[PATCH -next] arch/x86: Return value from notify_die should to be checked.

[PATCH -next] arch/x86: Return value from notify_die should to be checked.

[Bo YU]

This is detected by Coverity scan: #CID: 1464472(CHECKED_RETURN)

FIXES: c94082656dac7(x86: Use enum instead of literals for trap values)
Signed-off-by: Bo YU <tsu.yubo@gmail.com>
---
 arch/x86/kernel/traps.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index af75109485c2..bf014fb59017 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -401,7 +401,8 @@ DEFINE_IDTENTRY_DF(exc_double_fault)

 	nmi_enter();
 	instrumentation_begin();
-	notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);
+	if (notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV))
+		return;

 	tsk->thread.error_code = error_code;
 	tsk->thread.trap_nr = X86_TRAP_DF;
--
2.11.0

Re: [PATCH -next] arch/x86: Return value from notify_die should to be checked.

[Alexandre Chartre]

On 6/21/20 12:26 PM, Bo YU wrote:
> This is detected by Coverity scan: #CID: 1464472(CHECKED_RETURN)
> 
> FIXES: c94082656dac7(x86: Use enum instead of literals for trap values)
> Signed-off-by: Bo YU <tsu.yubo@gmail.com>
> ---
> arch/x86/kernel/traps.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
> index af75109485c2..bf014fb59017 100644
> --- a/arch/x86/kernel/traps.c
> +++ b/arch/x86/kernel/traps.c
> @@ -401,7 +401,8 @@ DEFINE_IDTENTRY_DF(exc_double_fault)
> 
>      nmi_enter();
>      instrumentation_begin();
> -    notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);
> +    if (notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV))
> +        return;
> 

This change is not correct, if there's a double fault then we should die even if
notify_die() fails. So the appropriate change to make Coverity happy is probably:

  (void) notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);

alex.

Re: [PATCH -next] arch/x86: Return value from notify_die should to be checked.

[Boris Petkov]

On June 22, 2020 10:52:23 AM GMT+02:00, Alexandre Chartre <alexandre.chartre@oracle.com> wrote:
> So the appropriate change to make Coverity happy 

Or we can stop "fixing" the kernel in order to shut up tools and not do anything.


-- 
Sent from a small device: formatting sux and brevity is inevitable.

Re: [PATCH -next] arch/x86: Return value from notify_die should to be checked.

[Peter Zijlstra]

On Mon, Jun 22, 2020 at 11:17:51AM +0200, Boris Petkov wrote:
> On June 22, 2020 10:52:23 AM GMT+02:00, Alexandre Chartre <alexandre.chartre@oracle.com> wrote:
> > So the appropriate change to make Coverity happy 
> 
> Or we can stop "fixing" the kernel in order to shut up tools and not do anything.

Agreed, no change required here.

Re: [PATCH -next] arch/x86: Return value from notify_die should to be checked.

[Bo YU]

On Mon, Jun 22, 2020 at 5:46 PM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Mon, Jun 22, 2020 at 11:17:51AM +0200, Boris Petkov wrote:
> > On June 22, 2020 10:52:23 AM GMT+02:00, Alexandre Chartre <alexandre.chartre@oracle.com> wrote:
> > > So the appropriate change to make Coverity happy
> >
> > Or we can stop "fixing" the kernel in order to shut up tools and not do anything.
>
> Agreed, no change required here.
Ok, thanks for everyone.