1f578a49a951d1e239..b4c208ae4041c1f4e32c2a158322422ce7353d06 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -510,6 +510,7 @@ OBJCOPY		= $(LLVM_PREFIX)llvm-objcopy$(LLVM_SUFFIX)
>  OBJDUMP		= $(LLVM_PREFIX)llvm-objdump$(LLVM_SUFFIX)
>  READELF		= $(LLVM_PREFIX)llvm-readelf$(LLVM_SUFFIX)
>  STRIP		= $(LLVM_PREFIX)llvm-strip$(LLVM_SUFFIX)
> +KBUILD_USERLDFLAGS += -fuse-ld=lld

Now that our minimum supported version upstream is 13.0.1, I think we
can make this

  KBUILD_USERLDFLAGS += --ld-path=$(LD)

as it should respect the user's choice of linker a little bit more, such
as if they specific LLVM=<prefix>/bin/ or LLVM=-20. That reminds me that
I can clean up what I did in commit 4406b12214f6 ("powerpc/vdso: Link
with ld.lld when requested").

Additionally, this would not fix someone using CC=clang and LD=ld.lld
(it is uncommon but still techincally supported) so could we use a check
like

  ifeq ($(CONFIG_CC_IS_CLANG)$(CONFIG_LD_IS_LLD),yy)
  KBUILD_USERLDFLAGS += --ld-path=$(LD)
  endif

further down in Makefile to make it more robust?

The stable backport may want to use cc-option like I did for the powerpc
vdso since there is a lower minimum supported version of LLVM there.

>  else
>  CC		= $(CROSS_COMPILE)gcc
>  LD		= $(CROSS_COMPILE)ld
> 
> -- 
> 2.48.1
> 

Cheers,
Nathan