diff mbox

Applied "spi: dw: explicitly free IRQ handler in dw_spi_remove_host()" to the spi tree

Message ID E1ZpP0f-0000MP-5w@finisterre (mailing list archive)
State Not Applicable
Commit 9f89566dac9ad776ad4d919922e99d799731e513
Headers show

Commit Message

Mark Brown Oct. 22, 2015, 11:11 p.m. UTC
The patch

   spi: dw: explicitly free IRQ handler in dw_spi_remove_host()

has been applied to the spi tree at

   git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git 

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From 9f89566dac9ad776ad4d919922e99d799731e513 Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Tue, 20 Oct 2015 11:39:36 +0300
Subject: [PATCH] spi: dw: explicitly free IRQ handler in dw_spi_remove_host()

The following warning occurs when DW SPI is compiled as a module and it's a PCI
device. On the removal stage pcibios_free_irq() is called earlier than
free_irq() due to the latter is called at managed resources free strage.

------------[ cut here ]------------
WARNING: CPU: 1 PID: 1003 at /home/andy/prj/linux/fs/proc/generic.c:575 remove_proc_entry+0x118/0x150()
remove_proc_entry: removing non-empty directory 'irq/38', leaking at least 'dw_spi1'
Modules linked in: spi_dw_midpci(-) spi_dw [last unloaded: dw_dmac_core]
CPU: 1 PID: 1003 Comm: modprobe Not tainted 4.3.0-rc5-next-20151013+ #32
 00000000 00000000 f5535d70 c12dc220 f5535db0 f5535da0 c104e912 c198a6bc
 f5535dcc 000003eb c198a638 0000023f c11b4098 c11b4098 f54f1ec8 f54f1ea0
 f642ba20 f5535db8 c104e96e 00000009 f5535db0 c198a6bc f5535dcc f5535df0
Call Trace:
 [<c12dc220>] dump_stack+0x41/0x61
 [<c104e912>] warn_slowpath_common+0x82/0xb0
 [<c11b4098>] ? remove_proc_entry+0x118/0x150
 [<c11b4098>] ? remove_proc_entry+0x118/0x150
 [<c104e96e>] warn_slowpath_fmt+0x2e/0x30
 [<c11b4098>] remove_proc_entry+0x118/0x150
 [<c109b96a>] unregister_irq_proc+0xaa/0xc0
 [<c109575e>] free_desc+0x1e/0x60
 [<c10957d2>] irq_free_descs+0x32/0x70
 [<c109b1a0>] irq_domain_free_irqs+0x120/0x150
 [<c1039e8c>] mp_unmap_irq+0x5c/0x60
 [<c16277b0>] intel_mid_pci_irq_disable+0x20/0x40
 [<c1627c7f>] pcibios_free_irq+0xf/0x20
 [<c13189f2>] pci_device_remove+0x52/0xb0
 [<c13f6367>] __device_release_driver+0x77/0x100
 [<c13f6da7>] driver_detach+0x87/0x90
 [<c13f5eaa>] bus_remove_driver+0x4a/0xc0
 [<c128bf0d>] ? selinux_capable+0xd/0x10
 [<c13f7483>] driver_unregister+0x23/0x60
 [<c10bad8a>] ? find_module_all+0x5a/0x80
 [<c1317413>] pci_unregister_driver+0x13/0x60
 [<f80ac654>] dw_spi_driver_exit+0xd/0xf [spi_dw_midpci]
 [<c10bce9a>] SyS_delete_module+0x17a/0x210

Explicitly call free_irq() at removal stage of the DW SPI driver.

Fixes: 04f421e7b0b1 (spi: dw: use managed resources)
Cc: stable@vger.kernel.org
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-dw.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Andy Shevchenko Oct. 23, 2015, 7:40 a.m. UTC | #1
On Fri, 2015-10-23 at 08:11 +0900, Mark Brown wrote:
> The patch
> 
>    spi: dw: explicitly free IRQ handler in dw_spi_remove_host()

This one seems incomplete.

Yes, we already have a fix 02f20387e1bc in your tree.

This makes free_irq() call duplicate. Please, remove it from your tree
or do the revert.


