linux-fpga.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [RESEND PATCH 0/2] Bug fixes for FPGA DFL
@ 2020-07-09  8:12 Xu Yilun
  2020-07-09  8:12 ` [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret' Xu Yilun
  2020-07-09  8:12 ` [RESEND PATCH 2/2] fpga: dfl: fix bug in port reset handshake Xu Yilun
  0 siblings, 2 replies; 7+ messages in thread
From: Xu Yilun @ 2020-07-09  8:12 UTC (permalink / raw)
  To: mdf, linux-fpga, linux-kernel; +Cc: trix, lgoncalv, Xu Yilun

Resend these 2 fix patches since the to-be-fixed patches have been
merged to mainline.

Matthew Gerlach (1):
  fpga: dfl: fix bug in port reset handshake

Xu Yilun (1):
  fpga: dfl: pci: reduce the scope of variable 'ret'

 drivers/fpga/dfl-afu-main.c | 3 ++-
 drivers/fpga/dfl-pci.c      | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

-- 
2.7.4

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret'
  2020-07-09  8:12 [RESEND PATCH 0/2] Bug fixes for FPGA DFL Xu Yilun
@ 2020-07-09  8:12 ` Xu Yilun
  2020-07-09 13:18   ` Tom Rix
  2020-07-09  8:12 ` [RESEND PATCH 2/2] fpga: dfl: fix bug in port reset handshake Xu Yilun
  1 sibling, 1 reply; 7+ messages in thread
From: Xu Yilun @ 2020-07-09  8:12 UTC (permalink / raw)
  To: mdf, linux-fpga, linux-kernel; +Cc: trix, lgoncalv, Xu Yilun

This is to fix lkp cppcheck warnings:

 drivers/fpga/dfl-pci.c:230:6: warning: The scope of the variable 'ret' can be reduced. [variableScope]
    int ret = 0;
        ^

 drivers/fpga/dfl-pci.c:230:10: warning: Variable 'ret' is assigned a value that is never used. [unreadVariable]
    int ret = 0;
            ^

Fixes: 3c2760b78f90 ("fpga: dfl: pci: fix return value of cci_pci_sriov_configure")
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Xu Yilun <yilun.xu@intel.com>
Acked-by: Wu Hao <hao.wu@intel.com>
---
 drivers/fpga/dfl-pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/fpga/dfl-pci.c b/drivers/fpga/dfl-pci.c
index 4a14a24..73b5153 100644
--- a/drivers/fpga/dfl-pci.c
+++ b/drivers/fpga/dfl-pci.c
@@ -285,7 +285,6 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
 {
 	struct cci_drvdata *drvdata = pci_get_drvdata(pcidev);
 	struct dfl_fpga_cdev *cdev = drvdata->cdev;
-	int ret = 0;
 
 	if (!num_vfs) {
 		/*
@@ -297,6 +296,8 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
 		dfl_fpga_cdev_config_ports_pf(cdev);
 
 	} else {
+		int ret;
+
 		/*
 		 * before enable SRIOV, put released ports into VF access mode
 		 * first of all.
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 7+ messages in thread

* [RESEND PATCH 2/2] fpga: dfl: fix bug in port reset handshake
  2020-07-09  8:12 [RESEND PATCH 0/2] Bug fixes for FPGA DFL Xu Yilun
  2020-07-09  8:12 ` [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret' Xu Yilun
@ 2020-07-09  8:12 ` Xu Yilun
  2020-07-09 13:20   ` Tom Rix
  1 sibling, 1 reply; 7+ messages in thread
From: Xu Yilun @ 2020-07-09  8:12 UTC (permalink / raw)
  To: mdf, linux-fpga, linux-kernel; +Cc: trix, lgoncalv, Matthew Gerlach, Xu Yilun

From: Matthew Gerlach <matthew.gerlach@linux.intel.com>

When putting the port in reset, driver must wait for the soft reset
acknowledgment bit instead of the soft reset bit.

