2019-11-05 07:28:34

by Alexandru Ardelean

Subject: [PATCH] uio: fix irq init with dt support & irq not defined

This change also does a bit of a unification for the IRQ init code.

But the actual problem is that UIO_IRQ_NONE == 0, so for the DT case where
UIO_IRQ_NONE gets assigned to `uioinfo->irq`, a 2nd initialization will get
triggered (for the IRQ) and this one will exit via `goto bad1`.

As far as things seem to go, the only case where UIO_IRQ_NONE seems valid,
is when using a device-tree. The driver has some legacy support for old
platform_data structures. It looks like, for platform_data a non-existent
IRQ is an invalid case (or was considered an invalid case).
Which is why -ENXIO is treated only when a DT is used.

Signed-off-by: Dragos Bogdan <[email protected]>
Signed-off-by: Alexandru Ardelean <[email protected]>
---
drivers/uio/uio_dmem_genirq.c | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/uio/uio_dmem_genirq.c b/drivers/uio/uio_dmem_genirq.c
index ebcf1434e296..0a7d71d4a4f0 100644
--- a/drivers/uio/uio_dmem_genirq.c
+++ b/drivers/uio/uio_dmem_genirq.c
@@ -163,13 +163,6 @@ static int uio_dmem_genirq_probe(struct platform_device *pdev)
uioinfo->name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%pOFn",
pdev->dev.of_node);
uioinfo->version = "devicetree";
-
- /* Multiple IRQs are not supported */
- irq = platform_get_irq(pdev, 0);
- if (irq == -ENXIO)
- uioinfo->irq = UIO_IRQ_NONE;
- else
- uioinfo->irq = irq;
}

if (!uioinfo || !uioinfo->name || !uioinfo->version) {
@@ -199,8 +192,11 @@ static int uio_dmem_genirq_probe(struct platform_device *pdev)
mutex_init(&priv->alloc_lock);

if (!uioinfo->irq) {
+ /* Multiple IRQs are not supported */
ret = platform_get_irq(pdev, 0);
- if (ret < 0)
+ if (ret == -ENXIO && pdev->dev.of_node)
+ ret = UIO_IRQ_NONE;
+ else if (ret < 0)
goto bad1;
uioinfo->irq = ret;
}
--
2.20.1

2019-11-05 07:33:08

by Alexandru Ardelean

Subject: [PATCH v2] uio: fix irq init with dt support & irq not defined

This change also does a bit of a unification for the IRQ init code.

But the actual problem is that UIO_IRQ_NONE == 0, so for the DT case where
UIO_IRQ_NONE gets assigned to `uioinfo->irq`, a 2nd initialization will get
triggered (for the IRQ) and this one will exit via `goto bad1`.

As far as things seem to go, the only case where UIO_IRQ_NONE seems valid,
is when using a device-tree. The driver has some legacy support for old
platform_data structures. It looks like, for platform_data a non-existent
IRQ is an invalid case (or was considered an invalid case).
Which is why -ENXIO is treated only when a DT is used.

Signed-off-by: Dragos Bogdan <[email protected]>
Signed-off-by: Alexandru Ardelean <[email protected]>
---

Changelog v1 -> v2:
* removed `int irq` variable ; was omitted when porting the patch to a
newer kernel base

drivers/uio/uio_dmem_genirq.c | 14 ++++----------
1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/uio/uio_dmem_genirq.c b/drivers/uio/uio_dmem_genirq.c
index ebcf1434e296..81c88f7bbbcb 100644
--- a/drivers/uio/uio_dmem_genirq.c
+++ b/drivers/uio/uio_dmem_genirq.c
@@ -151,8 +151,6 @@ static int uio_dmem_genirq_probe(struct platform_device *pdev)
int i;

if (pdev->dev.of_node) {
- int irq;
-
/* alloc uioinfo for one device */
uioinfo = kzalloc(sizeof(*uioinfo), GFP_KERNEL);
if (!uioinfo) {
@@ -163,13 +161,6 @@ static int uio_dmem_genirq_probe(struct platform_device *pdev)
uioinfo->name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%pOFn",
pdev->dev.of_node);
uioinfo->version = "devicetree";
-
- /* Multiple IRQs are not supported */
- irq = platform_get_irq(pdev, 0);
- if (irq == -ENXIO)
- uioinfo->irq = UIO_IRQ_NONE;
- else
- uioinfo->irq = irq;
}

if (!uioinfo || !uioinfo->name || !uioinfo->version) {
@@ -199,8 +190,11 @@ static int uio_dmem_genirq_probe(struct platform_device *pdev)
mutex_init(&priv->alloc_lock);

if (!uioinfo->irq) {
+ /* Multiple IRQs are not supported */
ret = platform_get_irq(pdev, 0);
- if (ret < 0)
+ if (ret == -ENXIO && pdev->dev.of_node)
+ ret = UIO_IRQ_NONE;
+ else if (ret < 0)
goto bad1;
uioinfo->irq = ret;
}
--
2.20.1

2019-11-06 06:43:00

by Damian Hobson-Garcia

Subject: Re: [PATCH v2] uio: fix irq init with dt support & irq not defined

On 2019-11-05 4:32 p.m., Alexandru Ardelean wrote:
> This change also does a bit of a unification for the IRQ init code.
>
> But the actual problem is that UIO_IRQ_NONE == 0, so for the DT case where
> UIO_IRQ_NONE gets assigned to `uioinfo->irq`, a 2nd initialization will get
> triggered (for the IRQ) and this one will exit via `goto bad1`.
>
> As far as things seem to go, the only case where UIO_IRQ_NONE seems valid,
> is when using a device-tree. The driver has some legacy support for old
> platform_data structures. It looks like, for platform_data a non-existent
> IRQ is an invalid case (or was considered an invalid case).
> Which is why -ENXIO is treated only when a DT is used.
>
> Signed-off-by: Dragos Bogdan <[email protected]>
> Signed-off-by: Alexandru Ardelean <[email protected]>
> ---
>
> Changelog v1 -> v2:
> * removed `int irq` variable ; was omitted when porting the patch to a
> newer kernel base
>

This also brings the implementation in line with the what is done in
uio_pdrv_genirq.c

Acked-by: Damian Hobson-Garcia <[email protected]>