Re: [PATCH] compat: Fix endian issue in union sigval

[Catalin Marinas <catalin.marinas@arm.com>]

On Fri, Feb 13, 2015 at 04:00:43PM +0800, Bamvor Jian Zhang wrote:
> On 2015/2/11 23:40, Catalin Marinas wrote:
> > On Wed, Feb 11, 2015 at 07:22:08PM +0800, Bamvor Jian Zhang wrote:
> >> On 2015/2/10 20:27, Catalin Marinas wrote:
> >>> On Tue, Feb 10, 2015 at 10:10:11AM +0000, Zhang Jian(Bamvor) wrote:
> ...
> > The native sigval_t is also a union but on 64-bit big endian, the
> > sival_int overlaps with the most significant 32-bit of the sival_ptr.
> > So reading sival_int would always be 0. When the compat siginfo is
> > copied to user, arm64 reads the native sival_ptr (si_ptr) and converts
> > it to the compat one, getting the correct 32-bit value. However, other
> > architectures access sival_int (si_int) instead which breaks with your
> > get_compat_sigevent() changes.
> > 
> >>> I think the correct fix is in the arm64 code:
> >>
> >> The following code could fix my issue.
> > 
> > Without any parts of your patch?
> 
> Yes. As you mentioned above, sival_int overlaps the most significant 32bit
> of the sival_ptr, it seems that your patch is right if sival_ptr is less than
> 32bit.

I don't follow you here. sival_ptr is greater than 32-bit on a 64-bit
kernel.

> > I think that's correct fix since in the SIGEV_THREAD mq_notify case, we
> > would not deliver a signal as notification, so the sival_int value is
> > irrelevant (it would be 0 for big-endian compat tasks because of the
> > sigval_t union on 64-bit).
> > 
> > Your patch would work as well but you have to change all the other
> > architectures to use si_ptr when copying to a compat siginfo.
> 
> Yeah, it seems that my patch is useful only if the sival_ptr is bigger
> than 32bit. It need the similar changes with following catalin's patch
> in the following 64bit architecture:
> 
> x86:                 arch/x86/ia32/ia32_signal.c

That's a little endian architecture and the use of ptr_to_compat() looks
fine to me.

> tile, s390:          arch/xxx/kernel/compat_signal.c

s390 uses si_int already, as in my proposed arm64 patch.

tile seems to be bi-endian, though I couldn't see a Kconfig option, nor
something defining __BIG_ENDIAN__ in headers or Makefile. I guess it's
coming from the compiler directly. Anyway, on big endian tile, I we have
the same issue as on big endian arm64.

> parisc, sparc, mips: arch/xxx/kernel/signal32.c

paris, sparc and mips all use si_int instead of si_ptr, so that's fine.

> powerpc:             arch/xxx/kernel/signal_32.c

Same for powerpc, it uses si_int instead of si_ptr.

> cc these maintainers for input.

I think it's only tile that needs fixing for big endian, something like
the arm64 patch below:

> >>> diff --git a/arch/arm64/kernel/signal32.c b/arch/arm64/kernel/signal32.c
> >>> index e299de396e9b..32601939a3c8 100644
> >>> --- a/arch/arm64/kernel/signal32.c
> >>> +++ b/arch/arm64/kernel/signal32.c
> >>> @@ -154,8 +154,7 @@ int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from)
> >>>  	case __SI_TIMER:
> >>>  		 err |= __put_user(from->si_tid, &to->si_tid);
> >>>  		 err |= __put_user(from->si_overrun, &to->si_overrun);
> >>> -		 err |= __put_user((compat_uptr_t)(unsigned long)from->si_ptr,
> >>> -				   &to->si_ptr);
> >>> +		 err |= __put_user(from->si_int, &to->si_int);
> >>>  		break;
> >>>  	case __SI_POLL:
> >>>  		err |= __put_user(from->si_band, &to->si_band);
> >>> @@ -184,7 +183,7 @@ int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from)
> >>>  	case __SI_MESGQ: /* But this is */
> >>>  		err |= __put_user(from->si_pid, &to->si_pid);
> >>>  		err |= __put_user(from->si_uid, &to->si_uid);
> >>> -		err |= __put_user((compat_uptr_t)(unsigned long)from->si_ptr, &to->si_ptr);
> >>> +		err |= __put_user(from->si_int, &to->si_int);
> >>>  		break;
> >>>  	case __SI_SYS:
> >>>  		err |= __put_user((compat_uptr_t)(unsigned long)

-- 
Catalin