From: Muchun Song <songmuchun@bytedance.com>
To: "HORIGUCHI NAOYA(堀口 直也)" <naoya.horiguchi@nec.com>
Cc: Naoya Horiguchi <nao.horiguchi@gmail.com>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
David Hildenbrand <david@redhat.com>,
Mike Kravetz <mike.kravetz@oracle.com>,
Miaohe Lin <linmiaohe@huawei.com>,
Liu Shixin <liushixin2@huawei.com>,
Yang Shi <shy828301@gmail.com>,
Oscar Salvador <osalvador@suse.de>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 4/9] mm, hwpoison, hugetlb: support saving mechanism of raw error pages
Date: Tue, 28 Jun 2022 18:37:18 +0800 [thread overview]
Message-ID: <YrrZ3hZlqEb3rlM0@FVFYT0MHHV2J> (raw)
In-Reply-To: <20220628081754.GA2206088@hori.linux.bs1.fc.nec.co.jp>
On Tue, Jun 28, 2022 at 08:17:55AM +0000, HORIGUCHI NAOYA(堀口 直也) wrote:
> On Tue, Jun 28, 2022 at 02:26:47PM +0800, Muchun Song wrote:
> > On Tue, Jun 28, 2022 at 02:41:22AM +0000, HORIGUCHI NAOYA(堀口 直也) wrote:
> > > On Mon, Jun 27, 2022 at 05:26:01PM +0800, Muchun Song wrote:
> > > > On Fri, Jun 24, 2022 at 08:51:48AM +0900, Naoya Horiguchi wrote:
> > > > > From: Naoya Horiguchi <naoya.horiguchi@nec.com>
> ...
> > > > > + } else {
> > > > > + /*
> > > > > + * Failed to save raw error info. We no longer trace all
> > > > > + * hwpoisoned subpages, and we need refuse to free/dissolve
> > > > > + * this hwpoisoned hugepage.
> > > > > + */
> > > > > + set_raw_hwp_unreliable(hpage);
> > > > > + return ret;
> > > > > + }
> > > > > + return ret;
> > > > > +}
> > > > > +
> > > > > +inline int hugetlb_clear_page_hwpoison(struct page *hpage)
> > > > > +{
> > > > > + struct llist_head *head;
> > > > > + struct llist_node *t, *tnode;
> > > > > +
> > > > > + if (raw_hwp_unreliable(hpage))
> > > > > + return -EBUSY;
> > > >
> > > > IIUC, we use head page's PageHWPoison to synchronize hugetlb_clear_page_hwpoison()
> > > > and hugetlb_set_page_hwpoison(), right? If so, who can set hwp_unreliable here?
> > >
> > > Sorry if I might miss your point, but raw_hwp_unreliable is set when
> > > allocating raw_hwp_page failed. hugetlb_set_page_hwpoison() can be called
> >
> > Sorry. I have missed this. Thanks for your clarification.
> >
> > > multiple times on a hugepage and if one of the calls fails, the hwpoisoned
> > > hugepage becomes unreliable.
> > >
> > > BTW, as you pointed out above, if we switch to passing GFP_ATOMIC to kmalloc(),
> > > the kmalloc() never fails, so we no longer have to implement this unreliable
> >
> > No. kmalloc() with GFP_ATOMIC can fail unless I miss something important.
>
> OK, I've interpretted the comment about GFP_ATOMIC wrongly.
>
> * %GFP_ATOMIC users can not sleep and need the allocation to succeed. A lower
> * watermark is applied to allow access to "atomic reserves".
>
>
> > > flag, so things get simpler.
> > >
> > > >
> > > > > + ClearPageHWPoison(hpage);
> > > > > + head = raw_hwp_list_head(hpage);
> > > > > + llist_for_each_safe(tnode, t, head->first) {
> > > >
> > > > Is it possible that a new item is added hugetlb_set_page_hwpoison() and we do not
> > > > traverse it (we have cleared page's PageHWPoison)? Then we ignored a real hwpoison
> > > > page, right?
> > >
> > > Maybe you are mentioning the race like below. Yes, that's possible.
> > >
> >
> > Sorry, ignore my previous comments, I'm thinking something wrong.
> >
> > > CPU 0 CPU 1
> > >
> > > free_huge_page
> > > lock hugetlb_lock
> > > ClearHPageMigratable
> > remove_hugetlb_page()
> > // the page is non-HugeTLB now
>
> Oh, I missed that.
>
> > > unlock hugetlb_lock
> > > get_huge_page_for_hwpoison
> > > lock hugetlb_lock
> > > __get_huge_page_for_hwpoison
> >
> > // cannot reach here since it is not a HugeTLB page now.
> > // So this race is impossible. Then we fallback to normal
> > // page handling. Seems there is a new issue here.
> > //
> > // memory_failure()
> > // try_memory_failure_hugetlb()
> > // if (hugetlb)
> > // goto unlock_mutex;
> > // if (TestSetPageHWPoison(p)) {
> > // // This non-HugeTLB page's vmemmap is still optimized.
