forcedeth: tx max work
This patch adds a limit to how much tx work can be done in each iteration of tx processing. If the max limit is reached, remaining tx completions will be handled by timer interrupt. Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
This commit is contained in:
parent
f0734ab658
commit
4e16ed1b0e
|
@ -210,7 +210,7 @@ enum {
|
||||||
* NVREG_POLL_DEFAULT=97 would result in an interval length of 1 ms
|
* NVREG_POLL_DEFAULT=97 would result in an interval length of 1 ms
|
||||||
*/
|
*/
|
||||||
NvRegPollingInterval = 0x00c,
|
NvRegPollingInterval = 0x00c,
|
||||||
#define NVREG_POLL_DEFAULT_THROUGHPUT 970
|
#define NVREG_POLL_DEFAULT_THROUGHPUT 970 /* backup tx cleanup if loop max reached */
|
||||||
#define NVREG_POLL_DEFAULT_CPU 13
|
#define NVREG_POLL_DEFAULT_CPU 13
|
||||||
NvRegMSIMap0 = 0x020,
|
NvRegMSIMap0 = 0x020,
|
||||||
NvRegMSIMap1 = 0x024,
|
NvRegMSIMap1 = 0x024,
|
||||||
|
@ -1859,14 +1859,15 @@ static void nv_tx_done(struct net_device *dev)
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
static void nv_tx_done_optimized(struct net_device *dev)
|
static void nv_tx_done_optimized(struct net_device *dev, int limit)
|
||||||
{
|
{
|
||||||
struct fe_priv *np = netdev_priv(dev);
|
struct fe_priv *np = netdev_priv(dev);
|
||||||
u32 flags;
|
u32 flags;
|
||||||
struct ring_desc_ex* orig_get_tx = np->get_tx.ex;
|
struct ring_desc_ex* orig_get_tx = np->get_tx.ex;
|
||||||
|
|
||||||
while ((np->get_tx.ex != np->put_tx.ex) &&
|
while ((np->get_tx.ex != np->put_tx.ex) &&
|
||||||
!((flags = le32_to_cpu(np->get_tx.ex->flaglen)) & NV_TX_VALID)) {
|
!((flags = le32_to_cpu(np->get_tx.ex->flaglen)) & NV_TX_VALID) &&
|
||||||
|
(limit-- > 0)) {
|
||||||
|
|
||||||
dprintk(KERN_DEBUG "%s: nv_tx_done_optimized: flags 0x%x.\n",
|
dprintk(KERN_DEBUG "%s: nv_tx_done_optimized: flags 0x%x.\n",
|
||||||
dev->name, flags);
|
dev->name, flags);
|
||||||
|
@ -1973,7 +1974,7 @@ static void nv_tx_timeout(struct net_device *dev)
|
||||||
if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2)
|
if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2)
|
||||||
nv_tx_done(dev);
|
nv_tx_done(dev);
|
||||||
else
|
else
|
||||||
nv_tx_done_optimized(dev);
|
nv_tx_done_optimized(dev, np->tx_ring_size);
|
||||||
|
|
||||||
/* 3) if there are dead entries: clear everything */
|
/* 3) if there are dead entries: clear everything */
|
||||||
if (np->get_tx_ctx != np->put_tx_ctx) {
|
if (np->get_tx_ctx != np->put_tx_ctx) {
|
||||||
|
@ -2899,7 +2900,7 @@ static irqreturn_t nv_nic_irq_optimized(int foo, void *data)
|
||||||
break;
|
break;
|
||||||
|
|
||||||
spin_lock(&np->lock);
|
spin_lock(&np->lock);
|
||||||
nv_tx_done_optimized(dev);
|
nv_tx_done_optimized(dev, TX_WORK_PER_LOOP);
|
||||||
spin_unlock(&np->lock);
|
spin_unlock(&np->lock);
|
||||||
|
|
||||||
#ifdef CONFIG_FORCEDETH_NAPI
|
#ifdef CONFIG_FORCEDETH_NAPI
|
||||||
|
@ -3006,7 +3007,7 @@ static irqreturn_t nv_nic_irq_tx(int foo, void *data)
|
||||||
break;
|
break;
|
||||||
|
|
||||||
spin_lock_irqsave(&np->lock, flags);
|
spin_lock_irqsave(&np->lock, flags);
|
||||||
nv_tx_done_optimized(dev);
|
nv_tx_done_optimized(dev, TX_WORK_PER_LOOP);
|
||||||
spin_unlock_irqrestore(&np->lock, flags);
|
spin_unlock_irqrestore(&np->lock, flags);
|
||||||
|
|
||||||
if (unlikely(events & (NVREG_IRQ_TX_ERR))) {
|
if (unlikely(events & (NVREG_IRQ_TX_ERR))) {
|
||||||
|
@ -3163,6 +3164,11 @@ static irqreturn_t nv_nic_irq_other(int foo, void *data)
|
||||||
if (!(events & np->irqmask))
|
if (!(events & np->irqmask))
|
||||||
break;
|
break;
|
||||||
|
|
||||||
|
/* check tx in case we reached max loop limit in tx isr */
|
||||||
|
spin_lock_irqsave(&np->lock, flags);
|
||||||
|
nv_tx_done_optimized(dev, TX_WORK_PER_LOOP);
|
||||||
|
spin_unlock_irqrestore(&np->lock, flags);
|
||||||
|
|
||||||
if (events & NVREG_IRQ_LINK) {
|
if (events & NVREG_IRQ_LINK) {
|
||||||
spin_lock_irqsave(&np->lock, flags);
|
spin_lock_irqsave(&np->lock, flags);
|
||||||
nv_link_irq(dev);
|
nv_link_irq(dev);
|
||||||
|
|
Loading…
Reference in a new issue