public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: Joseph Myers <joseph@codesourcery.com>
Cc: libc-alpha@sourceware.org
Subject: Re: [PATCH 08/19] nptl: arm: Fix Race conditions in pthread cancellation (BZ#12683)
Date: Tue, 12 Dec 2017 14:03:00 -0000	[thread overview]
Message-ID: <4cab20d3-fa04-ab73-1275-025ee6d8580c@linaro.org> (raw)
In-Reply-To: <alpine.DEB.2.20.1712121348200.4596@digraph.polyomino.org.uk>



On 12/12/2017 11:51, Joseph Myers wrote:
> On Tue, 12 Dec 2017, Adhemerval Zanella wrote:
> 
>> So I am not sure we need all the CFI directives to enable cancellation work
>> on ARM (at least current syscall wrappers which use C version of 
>> {INLINE,INTERNAL}_SYSCALL are not generating them).
> 
> The cfi_* on ARM are for debugging, not for cancellation (and so to get 
> corresponding output from the compiler you'd need to compile with -g).  
> ARM uses ".cfi_sections .debug_frame".
> 

Right, I think we can let the compiler generate the correct info in
such cases.  Below there is an updated patch that removes the arch
specific implementation.  I need to adjust the syscall_cancel from
02/19 patch to uses a '%' mark for function symbol declaration
(the version I sent uses a '@' which ARM gas interprets as a inline
comment).

---

This patch adds the ARM modifications required for the BZ#12683.
It basically adds the required ucontext_get_pc function and adjust
the generic syscall_cancel build.

For ARM we need to build syscall_cancel in ARM mode (-marm) to avoid
INTERNAL_SYSCALL to issue the syscall through the helper gate
__libc_do_syscall (which invalidates the mark checks on SIGCANCEL
handler).

Checked on arm-linux-gnueabihf.

	* sysdeps/unix/sysv/linux/arm/Makefile (CFLAGS-syscall_cancel.c): New
	rule.
	* sysdeps/unix/sysv/linux/arm/sigcontextinfo.h (ucontext_get_pc):
	New function.

Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
---
 ChangeLog                                    |  5 +++++
 sysdeps/unix/sysv/linux/arm/Makefile         |  3 +++
 sysdeps/unix/sysv/linux/arm/sigcontextinfo.h | 12 ++++++++++++
 3 files changed, 20 insertions(+)

diff --git a/sysdeps/unix/sysv/linux/arm/Makefile b/sysdeps/unix/sysv/linux/arm/Makefile
index 4adc35d..8f01b52 100644
--- a/sysdeps/unix/sysv/linux/arm/Makefile
+++ b/sysdeps/unix/sysv/linux/arm/Makefile
@@ -30,6 +30,9 @@ endif
 ifeq ($(subdir),nptl)
 libpthread-sysdep_routines += libc-do-syscall
 libpthread-shared-only-routines += libc-do-syscall
+
+# INLINE_SYSCALL uses the helper __libc_do_syscall in thumb mode.
+CFLAGS-syscall_cancel.c += -marm
 endif
 
 ifeq ($(subdir),resolv)
diff --git a/sysdeps/unix/sysv/linux/arm/sigcontextinfo.h b/sysdeps/unix/sysv/linux/arm/sigcontextinfo.h
index d3313af..8132a95 100644
--- a/sysdeps/unix/sysv/linux/arm/sigcontextinfo.h
+++ b/sysdeps/unix/sysv/linux/arm/sigcontextinfo.h
@@ -16,6 +16,10 @@
    License along with the GNU C Library.  If not, see
    <http://www.gnu.org/licenses/>.  */
 
+#ifndef _SIGCONTEXTINFO_H
+#define _SIGCONTEXTINFO_H
+
+#include <stdint.h>
 #include <sys/ucontext.h>
 
 #define SIGCONTEXT siginfo_t *_si, ucontext_t *
@@ -46,3 +50,11 @@
   (act)->sa_flags |= SA_SIGINFO; \
   (sigaction) (sig, act, oact); \
 })
+
+static inline uintptr_t
+ucontext_get_pc (const ucontext_t *uc)
+{
+  return uc->uc_mcontext.arm_pc;
+}
+
+#endif /* _SIGCONTEXTINFO_H  */
-- 
2.7.4

  reply	other threads:[~2017-12-12 14:03 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-11 19:07 [PATCH 00/19] nptl: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 18/19] nptl: mips: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 09/19] nptl: s390: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 04/19] nptl: x32: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 13/19] nptl: microblaze: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 17/19] nptl: sh: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 03/19] nptl: x86_64: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 11/19] nptl: alpha: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 08/19] nptl: arm: " Adhemerval Zanella
2017-12-11 20:16   ` Phil Blundell
2017-12-11 20:29     ` Andreas Schwab
2017-12-11 21:03       ` Phil Blundell
2017-12-11 23:57   ` Joseph Myers
2017-12-12 12:28     ` Adhemerval Zanella
2017-12-12 13:51       ` Joseph Myers
2017-12-12 14:03         ` Adhemerval Zanella [this message]
2017-12-11 19:07 ` [PATCH 15/19] nptl: sparc: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 05/19] nptl: i386: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 01/19] nptl: Fix testcases for new pthread cancellation mechanism Adhemerval Zanella
2017-12-12 12:18   ` Florian Weimer
2017-12-12 12:27     ` Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 16/19] nptl: nios2: Fix Race conditions in pthread cancellation (BZ#12683) Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 19/19] nptl: hppa: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 07/19] nptl: aarch64: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 02/19] nptl: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 10/19] nptl: ia64: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 06/19] nptl: powerpc: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 12/19] nptl: m68k: " Adhemerval Zanella
2017-12-11 19:07 ` [PATCH 14/19] nptl: tile: " Adhemerval Zanella
2017-12-11 19:18 ` [PATCH 00/19] nptl: " John Paul Adrian Glaubitz
2017-12-11 20:31   ` Adhemerval Zanella

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=4cab20d3-fa04-ab73-1275-025ee6d8580c@linaro.org \
    --to=adhemerval.zanella@linaro.org \
    --cc=joseph@codesourcery.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).