On 7/29/25 02:04, Byungchul Park wrote:
On Mon, Jul 28, 2025 at 07:49:30PM +0100, Pavel Begunkov wrote:
On 7/28/25 19:39, Mina Almasry wrote:
On Mon, Jul 28, 2025 at 11:35 AM Pavel Begunkov <asml.silence@xxxxxxxxx> wrote:
...>>> This may be my bad since I said we should check if it's 0 initialized.
It looks like on the devmem side as well we kvmalloc_array the niovs,
and if I'm checking through the helpers right, kvmalloc_array does
0-initialize indeed.
I wouldn't rely on that, it's just for zcrx I do:
kvmalloc_array(..., GFP_KERNEL | __GFP_ZERO);
For net_devmem_bind_dmabuf(), __GFP_ZERO will add bigger overhead than
just assignment, 'niov->pp = NULL'.
That's not a place where you should care about zeroing overhead.
I'd like to ask you if you are still good with __GFP_ZERO overhead
before going ahead.
However, it might be easier for you to just assign it directly
for devmem and let Mina add GFP_ZERO if he fancy the idea.
--
Pavel Begunkov
