Merge remote branch 'rmk/for-linus' into for-linus

* rmk/for-linus: (1557 commits)
  ARM: 6806/1: irq: introduce entry and exit functions for chained handlers
  ARM: 6781/1: Thumb-2: Work around buggy Thumb-2 short branch relocations in gas
  ARM: 6747/1: P2V: Thumb2 support
  ARM: 6798/1: aout-core: zero thread debug registers in a.out core dump
  ARM: 6796/1: Footbridge: Fix I/O mappings for NOMMU mode
  ARM: 6784/1: errata: no automatic Store Buffer drain on Cortex-A9
  ARM: 6772/1: errata: possible fault MMU translations following an ASID switch
  ARM: 6776/1: mach-ux500: activate fix for errata 753970
  ARM: 6794/1: SPEAr: Append UL to device address macros.
  ARM: 6793/1: SPEAr: Remove unused *_SIZE macros from spear*.h files
  ARM: 6792/1: SPEAr: Replace SIZE macro's with SZ_4K macros
  ARM: 6791/1: SPEAr3xx: Declare device structures after shirq code
  ARM: 6790/1: SPEAr: Clock Framework: Rename usbd clock and align apb_clk entry
  ARM: 6789/1: SPEAr3xx: Rename sdio to sdhci
  ARM: 6788/1: SPEAr: Include mach/hardware.h instead of mach/spear.h
  ARM: 6787/1: SPEAr: Reorder #includes in .h & .c files.
  ARM: 6681/1: SPEAr: add debugfs support to clk API
  ARM: 6703/1: SPEAr: update clk API support
  ARM: 6679/1: SPEAr: make clk API functions more generic
  ARM: 6737/1: SPEAr: formalized timer support
  ...

Conflicts:
	arch/arm/mach-msm/board-msm7x27.c
	arch/arm/mach-msm/board-msm7x30.c
	arch/arm/mach-msm/board-qsd8x50.c
	arch/arm/mach-msm/board-sapphire.c
	arch/arm/mach-msm/include/mach/memory.h

1 files changed, 41 insertions, 11 deletions

diff --git a/fs/aio.c b/fs/aio.c
index fc557a3be0a9..26869cde3953 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -239,15 +239,23 @@ static void __put_ioctx(struct kioctx *ctx)
         call_rcu(&ctx->rcu_head, ctx_rcu_free);
 }
 
-#define get_ioctx(kioctx) do {                                          \
-        BUG_ON(atomic_read(&(kioctx)->users) <= 0);                     \
-        atomic_inc(&(kioctx)->users);                                   \
-} while (0)
-#define put_ioctx(kioctx) do {                                          \
-        BUG_ON(atomic_read(&(kioctx)->users) <= 0);                     \
-        if (unlikely(atomic_dec_and_test(&(kioctx)->users)))            \
-                __put_ioctx(kioctx);                                    \
-} while (0)
+static inline void get_ioctx(struct kioctx *kioctx)
+{
+        BUG_ON(atomic_read(&kioctx->users) <= 0);
+        atomic_inc(&kioctx->users);
+}
+
+static inline int try_get_ioctx(struct kioctx *kioctx)
+{
+        return atomic_inc_not_zero(&kioctx->users);
+}
+
+static inline void put_ioctx(struct kioctx *kioctx)
+{
+        BUG_ON(atomic_read(&kioctx->users) <= 0);
+        if (unlikely(atomic_dec_and_test(&kioctx->users)))
+                __put_ioctx(kioctx);
+}
 
 /* ioctx_alloc
  *      Allocates and initializes an ioctx.  Returns an ERR_PTR if it failed.
@@ -601,8 +609,13 @@ static struct kioctx *lookup_ioctx(unsigned long ctx_id)
         rcu_read_lock();
 
         hlist_for_each_entry_rcu(ctx, n, &mm->ioctx_list, list) {
-                if (ctx->user_id == ctx_id && !ctx->dead) {
-                        get_ioctx(ctx);
+                /*
+                 * RCU protects us against accessing freed memory but
+                 * we have to be careful not to get a reference when the
+                 * reference count already dropped to 0 (ctx->dead test
+                 * is unreliable because of races).
+                 */
+                if (ctx->user_id == ctx_id && !ctx->dead && try_get_ioctx(ctx)){
                         ret = ctx;
                         break;
                 }
@@ -1629,6 +1642,23 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
                 goto out_put_req;
 
         spin_lock_irq(&ctx->ctx_lock);
+        /*
+         * We could have raced with io_destroy() and are currently holding a
+         * reference to ctx which should be destroyed. We cannot submit IO
+         * since ctx gets freed as soon as io_submit() puts its reference.  The
+         * check here is reliable: io_destroy() sets ctx->dead before waiting
+         * for outstanding IO and the barrier between these two is realized by
+         * unlock of mm->ioctx_lock and lock of ctx->ctx_lock.  Analogously we
+         * increment ctx->reqs_active before checking for ctx->dead and the
+         * barrier is realized by unlock and lock of ctx->ctx_lock. Thus if we
+         * don't see ctx->dead set here, io_destroy() waits for our IO to
+         * finish.
+         */
+        if (ctx->dead) {
+                spin_unlock_irq(&ctx->ctx_lock);
+                ret = -EINVAL;
+                goto out_put_req;
+        }
         aio_run_iocb(req);
         if (!list_empty(&ctx->run_list)) {
                 /* drain the run list */