Ensure find_vma() callers do so with the mmap_sem held. I'm sure there are a few more places left to fix, but this is a pretty good start. Following the call chain, some users become all tangled up, but I believe these fixes are correct. Furthermore, the bulk of the callers of find_vma are in a lot of functions where it is well known that the mmap_sem is taken way before, such as get_unmapped_area() family. Please note that none of the patches are tested. Thanks! blackfin/ptrace: call find_vma with the mmap_sem held m68k: call find_vma with the mmap_sem held in sys_cacheflush() mips: call find_vma with the mmap_sem held arc: call find_vma with the mmap_sem held drivers/misc/sgi-gru/grufault.c: call find_vma with the mmap_sem held drm/exynos: call find_vma with the mmap_sem held arch/arc/kernel/troubleshoot.c | 7 ++++--- arch/blackfin/kernel/ptrace.c | 8 ++++++-- arch/m68k/kernel/sys_m68k.c | 18 ++++++++++++------ arch/mips/kernel/traps.c | 2 ++ arch/mips/mm/c-octeon.c | 2 ++ drivers/gpu/drm/exynos/exynos_drm_g2d.c | 6 ++++++ drivers/misc/sgi-gru/grufault.c | 13 +++++++++---- 7 files changed, 41 insertions(+), 15 deletions(-) -- 1.8.1.4 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>