soc: ti: reset irq affinity before freeing irq

When using accumulator queue for rx side for network driver, following
warning is seen when doing a reboot command from Linux console. This
is because, affinity value is not reset before calling free_irq(). This
patch fixes this.

Deconfiguring network interfaces...
 ------------[ cut here ]-----------
 WARNING: CPU: 0 PID: 2081 at kernel/irq/manage.c:1370
__free_irq+0x208/0x214
 Modules linked in:
 CPU: 0 PID: 2081 Comm: ifconfig Not tainted 4.1.5-00908-g1049e206-dirty #1
 Hardware name: Keystone
 Backtrace:
 [<c0012b98>] (dump_backtrace) from [<c0012dbc>] (show_stack+0x18/0x1c)
 r7:c005d0a8 r6:c06e2184 r5:c06e2184 r4:00000000
 [<c0012da4>] (show_stack) from [<c04e7168>] (dump_stack+0x8c/0xcc)
 [<c04e70dc>] (dump_stack) from [<c0025944>] (warn_slowpath_common+0x88/0xb8)
 r7:c005d0a8 r6:0000055a r5:00000009 r4:0000000
 [<c00258bc>] (warn_slowpath_common) from [<c0025a18>] (warn_slowpath_null+0x
24/0x2c)
 r8:0000006c r7:ee513f60 r6:ee513f00 r5:ee611010 r4:cc873a00
 (warn_slowpath_null) from [<c005d0a8>] (__free_irq+0x208/0x214)
 [<c005cea0>] (__free_irq) from [<c005d158>] (free_irq+0x54/0xac)
 r10:00000002 r9:00000000 r8:00000000 r7:ee611010
r6:0000006c r5:00000000
  r4:ee513f00 r3:00000000
  [<c005d104>] (free_irq) from [<c02a81b0>] (knav_range_setup_acc_irq+0xb0/0x1
28)
  r7:00000001 r6:0000006c r5:ee611010 r4:00000001
  [<c02a8100>] (knav_range_setup_acc_irq) from [<c02a8248>] (knav_acc_close_qu
eue+0x20/0x24)
  r8:edd1a4c8 r7:00001000 r6:eed89980 r5:ee616650 r4:edf9d990
  [<c02a8228>] (knav_acc_close_queue) from [<c02a6160>] (knav_queue_close+0xb4
/0xb8)
  [<c02a60ac>] (knav_queue_close) from [<c0336270>] (netcp_free_navigator_reso
urces+0x1d4/0x2c0)
 r5:edd1a480 r4:00000400
 [<c033609c>] (netcp_free_navigator_resources) from
 [<c033657c>] (netcp_ndo_stop+0x220/0x230)
 r10:00008914 r9:edf34400 r8:00000000 r7:edd1a5d8 r6:edd1a480 r5:00000400
 r4:edd1a000
 [<c033635c>] (netcp_ndo_stop) from [<c03d9c80> (__dev_close_many+0x90/0xd8)
 r7:00001003 r6:00001042 r5:edb0de20 r4:edd1a000
 [<c03d9bf0>] (__dev_close_many) from [<c03d9df4>] (__dev_close+0x30/0x48)
  r5:00000001 r4:edd1a000
 [<c03d9dc4>] (__dev_close) from [<c03e175c>] (__dev_change_flags+0x9c/0x14c)
 [<c03e16c0>] (__dev_change_flags) from [<c03e182c>] (dev_change_flags+0x20/0

  r9:edf34400 r8:00000000 r7:00000000 r6:00001003 r5:edd1a138 r4:edd1a000
 [<c03e180c>] (dev_change_flags) from [<c0446f94>] (devinet_ioctl+0x680/0x754

  r9:edf34400 r8:bef6af3c r7:00000000 r6:bef6ac64 r5:edf3440c r4:00000000
 [<c0446914>] (devinet_ioctl) from [<c0448760>] (inet_ioctl+0x1a8/0x1d0)
  r10:00000000 r9:edb0c000 r8:bef6ac64 r7:00000003 r6:bef6ac64 r5:bef6ac64
  r4:00008914
 [<c04485b8>] (inet_ioctl) from [<c03c72fc>] (sock_ioctl+0x1d0/0x2a8)
 [<c03c712c>] (sock_ioctl) from [<c00eb608>] (do_vfs_ioctl+0x414/0x604)
  r7:00000003 r6:ed8b60c0 r5:bef6ac64 r4:ccc88f20
 [<c00eb1f4>] (do_vfs_ioctl) from [<c00eb834>] (SyS_ioctl+0x3c/0x64)
  r9:edb0c000 r8:bef6ac64 r7:00008914 r6:ed8b60c0 r5:00000003 r4:ed8b60c0
 [<c00eb7f8>] (SyS_ioctl) from [<c000f780>] (ret_fast_syscall+0x0/0x3c)
  r9:edb0c000 r8:c000f924 r7:00000036 r6:0007e77c r5:bef6ac64 r4:0007e7d0
 ---[ end trace f565594c905af0b4 ]---

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

1 files changed, 4 insertions, 0 deletions

diff --git a/drivers/soc/ti/knav_qmss_acc.c b/drivers/soc/ti/knav_qmss_acc.c
index ef6f69db0bd0..b98fe56598dd 100644
--- a/drivers/soc/ti/knav_qmss_acc.c
+++ b/drivers/soc/ti/knav_qmss_acc.c
@@ -261,6 +261,10 @@ static int knav_range_setup_acc_irq(struct knav_range_info *range,
 	if (old && !new) {
 		dev_dbg(kdev->dev, "setup-acc-irq: freeing %s for channel %s\n",
 			acc->name, acc->name);
+		ret = irq_set_affinity_hint(irq, NULL);
+		if (ret)
+			dev_warn(range->kdev->dev,
+				 "Failed to set IRQ affinity\n");
 		free_irq(irq, range);
 	}