Re: [PATCH v2 4/5] pselect6: use __kernel_timespec

On Sat, Sep 15, 2018 at 8:28 AM Arnd Bergmann [off-list ref] wrote:

On Sat, Sep 15, 2018 at 7:09 AM Deepa Dinamani [off-list ref] wrote:

quoted

+#if defined(CONFIG_64BIT_TIME)
+
+COMPAT_SYSCALL_DEFINE6(pselect6_time64, int, n, compat_ulong_t __user *, inp,
+       compat_ulong_t __user *, outp, compat_ulong_t __user *, exp,
+       struct __kernel_timespec __user *, tsp, void __user *, sig)

I got a link error here since compat_sys_pselect6_time64 and
compat_sys_ppoll_time64 are only defined when CONFIG_64BIT_TIME
is set.

I did not think we would select this symbol on arm64, is that a mistake
on my side, or should the #ifdef check be removed?

But, this is a compat syscall.
When we introduced this CONFIG_64BIT_TIME we planned to use it for
compat syscalls also is my understanding.

config 64BIT_TIME
         def_bool ARCH_HAS_64BIT_TIME
         help
          This should be selected by all architectures that need to support
           new system calls with a 64-bit time_t. This is relevant on all 32-bit
           architectures, and 64-bit architectures as part of compat syscall
           handling.

This means it should be set on 64 bit architechtures also right?

If we don't have the #ifdef here then we have an entry point from
userspace defined and Thomas had pointed out that it was a security
hole.

-Deepa

--
To unsubscribe, send a message with 'unsubscribe linux-aio' in
the body to majordomo@kvack.org.  For more info on Linux AIO,
see: http://www.kvack.org/aio/
Don't email: <a href=mailto:"aart@kvack.org">aart@kvack.org</a>