On Thu, 16 Feb 2023, Matthew Wilcox wrote:
> > - len = (remaining_size > PAGE_SIZE) ? PAGE_SIZE : remaining_size;
> > -
> > - bio_add_page(clone, page, len, 0);
> > +have_pages:
> > + page->compound_order = order;
>
> No. You'll corrupt the next page if page is o
On Thu, Feb 16, 2023 at 12:47:08PM -0500, Mikulas Patocka wrote:
> + while (order > 0) {
> + page = alloc_pages(gfp_mask
> + | __GFP_NOMEMALLOC | __GFP_NORETRY |
> __GFP_NOWARN, order);
... | __GFP_COMP
> page = mempool_al
It was reported that allocating pages for the write buffer in dm-crypt
causes measurable overhead [1].
This patch changes dm-crypt to allocate compound pages if they are
available. If not, we fall back to the mempool.
[1] https://listman.redhat.com/archives/dm-devel/2023-February/053284.html
Sig
