diff mbox

[net-next] r8169: Disable some bits on pcie

Message ID 1440065553-15028-1-git-send-email-corcodel.marian@gmail.com
State Rejected, archived
Delegated to: David Miller
Headers show

Commit Message

Corcodel Marian Aug. 20, 2015, 10:12 a.m. UTC
Disable legacy interrupt on pci express interface use msi 
 disable some bits from pci express interface wich is not   need on this nic.


Signed-off-by: Corcodel Marian <corcodel.marian@gmail.com>

Comments

Sergei Shtylyov Aug. 20, 2015, 12:57 p.m. UTC | #1
Hello.

On 8/20/2015 1:12 PM, Corcodel Marian wrote:

> Disable legacy interrupt on pci express interface use msi
>   disable some bits from pci express interface wich is not   need on this nic.
>
>
> Signed-off-by: Corcodel Marian <corcodel.marian@gmail.com>
>
> diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
> index 6d16de3..b1fb54f 100644
> --- a/drivers/net/ethernet/realtek/r8169.c
> +++ b/drivers/net/ethernet/realtek/r8169.c
> @@ -8164,6 +8164,13 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
>   	if (!pci_is_pcie(pdev))
>   		netif_info(tp, probe, dev, "not PCI Express\n");
>
> +	if (pci_is_pcie(pdev))
> +		pci_write_config_word(pdev, PCI_COMMAND, ~(PCI_COMMAND_FAST_BACK | PCI_COMMAND_WAIT |
> +					PCI_COMMAND_VGA_PALETTE | PCI_COMMAND_INVALIDATE | PCI_COMMAND_SPECIAL));

    Please start the continuation line under 'pdev' on the first line.

> +
> +	if (pci_is_pcie(pdev))
> +		pci_intx(pdev, 0);
> +

    Why two identical *if* statements? Please make it one.

MBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Sergei Shtylyov Aug. 20, 2015, 1:29 p.m. UTC | #2
On 8/20/2015 3:57 PM, Sergei Shtylyov wrote:

>> Disable legacy interrupt on pci express interface use msi
>>   disable some bits from pci express interface wich is not   need on this nic.
>>
>>
>> Signed-off-by: Corcodel Marian <corcodel.marian@gmail.com>
>>
>> diff --git a/drivers/net/ethernet/realtek/r8169.c
>> b/drivers/net/ethernet/realtek/r8169.c
>> index 6d16de3..b1fb54f 100644
>> --- a/drivers/net/ethernet/realtek/r8169.c
>> +++ b/drivers/net/ethernet/realtek/r8169.c
>> @@ -8164,6 +8164,13 @@ static int rtl_init_one(struct pci_dev *pdev, const
>> struct pci_device_id *ent)
>>       if (!pci_is_pcie(pdev))
>>           netif_info(tp, probe, dev, "not PCI Express\n");
>>
>> +    if (pci_is_pcie(pdev))
>> +        pci_write_config_word(pdev, PCI_COMMAND, ~(PCI_COMMAND_FAST_BACK |
>> PCI_COMMAND_WAIT |
>> +                    PCI_COMMAND_VGA_PALETTE | PCI_COMMAND_INVALIDATE |
>> PCI_COMMAND_SPECIAL));
>
>     Please start the continuation line under 'pdev' on the first line.

    Can all those bits indeed be set for a networking device? 
PCI_COMMAND_VGA_PALETTE is the most incredible...

>> +
>> +    if (pci_is_pcie(pdev))
>> +        pci_intx(pdev, 0);
>> +
>
>     Why two identical *if* statements? Please make it one.

    Ah, I missed the *if* at the top. You need just an *else* clause here.

MBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Eric Dumazet Aug. 20, 2015, 1:44 p.m. UTC | #3
On Thu, 2015-08-20 at 13:12 +0300, Corcodel Marian wrote:
> Disable legacy interrupt on pci express interface use msi 
>  disable some bits from pci express interface wich is not   need on this nic.
> 
> 
> Signed-off-by: Corcodel Marian <corcodel.marian@gmail.com>
> 
> diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
> index 6d16de3..b1fb54f 100644
> --- a/drivers/net/ethernet/realtek/r8169.c
> +++ b/drivers/net/ethernet/realtek/r8169.c
> @@ -8164,6 +8164,13 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
>  	if (!pci_is_pcie(pdev))
>  		netif_info(tp, probe, dev, "not PCI Express\n");
>  
> +	if (pci_is_pcie(pdev))
> +		pci_write_config_word(pdev, PCI_COMMAND, ~(PCI_COMMAND_FAST_BACK | PCI_COMMAND_WAIT |
> +					PCI_COMMAND_VGA_PALETTE | PCI_COMMAND_INVALIDATE | PCI_COMMAND_SPECIAL));
> +
> +	if (pci_is_pcie(pdev))
> +		pci_intx(pdev, 0);
> +
>  	/* Identify chip attached to board */
>  	rtl8169_get_mac_version(tp, dev, cfg->default_ver);
>  

Why is it needed ?

PCI_COMMAND_VGA_PALETTE is never used in a networking driver, why should
we eventually take care ?

What exact problem are you solving ?

We wont accept a patch without a clear description, given all the crap
you have sent.


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index 6d16de3..b1fb54f 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -8164,6 +8164,13 @@  static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!pci_is_pcie(pdev))
 		netif_info(tp, probe, dev, "not PCI Express\n");
 
+	if (pci_is_pcie(pdev))
+		pci_write_config_word(pdev, PCI_COMMAND, ~(PCI_COMMAND_FAST_BACK | PCI_COMMAND_WAIT |
+					PCI_COMMAND_VGA_PALETTE | PCI_COMMAND_INVALIDATE | PCI_COMMAND_SPECIAL));
+
+	if (pci_is_pcie(pdev))
+		pci_intx(pdev, 0);
+
 	/* Identify chip attached to board */
 	rtl8169_get_mac_version(tp, dev, cfg->default_ver);