# \[PATCH 00/20\] exit cleanups

- Subsystems: LINUX FOR POWERPC (32-BIT AND 64-BIT), BATMAN ADVANCED, FILESYSTEMS (VFS and infrastructure), GENERIC ENTRY CODE, INTEL STRATIX10 FIRMWARE DRIVERS, MIPS, OPENRISC ARCHITECTURE, ORACLE CLUSTER FILESYSTEM 2 (OCFS2), S390 ARCHITECTURE, SPARC + UltraSPARC (sparc/sparc64), STAGING SUBSYSTEM, SUPERH, TENSILICA XTENSA PORT (xtensa), THE REST, TI KEYSTONE MULTICORE NAVIGATOR DRIVERS, X86 ARCHITECTURE (32-BIT AND 64-BIT), X86 ENTRY CODE
- Maintainers: Madhavan Srinivasan, Marek Lindner, Simon Wunderlich, Antonio Quartulli, Sven Eckelmann, Alexander Viro, Christian Brauner, Thomas Gleixner, Peter Zijlstra, Andy Lutomirski, Dinh Nguyen, Thomas Bogendoerfer, Jonas Bonn, Stefan Kristiansson, Stafford Horne, Mark Fasheh, Joel Becker, Joseph Qi, Heiko Carstens, Vasily Gorbik, Alexander Gordeev, "David S. Miller", Andreas Larsson, Greg Kroah-Hartman, Yoshinori Sato, Rich Felker, John Paul Adrian Glaubitz, Chris Zankel, Max Filippov, Linus Torvalds, Nishanth Menon, Santosh Shilimkar, Ingo Molnar, Borislav Petkov, Dave Hansen
- Series: 20 patches touching 43 files, +83/-91
- Status: PENDING · 19 of 20 landed
- Pending, 19 of 20 landed.
- List: linuxppc-dev
- 3 messages, 2 authors
- Page: https://ratatoskr.run/linuxppc-dev/2021/10/1548606/t