Fixes: 47c1b19c160f (fpga: dfl: afu: add port ops support)
Signed-off-by: Matthew Gerlach <matthew.gerlach@linux.intel.com>
Signed-off-by: Xu Yilun <yilun.xu@intel.com>
Acked-by: Wu Hao <hao.wu@intel.com>
---
 drivers/fpga/dfl-afu-main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/fpga/dfl-afu-main.c b/drivers/fpga/dfl-afu-main.c
index 7c84fee..753cda4 100644
--- a/drivers/fpga/dfl-afu-main.c
+++ b/drivers/fpga/dfl-afu-main.c
@@ -83,7 +83,8 @@ int __afu_port_disable(struct platform_device *pdev)
 	 * on this port and minimum soft reset pulse width has elapsed.
 	 * Driver polls port_soft_reset_ack to determine if reset done by HW.
 	 */
-	if (readq_poll_timeout(base + PORT_HDR_CTRL, v, v & PORT_CTRL_SFTRST,
+	if (readq_poll_timeout(base + PORT_HDR_CTRL, v,
+			       v & PORT_CTRL_SFTRST_ACK,
 			       RST_POLL_INVL, RST_POLL_TIMEOUT)) {
 		dev_err(&pdev->dev, "timeout, fail to reset device\n");
 		return -ETIMEDOUT;
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 7+ messages in thread

* Re: [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret'
  2020-07-09  8:12 ` [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret' Xu Yilun
@ 2020-07-09 13:18   ` Tom Rix
  2020-07-10  5:16     ` Xu Yilun
  0 siblings, 1 reply; 7+ messages in thread
From: Tom Rix @ 2020-07-09 13:18 UTC (permalink / raw)
  To: Xu Yilun, mdf, linux-fpga, linux-kernel; +Cc: lgoncalv

I think a better change is to use the ret variable, like this

--- a/drivers/fpga/dfl-pci.c
+++ b/drivers/fpga/dfl-pci.c
@@ -312,7 +312,7 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
                }
        }
 
-       return num_vfs;
+       return ret;
 }

The existing use of returning num_vfs is not right, the function should return 0/err not num_vfs. currently it is reusing the 0 passed in with num_vfs to mean disable as the 0 return status.  it should be properly returning ret.

Tom

On 7/9/20 1:12 AM, Xu Yilun wrote:
> This is to fix lkp cppcheck warnings:
>
>  drivers/fpga/dfl-pci.c:230:6: warning: The scope of the variable 'ret' can be reduced. [variableScope]
>     int ret = 0;
>         ^
>
>  drivers/fpga/dfl-pci.c:230:10: warning: Variable 'ret' is assigned a value that is never used. [unreadVariable]
>     int ret = 0;
>             ^
>
> Fixes: 3c2760b78f90 ("fpga: dfl: pci: fix return value of cci_pci_sriov_configure")
> Reported-by: kbuild test robot <lkp@intel.com>
> Signed-off-by: Xu Yilun <yilun.xu@intel.com>
> Acked-by: Wu Hao <hao.wu@intel.com>
> ---
>  drivers/fpga/dfl-pci.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/fpga/dfl-pci.c b/drivers/fpga/dfl-pci.c
> index 4a14a24..73b5153 100644
> --- a/drivers/fpga/dfl-pci.c
> +++ b/drivers/fpga/dfl-pci.c
> @@ -285,7 +285,6 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
>  {
>  	struct cci_drvdata *drvdata = pci_get_drvdata(pcidev);
>  	struct dfl_fpga_cdev *cdev = drvdata->cdev;
> -	int ret = 0;
>  
>  	if (!num_vfs) {
>  		/*
> @@ -297,6 +296,8 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
>  		dfl_fpga_cdev_config_ports_pf(cdev);
>  
>  	} else {
> +		int ret;
> +
>  		/*
>  		 * before enable SRIOV, put released ports into VF access mode
>  		 * first of all.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [RESEND PATCH 2/2] fpga: dfl: fix bug in port reset handshake
  2020-07-09  8:12 ` [RESEND PATCH 2/2] fpga: dfl: fix bug in port reset handshake Xu Yilun
@ 2020-07-09 13:20   ` Tom Rix
  0 siblings, 0 replies; 7+ messages in thread
From: Tom Rix @ 2020-07-09 13:20 UTC (permalink / raw)
  To: Xu Yilun, mdf, linux-fpga, linux-kernel; +Cc: lgoncalv, Matthew Gerlach


