public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Szabolcs Nagy <szabolcs.nagy@arm.com>
To: libc-alpha@sourceware.org
Subject: [PATCH v6 12/14] aarch64: fix pac-ret support in _mcount
Date: Wed,  1 Jul 2020 15:40:36 +0100	[thread overview]
Message-ID: <a688767ffbe00df6bac7e2f2d38801aace571927.1593612309.git.szabolcs.nagy@arm.com> (raw)
In-Reply-To: <cover.1593612309.git.szabolcs.nagy@arm.com>

Currently gcc -pg -mbranch-protection=pac-ret passes signed return
address to _mcount, so _mcount now has to always strip pac from the
frompc since that's from user code that may be built with pac-ret.

This is gcc PR target/94791: signed pointers should not escape and get
passed across extern call boundaries, since that's an ABI break, but
because existing gcc has this issue we work it around in glibc until
that is resolved. This is compatible with a fixed gcc and it is a nop
on systems without PAuth support. The bug was introduced in gcc-7 with
-msign-return-address=non-leaf|all support which in gcc-9 got renamed
to -mbranch-protection=pac-ret|pac-ret+leaf|standard.

strip_pac uses inline asm instead of __builtin_aarch64_xpaclri since
that is not a documented api and not available in all supported gccs.
---
 sysdeps/aarch64/machine-gmon.h |  3 ++-
 sysdeps/aarch64/sysdep.h       | 11 +++++++++++
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/sysdeps/aarch64/machine-gmon.h b/sysdeps/aarch64/machine-gmon.h
index 730a23b781..a687298b1c 100644
--- a/sysdeps/aarch64/machine-gmon.h
+++ b/sysdeps/aarch64/machine-gmon.h
@@ -27,8 +27,9 @@ static void mcount_internal (u_long frompc, u_long selfpc);
 #define _MCOUNT_DECL(frompc, selfpc) \
 static inline void mcount_internal (u_long frompc, u_long selfpc)
 
+/* Note: strip_pac is needed for frompc because of gcc PR target/94791.  */
 #define MCOUNT                                                    \
 void __mcount (void *frompc)                                      \
 {                                                                 \
-  mcount_internal ((u_long) frompc, (u_long) RETURN_ADDRESS (0)); \
+  mcount_internal ((u_long) strip_pac (frompc), (u_long) RETURN_ADDRESS (0)); \
 }
diff --git a/sysdeps/aarch64/sysdep.h b/sysdeps/aarch64/sysdep.h
index 500c272745..2879aeaa5c 100644
--- a/sysdeps/aarch64/sysdep.h
+++ b/sysdeps/aarch64/sysdep.h
@@ -35,6 +35,17 @@
 
 #define PTR_SIZE	(1<<PTR_LOG_SIZE)
 
+#ifndef __ASSEMBLER__
+/* Strip pointer authentication code from pointer p.  */
+static inline void *
+strip_pac (void *p)
+{
+  register void *ra asm ("x30") = (p);
+  asm ("hint 7 // xpaclri" : "+r"(ra));
+  return ra;
+}
+#endif
+
 #ifdef	__ASSEMBLER__
 
 /* Syntactic details of assembler.  */
-- 
2.17.1


  parent reply	other threads:[~2020-07-01 14:40 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-01 14:37 [PATCH v6 00/14] aarch64: branch protection support Szabolcs Nagy
2020-07-01 14:37 ` [PATCH v6 01/14] Rewrite abi-note.S in C Szabolcs Nagy
2020-07-01 14:41   ` H.J. Lu
2020-07-01 17:31     ` Szabolcs Nagy
2020-07-01 17:43       ` H.J. Lu
2020-07-02  8:39         ` Szabolcs Nagy
2020-07-01 14:38 ` [PATCH v6 02/14] aarch64: configure test for BTI support Szabolcs Nagy
2020-07-06 14:11   ` Adhemerval Zanella
2020-07-06 18:07     ` Szabolcs Nagy
2020-07-06 18:12       ` Adhemerval Zanella
2020-07-07 14:26         ` Szabolcs Nagy
2020-07-07 14:39           ` H.J. Lu
2020-07-07 16:58             ` Szabolcs Nagy
2020-07-07 17:24               ` H.J. Lu
2020-07-01 14:38 ` [PATCH v6 03/14] aarch64: Rename place holder .S files to .c Szabolcs Nagy
2020-07-01 14:38 ` [PATCH v6 04/14] aarch64: Add BTI support to assembly files Szabolcs Nagy
2020-07-03 16:19   ` Szabolcs Nagy
2020-07-01 14:38 ` [PATCH v6 05/14] aarch64: fix swapcontext for BTI Szabolcs Nagy
2020-07-01 14:39 ` [PATCH v6 06/14] aarch64: fix RTLD_START " Szabolcs Nagy
2020-07-01 14:39 ` [PATCH v6 07/14] rtld: Clean up PT_NOTE and add PT_GNU_PROPERTY handling Szabolcs Nagy
2020-07-06 16:11   ` Adhemerval Zanella
2020-07-01 14:39 ` [PATCH v6 08/14] aarch64: enable BTI at runtime Szabolcs Nagy
2020-07-06 17:28   ` Adhemerval Zanella
2020-07-11 15:58   ` Richard Henderson
2020-07-13  8:32     ` Szabolcs Nagy
2020-07-13 13:14     ` Szabolcs Nagy
2020-07-13 13:28       ` Szabolcs Nagy
2020-07-13 16:55         ` Richard Henderson
2020-07-01 14:39 ` [PATCH v6 09/14] aarch64: ensure objects are BTI compatible Szabolcs Nagy
2020-07-06 17:37   ` Adhemerval Zanella
2020-07-06 18:01     ` Szabolcs Nagy
2020-07-06 18:17       ` Adhemerval Zanella
2020-07-01 14:40 ` [PATCH v6 10/14] aarch64: configure check for pac-ret code generation Szabolcs Nagy
2020-07-01 14:40 ` [PATCH v6 11/14] aarch64: Add pac-ret support to assembly files Szabolcs Nagy
2020-07-01 14:40 ` Szabolcs Nagy [this message]
2020-07-06 18:33   ` [PATCH v6 12/14] aarch64: fix pac-ret support in _mcount Adhemerval Zanella
2020-07-01 14:40 ` [PATCH v6 13/14] aarch64: redefine RETURN_ADDRESS to strip PAC Szabolcs Nagy
2020-07-06 18:34   ` Adhemerval Zanella
2020-07-01 14:41 ` [PATCH v6 14/14] aarch64: add NEWS entry about branch protection support Szabolcs Nagy
2020-07-06 18:41   ` Adhemerval Zanella
2020-07-08 10:04     ` Szabolcs Nagy

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=a688767ffbe00df6bac7e2f2d38801aace571927.1593612309.git.szabolcs.nagy@arm.com \
    --to=szabolcs.nagy@arm.com \
    --cc=libc-alpha@sourceware.org \
    /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 read-only IMAP folder(s) and NNTP newsgroup(s).