gpu: nvgpu: fix clk_arb completion file private data access race

clk_arb completion file descriptor can get closed immediately after
poll finishes in the work item gp10b_clk_arb_run_arbiter_cb. In
that case, the refcount for nvgpu_clk_dev can become zero in
the work item and can lead to invalid access while removing
nvgpu_clk_dev from the lists.

Remove nvgpu_clk_dev from the list before dropping the reference to
it.

Also, delete the nvgpu_clk_dev in completion file release handler
within the session and requests spinlocks to avoid race with
gp10b_clk_arb_run_arbiter_cb using it.

bug 200757277

Change-Id: I054eee547f2a6fa633d7ef55df216ec36647a826
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2569522
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 13 insertions, 4 deletions

diff --git a/drivers/gpu/nvgpu/os/linux/ioctl_clk_arb.c b/drivers/gpu/nvgpu/os/linux/ioctl_clk_arb.c
index 477222dc..9f321021 100644
--- a/drivers/gpu/nvgpu/os/linux/ioctl_clk_arb.c
+++ b/drivers/gpu/nvgpu/os/linux/ioctl_clk_arb.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2018, NVIDIA CORPORATION.  All rights reserved.
+ * Copyright (c) 2016-2021, NVIDIA CORPORATION.  All rights reserved.
  *
  * This software is licensed under the terms of the GNU General Public
  * License version 2, as published by the Free Software Foundation, and
@@ -51,19 +51,28 @@ static int nvgpu_clk_arb_release_completion_dev(struct inode *inode,
 {
         struct nvgpu_clk_dev *dev = filp->private_data;
         struct nvgpu_clk_session *session = dev->session;
+        struct gk20a *g = session->g;
+        struct nvgpu_clk_arb *arb = g->clk_arb;
 
+        clk_arb_dbg(g, " ");
 
-        clk_arb_dbg(session->g, " ");
+        nvgpu_spinlock_acquire(&session->session_lock);
+        nvgpu_spinlock_acquire(&arb->requests_lock);
+
+        nvgpu_list_del(&dev->node);
+
+        nvgpu_spinlock_release(&arb->requests_lock);
+        nvgpu_spinlock_release(&session->session_lock);
 
         /* This is done to account for the extra refcount taken in
          * nvgpu_clk_arb_commit_request_fd without events support in iGPU
          */
-        if (!session->g->clk_arb->clk_arb_events_supported) {
+        if (!arb->clk_arb_events_supported) {
                 nvgpu_ref_put(&dev->refcount, nvgpu_clk_arb_free_fd);
         }
 
-        nvgpu_ref_put(&session->refcount, nvgpu_clk_arb_free_session);
         nvgpu_ref_put(&dev->refcount, nvgpu_clk_arb_free_fd);
+        nvgpu_ref_put(&session->refcount, nvgpu_clk_arb_free_session);
         return 0;
 }