On 7/9/20 1:12 AM, Xu Yilun wrote:
> From: Matthew Gerlach <matthew.gerlach@linux.intel.com>
>
> When putting the port in reset, driver must wait for the soft reset
> acknowledgment bit instead of the soft reset bit.
>
> Fixes: 47c1b19c160f (fpga: dfl: afu: add port ops support)
> Signed-off-by: Matthew Gerlach <matthew.gerlach@linux.intel.com>
> Signed-off-by: Xu Yilun <yilun.xu@intel.com>
> Acked-by: Wu Hao <hao.wu@intel.com>
> ---
>  drivers/fpga/dfl-afu-main.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/fpga/dfl-afu-main.c b/drivers/fpga/dfl-afu-main.c
> index 7c84fee..753cda4 100644
> --- a/drivers/fpga/dfl-afu-main.c
> +++ b/drivers/fpga/dfl-afu-main.c
> @@ -83,7 +83,8 @@ int __afu_port_disable(struct platform_device *pdev)
>  	 * on this port and minimum soft reset pulse width has elapsed.
>  	 * Driver polls port_soft_reset_ack to determine if reset done by HW.
>  	 */
> -	if (readq_poll_timeout(base + PORT_HDR_CTRL, v, v & PORT_CTRL_SFTRST,
> +	if (readq_poll_timeout(base + PORT_HDR_CTRL, v,
> +			       v & PORT_CTRL_SFTRST_ACK,
>  			       RST_POLL_INVL, RST_POLL_TIMEOUT)) {
>  		dev_err(&pdev->dev, "timeout, fail to reset device\n");
>  		return -ETIMEDOUT;

Looks ok to me.

Reviewed-by: Tom Rix <trix@redhat.com>

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret'
  2020-07-09 13:18   ` Tom Rix
@ 2020-07-10  5:16     ` Xu Yilun
  2020-07-11 16:21       ` Tom Rix
  0 siblings, 1 reply; 7+ messages in thread
From: Xu Yilun @ 2020-07-10  5:16 UTC (permalink / raw)
  To: Tom Rix; +Cc: mdf, linux-fpga, linux-kernel, lgoncalv

On Thu, Jul 09, 2020 at 06:18:18AM -0700, Tom Rix wrote:
> I think a better change is to use the ret variable, like this
> 
> --- a/drivers/fpga/dfl-pci.c
> +++ b/drivers/fpga/dfl-pci.c
> @@ -312,7 +312,7 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
> ��������������� }
> ������� }
> �
> -������ return num_vfs;
> +������ return ret;
> �}
> 
> The existing use of returning num_vfs is not right, the function should return 0/err not num_vfs. currently it is reusing the 0 passed in with num_vfs to mean disable as the 0 return status.� it should be properly returning ret.

The sriov_configure callback should return negative value for error, and
return num_vfs if success.

See the Documentation/PCI/pci-iov-howto.rst

also in drivers/pci/iov.c:

  static ssize_t sriov_numvfs_store(struct device *dev, ...)
  {
	...

        ret = pdev->driver->sriov_configure(pdev, num_vfs);
        if (ret < 0) 
                goto exit;

        if (ret != num_vfs)
                pci_warn(pdev, "%d VFs requested; only %d enabled\n",
                         num_vfs, ret);

	...
  }

> 
> Tom
> 
> On 7/9/20 1:12 AM, Xu Yilun wrote:
> > This is to fix lkp cppcheck warnings:
> >
> >  drivers/fpga/dfl-pci.c:230:6: warning: The scope of the variable 'ret' can be reduced. [variableScope]
> >     int ret = 0;
> >         ^
> >
> >  drivers/fpga/dfl-pci.c:230:10: warning: Variable 'ret' is assigned a value that is never used. [unreadVariable]
> >     int ret = 0;
> >             ^
> >
> > Fixes: 3c2760b78f90 ("fpga: dfl: pci: fix return value of cci_pci_sriov_configure")
> > Reported-by: kbuild test robot <lkp@intel.com>
> > Signed-off-by: Xu Yilun <yilun.xu@intel.com>
> > Acked-by: Wu Hao <hao.wu@intel.com>
> > ---
> >  drivers/fpga/dfl-pci.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/fpga/dfl-pci.c b/drivers/fpga/dfl-pci.c
> > index 4a14a24..73b5153 100644
> > --- a/drivers/fpga/dfl-pci.c
> > +++ b/drivers/fpga/dfl-pci.c
> > @@ -285,7 +285,6 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
> >  {
> >  	struct cci_drvdata *drvdata = pci_get_drvdata(pcidev);
> >  	struct dfl_fpga_cdev *cdev = drvdata->cdev;
> > -	int ret = 0;
> >  
> >  	if (!num_vfs) {
> >  		/*
> > @@ -297,6 +296,8 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
> >  		dfl_fpga_cdev_config_ports_pf(cdev);
> >  
> >  	} else {
> > +		int ret;
> > +
> >  		/*
> >  		 * before enable SRIOV, put released ports into VF access mode
> >  		 * first of all.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret'
  2020-07-10  5:16     ` Xu Yilun
@ 2020-07-11 16:21       ` Tom Rix
  0 siblings, 0 replies; 7+ messages in thread
From: Tom Rix @ 2020-07-11 16:21 UTC (permalink / raw)
  To: Xu Yilun; +Cc: mdf, linux-fpga, linux-kernel, lgoncalv

Ok, i take your point on the api.

