The patch titled rmap: annotate lock context change on page_[un]lock_anon_vma() has been added to the -mm tree. Its filename is rmap-annotate-lock-context-change-on-page_lock_anon_vma.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: rmap: annotate lock context change on page_[un]lock_anon_vma() From: Namhyung Kim <namhyung@xxxxxxxxx> The page_lock_anon_vma() conditionally grabs RCU and anon_vma lock but page_unlock_anon_vma() releases them unconditionally. This leads sparse to complain about context imbalance. Annotate them. Signed-off-by: Namhyung Kim <namhyung@xxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- include/linux/rmap.h | 15 ++++++++++++++- mm/rmap.c | 4 +++- 2 files changed, 17 insertions(+), 2 deletions(-) diff -puN include/linux/rmap.h~rmap-annotate-lock-context-change-on-page_lock_anon_vma include/linux/rmap.h --- a/include/linux/rmap.h~rmap-annotate-lock-context-change-on-page_lock_anon_vma +++ a/include/linux/rmap.h @@ -230,7 +230,20 @@ int try_to_munlock(struct page *); /* * Called by memory-failure.c to kill processes. */ -struct anon_vma *page_lock_anon_vma(struct page *page); +struct anon_vma *__page_lock_anon_vma(struct page *page); + +static inline struct anon_vma *page_lock_anon_vma(struct page *page) +{ + struct anon_vma *anon_vma; + + __cond_lock(RCU, anon_vma = __page_lock_anon_vma(page)); + + /* (void) is needed to make gcc happy */ + (void) __cond_lock(&anon_vma->root->lock, anon_vma); + + return anon_vma; +} + void page_unlock_anon_vma(struct anon_vma *anon_vma); int page_mapped_in_vma(struct page *page, struct vm_area_struct *vma); diff -puN mm/rmap.c~rmap-annotate-lock-context-change-on-page_lock_anon_vma mm/rmap.c --- a/mm/rmap.c~rmap-annotate-lock-context-change-on-page_lock_anon_vma +++ a/mm/rmap.c @@ -314,7 +314,7 @@ void __init anon_vma_init(void) * Getting a lock on a stable anon_vma from a page off the LRU is * tricky: page_lock_anon_vma rely on RCU to guard against the races. */ -struct anon_vma *page_lock_anon_vma(struct page *page) +struct anon_vma *__page_lock_anon_vma(struct page *page) { struct anon_vma *anon_vma, *root_anon_vma; unsigned long anon_mapping; @@ -348,6 +348,8 @@ out: } void page_unlock_anon_vma(struct anon_vma *anon_vma) + __releases(&anon_vma->root->lock) + __releases(RCU) { anon_vma_unlock(anon_vma); rcu_read_unlock(); _ Patches currently in -mm which might be from namhyung@xxxxxxxxx are origin.patch linux-next.patch vfs-remove-a-warning-on-open_fmode.patch vfs-add-__fmode_exec.patch mm-cleanup-gfp_zone-fix-sparse-warnings.patch mm-remove-temporary-variable-on-generic_file_direct_write.patch mm-add-casts-to-from-gfp_t-in-gfp_to_alloc_flags.patch mm-wrap-get_locked_pte-using-__cond_lock.patch mm-add-lock-release-annotation-on-do_wp_page.patch mm-wrap-follow_pte-using-__cond_lock.patch rmap-annotate-lock-context-change-on-page_lock_anon_vma.patch rmap-wrap-page_check_address-using-__cond_lock.patch rmap-make-anon_vma_free-static.patch vmalloc-rename-temporary-variable-in-__insert_vmap_area.patch vmalloc-annotate-lock-context-change-on-s_start-stop.patch mm-declare-some-external-symbols.patch vmstat-include-compactionh-when-config_compaction.patch vmstat-include-compactionh-when-config_compaction-fix.patch init-mark-__user-address-space-on-string-literals.patch kernel-userc-add-lock-release-annotation-on-free_user.patch printk-fixup-declaration-of-kmsg_reasons.patch printk-add-lock-context-annotation.patch printk-change-type-of-boot_delay-to-int.patch printk-declare-printk_ratelimit_state-in-ratelimith.patch printk-declare-printk_ratelimit_state-in-ratelimith-fix.patch ptrace-annotate-lock-context-change-on-exit_ptrace.patch ptrace-change-signature-of-sys_ptrace-and-friends.patch ptrace-cleanup-ptrace_request.patch ptrace-change-signature-of-arch_ptrace.patch ptrace-cleanup-arch_ptrace-on-x86.patch ptrace-cleanup-arch_ptrace-on-arm.patch ptrace-cleanup-arch_ptrace-on-avr32.patch ptrace-cleanup-arch_ptrace-and-friends-on-blackfin.patch ptrace-cleanup-arch_ptrace-on-cris.patch ptrace-cleanup-arch_ptrace-on-frv.patch ptrace-cleanup-arch_ptrace-on-h8300.patch ptrace-cleanup-arch_ptrace-on-m32r.patch ptrace-cleanup-arch_ptrace-on-m68k.patch ptrace-cleanup-arch_ptrace-on-m68knommu.patch ptrace-cleanup-arch_ptrace-on-microblaze.patch ptrace-cleanup-arch_ptrace-on-mips.patch ptrace-cleanup-arch_ptrace-on-mn10300.patch ptrace-cleanup-arch_ptrace-on-parisc.patch ptrace-cleanup-arch_ptrace-on-powerpc.patch ptrace-cleanup-arch_ptrace-on-score.patch ptrace-cleanup-arch_ptrace-on-sh.patch ptrace-cleanup-arch_ptrace-on-sparc.patch ptrace-cleanup-arch_ptrace-on-tile.patch ptrace-cleanup-arch_ptrace-on-um.patch ptrace-cleanup-arch_ptrace-on-xtensa.patch signals-annotate-lock_task_sighand.patch signals-annotate-lock-context-change-on-ptrace_stop.patch exit-add-lock-context-annotation-on-find_new_reaper.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html