2022-07-12 03:47:07

by Naoya Horiguchi

[permalink] [raw]
Subject: [mm-unstable PATCH v6 4/8] mm, hwpoison: make unpoison aware of raw error info in hwpoisoned hugepage

From: Naoya Horiguchi <[email protected]>

Raw error info list needs to be removed when hwpoisoned hugetlb is
unpoisoned. And unpoison handler needs to know how many errors there
are in the target hugepage. So add them.

HPageVmemmapOptimized(hpage) and HPageRawHwpUnreliable(hpage)) can't be
unpoisoned, so let's skip them.

Signed-off-by: Naoya Horiguchi <[email protected]>
Reported-by: kernel test robot <[email protected]>
---
v5 -> v6:
- set type of return value of hugetlb_clear_page_hwpoison() to void
- change type of return value of hugetlb_clear_page_hwpoison() to unsigned long

v4 -> v5:
- fix type of return value of free_raw_hwp_pages()
(found by kernel test robot),
- prevent unpoison for HPageVmemmapOptimized and HPageRawHwpUnreliable.
---
include/linux/swapops.h | 9 ++++++++
mm/memory-failure.c | 46 ++++++++++++++++++++++++++++++++++++-----
2 files changed, 50 insertions(+), 5 deletions(-)

diff --git a/include/linux/swapops.h b/include/linux/swapops.h
index a01aeb3fcc0b..ddc98f96ad2c 100644
--- a/include/linux/swapops.h
+++ b/include/linux/swapops.h
@@ -498,6 +498,11 @@ static inline void num_poisoned_pages_dec(void)
atomic_long_dec(&num_poisoned_pages);
}

+static inline void num_poisoned_pages_sub(long i)
+{
+ atomic_long_sub(i, &num_poisoned_pages);
+}
+
#else

static inline swp_entry_t make_hwpoison_entry(struct page *page)
@@ -518,6 +523,10 @@ static inline struct page *hwpoison_entry_to_page(swp_entry_t entry)
static inline void num_poisoned_pages_inc(void)
{
}
+
+static inline void num_poisoned_pages_sub(long i)
+{
+}
#endif

static inline int non_swap_entry(swp_entry_t entry)
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 74195c181d69..56fd9d809013 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1720,21 +1720,38 @@ static int hugetlb_set_page_hwpoison(struct page *hpage, struct page *page)
return ret;
}

-void hugetlb_clear_page_hwpoison(struct page *hpage)
+static unsigned long free_raw_hwp_pages(struct page *hpage, bool move_flag)
{
struct llist_head *head;
struct llist_node *t, *tnode;
+ unsigned long count = 0;

- if (!HPageRawHwpUnreliable(hpage))
- ClearPageHWPoison(hpage);
+ /*
+ * HPageVmemmapOptimized hugepages can't be unpoisoned because
+ * struct pages for tail pages are required to free hwpoisoned
+ * hugepages. HPageRawHwpUnreliable hugepages shouldn't be
+ * unpoisoned by definition.
+ */
+ if (HPageVmemmapOptimized(hpage) || HPageRawHwpUnreliable(hpage))
+ return 0;
head = raw_hwp_list_head(hpage);
llist_for_each_safe(tnode, t, head->first) {
struct raw_hwp_page *p = container_of(tnode, struct raw_hwp_page, node);

- SetPageHWPoison(p->page);
+ if (move_flag)
+ SetPageHWPoison(p->page);
kfree(p);
+ count++;
}
llist_del_all(head);
+ return count;
+}
+
+void hugetlb_clear_page_hwpoison(struct page *hpage)
+{
+ if (!HPageRawHwpUnreliable(hpage))
+ ClearPageHWPoison(hpage);
+ free_raw_hwp_pages(hpage, true);
}

/*
@@ -1878,6 +1895,10 @@ static inline int try_memory_failure_hugetlb(unsigned long pfn, int flags, int *
return 0;
}

+static inline unsigned long free_raw_hwp_pages(struct page *hpage, bool flag)
+{
+ return 0;
+}
#endif /* CONFIG_HUGETLB_PAGE */

static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
@@ -2283,6 +2304,7 @@ int unpoison_memory(unsigned long pfn)
struct page *p;
int ret = -EBUSY;
int freeit = 0;
+ unsigned long count = 1;
static DEFINE_RATELIMIT_STATE(unpoison_rs, DEFAULT_RATELIMIT_INTERVAL,
DEFAULT_RATELIMIT_BURST);

