Add PM support to enable suspend to RAM support for Tegra virtual Ethernet over PCIe in RP system. Bug 3216243 Change-Id: I39e023c6c890699b082f6761dd2955b19659da16 Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2469498 Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit
replace smp_mb() memory barrier with mb() to take care of IO and Memory access synchronization as well. Bug 200600954 Change-Id: I2526c1b972166f241339e25f5ae8722a8d532bd6 Signed-off-by: Om Prakash Singh <omp@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2327768 Reviewed-by: automaticguardword <automaticguardword@nvidia.com> Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit
add tegra_vnet.h for tegra virtual network host and endpoint driver Bug 200585814 Change-Id: I3a8070983885cbb3a1680f0f3d356cf027720419 Signed-off-by: Om Prakash Singh <omp@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2283952 Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit
host and ep driver has functions with same name. It can create confusion is debugging. ie. using ftrace. Use below convention for host and ep driver function: - host driver function should start with tvnet_host_ - EP driver should start with tvnet_ep_ Bug 200585814 Change-Id: I2776bd3cb162d593297bfced50d189bb7a7aa55b Signed-off-by: Om Prakash Singh <omp@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2283525 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit
a) add change_mtu call back to toggle interface while changing mtu b) set default size of 64512 to get maximum throughput Bug 200581381 Bug 200581376 Change-Id: If06e4756db4a7da4ce1840855fdd45ea736b1048 Signed-off-by: Om Prakash Singh <omp@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2280044 Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Process control messages in the same kthread context instead of scheduling a separate work for the same to improve performance. Bug 200495545 Change-Id: Iac004bd313754e61d789dff233cf487b1f4710d2 Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2183364 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Tested-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Raise interrupt only when full buffer is processed to avoid Tx queue hang. Bug 200495545 Change-Id: Ibd47870707e0eb4fccdb9747e0b8e4dd2a27540f Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2183363 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Tested-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Optimizes read-write counter accesses by splitting the single API that currently caters to all kinds of accesses to multiple small APIs to cater to specific purpose. Bug 200495545 Change-Id: I07da794bd75ce9c968d7882542739c3faca3a565 Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2183362 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Tested-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Enables NAPI in the drivers of T2T virtual ethernet framework to improve performance. Bug 200495545 Change-Id: I656c1897f6dd74738c0d493c4ddc94af410cf41b Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2183361 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Tested-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Pre allocates SKB buffers and DMA maps them to avoid memcpy to improve performance Bug 200495545 Change-Id: I973067041e49d1581eaaf6f93cdda7d02fd74461 Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2183359 GVS: Gerrit_Virtual_Submit Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Tested-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>