On Sat, 20 Mar 2021 00:35:16 +0000
Matthew Wilcox [off-list ref] wrote:

On Fri, Mar 19, 2021 at 10:44:37AM +0800, Aili Yao wrote:

quoted

+++ b/mm/gup.c

@@ -1536,6 +1536,10 @@ struct page *get_dump_page(unsigned long addr)
 				      FOLL_FORCE | FOLL_DUMP | FOLL_GET);
 	if (locked)
 		mmap_read_unlock(mm);
+
+	if (ret == 1 && is_page_poisoned(page))
+		return NULL;
+
 	return (ret == 1) ? page : NULL;
 }
 #endif /* CONFIG_ELF_CORE */

diff --git a/mm/internal.h b/mm/internal.h
index 25d2b2439..902d993 100644
--- a/mm/internal.h
+++ b/mm/internal.h

@@ -97,6 +97,27 @@ static inline void set_page_refcounted(struct page *page)
 	set_page_count(page, 1);
 }
 
+/*
+ * When kernel touch the user page, the user page may be have been marked
+ * poison but still mapped in user space, if without this page, the kernel
+ * can guarantee the data integrity and operation success, the kernel is
+ * better to check the posion status and avoid touching it, be good not to
+ * panic, coredump for process fatal signal is a sample case matching this
+ * scenario. Or if kernel can't guarantee the data integrity, it's better
+ * not to call this function, let kernel touch the poison page and get to
+ * panic.
+ */
+static inline bool is_page_poisoned(struct page *page)
+{
+	if (page != NULL) {  

Why are you checking page for NULL here?  How can it possibly be NULL?

For this get_dump_page() case, it can't be NULL, I thougt may other place
will call this function and may not guarantee this, But yes, kernel is a more
safer place and checking page NULL is not a common behavior.

Better to remove it, Thanks you very much for pointing this!

quoted

+		if (PageHWPoison(page))
+			return true;
+		else if (PageHuge(page) && PageHWPoison(compound_head(page)))
+			return true;
+	}
+	return 0;
+}
+
 extern unsigned long highest_memmap_pfn;
 
 /*
-- 
1.8.3.1

  

-- 
Thanks!
Aili Yao