linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] PCI/VFIO: use pcie_flags_reg instead of access PCI-E Capabilities Register
@ 2013-03-27  9:05 Yijing Wang
  2013-04-01 15:16 ` Alex Williamson
  0 siblings, 1 reply; 3+ messages in thread
From: Yijing Wang @ 2013-03-27  9:05 UTC (permalink / raw)
  To: Alex Williamson
  Cc: Bjorn Helgaas, linux-pci, Arnd Bergmann, linux-kernel, Hanjun Guo,
	jiang.liu, Yijing Wang

Currently, we use pcie_flags_reg to cache PCI-E Capabilities Register,
because PCI-E Capabilities Register bits are almost read-only. This patch
use pcie_flags_reg instead of another access PCI-E Capabilities Register.

Signed-off-by: Yijing Wang <wangyijing@huawei.com>
---
 drivers/vfio/pci/vfio_pci_config.c |    6 +-----
 1 files changed, 1 insertions(+), 5 deletions(-)

diff --git a/drivers/vfio/pci/vfio_pci_config.c b/drivers/vfio/pci/vfio_pci_config.c
index aeb00fc..afc6ca4 100644
--- a/drivers/vfio/pci/vfio_pci_config.c
+++ b/drivers/vfio/pci/vfio_pci_config.c
@@ -1017,13 +1017,9 @@ static int vfio_cap_len(struct vfio_pci_device *vdev, u8 cap, u8 pos)
 		return byte;
 	case PCI_CAP_ID_EXP:
 		/* length based on version */
-		ret = pci_read_config_word(pdev, pos + PCI_EXP_FLAGS, &word);
-		if (ret)
-			return pcibios_err_to_errno(ret);
-
 		vdev->extended_caps = true;
 
-		if ((word & PCI_EXP_FLAGS_VERS) == 1)
+		if ((pdev->pcie_flags_reg & PCI_EXP_FLAGS_VERS) == 1)
 			return PCI_CAP_EXP_ENDPOINT_SIZEOF_V1;
 		else
 			return PCI_CAP_EXP_ENDPOINT_SIZEOF_V2;
-- 
1.7.1



^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] PCI/VFIO: use pcie_flags_reg instead of access PCI-E Capabilities Register
  2013-03-27  9:05 [PATCH] PCI/VFIO: use pcie_flags_reg instead of access PCI-E Capabilities Register Yijing Wang
@ 2013-04-01 15:16 ` Alex Williamson
  2013-04-02  1:47   ` Yijing Wang
  0 siblings, 1 reply; 3+ messages in thread
From: Alex Williamson @ 2013-04-01 15:16 UTC (permalink / raw)
  To: Yijing Wang
  Cc: Bjorn Helgaas, linux-pci, Arnd Bergmann, linux-kernel, Hanjun Guo,
	jiang.liu

On Wed, 2013-03-27 at 17:05 +0800, Yijing Wang wrote:
> Currently, we use pcie_flags_reg to cache PCI-E Capabilities Register,
> because PCI-E Capabilities Register bits are almost read-only. This patch
> use pcie_flags_reg instead of another access PCI-E Capabilities Register.
> 
> Signed-off-by: Yijing Wang <wangyijing@huawei.com>
> ---
>  drivers/vfio/pci/vfio_pci_config.c |    6 +-----
>  1 files changed, 1 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/vfio/pci/vfio_pci_config.c b/drivers/vfio/pci/vfio_pci_config.c
> index aeb00fc..afc6ca4 100644
> --- a/drivers/vfio/pci/vfio_pci_config.c
> +++ b/drivers/vfio/pci/vfio_pci_config.c
> @@ -1017,13 +1017,9 @@ static int vfio_cap_len(struct vfio_pci_device *vdev, u8 cap, u8 pos)
>  		return byte;
>  	case PCI_CAP_ID_EXP:
>  		/* length based on version */
> -		ret = pci_read_config_word(pdev, pos + PCI_EXP_FLAGS, &word);
> -		if (ret)
> -			return pcibios_err_to_errno(ret);
> -
>  		vdev->extended_caps = true;
>  
> -		if ((word & PCI_EXP_FLAGS_VERS) == 1)
> +		if ((pdev->pcie_flags_reg & PCI_EXP_FLAGS_VERS) == 1)

It looks like we should be using pcie_caps_reg(pdev) here instead of
directly accessing pcie_flags_reg.  Thanks,

Alex

>  			return PCI_CAP_EXP_ENDPOINT_SIZEOF_V1;
>  		else
>  			return PCI_CAP_EXP_ENDPOINT_SIZEOF_V2;




^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] PCI/VFIO: use pcie_flags_reg instead of access PCI-E Capabilities Register
  2013-04-01 15:16 ` Alex Williamson
@ 2013-04-02  1:47   ` Yijing Wang
  0 siblings, 0 replies; 3+ messages in thread
From: Yijing Wang @ 2013-04-02  1:47 UTC (permalink / raw)
  To: Alex Williamson
  Cc: Bjorn Helgaas, linux-pci, Arnd Bergmann, linux-kernel, Hanjun Guo,
	jiang.liu

>> @@ -1017,13 +1017,9 @@ static int vfio_cap_len(struct vfio_pci_device *vdev, u8 cap, u8 pos)
>>  		return byte;
>>  	case PCI_CAP_ID_EXP:
>>  		/* length based on version */
>> -		ret = pci_read_config_word(pdev, pos + PCI_EXP_FLAGS, &word);
>> -		if (ret)
>> -			return pcibios_err_to_errno(ret);
>> -
>>  		vdev->extended_caps = true;
>>  
>> -		if ((word & PCI_EXP_FLAGS_VERS) == 1)
>> +		if ((pdev->pcie_flags_reg & PCI_EXP_FLAGS_VERS) == 1)
> 
> It looks like we should be using pcie_caps_reg(pdev) here instead of
> directly accessing pcie_flags_reg.  Thanks,

That's right, I forget Myron Stowe introduced the PCIe Cap accessor function  pcie_caps_reg().
I will resend the updated patch.

Thanks!
Yijing

> 
> Alex
> 
>>  			return PCI_CAP_EXP_ENDPOINT_SIZEOF_V1;
>>  		else
>>  			return PCI_CAP_EXP_ENDPOINT_SIZEOF_V2;
> 
> 
> 
> 
> .
> 


-- 
Thanks!
Yijing


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2013-04-02  1:48 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-03-27  9:05 [PATCH] PCI/VFIO: use pcie_flags_reg instead of access PCI-E Capabilities Register Yijing Wang
2013-04-01 15:16 ` Alex Williamson
2013-04-02  1:47   ` Yijing Wang

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).