## \[PATCH 00/20\] exit cleanups

- From: Eric W. Biederman \<hidden\>
- Date: 2021-10-20 17:32:40 
- Message: https://ratatoskr.run/linuxppc-dev/2021/10/1548606.md

```text

While looking at some issues related to the exit path in the kernel I
found several instances where the code is not using the existing
abstractions properly.

This set of changes introduces force_fatal_sig a way of sending
a signal and not allowing it to be caught, and corrects the
misuse of the existing abstractions that I found.

A lot of the misuse of the existing abstractions are silly things such
as doing something after calling a no return function, rolling BUG by
hand, doing more work than necessary to terminate a kernel thread, or
calling do_exit(SIGKILL) instead of calling force_sig(SIGKILL).

It is my plan after sending all of these changes out for review to place
them in a topic branch for sending Linus.  Especially for the changes
that depend upon the new helper force_fatal_sig this is important.

Eric W. Biederman (20):
      exit/doublefault: Remove apparently bogus comment about rewind_stack_do_exit
      exit: Remove calls of do_exit after noreturn versions of die
      reboot: Remove the unreachable panic after do_exit in reboot(2)
      signal/sparc32: Remove unreachable do_exit in do_sparc_fault
      signal/mips: Update (_save|_restore)_fp_context to fail with -EFAULT
      signal/sh: Use force_sig(SIGKILL) instead of do_group_exit(SIGKILL)
      signal/powerpc: On swapcontext failure force SIGSEGV
      signal/sparc: In setup_tsb_params convert open coded BUG into BUG
      signal/vm86_32: Replace open coded BUG_ON with an actual BUG_ON
      signal/vm86_32: Properly send SIGSEGV when the vm86 state cannot be saved.
      signal/s390: Use force_sigsegv in default_trap_handler
      exit/kthread: Have kernel threads return instead of calling do_exit
      signal: Implement force_fatal_sig
      exit/syscall_user_dispatch: Send ordinary signals on failure
      signal/sparc32: Exit with a fatal signal when try_to_clear_window_buffer fails
      signal/sparc32: In setup_rt_frame and setup_fram use force_fatal_sig
      signal/x86: In emulate_vsyscall force a signal instead of calling do_exit
      exit/rtl8723bs: Replace the macro thread_exit with a simple return 0
      exit/rtl8712: Replace the macro thread_exit with a simple return 0
      exit/r8188eu: Replace the macro thread_exit with a simple return 0

 arch/mips/kernel/r2300_fpu.S                       |  4 ++--
 arch/mips/kernel/syscall.c                         |  9 --------
 arch/nds32/kernel/traps.c                          |  2 +-
 arch/nds32/mm/fault.c                              |  6 +----
 arch/openrisc/kernel/traps.c                       |  2 +-
 arch/openrisc/mm/fault.c                           |  4 +---
 arch/powerpc/kernel/signal_32.c                    |  6 +++--
 arch/powerpc/kernel/signal_64.c                    |  9 +++++---
 arch/s390/include/asm/kdebug.h                     |  2 +-
 arch/s390/kernel/dumpstack.c                       |  2 +-
 arch/s390/kernel/traps.c                           |  2 +-
 arch/s390/mm/fault.c                               |  2 --
 arch/sh/kernel/cpu/fpu.c                           | 10 +++++----
 arch/sh/kernel/traps.c                             |  2 +-
 arch/sh/mm/fault.c                                 |  2 --
 arch/sparc/kernel/signal_32.c                      |  4 ++--
 arch/sparc/kernel/windows.c                        |  6 +++--
 arch/sparc/mm/fault_32.c                           |  1 -
 arch/sparc/mm/tsb.c                                |  2 +-
 arch/x86/entry/vsyscall/vsyscall_64.c              |  3 ++-
 arch/x86/kernel/doublefault_32.c                   |  3 ---
 arch/x86/kernel/signal.c                           |  6 ++++-
 arch/x86/kernel/vm86_32.c                          |  8 +++----
 arch/xtensa/kernel/traps.c                         |  2 +-
 arch/xtensa/mm/fault.c                             |  3 +--
 drivers/firmware/stratix10-svc.c                   |  4 ++--
 drivers/soc/ti/wkup_m3_ipc.c                       |  2 +-
 drivers/staging/r8188eu/core/rtw_cmd.c             |  2 +-
 drivers/staging/r8188eu/core/rtw_mp.c              |  2 +-
 drivers/staging/r8188eu/include/osdep_service.h    |  2 --
 drivers/staging/rtl8712/osdep_service.h            |  1 -
 drivers/staging/rtl8712/rtl8712_cmd.c              |  2 +-
 drivers/staging/rtl8723bs/core/rtw_cmd.c           |  2 +-
 drivers/staging/rtl8723bs/core/rtw_xmit.c          |  2 +-
 drivers/staging/rtl8723bs/hal/rtl8723bs_xmit.c     |  2 +-
 .../rtl8723bs/include/osdep_service_linux.h        |  2 --
 fs/ocfs2/journal.c                                 |  5 +----
 include/linux/sched/signal.h                       |  1 +
 kernel/entry/syscall_user_dispatch.c               | 12 ++++++----
 kernel/kthread.c                                   |  2 +-
 kernel/reboot.c                                    |  1 -
 kernel/signal.c                                    | 26 ++++++++++++++--------
 net/batman-adv/tp_meter.c                          |  2 +-
 43 files changed, 83 insertions(+), 91 deletions(-)

Eric
```