However I still feel that this change should be improved so the return is not defined inside a block.

I think this could be done better with

int ret = num_vfs;

...

return ret;

There is no consistency on this check with the other drivers' sriov_configure.

I do not feel strongly about this change, your original change will work. So you may add.

Reviewed-by: Tom Rix <trix@redhat.com>

Tom

On 7/9/20 10:16 PM, Xu Yilun wrote:
> On Thu, Jul 09, 2020 at 06:18:18AM -0700, Tom Rix wrote:
>> I think a better change is to use the ret variable, like this
>>
>> --- a/drivers/fpga/dfl-pci.c
>> +++ b/drivers/fpga/dfl-pci.c
>> @@ -312,7 +312,7 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
>>                 }
>>         }
>>  
>> -       return num_vfs;
>> +       return ret;
>>  }
>>
>> The existing use of returning num_vfs is not right, the function should return 0/err not num_vfs. currently it is reusing the 0 passed in with num_vfs to mean disable as the 0 return status.  it should be properly returning ret.
> The sriov_configure callback should return negative value for error, and
> return num_vfs if success.
>
> See the Documentation/PCI/pci-iov-howto.rst
>
> also in drivers/pci/iov.c:
>
>   static ssize_t sriov_numvfs_store(struct device *dev, ...)
>   {
> 	...
>
>         ret = pdev->driver->sriov_configure(pdev, num_vfs);
>         if (ret < 0) 
>                 goto exit;
>
>         if (ret != num_vfs)
>                 pci_warn(pdev, "%d VFs requested; only %d enabled\n",
>                          num_vfs, ret);
>
> 	...
>   }
>
>> Tom
>>
>> On 7/9/20 1:12 AM, Xu Yilun wrote:
>>> This is to fix lkp cppcheck warnings:
>>>
>>>  drivers/fpga/dfl-pci.c:230:6: warning: The scope of the variable 'ret' can be reduced. [variableScope]
>>>     int ret = 0;
>>>         ^
>>>
>>>  drivers/fpga/dfl-pci.c:230:10: warning: Variable 'ret' is assigned a value that is never used. [unreadVariable]
>>>     int ret = 0;
>>>             ^
>>>
>>> Fixes: 3c2760b78f90 ("fpga: dfl: pci: fix return value of cci_pci_sriov_configure")
>>> Reported-by: kbuild test robot <lkp@intel.com>
>>> Signed-off-by: Xu Yilun <yilun.xu@intel.com>
>>> Acked-by: Wu Hao <hao.wu@intel.com>
>>> ---
>>>  drivers/fpga/dfl-pci.c | 3 ++-
>>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/fpga/dfl-pci.c b/drivers/fpga/dfl-pci.c
>>> index 4a14a24..73b5153 100644
>>> --- a/drivers/fpga/dfl-pci.c
>>> +++ b/drivers/fpga/dfl-pci.c
>>> @@ -285,7 +285,6 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
>>>  {
>>>  	struct cci_drvdata *drvdata = pci_get_drvdata(pcidev);
>>>  	struct dfl_fpga_cdev *cdev = drvdata->cdev;
>>> -	int ret = 0;
>>>  
>>>  	if (!num_vfs) {
>>>  		/*
>>> @@ -297,6 +296,8 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
>>>  		dfl_fpga_cdev_config_ports_pf(cdev);
>>>  
>>>  	} else {
>>> +		int ret;
>>> +
>>>  		/*
>>>  		 * before enable SRIOV, put released ports into VF access mode
>>>  		 * first of all.

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2020-07-11 16:21 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-07-09  8:12 [RESEND PATCH 0/2] Bug fixes for FPGA DFL Xu Yilun
2020-07-09  8:12 ` [RESEND PATCH 1/2] fpga: dfl: pci: reduce the scope of variable 'ret' Xu Yilun
2020-07-09 13:18   ` Tom Rix
2020-07-10  5:16     ` Xu Yilun
2020-07-11 16:21       ` Tom Rix
2020-07-09  8:12 ` [RESEND PATCH 2/2] fpga: dfl: fix bug in port reset handshake Xu Yilun
2020-07-09 13:20   ` Tom Rix

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).