Merge tag 'io_uring-5.11-2021-01-01' of git://git.kernel.dk/linux-block

Pull io_uring fixes from Jens Axboe:
 "A few fixes that should go into 5.11, all marked for stable as well:

   - Fix issue around identity COW'ing and users that share a ring
     across processes

   - Fix a hang associated with unregistering fixed files (Pavel)

   - Move the 'process is exiting' cancelation a bit earlier, so
     task_works aren't affected by it (Pavel)"

* tag 'io_uring-5.11-2021-01-01' of git://git.kernel.dk/linux-block:
  kernel/io_uring: cancel io_uring before task works
  io_uring: fix io_sqe_files_unregister() hangs
  io_uring: add a helper for setting a ref node
  io_uring: don't assume mm is constant across submits

diff --git a/fs/file.c b/fs/file.c
index c0b6096..dab120b 100644 (file)
--- a/fs/file.c
+++ b/fs/file.c
@@ -21,7 +21,6 @@
 #include <linux/rcupdate.h>
 #include <linux/close_range.h>
 #include <net/sock.h>
-#include <linux/io_uring.h>
 
 unsigned int sysctl_nr_open __read_mostly = 1024*1024;
 unsigned int sysctl_nr_open_min = BITS_PER_LONG;
@@ -428,7 +427,6 @@ void exit_files(struct task_struct *tsk)
        struct files_struct * files = tsk->files;
 
        if (files) {
-               io_uring_files_cancel(files);
                task_lock(tsk);
                tsk->files = NULL;
                task_unlock(tsk);

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 7e35283..ca46f31 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -992,6 +992,10 @@ enum io_mem_account {
        ACCT_PINNED,
 };
 
+static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node);
+static struct fixed_file_ref_node *alloc_fixed_file_ref_node(
+                       struct io_ring_ctx *ctx);
+
 static void __io_complete_rw(struct io_kiocb *req, long res, long res2,
                             struct io_comp_state *cs);
 static void io_cqring_fill_event(struct io_kiocb *req, long res);
@@ -1501,6 +1505,13 @@ static bool io_grab_identity(struct io_kiocb *req)
                spin_unlock_irq(&ctx->inflight_lock);
                req->work.flags |= IO_WQ_WORK_FILES;
        }
+       if (!(req->work.flags & IO_WQ_WORK_MM) &&
+           (def->work_flags & IO_WQ_WORK_MM)) {
+               if (id->mm != current->mm)
+                       return false;
+               mmgrab(id->mm);
+               req->work.flags |= IO_WQ_WORK_MM;
+       }
 
        return true;
 }
@@ -1525,13 +1536,6 @@ static void io_prep_async_work(struct io_kiocb *req)
                        req->work.flags |= IO_WQ_WORK_UNBOUND;
        }
 
-       /* ->mm can never change on us */
-       if (!(req->work.flags & IO_WQ_WORK_MM) &&
-           (def->work_flags & IO_WQ_WORK_MM)) {
-               mmgrab(id->mm);
-               req->work.flags |= IO_WQ_WORK_MM;
-       }
-
        /* if we fail grabbing identity, we must COW, regrab, and retry */
        if (io_grab_identity(req))
                return;
@@ -7231,14 +7235,28 @@ static void io_file_ref_kill(struct percpu_ref *ref)
        complete(&data->done);
 }
 
+static void io_sqe_files_set_node(struct fixed_file_data *file_data,
+                                 struct fixed_file_ref_node *ref_node)
+{
+       spin_lock_bh(&file_data->lock);
+       file_data->node = ref_node;
+       list_add_tail(&ref_node->node, &file_data->ref_list);
+       spin_unlock_bh(&file_data->lock);
+       percpu_ref_get(&file_data->refs);
+}
+
 static int io_sqe_files_unregister(struct io_ring_ctx *ctx)
 {
        struct fixed_file_data *data = ctx->file_data;
-       struct fixed_file_ref_node *ref_node = NULL;
+       struct fixed_file_ref_node *backup_node, *ref_node = NULL;
        unsigned nr_tables, i;
+       int ret;
 
        if (!data)
                return -ENXIO;
+       backup_node = alloc_fixed_file_ref_node(ctx);
+       if (!backup_node)
+               return -ENOMEM;
 
        spin_lock_bh(&data->lock);
        ref_node = data->node;
@@ -7250,7 +7268,18 @@ static int io_sqe_files_unregister(struct io_ring_ctx *ctx)
 
        /* wait for all refs nodes to complete */
        flush_delayed_work(&ctx->file_put_work);
-       wait_for_completion(&data->done);
+       do {
+               ret = wait_for_completion_interruptible(&data->done);
+               if (!ret)
+                       break;
+               ret = io_run_task_work_sig();
+               if (ret < 0) {
+                       percpu_ref_resurrect(&data->refs);
+                       reinit_completion(&data->done);
+                       io_sqe_files_set_node(data, backup_node);
+                       return ret;
+               }
+       } while (1);
 
        __io_sqe_files_unregister(ctx);
        nr_tables = DIV_ROUND_UP(ctx->nr_user_files, IORING_MAX_FILES_TABLE);
@@ -7261,6 +7290,7 @@ static int io_sqe_files_unregister(struct io_ring_ctx *ctx)
        kfree(data);
        ctx->file_data = NULL;
        ctx->nr_user_files = 0;
+       destroy_fixed_file_ref_node(backup_node);
        return 0;
 }
 
@@ -7758,11 +7788,7 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
                return PTR_ERR(ref_node);
        }
 
-       file_data->node = ref_node;
-       spin_lock_bh(&file_data->lock);
-       list_add_tail(&ref_node->node, &file_data->ref_list);
-       spin_unlock_bh(&file_data->lock);
-       percpu_ref_get(&file_data->refs);
+       io_sqe_files_set_node(file_data, ref_node);
        return ret;
 out_fput:
        for (i = 0; i < ctx->nr_user_files; i++) {
@@ -7918,11 +7944,7 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx,
 
        if (needs_switch) {
                percpu_ref_kill(&data->node->refs);
-               spin_lock_bh(&data->lock);
-               list_add_tail(&ref_node->node, &data->ref_list);
-               data->node = ref_node;
-               spin_unlock_bh(&data->lock);
-               percpu_ref_get(&ctx->file_data->refs);
+               io_sqe_files_set_node(data, ref_node);
        } else
                destroy_fixed_file_ref_node(ref_node);
 

diff --git a/kernel/exit.c b/kernel/exit.c
index 3594291..04029e3 100644 (file)
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -63,6 +63,7 @@
 #include <linux/random.h>
 #include <linux/rcuwait.h>
 #include <linux/compat.h>
+#include <linux/io_uring.h>
 
 #include <linux/uaccess.h>
 #include <asm/unistd.h>
@@ -776,6 +777,7 @@ void __noreturn do_exit(long code)
                schedule();
        }
 
+       io_uring_files_cancel(tsk->files);
        exit_signals(tsk);  /* sets PF_EXITING */
 
        /* sync mm's RSS info before statistics gathering */