public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: will.deacon@arm•com (Will Deacon)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH v8 2/6] arm64: ptrace: allow tracer to skip a system call
Date: Tue, 18 Nov 2014 14:04:25 +0000	[thread overview]
Message-ID: <20141118140425.GM18842@arm.com> (raw)
In-Reply-To: <1416273038-15590-3-git-send-email-takahiro.akashi@linaro.org>

On Tue, Nov 18, 2014 at 01:10:34AM +0000, AKASHI Takahiro wrote:
> If tracer specifies -1 as a syscall number, this traced system call should
> be skipped with a return value specified in x0.
> This patch implements this semantics, but there is one restriction here:
> 
>     syscall(-1) always return ENOSYS whatever value is stored in x0
>     (a return value) at syscall entry.
> 
> Normally, with ptrace off, syscall(-1) returns -ENOSYS. With ptrace on,
> however, if a tracer didn't pay any attention to user-issued syscall(-1)
> and just let it go, it would return a value in x0 as in other system call
> cases. This means that this system call might succeed and yet see any bogus
> return value. This should be definitely avoided.
> 
> Please also note:
> * syscall entry tracing and syscall exit tracing (ftrace tracepoint and
>   audit) are always executed, if enabled, even when skipping a system call
>   (that is, -1).
>   In this way, we can avoid a potential bug where audit_syscall_entry()
>   might be called without audit_syscall_exit() at the previous system call
>   being called, that would cause OOPs in audit_syscall_entry().
> 
> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro•org>
> ---
>  arch/arm64/kernel/entry.S  |    3 +++
>  arch/arm64/kernel/ptrace.c |   18 ++++++++++++++++++
>  2 files changed, 21 insertions(+)
> 
> diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
> index 726b910..01118b1 100644
> --- a/arch/arm64/kernel/entry.S
> +++ b/arch/arm64/kernel/entry.S
> @@ -670,6 +670,8 @@ ENDPROC(el0_svc)
>  __sys_trace:
>  	mov	x0, sp
>  	bl	syscall_trace_enter
> +	cmp	w0, #-1				// skip the syscall?
> +	b.eq	__sys_trace_return_skipped
>  	adr	lr, __sys_trace_return		// return address
>  	uxtw	scno, w0			// syscall number (possibly new)
>  	mov	x1, sp				// pointer to regs
> @@ -684,6 +686,7 @@ __sys_trace:
>  
>  __sys_trace_return:
>  	str	x0, [sp]			// save returned x0
> +__sys_trace_return_skipped:
>  	mov	x0, sp
>  	bl	syscall_trace_exit
>  	b	ret_to_user
> diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c
> index 8b98781..34b1e85 100644
> --- a/arch/arm64/kernel/ptrace.c
> +++ b/arch/arm64/kernel/ptrace.c
> @@ -1149,6 +1149,8 @@ static void tracehook_report_syscall(struct pt_regs *regs,
>  
>  asmlinkage int syscall_trace_enter(struct pt_regs *regs)
>  {
> +	int orig_syscallno = regs->syscallno;
> +
>  	if (test_thread_flag(TIF_SYSCALL_TRACE))
>  		tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER);
>  
> @@ -1158,6 +1160,22 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs)
>  	audit_syscall_entry(regs->syscallno, regs->orig_x0, regs->regs[1],
>  			    regs->regs[2], regs->regs[3]);
>  
> +	if (((int)regs->syscallno == -1) && (orig_syscallno == -1)) {
> +		/*
> +		 * user-issued syscall(-1):
> +		 * RESTRICTION: We always return ENOSYS whatever value is
> +		 *   stored in x0 (a return value) at this point.
> +		 * Normally, with ptrace off, syscall(-1) returns -ENOSYS.
> +		 * With ptrace on, however, if a tracer didn't pay any
> +		 * attention to user-issued syscall(-1) and just let it go
> +		 * without a hack here, it would return a value in x0 as in
> +		 * other system call cases. This means that this system call
> +		 * might succeed and see any bogus return value.
> +		 * This should be definitely avoided.
> +		 */
> +		regs->regs[0] = -ENOSYS;
> +	}

I'm still really uncomfortable with this, and it doesn't seem to match what
arch/arm/ does either. Doesn't it also prevent a tracer from skipping
syscall(-1)?

Will

  reply	other threads:[~2014-11-18 14:04 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-18  1:10 [PATCH v8 0/6] arm64: add seccomp support AKASHI Takahiro
2014-11-18  1:10 ` [PATCH v8 1/6] arm64: ptrace: add NT_ARM_SYSTEM_CALL regset AKASHI Takahiro
2014-11-18  1:10 ` [PATCH v8 2/6] arm64: ptrace: allow tracer to skip a system call AKASHI Takahiro
2014-11-18 14:04   ` Will Deacon [this message]
2014-11-19  8:46     ` AKASHI Takahiro
2014-11-19 19:06       ` Will Deacon
2014-11-20  5:13         ` AKASHI Takahiro
2014-11-20  5:52           ` AKASHI Takahiro
2014-11-25 13:56             ` Will Deacon
2014-11-20 19:17           ` Will Deacon
2014-11-25  7:42             ` AKASHI Takahiro
2014-11-25 10:30               ` Will Deacon
2014-11-25 14:14           ` Russell King - ARM Linux
2014-11-18  1:10 ` [PATCH v8 3/6] asm-generic: add generic seccomp.h for secure computing mode 1 AKASHI Takahiro
2014-11-18  1:10 ` [PATCH v8 4/6] arm64: add seccomp syscall for compat task AKASHI Takahiro
2014-11-18  1:10 ` [PATCH v8 5/6] arm64: add SIGSYS siginfo " AKASHI Takahiro
2014-11-18  1:10 ` [PATCH v8 6/6] arm64: add seccomp support AKASHI Takahiro

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=20141118140425.GM18842@arm.com \
    --to=will.deacon@arm$(echo .)com \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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