While handling hwpoison in a THP page, it is possible that
try_to_split_thp_page() fails. For example, when the THP page has
been RDMA pinned. At this point, the kernel cannot isolate the
poisoned THP page, all it could do is to send a SIGBUS to the user
process with meaningful payload to give user-level recovery a chance.
Signed-off-by: Jane Chu <jane.chu@oracle.com>
---
mm/memory-failure.c | 35 ++++++++++++++++++++++++++++++-----
1 file changed, 30 insertions(+), 5 deletions(-)
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 794196951a04..a14d56e66902 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1706,7 +1706,12 @@ static int identify_page_state(unsigned long pfn, struct page *p,
return page_action(ps, p, pfn);
}
-static int try_to_split_thp_page(struct page *page)
+/*
+ * When 'release' is 'false', it means that if thp split has failed,
+ * there is still more to do, hence the page refcount we took earlier
+ * is still needed.
+ */
+static int try_to_split_thp_page(struct page *page, bool release)
{
int ret;
@@ -1714,7 +1719,7 @@ static int try_to_split_thp_page(struct page *page)
ret = split_huge_page(page);
unlock_page(page);
- if (unlikely(ret))
+ if (ret && release)
put_page(page);
return ret;
@@ -2187,6 +2192,24 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
return rc;
}
+/*
+ * The calling condition is as such: thp split failed, page might have
+ * been RDMA pinned, not much can be done for recovery.
+ * But a SIGBUS should be delivered with vaddr provided so that the user
+ * application has a chance to recover. Also, application processes'
+ * election for MCE early killed will be honored.
+ */
+static int kill_procs_now(struct page *p, unsigned long pfn, int flags,
+ struct folio *folio)
+{
+ LIST_HEAD(tokill);
+
+ collect_procs(folio, p, &tokill, flags & MF_ACTION_REQUIRED);
+ kill_procs(&tokill, true, pfn, flags);
+
+ return -EHWPOISON;
+}
+
/**
* memory_failure - Handle memory failure of a page.
* @pfn: Page Number of the corrupted page
@@ -2328,8 +2351,10 @@ int memory_failure(unsigned long pfn, int flags)
* page is a valid handlable page.
*/
folio_set_has_hwpoisoned(folio);
- if (try_to_split_thp_page(p) < 0) {
- res = action_result(pfn, MF_MSG_UNSPLIT_THP, MF_IGNORED);
+ if (try_to_split_thp_page(p, false) < 0) {
+ res = kill_procs_now(p, pfn, flags, folio);
+ put_page(p);
+ action_result(pfn, MF_MSG_UNSPLIT_THP, MF_FAILED);
goto unlock_mutex;
}
VM_BUG_ON_PAGE(!page_count(p), p);
@@ -2703,7 +2728,7 @@ static int soft_offline_in_use_page(struct page *page)
};
if (!huge && folio_test_large(folio)) {
- if (try_to_split_thp_page(page)) {
+ if (try_to_split_thp_page(page, true)) {
pr_info("soft offline: %#lx: thp split failed\n", pfn);
return -EBUSY;
}
--
2.39.3On 2024/5/22 7:54, Jane Chu wrote:
> While handling hwpoison in a THP page, it is possible that
> try_to_split_thp_page() fails. For example, when the THP page has
> been RDMA pinned. At this point, the kernel cannot isolate the
> poisoned THP page, all it could do is to send a SIGBUS to the user
> process with meaningful payload to give user-level recovery a chance.
>
Thanks for your patch.
> Signed-off-by: Jane Chu <jane.chu@oracle.com>
> ---
> mm/memory-failure.c | 35 ++++++++++++++++++++++++++++++-----
> 1 file changed, 30 insertions(+), 5 deletions(-)
>
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index 794196951a04..a14d56e66902 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -1706,7 +1706,12 @@ static int identify_page_state(unsigned long pfn, struct page *p,
> return page_action(ps, p, pfn);
> }
>
> -static int try_to_split_thp_page(struct page *page)
> +/*
> + * When 'release' is 'false', it means that if thp split has failed,
> + * there is still more to do, hence the page refcount we took earlier
> + * is still needed.
> + */
> +static int try_to_split_thp_page(struct page *page, bool release)
> {
> int ret;
>
> @@ -1714,7 +1719,7 @@ static int try_to_split_thp_page(struct page *page)
> ret = split_huge_page(page);
> unlock_page(page);
>
> - if (unlikely(ret))
> + if (ret && release)
> put_page(page);
Is "unlikely" still needed?
>
> return ret;
> @@ -2187,6 +2192,24 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
> return rc;
> }
>
> +/*
> + * The calling condition is as such: thp split failed, page might have
> + * been RDMA pinned, not much can be done for recovery.
> + * But a SIGBUS should be delivered with vaddr provided so that the user
> + * application has a chance to recover. Also, application processes'
> + * election for MCE early killed will be honored.
