firmware: xilinx: Fix an error handling path in 'zynqmp_firmware_probe()'

If 'mfd_add_devices()' fails, we must undo 'zynqmp_pm_api_debugfs_init()'
otherwise some debugfs directory and files will be left.

Just move the call to 'zynqmp_pm_api_debugfs_init()' a few lines below to
fix the issue.

Fixes: e23d9c6d ("drivers: soc: xilinx: Add ZynqMP power domain driver")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Jolly Shah <jolly.shah@xilinx.com>
Link: https://lore.kernel.org/r/20200510130357.233364-1-christophe.jaillet@wanadoo.frSigned-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/firmware/xilinx/zynqmp.c

@@ -1235,8 +1235,6 @@ static int zynqmp_firmware_probe(struct platform_device *pdev)
 	pr_info("%s Trustzone version v%d.%d\n", __func__,
 		pm_tz_version >> 16, pm_tz_version & 0xFFFF);
 
-	zynqmp_pm_api_debugfs_init();
-
 	ret = mfd_add_devices(&pdev->dev, PLATFORM_DEVID_NONE, firmware_devs,
 			      ARRAY_SIZE(firmware_devs), NULL, 0, NULL);
 	if (ret) {
@@ -1244,6 +1242,8 @@ static int zynqmp_firmware_probe(struct platform_device *pdev)
 		return ret;
 	}
 
+	zynqmp_pm_api_debugfs_init();
+
 	return of_platform_populate(dev->of_node, NULL, NULL, dev);
 }