since commit 031bc5743f158 ("mm/debug-pagealloc: make debug-pagealloc
boottime configurable") CONFIG_DEBUG_PAGEALLOC is by default not
adding any page debugging.

This resulted in several unnoticed bugs, e.g.

https://lkml.kernel.org/g/<569f5e29.3090...@de.ibm.com>
or
https://lkml.kernel.org/g/<56a20f30.4050...@de.ibm.com>

as this behaviour change was not even documented in Kconfig.

Let's provide a new Kconfig symbol that allows to change the default
back to enabled, e.g. for debug kernels. This also makes the change
obvious to kernel packagers.

Let's also change the Kconfig description for CONFIG_DEBUG_PAGEALLOC,
to indicate that there are two stages of overhead.

Cc: Joonsoo Kim <iamjoonsoo....@lge.com>
Cc: Andrew Morton <a...@linux-foundation.org>
Signed-off-by: Christian Borntraeger <borntrae...@de.ibm.com>
---
V1->V2: change Kconfig help to indicate, that CONFIG_DEBUG_PAGEALLOC
        is not for free, even if disabled

 mm/Kconfig.debug | 20 ++++++++++++++++++--
 mm/page_alloc.c  |  6 +++++-
 2 files changed, 23 insertions(+), 3 deletions(-)

diff --git a/mm/Kconfig.debug b/mm/Kconfig.debug
index 957d3da..f15c1cd 100644
--- a/mm/Kconfig.debug
+++ b/mm/Kconfig.debug
@@ -16,8 +16,8 @@ config DEBUG_PAGEALLOC
        select PAGE_POISONING if !ARCH_SUPPORTS_DEBUG_PAGEALLOC
        ---help---
          Unmap pages from the kernel linear mapping after free_pages().
-         This results in a large slowdown, but helps to find certain types
-         of memory corruption.
+         Depending on runtime enablement, this results in a small or large
+         slowdown, but helps to find certain types of memory corruption.
 
          For architectures which don't enable ARCH_SUPPORTS_DEBUG_PAGEALLOC,
          fill the pages with poison patterns after free_pages() and verify
@@ -26,5 +26,21 @@ config DEBUG_PAGEALLOC
          that would result in incorrect warnings of memory corruption after
          a resume because free pages are not saved to the suspend image.
 
+         By default this option will have a small overhead, e.g. by not
+         allowing the kernel mapping to be backed by large pages on some
+         architectures. Even bigger overhead comes when the debugging is
+         enabled by DEBUG_PAGEALLOC_ENABLE_DEFAULT or the debug_pagealloc
+         command line parameter.
+
+config DEBUG_PAGEALLOC_ENABLE_DEFAULT
+       bool "Enable debug page memory allocations by default?"
+        default off
+        depends on DEBUG_PAGEALLOC
+        ---help---
+         Enable debug page memory allocations by default? This value
+         can be overridden by debug_pagealloc=off|on.
+
+         If unsure say no.
+
 config PAGE_POISONING
        bool
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9d666df..933def7 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -479,7 +479,8 @@ void prep_compound_page(struct page *page, unsigned int 
order)
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 unsigned int _debug_guardpage_minorder;
-bool _debug_pagealloc_enabled __read_mostly;
+bool _debug_pagealloc_enabled __read_mostly
+                       = IS_ENABLED(CONFIG_DEBUG_PAGEALLOC_ENABLE_DEFAULT);
 bool _debug_guardpage_enabled __read_mostly;
 
 static int __init early_debug_pagealloc(char *buf)
@@ -490,6 +491,9 @@ static int __init early_debug_pagealloc(char *buf)
        if (strcmp(buf, "on") == 0)
                _debug_pagealloc_enabled = true;
 
+       if (strcmp(buf, "off") == 0)
+               _debug_pagealloc_enabled = false;
+
        return 0;
 }
 early_param("debug_pagealloc", early_debug_pagealloc);
-- 
2.3.0

Reply via email to