iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

mm/memory-failure: send SIGBUS in the event of thp split fail

Browse Source 
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.

Link: https://lkml.kernel.org/r/20240524215306.2705454-6-jane.chu@oracle.com
Signed-off-by: Jane Chu <jane.chu@oracle.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <nao.horiguchi@gmail.com>
Cc: Oscar Salvador <oalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
Jane Chu 2024-05-24 15:53:06 -06:00 committed by Andrew Morton
parent 9b0ab153d7
commit 1a3798dece
1 changed files with 29 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
mm/memory-failure.c
View File

@ -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;
@ -2186,6 +2191,22 @@ out:
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 void 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);
}
/**
* memory_failure - Handle memory failure of a page.
* @pfn: Page Number of the corrupted page
@ -2327,8 +2348,11 @@ try_again:
* 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 = -EHWPOISON;
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);
@ -2709,7 +2733,7 @@ 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;
}