On Thu, Nov 19, 2015 at 01:39:51AM +0300, Sergei Shtylyov wrote:
> When the R8A7795 support was added to the driver, little attention was paid
> to the ravb_open() error path: free_irq() for the EMAC interrupt was called
> uncoditionally, unlike request_irq(), and in a wrong order as well...
> As a result, on the R-Car gen2 SoCs I started getting the following in case
> of a device opening error:
>
> WARNING: CPU: 0 PID: 1 at kernel/irq/manage.c:1448 __free_irq+0x8c/0x228()
> Trying to free already-free IRQ 0
> Modules linked in:
> CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.4.0-rc1-dirty #1005
> Hardware name: Generic R8A7791 (Flattened Device Tree)
> Backtrace:
> [<c0013818>] (dump_backtrace) from [<c00139b4>] (show_stack+0x18/0x1c)
> r6:c063cdd6 r5:00000009 r4:00000000 r3:00204140
> [<c001399c>] (show_stack) from [<c022a578>] (dump_stack+0x74/0x90)
> [<c022a504>] (dump_stack) from [<c0025f04>] (warn_slowpath_common+0x8c/0xb8)
> r4:ef04fd38 r3:c0714770
> [<c0025e78>] (warn_slowpath_common) from [<c0025fd4>]
> (warn_slowpath_fmt+0x38/0x40)
>
> r8:ee8ad800 r7:ef0030a0 r6:00000000 r5:00000000 r4:ef003040
> [<c0025fa0>] (warn_slowpath_fmt) from [<c0064cc0>] (__free_irq+0x8c/0x228)
> r3:00000000 r2:c063ce9f
> [<c0064c34>] (__free_irq) from [<c0064ecc>] (free_irq+0x70/0xa4)
> r10:0000016b r8:00000000 r7:00000000 r6:ee8ad800 r5:00000000 r4:ef003040
> [<c0064e5c>] (free_irq) from [<c033472c>] (ravb_open+0x224/0x274)
> r7:fffffffe r6:00000000 r5:fffffffe r4:ee8ad800
> [<c0334508>] (ravb_open) from [<c041ac78>] (__dev_open+0x84/0x104)
> r7:ee8ad830 r6:c0566334 r5:00000000 r4:ee8ad800
> [<c041abf4>] (__dev_open) from [<c041af08>] (__dev_change_flags+0x94/0x13c)
> r7:00001002 r6:00000001 r5:00001003 r4:ee8ad800
> [<c041ae74>] (__dev_change_flags) from [<c041afe8>]
> (dev_change_flags+0x20/0x50)
> r7:c072e6e0 r6:00000138 r5:00001002 r4:ee8ad800
> [<c041afc8>] (dev_change_flags) from [<c06ec06c>] (ip_auto_config+0x174/0xfb8)
> r8:00001002 r7:c072e6e0 r6:c0703344 r5:00000001 r4:ee8ad800 r3:00000101
> [<c06ebef8>] (ip_auto_config) from [<c000a810>] (do_one_initcall+0x100/0x1cc)
> r10:c06fb83c r9:00000000 r8:c06ebef8 r7:c0736000 r6:c0710918 r5:c0710918
> r4:ef2f8f80
> [<c000a710>] (do_one_initcall) from [<c06ccddc>]
> (kernel_init_freeable+0x11c/0x1
> ec)
> r10:c06fb83c r9:00000000 r8:0000009a r7:c0736000 r6:c0706bf0 r5:c06fb834
> r4:00000007
> [<c06cccc0>] (kernel_init_freeable) from [<c0514c54>] (kernel_init+0x14/0xec)
> r10:00000000 r8:00000000 r7:00000000 r6:00000000 r5:c0514c40 r4:c0736000
> [<c0514c40>] (kernel_init) from [<c0010458>] (ret_from_fork+0x14/0x3c)
> r4:00000000 r3:ef04e000
>
> Fix up the free_irq() call order and add a new label on the error path.
>
> Fixes: 22d4df8ff3a3 ("ravb: Add support for r8a7795 SoC")
> Signed-off-by: Sergei Shtylyov <sergei.shtyl...@cogentembedded.com>
Thanks for fixing my mistake.
Acked-by: Simon Horman <horms+rene...@verge.net.au>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