## \[PATCH 07/20\] signal/powerpc: On swapcontext failure force SIGSEGV

- From: Eric W. Biederman \<hidden\>
- Date: 2021-10-20 17:45:09 
- Message: https://ratatoskr.run/linuxppc-dev/2021/10/1548623.md

```text
If the register state may be partial and corrupted instead of calling
do_exit, call force_sigsegv(SIGSEGV).  Which properly kills the
process with SIGSEGV and does not let any more userspace code execute,
instead of just killing one thread of the process and potentially
confusing everything.

Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <redacted>
Cc: <redacted>
History-tree: git://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git
Fixes: 756f1ae8a44e ("PPC32: Rework signal code and add a swapcontext system call.")
Fixes: 04879b04bf50 ("[PATCH] ppc64: VMX (Altivec) support & signal32 rework, from Ben Herrenschmidt")
Signed-off-by: "Eric W. Biederman" <redacted>
---
 arch/powerpc/kernel/signal_32.c | 6 ++++--
 arch/powerpc/kernel/signal_64.c | 9 ++++++---
 2 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c
index 0608581967f0..666f3da41232 100644
--- a/arch/powerpc/kernel/signal_32.c
+++ b/arch/powerpc/kernel/signal_32.c
@@ -1062,8 +1062,10 @@ SYSCALL_DEFINE3(swapcontext, struct ucontext __user *, old_ctx,
 	 * or if another thread unmaps the region containing the context.
 	 * We kill the task with a SIGSEGV in this situation.
 	 */
-	if (do_setcontext(new_ctx, regs, 0))
-		do_exit(SIGSEGV);
+	if (do_setcontext(new_ctx, regs, 0)) {
+		force_sigsegv(SIGSEGV);
+		return -EFAULT;
+	}
 
 	set_thread_flag(TIF_RESTOREALL);
 	return 0;
diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c
index 1831bba0582e..d8de622c9e4a 100644
--- a/arch/powerpc/kernel/signal_64.c
+++ b/arch/powerpc/kernel/signal_64.c
@@ -703,15 +703,18 @@ SYSCALL_DEFINE3(swapcontext, struct ucontext __user *, old_ctx,
 	 * We kill the task with a SIGSEGV in this situation.
 	 */
 
-	if (__get_user_sigset(&set, &new_ctx->uc_sigmask))
-		do_exit(SIGSEGV);
+	if (__get_user_sigset(&set, &new_ctx->uc_sigmask)) {
+		force_sigsegv(SIGSEGV);
+		return -EFAULT;
+	}
 	set_current_blocked(&set);
 
 	if (!user_read_access_begin(new_ctx, ctx_size))
 		return -EFAULT;
 	if (__unsafe_restore_sigcontext(current, NULL, 0, &new_ctx->uc_mcontext)) {
 		user_read_access_end();
-		do_exit(SIGSEGV);
+		force_sigsegv(SIGSEGV);
+		return -EFAULT;
 	}
 	user_read_access_end();
 
-- 
2.20.1

```

## Re: \[PATCH 07/20\] signal/powerpc: On swapcontext failure force SIGSEGV

- From: Kees Cook \<hidden\>
- Date: 2021-10-21 16:09:47 
- Message: https://ratatoskr.run/linuxppc-dev/2021/10/1549868.md

```text
On Wed, Oct 20, 2021 at 12:43:53PM -0500, Eric W. Biederman wrote:
> If the register state may be partial and corrupted instead of calling
> do_exit, call force_sigsegv(SIGSEGV).  Which properly kills the
> process with SIGSEGV and does not let any more userspace code execute,
> instead of just killing one thread of the process and potentially
> confusing everything.
> 
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> Cc: Paul Mackerras <redacted>
> Cc: <redacted>
> History-tree: git://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git
> Fixes: 756f1ae8a44e ("PPC32: Rework signal code and add a swapcontext system call.")
> Fixes: 04879b04bf50 ("[PATCH] ppc64: VMX (Altivec) support & signal32 rework, from Ben Herrenschmidt")
> Signed-off-by: "Eric W. Biederman" <redacted>

This looks right to me.

Reviewed-by: Kees Cook <redacted>

-- 
Kees Cook
```
