[BUG Fix Patch 3/6] Bug fix: Do not free direct mapping pages twice.
From: Tang Chen <hidden>
Date: 2013-01-15 10:56:08
Also in:
linux-acpi, linux-mm, linux-s390, linux-sh, lkml, sparclinux
Subsystem:
the rest, x86 architecture (32-bit and 64-bit), x86 mm · Maintainers:
Linus Torvalds, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, Andy Lutomirski, Peter Zijlstra
Direct mapped pages were freed when they were offlined, or they were not allocated. So we only need to free vmemmap pages, no need to free direct mapped pages. Signed-off-by: Tang Chen <redacted> --- arch/x86/mm/init_64.c | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 368cc3f..e77d312 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c@@ -804,6 +804,10 @@ remove_pte_table(pte_t *pte_start, unsigned long addr, unsigned long end, if (IS_ALIGNED(addr, PAGE_SIZE) && IS_ALIGNED(next, PAGE_SIZE)) { + /* + * Do not free direct mapping pages since they were + * freed when offlining. + */ if (!direct) free_pagetable(pte_page(*pte), 0);
@@ -819,10 +823,11 @@ remove_pte_table(pte_t *pte_start, unsigned long addr, unsigned long end, * remove the page when it is wholly filled with 0xFD. */ memset((void *)addr, PAGE_INUSE, next - addr); - page_addr = page_address(pte_page(*pte)); + page_addr = page_address(pte_page(*pte)); if (!memchr_inv(page_addr, PAGE_INUSE, PAGE_SIZE)) { - free_pagetable(pte_page(*pte), 0); + if (!direct) + free_pagetable(pte_page(*pte), 0); spin_lock(&init_mm.page_table_lock); pte_clear(&init_mm, addr, pte);
--
1.7.1