From: David Hildenbrand <david@redhat.com>
To: Kefeng Wang <wangkefeng.wang@huawei.com>,
Andrew Morton <akpm@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>,
Muchun Song <muchun.song@linux.dev>,
"Huang, Ying" <ying.huang@intel.com>,
linux-mm@kvack.org
Subject: Re: [PATCH v2 2/2] mm: use aligned address in copy_user_gigantic_page()
Date: Mon, 28 Oct 2024 11:01:45 +0100 [thread overview]
Message-ID: <839d4c08-c3a9-4163-b74e-06d309d7d9dc@redhat.com> (raw)
In-Reply-To: <20241026054307.3896926-2-wangkefeng.wang@huawei.com>
On 26.10.24 07:43, Kefeng Wang wrote:
> When copying gigantic page, it copies page from the first page to the
> last page, if directly passing addr_hint which maybe not the address
> of the first page of folio, then some archs could flush the wrong cache
> if it does use the addr_hint as a hint. For non-gigantic page, it
> calculates the base address inside, even passed the wrong addr_hint, it
> only has performance impact as the process_huge_page() wants to process
> target page last to keep its cache lines hot), no functional impact.
>
> Let's pass the real accessed address to copy_user_large_folio() and use
> the aligned address in copy_user_gigantic_page() to fix it.
>
> Fixes: 530dd9926dc1 ("mm: memory: improve copy_user_large_folio()")
> Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
> ---
> v2:
> - update changelog to clarify the impact, per Andrew
>
> mm/hugetlb.c | 5 ++---
> mm/memory.c | 1 +
> 2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 2c8c5da0f5d3..15b5d46d49d2 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -5338,7 +5338,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
> break;
> }
> ret = copy_user_large_folio(new_folio, pte_folio,
> - ALIGN_DOWN(addr, sz), dst_vma);
> + addr, dst_vma);
> folio_put(pte_folio);
> if (ret) {
> folio_put(new_folio);
> @@ -6641,8 +6641,7 @@ int hugetlb_mfill_atomic_pte(pte_t *dst_pte,
> *foliop = NULL;
> goto out;
> }
> - ret = copy_user_large_folio(folio, *foliop,
> - ALIGN_DOWN(dst_addr, size), dst_vma);
> + ret = copy_user_large_folio(folio, *foliop, dst_addr, dst_vma);
> folio_put(*foliop);
> *foliop = NULL;
> if (ret) {
> diff --git a/mm/memory.c b/mm/memory.c
> index ef47b7ea5ddd..e5284bab659d 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -6860,6 +6860,7 @@ static int copy_user_gigantic_page(struct folio *dst, struct folio *src,
> struct page *dst_page;
> struct page *src_page;
>
> + addr = ALIGN_DOWN(addr, folio_size(dst));
Same thing, please make it clearer that there is an "addr_hint" and an
"addr".
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2024-10-28 10:01 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-26 5:43 [PATCH v2 1/2] mm: use aligned address in clear_gigantic_page() Kefeng Wang
2024-10-26 5:43 ` [PATCH v2 2/2] mm: use aligned address in copy_user_gigantic_page() Kefeng Wang
2024-10-28 10:01 ` David Hildenbrand [this message]
2024-10-28 6:17 ` [PATCH v2 1/2] mm: use aligned address in clear_gigantic_page() Huang, Ying
2024-10-28 6:35 ` Kefeng Wang
2024-10-28 7:03 ` Huang, Ying
2024-10-28 8:35 ` Kefeng Wang
2024-10-28 10:00 ` David Hildenbrand
2024-10-28 12:52 ` Kefeng Wang
2024-10-28 13:14 ` David Hildenbrand
2024-10-28 13:33 ` Kefeng Wang
2024-10-28 13:46 ` David Hildenbrand
2024-10-28 14:22 ` Kefeng Wang
2024-10-28 14:24 ` David Hildenbrand
2024-10-29 13:04 ` Kefeng Wang
2024-10-29 14:04 ` David Hildenbrand
2024-10-30 1:04 ` Huang, Ying
2024-10-30 3:04 ` Kefeng Wang
2024-10-30 3:21 ` Huang, Ying
2024-10-30 5:05 ` Kefeng Wang
2024-10-31 8:39 ` Huang, Ying
2024-11-01 7:43 ` Kefeng Wang
2024-11-01 8:16 ` Huang, Ying
2024-11-01 9:45 ` Kefeng Wang
2024-11-04 2:35 ` Huang, Ying
2024-11-05 2:06 ` Kefeng Wang
2024-12-01 2:15 ` Andrew Morton
2024-12-01 5:37 ` Huang, Ying
2024-12-02 1:03 ` Kefeng Wang
2024-12-06 1:47 ` Andrew Morton
2024-12-06 2:08 ` Kefeng Wang
2024-11-01 6:18 ` Huang, Ying
2024-11-01 7:51 ` Kefeng Wang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=839d4c08-c3a9-4163-b74e-06d309d7d9dc@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=linux-mm@kvack.org \
--cc=muchun.song@linux.dev \
--cc=wangkefeng.wang@huawei.com \
--cc=willy@infradead.org \
--cc=ying.huang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox