diff mbox series

[1/2] Add GSO UDP Offloading feature to OVS Internal Port

Message ID 20240625082924.775877-1-chengcheng.luo@smartx.com (mailing list archive)
State Changes Requested
Delegated to: Netdev Maintainers
Headers show
Series [1/2] Add GSO UDP Offloading feature to OVS Internal Port | expand

Checks

Context Check Description
netdev/series_format warning Single patches do not need cover letters; Target tree name not specified in the subject
netdev/tree_selection success Guessed tree name to be net-next
netdev/ynl success Generated files up to date; no warnings/errors; no diff in generated;
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 842 this patch: 842
netdev/build_tools success No tools touched, skip
netdev/cc_maintainers success CCed 6 of 6 maintainers
netdev/build_clang success Errors and warnings before: 849 this patch: 849
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 849 this patch: 849
netdev/checkpatch success total: 0 errors, 0 warnings, 0 checks, 9 lines checked
netdev/build_clang_rust success No Rust files in patch. Skipping build
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0
netdev/contest success net-next-2024-06-25--09-00 (tests: 662)

Commit Message

echken June 25, 2024, 8:29 a.m. UTC
The OVS internal port does not support UDP fragmentation offloading,
resulting in large packets sent through the OVS internal port to OVS
being prematurely fragmented. This increases the total number of packets
processed in the path from the vport to the OVS bridge output port,
affecting transmission efficiency.

Signed-off-by: echken <chengcheng.luo@smartx.com>
---
 net/openvswitch/vport-internal_dev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Mike Pattrick June 25, 2024, 1:27 p.m. UTC | #1
On Tue, Jun 25, 2024 at 4:30 AM echken <chengcheng.luo@smartx.com> wrote:
>
> The OVS internal port does not support UDP fragmentation offloading,
> resulting in large packets sent through the OVS internal port to OVS
> being prematurely fragmented. This increases the total number of packets
> processed in the path from the vport to the OVS bridge output port,
> affecting transmission efficiency.
>
> Signed-off-by: echken <chengcheng.luo@smartx.com>
> ---
>  net/openvswitch/vport-internal_dev.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
> index 74c88a6baa43..c5a72c4dc6fd 100644
> --- a/net/openvswitch/vport-internal_dev.c
> +++ b/net/openvswitch/vport-internal_dev.c
> @@ -110,7 +110,8 @@ static void do_setup(struct net_device *netdev)
>
>         netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST |
>                            NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
> -                          NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL;
> +                          NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL |
> +                          NETIF_F_GSO_UDP | NETIF_F_GSO_UDP_L4;

I'll try testing this out, but preliminarily, NETIF_F_GSO_SOFTWARE
already contains NETIF_F_GSO_UDP_L4.


Thanks,
Mike

>
>         netdev->vlan_features = netdev->features;
>         netdev->hw_enc_features = netdev->features;
> --
> 2.34.1
>
>
Ilya Maximets June 25, 2024, 1:55 p.m. UTC | #2
On 6/25/24 15:27, Mike Pattrick wrote:
> On Tue, Jun 25, 2024 at 4:30 AM echken <chengcheng.luo@smartx.com> wrote:
>>
>> The OVS internal port does not support UDP fragmentation offloading,
>> resulting in large packets sent through the OVS internal port to OVS
>> being prematurely fragmented. This increases the total number of packets
>> processed in the path from the vport to the OVS bridge output port,
>> affecting transmission efficiency.
>>
>> Signed-off-by: echken <chengcheng.luo@smartx.com>
>> ---
>>  net/openvswitch/vport-internal_dev.c | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
>> index 74c88a6baa43..c5a72c4dc6fd 100644
>> --- a/net/openvswitch/vport-internal_dev.c
>> +++ b/net/openvswitch/vport-internal_dev.c
>> @@ -110,7 +110,8 @@ static void do_setup(struct net_device *netdev)
>>
>>         netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST |
>>                            NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
>> -                          NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL;
>> +                          NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL |
>> +                          NETIF_F_GSO_UDP | NETIF_F_GSO_UDP_L4;
> 
> I'll try testing this out, but preliminarily, NETIF_F_GSO_SOFTWARE
> already contains NETIF_F_GSO_UDP_L4.

And as far as my understanding goes, NETIF_F_GSO_UDP is
deprecated for all use, except for tuntap.  So, we probably
shouldn't add it.  UFO generally creates more issues than
it solves.

Best regards, Ilya Maximets.
diff mbox series

Patch

diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index 74c88a6baa43..c5a72c4dc6fd 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -110,7 +110,8 @@  static void do_setup(struct net_device *netdev)
 
 	netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST |
 			   NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
-			   NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL;
+			   NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL |
+			   NETIF_F_GSO_UDP | NETIF_F_GSO_UDP_L4;
 
 	netdev->vlan_features = netdev->features;
 	netdev->hw_enc_features = netdev->features;