From: Matthew Wilcox (Oracle) willy@infradead.org
[ Upstream commit e320d3012d25b1fb5f3df4edb7bd44a1c362ec10 ]
Here is a very rare race which leaks memory:
Page P0 is allocated to the page cache. Page P1 is free.
Thread A Thread B Thread C find_get_entry(): xas_load() returns P0 Removes P0 from page cache P0 finds its buddy P1 alloc_pages(GFP_KERNEL, 1) returns P0 P0 has refcount 1 page_cache_get_speculative(P0) P0 has refcount 2 __free_pages(P0) P0 has refcount 1 put_page(P0) P1 is not freed
Fix this by freeing all the pages in __free_pages() that won't be freed by the call to put_page(). It's usually not a good idea to split a page, but this is a very unlikely scenario.
Fixes: e286781d5f2e ("mm: speculative page references") Signed-off-by: Matthew Wilcox (Oracle) willy@infradead.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Acked-by: Mike Rapoport rppt@linux.ibm.com Cc: Nick Piggin npiggin@gmail.com Cc: Hugh Dickins hughd@google.com Cc: Peter Zijlstra peterz@infradead.org Link: https://lkml.kernel.org/r/20200926213919.26642-1-willy@infradead.org Signed-off-by: Linus Torvalds torvalds@linux-foundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- lib/Kconfig.debug | 9 +++++++++ lib/Makefile | 1 + lib/test_free_pages.c | 42 ++++++++++++++++++++++++++++++++++++++++++ mm/page_alloc.c | 3 +++ 4 files changed, 55 insertions(+) create mode 100644 lib/test_free_pages.c
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 0c781f912f9f0..491789a793ae5 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -2367,6 +2367,15 @@ config TEST_HMM
If unsure, say N.
+config TEST_FREE_PAGES + tristate "Test freeing pages" + help + Test that a memory leak does not occur due to a race between + freeing a block of pages and a speculative page reference. + Loading this module is safe if your kernel has the bug fixed. + If the bug is not fixed, it will leak gigabytes of memory and + probably OOM your system. + config TEST_FPU tristate "Test floating point operations in kernel space" depends on X86 && !KCOV_INSTRUMENT_ALL diff --git a/lib/Makefile b/lib/Makefile index a4a4c6864f518..071b687b7363f 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -99,6 +99,7 @@ obj-$(CONFIG_TEST_BLACKHOLE_DEV) += test_blackhole_dev.o obj-$(CONFIG_TEST_MEMINIT) += test_meminit.o obj-$(CONFIG_TEST_LOCKUP) += test_lockup.o obj-$(CONFIG_TEST_HMM) += test_hmm.o +obj-$(CONFIG_TEST_FREE_PAGES) += test_free_pages.o
# # CFLAGS for compiling floating point code inside the kernel. x86/Makefile turns diff --git a/lib/test_free_pages.c b/lib/test_free_pages.c new file mode 100644 index 0000000000000..074e76bd76b2b --- /dev/null +++ b/lib/test_free_pages.c @@ -0,0 +1,42 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * test_free_pages.c: Check that free_pages() doesn't leak memory + * Copyright (c) 2020 Oracle + * Author: Matthew Wilcox willy@infradead.org + */ + +#include <linux/gfp.h> +#include <linux/mm.h> +#include <linux/module.h> + +static void test_free_pages(gfp_t gfp) +{ + unsigned int i; + + for (i = 0; i < 1000 * 1000; i++) { + unsigned long addr = __get_free_pages(gfp, 3); + struct page *page = virt_to_page(addr); + + /* Simulate page cache getting a speculative reference */ + get_page(page); + free_pages(addr, 3); + put_page(page); + } +} + +static int m_in(void) +{ + test_free_pages(GFP_KERNEL); + test_free_pages(GFP_KERNEL | __GFP_COMP); + + return 0; +} + +static void m_ex(void) +{ +} + +module_init(m_in); +module_exit(m_ex); +MODULE_AUTHOR("Matthew Wilcox willy@infradead.org"); +MODULE_LICENSE("GPL"); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c449d74f55842..d99c2db7f254f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4961,6 +4961,9 @@ void __free_pages(struct page *page, unsigned int order) { if (put_page_testzero(page)) free_the_page(page, order); + else if (!PageHead(page)) + while (order-- > 0) + free_the_page(page + (1 << order), order); } EXPORT_SYMBOL(__free_pages);