On 23.04.2022 21:35, Paran Lee wrote:
> It doesn't seem necessary to do that
> duplicate calculation of order shift 2^@order in the loop.
Once again I'm not convinced. As in the other patch the compiler could
do this transformation via its CSE pass if it sees fit.
Also (applicable as well to the
It doesn't seem necessary to do that
duplicate calculation of order shift 2^@order in the loop.
In addition, I fixed type of total_avail_pages from long
to unsigned long. because when total_avail_pages static variable
substitute in functions of page alloc local variable,
type of local variables is
