* [PATCH 0/2] *** altera-msgdma: make response port optional ***
@ 2021-06-17 19:51 Olivier Dautricourt
2021-06-17 19:52 ` [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional Olivier Dautricourt
2021-06-17 19:53 ` [PATCH 2/2] dmaengine: " Olivier Dautricourt
0 siblings, 2 replies; 7+ messages in thread
From: Olivier Dautricourt @ 2021-06-17 19:51 UTC (permalink / raw)
To: Rob Herring, Vinod Koul, Stefan Roese
Cc: Olivier Dautricourt, dmaengine, devicetree, linux-kernel
Response port can be disabled in the ip core configuration,
so allow not to specify one.
This patch serie is applicable on
git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine.git next
Olivier Dautricourt (2):
dt-bindings: dma: altera-msgdma: make response port optional
dmaengine: altera-msgdma: make response port optional
.../devicetree/bindings/dma/altr,msgdma.yaml | 4 +-
drivers/dma/altera-msgdma.c | 37 +++++++++++++------
2 files changed, 29 insertions(+), 12 deletions(-)
base-commit: 656758425f98693bd61a08f6b51c4c5aa26c9d50
--
2.31.0.rc2
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional
2021-06-17 19:51 [PATCH 0/2] *** altera-msgdma: make response port optional *** Olivier Dautricourt
@ 2021-06-17 19:52 ` Olivier Dautricourt
2021-06-18 5:54 ` Stefan Roese
2021-06-24 21:34 ` Rob Herring
2021-06-17 19:53 ` [PATCH 2/2] dmaengine: " Olivier Dautricourt
1 sibling, 2 replies; 7+ messages in thread
From: Olivier Dautricourt @ 2021-06-17 19:52 UTC (permalink / raw)
To: Rob Herring, Vinod Koul, Stefan Roese
Cc: Olivier Dautricourt, dmaengine, devicetree, linux-kernel
Response port is not required in some configuration of the IP core.
Signed-off-by: Olivier Dautricourt <olivier.dautricourt@orolia.com>
---
Documentation/devicetree/bindings/dma/altr,msgdma.yaml | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/Documentation/devicetree/bindings/dma/altr,msgdma.yaml b/Documentation/devicetree/bindings/dma/altr,msgdma.yaml
index a4f9fe23dcd9..b193ee2db4a7 100644
--- a/Documentation/devicetree/bindings/dma/altr,msgdma.yaml
+++ b/Documentation/devicetree/bindings/dma/altr,msgdma.yaml
@@ -24,13 +24,15 @@ properties:
items:
- description: Control and Status Register Slave Port
- description: Descriptor Slave Port
- - description: Response Slave Port
+ - description: Response Slave Port (Optional)
+ minItems: 2
reg-names:
items:
- const: csr
- const: desc
- const: resp
+ minItems: 2
interrupts:
maxItems: 1
--
2.31.0.rc2
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH 2/2] dmaengine: altera-msgdma: make response port optional
2021-06-17 19:51 [PATCH 0/2] *** altera-msgdma: make response port optional *** Olivier Dautricourt
2021-06-17 19:52 ` [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional Olivier Dautricourt
@ 2021-06-17 19:53 ` Olivier Dautricourt
2021-06-18 5:54 ` Stefan Roese
2021-07-28 6:54 ` Vinod Koul
1 sibling, 2 replies; 7+ messages in thread
From: Olivier Dautricourt @ 2021-06-17 19:53 UTC (permalink / raw)
To: Rob Herring, Vinod Koul, Stefan Roese
Cc: Olivier Dautricourt, dmaengine, devicetree, linux-kernel
The response slave port can be disabled in some configuration [1] and
csr + MSGDMA_CSR_RESP_FILL_LEVEL will be 0 even if transfer has suceeded.
We have to only rely on the interrupts in that scenario.
This was tested on cyclone V with the controller resp port disabled.
[1] https://www.intel.com/content/www/us/en/programmable/documentation/sfo1400787952932.html
30.3.1.2
30.3.1.3
30.5.5
Fixes:
https://forum.rocketboards.org/t/ip-msgdma-linux-driver/1919
Signed-off-by: Olivier Dautricourt <olivier.dautricourt@orolia.com>
---
drivers/dma/altera-msgdma.c | 37 ++++++++++++++++++++++++++-----------
1 file changed, 26 insertions(+), 11 deletions(-)
diff --git a/drivers/dma/altera-msgdma.c b/drivers/dma/altera-msgdma.c
index 0fe0676f8e1d..5a2c7573b692 100644
--- a/drivers/dma/altera-msgdma.c
+++ b/drivers/dma/altera-msgdma.c
@@ -691,10 +691,14 @@ static void msgdma_tasklet(struct tasklet_struct *t)
spin_lock_irqsave(&mdev->lock, flags);
- /* Read number of responses that are available */
- count = ioread32(mdev->csr + MSGDMA_CSR_RESP_FILL_LEVEL);
- dev_dbg(mdev->dev, "%s (%d): response count=%d\n",
- __func__, __LINE__, count);
+ if (mdev->resp) {
+ /* Read number of responses that are available */
+ count = ioread32(mdev->csr + MSGDMA_CSR_RESP_FILL_LEVEL);
+ dev_dbg(mdev->dev, "%s (%d): response count=%d\n",
+ __func__, __LINE__, count);
+ } else {
+ count = 1;
+ }
while (count--) {
/*
@@ -703,8 +707,12 @@ static void msgdma_tasklet(struct tasklet_struct *t)
* have any real values, like transferred bytes or error
* bits. So we need to just drop these values.
*/
- size = ioread32(mdev->resp + MSGDMA_RESP_BYTES_TRANSFERRED);
- status = ioread32(mdev->resp + MSGDMA_RESP_STATUS);
+ if (mdev->resp) {
+ size = ioread32(mdev->resp +
+ MSGDMA_RESP_BYTES_TRANSFERRED);
+ status = ioread32(mdev->resp +
+ MSGDMA_RESP_STATUS);
+ }
msgdma_complete_descriptor(mdev);
msgdma_chan_desc_cleanup(mdev);
@@ -757,14 +765,21 @@ static void msgdma_dev_remove(struct msgdma_device *mdev)
}
static int request_and_map(struct platform_device *pdev, const char *name,
- struct resource **res, void __iomem **ptr)
+ struct resource **res, void __iomem **ptr,
+ bool optional)
{
struct resource *region;
struct device *device = &pdev->dev;
*res = platform_get_resource_byname(pdev, IORESOURCE_MEM, name);
if (*res == NULL) {
- dev_err(device, "resource %s not defined\n", name);
+ if (optional) {
+ *ptr = NULL;
+ dev_info(device, "optional resource %s not defined\n",
+ name);
+ return 0;
+ }
+ dev_err(device, "mandatory resource %s not defined\n", name);
return -ENODEV;
}
@@ -805,17 +820,17 @@ static int msgdma_probe(struct platform_device *pdev)
mdev->dev = &pdev->dev;
/* Map CSR space */
- ret = request_and_map(pdev, "csr", &dma_res, &mdev->csr);
+ ret = request_and_map(pdev, "csr", &dma_res, &mdev->csr, false);
if (ret)
return ret;
/* Map (extended) descriptor space */
- ret = request_and_map(pdev, "desc", &dma_res, &mdev->desc);
+ ret = request_and_map(pdev, "desc", &dma_res, &mdev->desc, false);
if (ret)
return ret;
/* Map response space */
- ret = request_and_map(pdev, "resp", &dma_res, &mdev->resp);
+ ret = request_and_map(pdev, "resp", &dma_res, &mdev->resp, true);
if (ret)
return ret;
--
2.31.0.rc2
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional
2021-06-17 19:52 ` [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional Olivier Dautricourt
@ 2021-06-18 5:54 ` Stefan Roese
2021-06-24 21:34 ` Rob Herring
1 sibling, 0 replies; 7+ messages in thread
From: Stefan Roese @ 2021-06-18 5:54 UTC (permalink / raw)
To: Olivier Dautricourt, Rob Herring, Vinod Koul
Cc: dmaengine, devicetree, linux-kernel
On 17.06.21 21:52, Olivier Dautricourt wrote:
> Response port is not required in some configuration of the IP core.
