xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

net: bcmgenet: fix throughtput regression

Browse Source 
This patch adds bcmgenet_tx_poll for the tx_rings. This can reduce the
interrupt load and send xmit in network stack on time. This also
separated for the completion of tx_ring16 from bcmgenet_poll.

The bcmgenet_tx_reclaim of tx_ring[{0,1,2,3}] operative by an interrupt
is to be not more than a certain number TxBDs. It is caused by too
slowly reclaiming the transmitted skb. Therefore, performance
degradation of xmit after 605ad7f ("tcp: refine TSO autosizing").

Signed-off-by: Jaedon Shin <jaedon.shin@gmail.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Jaedon Shin
2015-02-28 11:48:26 +09:00
committed by David S. Miller
parent 2f1d8b9e8a
commit 4092e6acf5
2 changed files with 88 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
drivers/net/ethernet/broadcom/genet/bcmgenet.h
View File

@@ -520,6 +520,7 @@ struct bcmgenet_hw_params {
struct bcmgenet_tx_ring {
spinlock_t lock; /* ring lock */
struct napi_struct napi; /* NAPI per tx queue */
unsigned int index; /* ring index */
unsigned int queue; /* queue index */
struct enet_cb *cbs; /* tx ring buffer control block*/
@@ -534,6 +535,7 @@ struct bcmgenet_tx_ring {
struct bcmgenet_tx_ring *);
void (*int_disable)(struct bcmgenet_priv *priv,
struct bcmgenet_tx_ring *);
struct bcmgenet_priv *priv;
};
/* device context */