> + */
> +static int kill_procs_now(struct page *p, unsigned long pfn, int flags,
> + struct folio *folio)
> +{
> + LIST_HEAD(tokill);
> +
> + collect_procs(folio, p, &tokill, flags & MF_ACTION_REQUIRED);
> + kill_procs(&tokill, true, pfn, flags);
> +
> + return -EHWPOISON;
> +}
> +
> /**
> * memory_failure - Handle memory failure of a page.
> * @pfn: Page Number of the corrupted page
> @@ -2328,8 +2351,10 @@ int memory_failure(unsigned long pfn, int flags)
> * page is a valid handlable page.
> */
> folio_set_has_hwpoisoned(folio);
> - if (try_to_split_thp_page(p) < 0) {
> - res = action_result(pfn, MF_MSG_UNSPLIT_THP, MF_IGNORED);
> + if (try_to_split_thp_page(p, false) < 0) {
> + res = kill_procs_now(p, pfn, flags, folio);
No strong opinion but we might remove the return value of kill_procs_now as
it always return -EHWPOISON? We could simply set res to -EHWPOISON here.
Besides from above possible nits, this patch looks good to me.
Acked-by: Miaohe Lin <linmiaohe@huawei.com>
Thanks.
.
On 5/22/2024 8:02 PM, Miaohe Lin wrote:
> On 2024/5/22 7:54, Jane Chu wrote:
>> While handling hwpoison in a THP page, it is possible that
>> try_to_split_thp_page() fails. For example, when the THP page has
>> been RDMA pinned. At this point, the kernel cannot isolate the
>> poisoned THP page, all it could do is to send a SIGBUS to the user
>> process with meaningful payload to give user-level recovery a chance.
>>
> Thanks for your patch.
>
>> Signed-off-by: Jane Chu <jane.chu@oracle.com>
>> ---
>> mm/memory-failure.c | 35 ++++++++++++++++++++++++++++++-----
>> 1 file changed, 30 insertions(+), 5 deletions(-)
>>
>> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
>> index 794196951a04..a14d56e66902 100644
>> --- a/mm/memory-failure.c
>> +++ b/mm/memory-failure.c
>> @@ -1706,7 +1706,12 @@ static int identify_page_state(unsigned long pfn, struct page *p,
>> return page_action(ps, p, pfn);
>> }
>>
>> -static int try_to_split_thp_page(struct page *page)
>> +/*
>> + * When 'release' is 'false', it means that if thp split has failed,
>> + * there is still more to do, hence the page refcount we took earlier
>> + * is still needed.
>> + */
>> +static int try_to_split_thp_page(struct page *page, bool release)
>> {
>> int ret;
>>
>> @@ -1714,7 +1719,7 @@ static int try_to_split_thp_page(struct page *page)
>> ret = split_huge_page(page);
>> unlock_page(page);
>>
>> - if (unlikely(ret))
>> + if (ret && release)
>> put_page(page);
> Is "unlikely" still needed?
I'd say not, because this code is not on performance sensitive code path.
>>
>> return ret;
>> @@ -2187,6 +2192,24 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>> return rc;
>> }
>>
>> +/*
>> + * The calling condition is as such: thp split failed, page might have
>> + * been RDMA pinned, not much can be done for recovery.
>> + * But a SIGBUS should be delivered with vaddr provided so that the user
>> + * application has a chance to recover. Also, application processes'
>> + * election for MCE early killed will be honored.
>> + */
>> +static int kill_procs_now(struct page *p, unsigned long pfn, int flags,
>> + struct folio *folio)
>> +{
>> + LIST_HEAD(tokill);
>> +
>> + collect_procs(folio, p, &tokill, flags & MF_ACTION_REQUIRED);
>> + kill_procs(&tokill, true, pfn, flags);
>> +
>> + return -EHWPOISON;
>> +}
>> +
>> /**
>> * memory_failure - Handle memory failure of a page.
>> * @pfn: Page Number of the corrupted page
>> @@ -2328,8 +2351,10 @@ int memory_failure(unsigned long pfn, int flags)
>> * page is a valid handlable page.
>> */
>> folio_set_has_hwpoisoned(folio);
>> - if (try_to_split_thp_page(p) < 0) {
>> - res = action_result(pfn, MF_MSG_UNSPLIT_THP, MF_IGNORED);
>> + if (try_to_split_thp_page(p, false) < 0) {
>> + res = kill_procs_now(p, pfn, flags, folio);
> No strong opinion but we might remove the return value of kill_procs_now as
> it always return -EHWPOISON? We could simply set res to -EHWPOISON here.
I like that, will change.
>
> Besides from above possible nits, this patch looks good to me.
> Acked-by: Miaohe Lin <linmiaohe@huawei.com>
> Thanks.
Thank!
-jane
> .
>
On Tue, May 21, 2024 at 05:54:29PM -0600, Jane Chu wrote: > While handling hwpoison in a THP page, it is possible that > try_to_split_thp_page() fails. For example, when the THP page has > been RDMA pinned. At this point, the kernel cannot isolate the > poisoned THP page, all it could do is to send a SIGBUS to the user > process with meaningful payload to give user-level recovery a chance. > > Signed-off-by: Jane Chu <jane.chu@oracle.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> -- Oscar Salvador SUSE Labs
© 2016 - 2026 Red Hat, Inc.