From: Thomas Huth <thuth@redhat.com>
To: Amit Machhiwal <amachhiw@linux.ibm.com>,
Laurent Vivier <lvivier@redhat.com>,
Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org, kvm@vger.kernel.org,
Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH] powerpc: Link with -z notext
Date: Wed, 11 Nov 2026 09:30:26 +0100 [thread overview]
Message-ID: <620ad58f-50ed-4b0b-9d13-b4df34276448@redhat.com> (raw)
In-Reply-To: <20261110154415.46220-1-amachhiw@linux.ibm.com>
Am 10.11.26 um 16:44 schrieb Amit Machhiwal:
> Building the powerpc tests with GNU ld 2.46.1 fails when linking the
> test binaries:
>
> /usr/bin/ld.bfd: error: read-only segment has dynamic relocations
>
> The tests are linked with -pie and relocate themselves at boot (see
> relocate() in powerpc/reloc64.c, called from powerpc/cstart64.S).
> Newer GNU ld rejects relocations in read-only segments by default. The
> Makefile already checks after linking that only R_PPC64_RELATIVE
> relocations are present.
>
> Fix this by adding `-z notext` to the powerpc linker flags, as was done
> for arm64 in commit 5d67db4e1860 ("arm64: Link with -z notext").
>
> Signed-off-by: Amit Machhiwal <amachhiw@linux.ibm.com>
> ---
> powerpc/Makefile | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/powerpc/Makefile b/powerpc/Makefile
> index 41e752b070bc..35fe58835726 100644
> --- a/powerpc/Makefile
> +++ b/powerpc/Makefile
> @@ -78,7 +78,7 @@ include $(SRCDIR)/scripts/asm-offsets.mak
>
> FLATLIBS = $(libcflat) $(LIBFDT_archive)
> %.elf: CFLAGS += $(arch_CFLAGS)
> -%.elf: LDFLAGS += $(arch_LDFLAGS) -pie -n
> +%.elf: LDFLAGS += $(arch_LDFLAGS) -pie -n -z notext
> %.elf: %.o $(FLATLIBS) $(SRCDIR)/powerpc/flat.lds $(cstart.o) $(reloc.o) %.aux.o
> $(LD) $(LDFLAGS) -o $@ \
> -T $(SRCDIR)/powerpc/flat.lds --build-id=none \
Another thought: Why are we only using -pie for the LDFLAGS, but we rather
seem to compile with -fno-pic in the CFLAGS? That looks weird ... I guess
the -fno-pic is required for certain targets (s390x and the non-EFI builds
of the other targets?), but maybe we should only add it only where it is
required and compile the targets that have relocation code with -fpie in the
CFLAGS instead?
Thomas
prev parent reply other threads:[~2026-11-11 8:30 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-10 15:44 [PATCH] powerpc: Link with -z notext Amit Machhiwal
2026-11-11 7:54 ` Thomas Huth
2026-11-11 10:18 ` Amit Machhiwal
2026-11-11 8:30 ` Thomas Huth [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=620ad58f-50ed-4b0b-9d13-b4df34276448@redhat.com \
--to=thuth@redhat.com \
--cc=amachhiw@linux.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=lvivier@redhat.com \
--cc=maddy@linux.ibm.com \
--cc=pbonzini@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox