Thanks for the patches, please include akpm@xxxxxxxxxxxxxxxxxxxx in the
future, as we can merge the changes through Andrew as well.
Also please Cc yzaikin@xxxxxxxxxx, brendanhiggins@xxxxxxxxxx in follow
ups for now. They are looking at the sysctl testing code as well.
Some feedback below:
In-Reply-To: <20181227111231.12912-2-zev@xxxxxxxxxxxxxxxxx>
On Thu, Dec 27, 2018 at 05:12:29AM -0600, Zev Weiss wrote:
> +run_wideint_tests()
> +{
> + # check negative and positive 64-bit values, with and without
> + # bits set in the lower 31, and with and without bit 31 (sign
> + # bit of a 32-bit int) set. None of these are representable
> + # in 32 bits, and hence all should fail.
> + check_failure 0x0000010000000000
> + check_failure 0x0000010080000000
> + check_failure 0x000001ff7fffffff
> + check_failure 0x000001ffffffffff
> + check_failure 0xffffffff7fffffff
> + check_failure 0xffffffffffffffff
This s64 version of -1
> + check_failure 0xffffff0000000000
> + check_failure 0xffffff0080000000
> +}
It was still unclear from the comments and manually looking at the
values why they are clear candidates to always test from all respective
64-bit values. A comment per each would be useful.
Luis
