[SRU][Trusty][Xenial][PATCH 2/2] UBUNTU: SAUCE: mm: Respect FOLL_FORCE/FOLL_COW for thp
Joseph Salisbury
joseph.salisbury at canonical.com
Fri Jan 27 22:23:16 UTC 2017
On 01/27/2017 05:08 PM, Joseph Salisbury wrote:
> From: Keno Fischer <keno at juliacomputing.com>
>
> BugLink: http://bugs.launchpad.net/bugs/1658270
>
> In 19be0eaff ("mm: remove gup_flags FOLL_WRITE games from __get_user_pages()"),
> the mm code was changed from unsetting FOLL_WRITE after a COW was resolved to
> setting the (newly introduced) FOLL_COW instead. Simultaneously, the check in
> gup.c was updated to still allow writes with FOLL_FORCE set if FOLL_COW had
> also been set. However, a similar check in huge_memory.c was forgotten. As a
> result, remote memory writes to ro regions of memory backed by transparent huge
> pages cause an infinite loop in the kernel (handle_mm_fault sets FOLL_COW and
> returns 0 causing a retry, but follow_trans_huge_pmd bails out immidiately
> because `(flags & FOLL_WRITE) && !pmd_write(*pmd)` is true. While in this
> state the process is stil SIGKILLable, but little else works (e.g. no ptrace
> attach, no other signals). This is easily reproduced with the following
> code (assuming thp are set to always):
>
> #include <assert.h>
> #include <fcntl.h>
> #include <stdint.h>
> #include <stdio.h>
> #include <string.h>
> #include <sys/mman.h>
> #include <sys/stat.h>
> #include <sys/types.h>
> #include <sys/wait.h>
> #include <unistd.h>
>
> #define TEST_SIZE 5 * 1024 * 1024
>
> int main(void) {
> int status;
> pid_t child;
> int fd = open("/proc/self/mem", O_RDWR);
> void *addr = mmap(NULL, TEST_SIZE, PROT_READ,
> MAP_ANONYMOUS | MAP_PRIVATE, 0, 0);
> assert(addr != MAP_FAILED);
> pid_t parent_pid = getpid();
> if ((child = fork()) == 0) {
> void *addr2 = mmap(NULL, TEST_SIZE, PROT_READ | PROT_WRITE,
> MAP_ANONYMOUS | MAP_PRIVATE, 0, 0);
> assert(addr2 != MAP_FAILED);
> memset(addr2, 'a', TEST_SIZE);
> pwrite(fd, addr2, TEST_SIZE, (uintptr_t)addr);
> return 0;
> }
> assert(child == waitpid(child, &status, 0));
> assert(WIFEXITED(status) && WEXITSTATUS(status) == 0);
> return 0;
> }
>
> Fix this by updating follow_trans_huge_pmd in huge_memory.c analogously to
> the update in gup.c in the original commit. The same pattern exists in
> follow_devmap_pmd. However, we should not be able to reach that check
> with FOLL_COW set, so add WARN_ONCE to make sure we notice if we ever
> do.
>
> Signed-off-by: Keno Fischer <keno at juliacomputing.com>
> Acked-by: Kirill A. Shutemov <kirill.shutemov at linux.intel.com>
> Acked-by: Michal Hocko <mhocko at suse.com>
> Tested-by: David Rientjes <rientjes at google.com>
> Signed-off-by: Joseph Salisbury <joseph.salisbury at canonical.com>
> ---
> mm/huge_memory.c | 12 +++++++++++-
> 1 file changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 65aa131..4333213 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -1230,6 +1230,16 @@ out_unlock:
> return ret;
> }
>
> +/*
> + * FOLL_FORCE can write to even unwritable pmd's, but only
> + * after we've gone through a COW cycle and they are dirty.
> + */
> +static inline bool can_follow_write_pmd(pmd_t pmd, unsigned int flags)
> +{
> + return pmd_write(pmd) ||
> + ((flags & FOLL_FORCE) && (flags & FOLL_COW) && pmd_dirty(pmd));
> +}
> +
> struct page *follow_trans_huge_pmd(struct vm_area_struct *vma,
> unsigned long addr,
> pmd_t *pmd,
> @@ -1240,7 +1250,7 @@ struct page *follow_trans_huge_pmd(struct vm_area_struct *vma,
>
> assert_spin_locked(pmd_lockptr(mm, pmd));
>
> - if (flags & FOLL_WRITE && !pmd_write(*pmd))
> + if (flags & FOLL_WRITE && !can_follow_write_pmd(*pmd, flags))
> goto out;
>
> /* Avoid dumping huge zero page */
It looks like this patch will also be needed in Yakkety and Zesty, since
both will also have the mainline commit that introduced the bug:
19be0eaffa3a mm: remove gup_flags FOLL_WRITE games from __get_user_pages()
More information about the kernel-team
mailing list