>
> Signed-off-by: Olivier Dautricourt <olivier.dautricourt@orolia.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Thanks,
Stefan
> ---
> Documentation/devicetree/bindings/dma/altr,msgdma.yaml | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/devicetree/bindings/dma/altr,msgdma.yaml b/Documentation/devicetree/bindings/dma/altr,msgdma.yaml
> index a4f9fe23dcd9..b193ee2db4a7 100644
> --- a/Documentation/devicetree/bindings/dma/altr,msgdma.yaml
> +++ b/Documentation/devicetree/bindings/dma/altr,msgdma.yaml
> @@ -24,13 +24,15 @@ properties:
> items:
> - description: Control and Status Register Slave Port
> - description: Descriptor Slave Port
> - - description: Response Slave Port
> + - description: Response Slave Port (Optional)
> + minItems: 2
>
> reg-names:
> items:
> - const: csr
> - const: desc
> - const: resp
> + minItems: 2
>
> interrupts:
> maxItems: 1
> --
> 2.31.0.rc2
>
Viele Grüße,
Stefan
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr@denx.de
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH 2/2] dmaengine: altera-msgdma: make response port optional
2021-06-17 19:53 ` [PATCH 2/2] dmaengine: " Olivier Dautricourt
@ 2021-06-18 5:54 ` Stefan Roese
2021-07-28 6:54 ` Vinod Koul
1 sibling, 0 replies; 7+ messages in thread
From: Stefan Roese @ 2021-06-18 5:54 UTC (permalink / raw)
To: Olivier Dautricourt, Rob Herring, Vinod Koul
Cc: dmaengine, devicetree, linux-kernel
On 17.06.21 21:53, Olivier Dautricourt wrote:
> The response slave port can be disabled in some configuration [1] and
> csr + MSGDMA_CSR_RESP_FILL_LEVEL will be 0 even if transfer has suceeded.
> We have to only rely on the interrupts in that scenario.
> This was tested on cyclone V with the controller resp port disabled.
>
> [1] https://www.intel.com/content/www/us/en/programmable/documentation/sfo1400787952932.html
> 30.3.1.2
> 30.3.1.3
> 30.5.5
>
> Fixes:
> https://forum.rocketboards.org/t/ip-msgdma-linux-driver/1919
> Signed-off-by: Olivier Dautricourt <olivier.dautricourt@orolia.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Thanks,
Stefan
> ---
> drivers/dma/altera-msgdma.c | 37 ++++++++++++++++++++++++++-----------
> 1 file changed, 26 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/dma/altera-msgdma.c b/drivers/dma/altera-msgdma.c
> index 0fe0676f8e1d..5a2c7573b692 100644
> --- a/drivers/dma/altera-msgdma.c
> +++ b/drivers/dma/altera-msgdma.c
> @@ -691,10 +691,14 @@ static void msgdma_tasklet(struct tasklet_struct *t)
>
> spin_lock_irqsave(&mdev->lock, flags);
>
> - /* Read number of responses that are available */
> - count = ioread32(mdev->csr + MSGDMA_CSR_RESP_FILL_LEVEL);
> - dev_dbg(mdev->dev, "%s (%d): response count=%d\n",
> - __func__, __LINE__, count);
> + if (mdev->resp) {
> + /* Read number of responses that are available */
> + count = ioread32(mdev->csr + MSGDMA_CSR_RESP_FILL_LEVEL);
> + dev_dbg(mdev->dev, "%s (%d): response count=%d\n",
> + __func__, __LINE__, count);
> + } else {
> + count = 1;
> + }
>
> while (count--) {
> /*
> @@ -703,8 +707,12 @@ static void msgdma_tasklet(struct tasklet_struct *t)
> * have any real values, like transferred bytes or error
> * bits. So we need to just drop these values.