> 
> has been applied to the spi tree at
> 
>    git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git 
> 
> All being well this means that it will be integrated into the linux-
> next
> tree (usually sometime in the next 24 hours) and sent to Linus during
> the next merge window (or sooner if it is a bug fix), however if
> problems are discovered then the patch may be dropped or reverted.  
> 
> You may get further e-mails resulting from automated or manual
> testing
> and review of the tree, please engage with people reporting problems
> and
> send followup patches addressing any issues that are reported if
> needed.
> 
> If any updates are required or you are submitting further changes
> they
> should be sent as incremental updates against current git, existing
> patches will not be replaced.
> 
> Please add any relevant lists and maintainers to the CCs when
> replying
> to this mail.
> 
> Thanks,
> Mark
> 
> From 9f89566dac9ad776ad4d919922e99d799731e513 Mon Sep 17 00:00:00
> 2001
> From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> Date: Tue, 20 Oct 2015 11:39:36 +0300
> Subject: [PATCH] spi: dw: explicitly free IRQ handler in
> dw_spi_remove_host()
> 
> The following warning occurs when DW SPI is compiled as a module and
> it's a PCI
> device. On the removal stage pcibios_free_irq() is called earlier
> than
> free_irq() due to the latter is called at managed resources free
> strage.
> 
> ------------[ cut here ]------------
> WARNING: CPU: 1 PID: 1003 at
> /home/andy/prj/linux/fs/proc/generic.c:575
> remove_proc_entry+0x118/0x150()
> remove_proc_entry: removing non-empty directory 'irq/38', leaking at
> least 'dw_spi1'
> Modules linked in: spi_dw_midpci(-) spi_dw [last unloaded:
> dw_dmac_core]
> CPU: 1 PID: 1003 Comm: modprobe Not tainted 4.3.0-rc5-next-20151013+
> #32
>  00000000 00000000 f5535d70 c12dc220 f5535db0 f5535da0 c104e912
> c198a6bc
>  f5535dcc 000003eb c198a638 0000023f c11b4098 c11b4098 f54f1ec8
> f54f1ea0
>  f642ba20 f5535db8 c104e96e 00000009 f5535db0 c198a6bc f5535dcc
> f5535df0
> Call Trace:
>  [<c12dc220>] dump_stack+0x41/0x61
>  [<c104e912>] warn_slowpath_common+0x82/0xb0
>  [<c11b4098>] ? remove_proc_entry+0x118/0x150
>  [<c11b4098>] ? remove_proc_entry+0x118/0x150
>  [<c104e96e>] warn_slowpath_fmt+0x2e/0x30
>  [<c11b4098>] remove_proc_entry+0x118/0x150
>  [<c109b96a>] unregister_irq_proc+0xaa/0xc0
>  [<c109575e>] free_desc+0x1e/0x60
>  [<c10957d2>] irq_free_descs+0x32/0x70
>  [<c109b1a0>] irq_domain_free_irqs+0x120/0x150
>  [<c1039e8c>] mp_unmap_irq+0x5c/0x60
>  [<c16277b0>] intel_mid_pci_irq_disable+0x20/0x40
>  [<c1627c7f>] pcibios_free_irq+0xf/0x20
>  [<c13189f2>] pci_device_remove+0x52/0xb0
>  [<c13f6367>] __device_release_driver+0x77/0x100
>  [<c13f6da7>] driver_detach+0x87/0x90
>  [<c13f5eaa>] bus_remove_driver+0x4a/0xc0
>  [<c128bf0d>] ? selinux_capable+0xd/0x10
>  [<c13f7483>] driver_unregister+0x23/0x60
>  [<c10bad8a>] ? find_module_all+0x5a/0x80
>  [<c1317413>] pci_unregister_driver+0x13/0x60
>  [<f80ac654>] dw_spi_driver_exit+0xd/0xf [spi_dw_midpci]
>  [<c10bce9a>] SyS_delete_module+0x17a/0x210
> 
> Explicitly call free_irq() at removal stage of the DW SPI driver.
> 
> Fixes: 04f421e7b0b1 (spi: dw: use managed resources)
> Cc: stable@vger.kernel.org
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> Signed-off-by: Mark Brown <broonie@kernel.org>
> ---
>  drivers/spi/spi-dw.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/spi/spi-dw.c b/drivers/spi/spi-dw.c
> index cc2e980..ad66bc5 100644
> --- a/drivers/spi/spi-dw.c
> +++ b/drivers/spi/spi-dw.c
> @@ -542,6 +542,8 @@ void dw_spi_remove_host(struct dw_spi *dws)
>  {
>  	dw_spi_debugfs_remove(dws);
>  
> +	free_irq(dws->irq, dws->master);
> +
>  	if (dws->dma_ops && dws->dma_ops->dma_exit)
>  		dws->dma_ops->dma_exit(dws);
>
Mark Brown Oct. 23, 2015, 4:18 p.m. UTC | #2
On Fri, Oct 23, 2015 at 10:40:58AM +0300, Andy Shevchenko wrote:
> On Fri, 2015-10-23 at 08:11 +0900, Mark Brown wrote:
> > The patch