@@ -2330,6 +2352,13 @@ int unpoison_memory(unsigned long pfn)

ret = get_hwpoison_page(p, MF_UNPOISON);
if (!ret) {
+ if (PageHuge(p)) {
+ count = free_raw_hwp_pages(page, false);
+ if (count == 0) {
+ ret = -EBUSY;
+ goto unlock_mutex;
+ }
+ }
ret = TestClearPageHWPoison(page) ? 0 : -EBUSY;
} else if (ret < 0) {
if (ret == -EHWPOISON) {
@@ -2338,6 +2367,13 @@ int unpoison_memory(unsigned long pfn)
unpoison_pr_info("Unpoison: failed to grab page %#lx\n",
pfn, &unpoison_rs);
} else {
+ if (PageHuge(p)) {
+ count = free_raw_hwp_pages(page, false);
+ if (count == 0) {
+ ret = -EBUSY;
+ goto unlock_mutex;
+ }
+ }
freeit = !!TestClearPageHWPoison(p);

put_page(page);
@@ -2350,7 +2386,7 @@ int unpoison_memory(unsigned long pfn)
unlock_mutex:
mutex_unlock(&mf_mutex);
if (!ret || freeit) {
- num_poisoned_pages_dec();
+ num_poisoned_pages_sub(count);
unpoison_pr_info("Unpoison: Software-unpoisoned page %#lx\n",
page_to_pfn(p), &unpoison_rs);
}
--
2.25.1


Subject: Re: [mm-unstable PATCH v6 4/8] mm, hwpoison: make unpoison aware of raw error info in hwpoisoned hugepage

On Wed, Jul 13, 2022 at 12:24:46PM +0800, Miaohe Lin wrote:
> On 2022/7/12 11:28, Naoya Horiguchi wrote:
> > From: Naoya Horiguchi <[email protected]>
> >
> > Raw error info list needs to be removed when hwpoisoned hugetlb is
> > unpoisoned. And unpoison handler needs to know how many errors there
> > are in the target hugepage. So add them.
> >
> > HPageVmemmapOptimized(hpage) and HPageRawHwpUnreliable(hpage)) can't be
> > unpoisoned, so let's skip them.
> >
> > Signed-off-by: Naoya Horiguchi <[email protected]>
> > Reported-by: kernel test robot <[email protected]>
>
> This patch looks good to me with some nits below.
>
> > ---
> ...
> >
> > -void hugetlb_clear_page_hwpoison(struct page *hpage)
> > +static unsigned long free_raw_hwp_pages(struct page *hpage, bool move_flag)
> > {
> > struct llist_head *head;
> > struct llist_node *t, *tnode;
> > + unsigned long count = 0;
> >
> > - if (!HPageRawHwpUnreliable(hpage))
> > - ClearPageHWPoison(hpage);
> > + /*
> > + * HPageVmemmapOptimized hugepages can't be unpoisoned because
> > + * struct pages for tail pages are required to free hwpoisoned
> > + * hugepages. HPageRawHwpUnreliable hugepages shouldn't be
> > + * unpoisoned by definition.
> > + */
> > + if (HPageVmemmapOptimized(hpage) || HPageRawHwpUnreliable(hpage))
>
> If move_flag == false, i.e. in unpoison case, tail pages are not touched. So HPageVmemmapOptimized
> can be ignored in this case? Or leave it as above to keep the code simple?

Oh, right. We can make HPageVmemmapOptimized() more conditional.

>
> > + return 0;
> > head = raw_hwp_list_head(hpage);
> > llist_for_each_safe(tnode, t, head->first) {
> > struct raw_hwp_page *p = container_of(tnode, struct raw_hwp_page, node);
> >
> > - SetPageHWPoison(p->page);
> > + if (move_flag)
> > + SetPageHWPoison(p->page);
> > kfree(p);
> > + count++;
> > }
> > llist_del_all(head);
> > + return count;
> > +}
> > +
> > +void hugetlb_clear_page_hwpoison(struct page *hpage)
> > +{
> > + if (!HPageRawHwpUnreliable(hpage))
>
> It seems we can return here if HPageRawHwpUnreliable as there's nothing to do?

OK, I'll update this, too.

>
> Anyway, for what it worth,
>
> Reviewed-by: Miaohe Lin <[email protected]>

Thank you.
- Naoya Horiguchi

>
> Thanks.
>
> > + ClearPageHWPoison(hpage);
> > + free_raw_hwp_pages(hpage, true);
> > }
> >
> > /*