Skip to contentSkip to conversation

debug-pagealloc: depend on CONFIG_DEBUG_KERNEL

PATCH New Akinobu Mita 1 participant 2 messages Mar 10, 2009

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 debug-pagealloc: depend on CONFIG_DEBUG_KERNEL

The generic debug-pagealloc should depend on CONFIG_DEBUG_KERNEL as well as the architecture specific one.

---
 mm/Kconfig.debug |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
mm/Kconfig.debug +1 -1
Index: 2.6-mmotm/mm/Kconfig.debug
===================================================================
@@ -3,7 +3,7 @@ config WANT_PAGE_DEBUG_FLAGS
 
 config PAGE_POISONING
 	bool "Debug page memory allocations"
-	depends on !ARCH_SUPPORTS_DEBUG_PAGEALLOC
+	depends on DEBUG_KERNEL && !ARCH_SUPPORTS_DEBUG_PAGEALLOC
 	select DEBUG_PAGEALLOC
 	select WANT_PAGE_DEBUG_FLAGS
 	help
#1 lore raw

PATCH 2/2 debug-pagealloc: cannot depend on hibernation

The generic debug-pagealloc generates wrong messages for the page corruption after hibernation. Because the free pages are not saved to the suspend image.

---
 mm/Kconfig.debug |    5 +++++
 1 file changed, 5 insertions(+)
mm/Kconfig.debug +5 -0
Index: 2.6-mmotm/mm/Kconfig.debug
===================================================================
@@ -4,9 +4,14 @@ config WANT_PAGE_DEBUG_FLAGS
 config PAGE_POISONING
 	bool "Debug page memory allocations"
 	depends on DEBUG_KERNEL && !ARCH_SUPPORTS_DEBUG_PAGEALLOC
+	depends on !HIBERNATION
 	select DEBUG_PAGEALLOC
 	select WANT_PAGE_DEBUG_FLAGS
 	help
 	   Fill the pages with poison patterns after free_pages() and verify
 	   the patterns before alloc_pages(). This results in a large slowdown,
 	   but helps to find certain types of memory corruptions.
+
+	   This option cannot enalbe with hibernation. Otherwise, it will get
+	   wrong messages for memory corruption because the free pages are not
+	   saved to the suspend image.