soc: ti: wkup_m3_ipc: Fix race condition with rproc_boot
authorDave Gerlach <d-gerlach@ti.com>
Thu, 12 Dec 2019 04:03:14 +0000 (22:03 -0600)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 5 Feb 2020 14:43:41 +0000 (14:43 +0000)
[ Upstream commit 03729cfa0d543bc996bf959e762ec999afc8f3d2 ]

Any user of wkup_m3_ipc calls wkup_m3_ipc_get to get a handle and this
checks the value of the static variable m3_ipc_state to see if the
wkup_m3 is ready. Currently this is populated during probe before
rproc_boot has been called, meaning there is a window of time that
wkup_m3_ipc_get can return a valid handle but the wkup_m3 itself is not
ready, leading to invalid IPC calls to the wkup_m3 and system
instability.

To avoid this, move the population of the m3_ipc_state variable until
after rproc_boot has succeeded to guarantee a valid and usable handle
is always returned.

Reported-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Dave Gerlach <d-gerlach@ti.com>
Acked-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/soc/ti/wkup_m3_ipc.c

index f5cb8c0..c1fda6a 100644 (file)
@@ -426,6 +426,8 @@ static void wkup_m3_rproc_boot_thread(struct wkup_m3_ipc *m3_ipc)
        ret = rproc_boot(m3_ipc->rproc);
        if (ret)
                dev_err(dev, "rproc_boot failed\n");
+       else
+               m3_ipc_state = m3_ipc;
 
        do_exit(0);
 }
@@ -512,8 +514,6 @@ static int wkup_m3_ipc_probe(struct platform_device *pdev)
                goto err_put_rproc;
        }
 
-       m3_ipc_state = m3_ipc;
-
        return 0;
 
 err_put_rproc: