Commit a4866aa81251 ("mm: Tighten x86 /dev/mem with zeroing reads") adds a
comment to the function devmem_is_allowed() referring to a non-existing
config STRICT_IOMEM, whereas the comment very likely intended to refer to
the config STRICT_DEVMEM, as the commit adds some behavior for the config
STRICT_DEVMEM.
Most of the initial analysis was actually done by Dave Hansen in the
email thread below (see Link).
Refer to the intended and existing config STRICT_DEVMEM.
Link: https://lore.kernel.org/r/f9074e8d-9314-9d7d-7bf5-5b5538c8be8d@xxxxxxxxx/
Suggested-by: Dave Hansen <dave.hansen@xxxxxxxxx>
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@xxxxxxxxx>
---
v1: https://lore.kernel.org/all/20211222125347.13054-1-lukas.bulwahn@xxxxxxxxx/
arch/x86/mm/init.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
index d8cfce221275..5d747a1923cb 100644
--- a/arch/x86/mm/init.c
+++ b/arch/x86/mm/init.c
@@ -846,7 +846,7 @@ int devmem_is_allowed(unsigned long pagenr)
/*
* This must follow RAM test, since System RAM is considered a
- * restricted resource under CONFIG_STRICT_IOMEM.
+ * restricted resource under CONFIG_STRICT_DEVMEM.
*/
if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) {
/* Low 1MB bypasses iomem restrictions. */
--
2.17.1