> >
> > Setting COMPOUND_PAGE_DTOR after hugetlb_vmemmap_restore() might fix this
> > issue, but we will encounter this race as you mentioned below.
>
> I don't have clear ideas about this now (I don't test vmemmap-optimized case
> yet), so I will think more about this case. Maybe memory_failure() need
> detect it because memory_failure() heaviliy depends on the status of struct
> page.
>
Because HVO (HugeTLB Vmemmap Optimization) will map all tail vmemmap pages
with read-only, we cannot write any data to some tail struct pages. It is
a new issue unrelated to this patch.
Thanks.
> Thanks,
> Naoya Horiguchi
>
> >
> > Thanks.
> >
> > > hugetlb_set_page_hwpoison
> > > allocate raw_hwp_page
> > > TestSetPageHWPoison
> > > update_and_free_page
> > > __update_and_free_page
> > > if (PageHWPoison)
> > > hugetlb_clear_page_hwpoison
> > > TestClearPageHWPoison
> > > // remove all list items
> > > llist_add
> > > unlock hugetlb_lock
> > >
> > >
> > > The end result seems not critical (leaking raced raw_hwp_page?), but
> > > we need fix.
next prev parent reply other threads:[~2022-06-28 10:37 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-23 23:51 [PATCH v2 0/9] mm, hwpoison: enable 1GB hugepage support (v2) Naoya Horiguchi
2022-06-23 23:51 ` [PATCH v2 1/9] mm/hugetlb: remove checking hstate_is_gigantic() in return_unused_surplus_pages() Naoya Horiguchi
2022-06-24 2:25 ` Miaohe Lin
2022-06-24 8:03 ` Muchun Song
2022-06-24 8:15 ` Miaohe Lin
2022-06-24 8:34 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-24 19:11 ` Mike Kravetz
2022-06-27 6:02 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-27 17:25 ` Mike Kravetz
2022-06-28 3:01 ` Miaohe Lin
2022-06-28 8:38 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-30 2:27 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-23 23:51 ` [PATCH v2 2/9] mm/hugetlb: separate path for hwpoison entry in copy_hugetlb_page_range() Naoya Horiguchi
2022-06-24 9:23 ` Miaohe Lin
2022-06-27 6:06 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-24 20:57 ` Mike Kravetz
2022-06-27 6:48 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-27 7:57 ` Muchun Song
2022-06-23 23:51 ` [PATCH v2 3/9] mm/hugetlb: make pud_huge() and huge_pud() aware of non-present pud entry Naoya Horiguchi
2022-06-24 8:40 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-25 0:02 ` Mike Kravetz
2022-06-27 7:07 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-25 9:42 ` Miaohe Lin
2022-06-27 7:24 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-23 23:51 ` [PATCH v2 4/9] mm, hwpoison, hugetlb: support saving mechanism of raw error pages Naoya Horiguchi
2022-06-27 3:16 ` Miaohe Lin
2022-06-27 7:56 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-27 9:26 ` Muchun Song
2022-06-28 2:41 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-28 6:26 ` Muchun Song
2022-06-28 7:51 ` Muchun Song
2022-06-28 8:17 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-28 10:37 ` Muchun Song [this message]
2022-06-23 23:51 ` [PATCH v2 5/9] mm, hwpoison: make unpoison aware of raw error info in hwpoisoned hugepage Naoya Horiguchi
2022-06-27 8:32 ` Miaohe Lin
2022-06-27 11:48 ` Miaohe Lin
2022-06-23 23:51 ` [PATCH v2 6/9] mm, hwpoison: set PG_hwpoison for busy hugetlb pages Naoya Horiguchi
2022-06-27 8:39 ` Miaohe Lin
2022-06-23 23:51 ` [PATCH v2 7/9] mm, hwpoison: make __page_handle_poison returns int Naoya Horiguchi
2022-06-27 9:02 ` Miaohe Lin
2022-06-28 6:02 ` HORIGUCHI NAOYA(堀口 直也)
2022-06-23 23:51 ` [PATCH v2 8/9] mm, hwpoison: skip raw hwpoison page in freeing 1GB hugepage Naoya Horiguchi
2022-06-27 12:24 ` Miaohe Lin
2022-06-23 23:51 ` [PATCH v2 9/9] mm, hwpoison: enable memory error handling on " Naoya Horiguchi
2022-06-28 2:06 ` Miaohe Lin
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=YrrZ3hZlqEb3rlM0@FVFYT0MHHV2J \
--to=songmuchun@bytedance.com \
--cc=akpm@linux-foundation.org \
--cc=david@redhat.com \
--cc=linmiaohe@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=liushixin2@huawei.com \
--cc=mike.kravetz@oracle.com \
--cc=nao.horiguchi@gmail.com \
--cc=naoya.horiguchi@nec.com \
--cc=osalvador@suse.de \
--cc=shy828301@gmail.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;
as well as URLs for NNTP newsgroup(s).