kernel-hardening.lists.openwall.com archive mirror
 help / color / mirror / Atom feed
From: Thomas Garnier <thgarnie@chromium.org>
To: kernel-hardening@lists.openwall.com
Cc: kristen@linux.intel.com, Thomas Garnier <thgarnie@chromium.org>,
	Masahiro Yamada <yamada.masahiro@socionext.com>,
	"Peter Zijlstra (Intel)" <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Kees Cook <keescook@chromium.org>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Thomas Garnier <thgarnie@google.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	Sam Ravnborg <sam@ravnborg.org>,
	Palmer Dabbelt <palmer@sifive.com>,
	Michael Forney <forney@google.com>,
	Cao jin <caoj.fnst@cn.fujitsu.com>,
	linux-kernel@vger.kernel.org
Subject: [PATCH v6 16/27] compiler: Option to add PROVIDE_HIDDEN replacement for weak symbols
Date: Thu, 31 Jan 2019 11:24:23 -0800	[thread overview]
Message-ID: <20190131192533.34130-17-thgarnie@chromium.org> (raw)
In-Reply-To: <20190131192533.34130-1-thgarnie@chromium.org>

Provide an option to have a PROVIDE_HIDDEN (linker script) entry for
each weak symbol. This option solves an error in x86_64 where the linker
optimizes PIE generated code to be non-PIE because --emit-relocs was used
instead of -pie (to reduce dynamic relocations).

Signed-off-by: Thomas Garnier <thgarnie@chromium.org>
---
 init/Kconfig            |  7 +++++++
 scripts/link-vmlinux.sh | 14 ++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/init/Kconfig b/init/Kconfig
index 116e0de4817f..2261c9891631 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -2104,6 +2104,13 @@ config ASN1
 	  inform it as to what tags are to be expected in a stream and what
 	  functions to call on what tags.
 
+config WEAK_PROVIDE_HIDDEN
+	bool
+	help
+	  Generate linker script PROVIDE_HIDDEN entries for all weak symbols. It
+	  allows to prevent non-PIE code being replaced by the linker if the
+	  emit-relocs option is used instead of PIE (useful for x86_64 PIE).
+
 source "kernel/Kconfig.locks"
 
 config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE
diff --git a/scripts/link-vmlinux.sh b/scripts/link-vmlinux.sh
index bc7f1fc1f55b..abf44a804c79 100755
--- a/scripts/link-vmlinux.sh
+++ b/scripts/link-vmlinux.sh
@@ -119,6 +119,17 @@ kallsyms()
 	${CC} ${aflags} -c -o ${2} ${afile}
 }
 
+gen_weak_provide_hidden()
+{
+        if [ -n "${CONFIG_WEAK_PROVIDE_HIDDEN}" ]; then
+                local pattern="s/^\s\+ w \(\w\+\)$/PROVIDE_HIDDEN(\1 = .);/gp"
+                echo -e "SECTIONS {\n. = _end;" > .tmp_vmlinux_hiddenld
+                ${NM} ${1} | sed -n "${pattern}" >> .tmp_vmlinux_hiddenld
+                echo "}" >> .tmp_vmlinux_hiddenld
+                LDFLAGS_vmlinux="${LDFLAGS_vmlinux} -T .tmp_vmlinux_hiddenld"
+        fi
+}
+
 # Create map file with all symbols from ${1}
 # See mksymap for additional details
 mksysmap()
@@ -200,6 +211,9 @@ modpost_link vmlinux.o
 # modpost vmlinux.o to check for section mismatches
 ${MAKE} -f "${srctree}/scripts/Makefile.modpost" vmlinux.o
 