> >    spi: dw: explicitly free IRQ handler in dw_spi_remove_host()

> This one seems incomplete.

> Yes, we already have a fix 02f20387e1bc in your tree.

This *appears* to be the same commit you are following up on (please
always include human readable descriptions)

> This makes free_irq() call duplicate. Please, remove it from your tree
> or do the revert.

Are you saying that you have realised that your patch is broken since
sending it?  I've dropped this commit...
Andy Shevchenko Oct. 23, 2015, 4:25 p.m. UTC | #3
On Sat, 2015-10-24 at 01:18 +0900, Mark Brown wrote:
> On Fri, Oct 23, 2015 at 10:40:58AM +0300, Andy Shevchenko wrote:
> > On Fri, 2015-10-23 at 08:11 +0900, Mark Brown wrote:
> > > The patch
> 
> > >    spi: dw: explicitly free IRQ handler in dw_spi_remove_host()
> 
> > This one seems incomplete.
> 
> > Yes, we already have a fix 02f20387e1bc in your tree.
> 
> This *appears* to be the same commit you are following up on (please
> always include human readable descriptions)

This commit (9f89566) is a redundancy made by mistake.

> 
> > This makes free_irq() call duplicate.

Currently in your for-next branch

void dw_spi_remove_host(struct dw_spi *dws)
{
        dw_spi_debugfs_remove(dws);

        free_irq(dws->irq, dws->master);
^^^ (1)


        if (dws->dma_ops && dws->dma_ops->dma_exit)
                dws->dma_ops->dma_exit(dws);

        spi_shutdown_chip(dws);

        free_irq(dws->irq, dws->master);
^^^ (2)
}
EXPORT_SYMBOL_GPL(dw_spi_remove_host);

> >  Please, remove it from your tree
> > or do the revert.
> 
> Are you saying that you have realised that your patch is broken since
> sending it?  I've dropped this commit...

No, the commit 02f20387e1bc is rightful and should be present in the
tree.

P.S. Commit IDs as they appear in your for-next branch of SPI tree.
Andy Shevchenko Oct. 23, 2015, 4:30 p.m. UTC | #4
On Fri, 2015-10-23 at 19:25 +0300, Andy Shevchenko wrote:
> On Sat, 2015-10-24 at 01:18 +0900, Mark Brown wrote:

> > On Fri, Oct 23, 2015 at 10:40:58AM +0300, Andy Shevchenko wrote:

> > > On Fri, 2015-10-23 at 08:11 +0900, Mark Brown wrote:


> > > > Are you saying that you have realised that your patch is broken

> > since

> > sending it?  I've dropped this commit...


Just checked updated topic/dw branch.
It'is correct right now!

I have no idea from which point the other commit appeared (I didn't
send a such).

-- 
Andy Shevchenko <andriy.shevchenko@intel.com>
Intel Finland Oy
---------------------------------------------------------------------
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
diff mbox

Patch

diff --git a/drivers/spi/spi-dw.c b/drivers/spi/spi-dw.c
index cc2e980..ad66bc5 100644
--- a/drivers/spi/spi-dw.c
+++ b/drivers/spi/spi-dw.c
@@ -542,6 +542,8 @@  void dw_spi_remove_host(struct dw_spi *dws)
 {
 	dw_spi_debugfs_remove(dws);
 
+	free_irq(dws->irq, dws->master);
+
 	if (dws->dma_ops && dws->dma_ops->dma_exit)
 		dws->dma_ops->dma_exit(dws);