> */
> - size = ioread32(mdev->resp + MSGDMA_RESP_BYTES_TRANSFERRED);
> - status = ioread32(mdev->resp + MSGDMA_RESP_STATUS);
> + if (mdev->resp) {
> + size = ioread32(mdev->resp +
> + MSGDMA_RESP_BYTES_TRANSFERRED);
> + status = ioread32(mdev->resp +
> + MSGDMA_RESP_STATUS);
> + }
>
> msgdma_complete_descriptor(mdev);
> msgdma_chan_desc_cleanup(mdev);
> @@ -757,14 +765,21 @@ static void msgdma_dev_remove(struct msgdma_device *mdev)
> }
>
> static int request_and_map(struct platform_device *pdev, const char *name,
> - struct resource **res, void __iomem **ptr)
> + struct resource **res, void __iomem **ptr,
> + bool optional)
> {
> struct resource *region;
> struct device *device = &pdev->dev;
>
> *res = platform_get_resource_byname(pdev, IORESOURCE_MEM, name);
> if (*res == NULL) {
> - dev_err(device, "resource %s not defined\n", name);
> + if (optional) {
> + *ptr = NULL;
> + dev_info(device, "optional resource %s not defined\n",
> + name);
> + return 0;
> + }
> + dev_err(device, "mandatory resource %s not defined\n", name);
> return -ENODEV;
> }
>
> @@ -805,17 +820,17 @@ static int msgdma_probe(struct platform_device *pdev)
> mdev->dev = &pdev->dev;
>
> /* Map CSR space */
> - ret = request_and_map(pdev, "csr", &dma_res, &mdev->csr);
> + ret = request_and_map(pdev, "csr", &dma_res, &mdev->csr, false);
> if (ret)
> return ret;
>
> /* Map (extended) descriptor space */
> - ret = request_and_map(pdev, "desc", &dma_res, &mdev->desc);
> + ret = request_and_map(pdev, "desc", &dma_res, &mdev->desc, false);
> if (ret)
> return ret;
>
> /* Map response space */
> - ret = request_and_map(pdev, "resp", &dma_res, &mdev->resp);
> + ret = request_and_map(pdev, "resp", &dma_res, &mdev->resp, true);
> if (ret)
> return ret;
>
> --
> 2.31.0.rc2
>
Viele Grüße,
Stefan
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr@denx.de
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional
2021-06-17 19:52 ` [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional Olivier Dautricourt
2021-06-18 5:54 ` Stefan Roese
@ 2021-06-24 21:34 ` Rob Herring
1 sibling, 0 replies; 7+ messages in thread
From: Rob Herring @ 2021-06-24 21:34 UTC (permalink / raw)
To: Olivier Dautricourt
Cc: Stefan Roese, devicetree, linux-kernel, Vinod Koul, dmaengine,
Rob Herring
On Thu, 17 Jun 2021 21:52:32 +0200, Olivier Dautricourt wrote:
> Response port is not required in some configuration of the IP core.
>
> Signed-off-by: Olivier Dautricourt <olivier.dautricourt@orolia.com>
> ---
> Documentation/devicetree/bindings/dma/altr,msgdma.yaml | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
Reviewed-by: Rob Herring <robh@kernel.org>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH 2/2] dmaengine: altera-msgdma: make response port optional
2021-06-17 19:53 ` [PATCH 2/2] dmaengine: " Olivier Dautricourt
2021-06-18 5:54 ` Stefan Roese
@ 2021-07-28 6:54 ` Vinod Koul
1 sibling, 0 replies; 7+ messages in thread
From: Vinod Koul @ 2021-07-28 6:54 UTC (permalink / raw)
To: Olivier Dautricourt
Cc: Rob Herring, Stefan Roese, dmaengine, devicetree, linux-kernel
On 17-06-21, 21:53, Olivier Dautricourt wrote:
> The response slave port can be disabled in some configuration [1] and
> csr + MSGDMA_CSR_RESP_FILL_LEVEL will be 0 even if transfer has suceeded.
> We have to only rely on the interrupts in that scenario.
> This was tested on cyclone V with the controller resp port disabled.
>
> [1] https://www.intel.com/content/www/us/en/programmable/documentation/sfo1400787952932.html
> 30.3.1.2
> 30.3.1.3
> 30.5.5
>
> Fixes:
> https://forum.rocketboards.org/t/ip-msgdma-linux-driver/1919
This should be single line, fixed it up while applying the series
--
~Vinod
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2021-07-28 6:54 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-06-17 19:51 [PATCH 0/2] *** altera-msgdma: make response port optional *** Olivier Dautricourt
2021-06-17 19:52 ` [PATCH 1/2] dt-bindings: dma: altera-msgdma: make response port optional Olivier Dautricourt
2021-06-18 5:54 ` Stefan Roese
2021-06-24 21:34 ` Rob Herring
2021-06-17 19:53 ` [PATCH 2/2] dmaengine: " Olivier Dautricourt
2021-06-18 5:54 ` Stefan Roese
2021-07-28 6:54 ` Vinod Koul
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).