Re: [PATCH net-next 4/8] be2net: SRIOV Queue distribution should factor in EQ-count of VFs
From: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Date: 2016-02-02 15:41:34
On Tue, Feb 2, 2016 at 8:00 PM, Sergei Shtylyov [off-list ref] wrote:
Hello. On 2/2/2016 1:20 PM, Sriharsha Basavapatna wrote:quoted
The SRIOV resource distribution logic for RX/TX queue counts is not optimal when a small number of VFs are enabled. It does not take into account the VF's EQ count while computing the queue counts. Because of this, the VF gets a large number of queues, though it doesn't have sufficient EQs, resulting in wasted queue resources. And the PF gets a smaller share of queues though it has more EQs. Fix this by capping the VF queue count at its EQ count. Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com> --- drivers/net/ethernet/emulex/benet/be.h | 4 ++++ drivers/net/ethernet/emulex/benet/be_main.c | 15 ++++++--------- 2 files changed, 10 insertions(+), 9 deletions(-)diff --git a/drivers/net/ethernet/emulex/benet/be.hb/drivers/net/ethernet/emulex/benet/be.h index cf83783..f33d73c 100644--- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h@@ -89,6 +89,10 @@ #define BE3_MAX_TX_QS 16 #define BE3_MAX_EVT_QS 16 #define BE3_SRIOV_MAX_EVT_QS 8 +#define SH_VF_MAX_NIC_EQS 3 /* Skyhawk VFs can have amax of 4 EQsOther #define's seem to have a space after them, not tab...
Thanks for catching it, I'll fix it in a subsequent patch ? Thanks, -Harsha
quoted
+ * and at least 1 is granted to either + * SURF/DPDK + */ #define MAX_RSS_IFACES 15 #define MAX_RX_QS 32[...] MBR, Sergei