| To: | Stephen Hemminger <shemminger@xxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH] acenic - don't spin in hard_start_xmit when ring fills |
| From: | "David S. Miller" <davem@xxxxxxxxxxxxx> |
| Date: | Thu, 16 Sep 2004 16:22:50 -0700 |
| Cc: | jes@xxxxxxxxxxxxxxxxxx, jgarzik@xxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <20040916161753.37254cbd@dell_ss3.pdx.osdl.net> |
| References: | <20040916161753.37254cbd@dell_ss3.pdx.osdl.net> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Thu, 16 Sep 2004 16:17:53 -0700
Stephen Hemminger <shemminger@xxxxxxxx> wrote:
> Running performance tests on the acenic, I noticed that the driver
> spins when the transmit ring gets full. This might have been okay when
> CPU's were slower, but it isn't the right thing to do. Better to
> return TX_BUSY and let network scheduling handle it.
Acenic does completely lockless processing, are you sure you didn't
add a race or bug?
Also, NETDEV_TX_BUSY is not a valid return value for non-LLTX drivers.
From Documentation/networking/netdevices.txt:
o NETDEV_TX_LOCKED Locking failed, please retry quickly.
Only valid when NETIF_F_LLTX is set.
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] pktgen handle netdev device getting full., David S. Miller |
|---|---|
| Next by Date: | Re: [PATCH] acenic - don't spin in hard_start_xmit when ring fills, Stephen Hemminger |
| Previous by Thread: | [PATCH] acenic - don't spin in hard_start_xmit when ring fills, Stephen Hemminger |
| Next by Thread: | Re: [PATCH] acenic - don't spin in hard_start_xmit when ring fills, Stephen Hemminger |
| Indexes: | [Date] [Thread] [Top] [All Lists] |