+# Generate weak linker script
+gen_weak_provide_hidden vmlinux.o
+
 kallsymso=""
 kallsyms_vmlinux=""
 if [ -n "${CONFIG_KALLSYMS}" ]; then
-- 
2.20.1.495.gaa96b0ce6b-goog

  parent reply	other threads:[~2019-01-31 19:24 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-31 19:24 [PATCH v6 00/27] x86: PIE support and option to extend KASLR randomization Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 01/27] x86/crypto: Adapt assembly for PIE support Thomas Garnier
2019-02-07 11:48   ` Borislav Petkov
2019-02-07 17:01     ` Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 02/27] x86: Use symbol name in jump table " Thomas Garnier
2019-02-07 12:17   ` Borislav Petkov
2019-02-07 17:04     ` Thomas Garnier
2019-02-07 17:11       ` Borislav Petkov
2019-02-07 23:55         ` Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 03/27] x86: Add macro to get symbol address " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 04/27] x86: relocate_kernel - Adapt assembly " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 05/27] x86/entry/64: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 06/27] x86: pm-trace - " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 07/27] x86/CPU: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 08/27] x86/acpi: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 09/27] x86/boot/64: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 10/27] x86/power/64: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 11/27] x86/paravirt: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 12/27] x86/alternatives: " Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 13/27] x86/boot/64: Build head64.c as mcmodel large when PIE is enabled Thomas Garnier
2019-02-01 11:15   ` Kirill A. Shutemov
2019-02-01 17:11     ` Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 14/27] x86/percpu: Adapt percpu for PIE support Thomas Garnier
2019-01-31 20:57   ` Christopher Lameter
2019-01-31 22:49     ` Thomas Garnier
2019-02-01  2:31       ` Christopher Lameter
2019-02-01 17:13         ` Thomas Garnier
2019-04-08 15:58           ` Thomas Garnier
2019-04-08 17:56             ` Christopher Lameter
2019-04-08 18:08               ` Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 15/27] compiler: Option to default to hidden symbols Thomas Garnier
2019-02-01  7:12   ` Dan Carpenter
2019-02-01 17:00     ` Thomas Garnier
2019-02-01  8:22   ` Adrian Hunter
2019-02-01 17:35     ` Thomas Garnier
2019-01-31 19:24 ` Thomas Garnier [this message]
2019-01-31 19:24 ` [PATCH v6 17/27] x86/relocs: Handle PIE relocations Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 18/27] xen: Adapt assembly for PIE support Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 19/27] kvm: " Thomas Garnier
2019-02-06 19:56   ` Sean Christopherson
2019-02-06 21:23     ` Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 20/27] x86: Support global stack cookie Thomas Garnier
2019-02-01 19:27   ` Andy Lutomirski
2019-02-01 20:21     ` Thomas Garnier
2019-02-01 22:36       ` Andy Lutomirski
2019-02-01 23:56         ` Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 21/27] x86/ftrace: Adapt function tracing for PIE support Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 22/27] x86/modules: Add option to start module section after kernel Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 23/27] x86/modules: Adapt module loading for PIE support Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 24/27] x86/mm: Make the x86 GOT read-only Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 25/27] x86/pie: Add option to build the kernel as PIE Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 26/27] x86/relocs: Add option to generate 64-bit relocations Thomas Garnier
2019-01-31 19:24 ` [PATCH v6 27/27] x86/kaslr: Add option to extend KASLR range from 1GB to 3GB Thomas Garnier
2019-01-31 19:59 ` [PATCH v6 00/27] x86: PIE support and option to extend KASLR randomization Kees Cook
2019-01-31 21:40 ` Konrad Rzeszutek Wilk
2019-01-31 22:42   ` Thomas Garnier

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=20190131192533.34130-17-thgarnie@chromium.org \
    --to=thgarnie@chromium.org \
    --cc=akpm@linux-foundation.org \
    --cc=caoj.fnst@cn.fujitsu.com \
    --cc=forney@google.com \
    --cc=keescook@chromium.org \
    --cc=kernel-hardening@lists.openwall.com \
    --cc=kristen@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mingo@kernel.org \
    --cc=npiggin@gmail.com \
    --cc=palmer@sifive.com \
    --cc=peterz@infradead.org \
    --cc=sam@ravnborg.org \
    --cc=thgarnie@google.com \
    --cc=yamada.masahiro@socionext.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;
as well as URLs for NNTP newsgroup(s).