Thread (20 messages) 20 messages, 5 authors, 2009-03-30

Re: [PATCH 1/2] ucc_geth: Move freeing of TX packets to NAPI context.

From: Joakim Tjernlund <hidden>
Date: 2009-03-30 07:48:06
Also in: linuxppc-dev

pku.leo@gmail.com wrote on 30/03/2009 09:36:21:
On Fri, Mar 27, 2009 at 7:52 PM, Joakim Tjernlund
[off-list ref] wrote:
quoted
pku.leo@gmail.com wrote on 27/03/2009 11:50:09:
quoted
On Thu, Mar 26, 2009 at 8:54 PM, Joakim Tjernlund
[off-list ref] wrote:
quoted
Also set NAPI weight to 64 as this is a common value.
This will make the system alot more responsive while
ping flooding the ucc_geth ethernet interaface.

Signed-off-by: Joakim Tjernlund <redacted>
---
       /* Errors and other events */
       if (ucce & UCCE_OTHER) {
               if (ucce & UCC_GETH_UCCE_BSY)
@@ -3733,7 +3725,7 @@ static int ucc_geth_probe(struct of_device*
ofdev, const struct of_device_id *ma
quoted
quoted
       dev->netdev_ops = &ucc_geth_netdev_ops;
       dev->watchdog_timeo = TX_TIMEOUT;
       INIT_WORK(&ugeth->timeout_work, ucc_geth_timeout_work);
-       netif_napi_add(dev, &ugeth->napi, ucc_geth_poll,
UCC_GETH_DEV_WEIGHT);
quoted
quoted
+       netif_napi_add(dev, &ugeth->napi, ucc_geth_poll, 64);
It doesn't make sense to have larger napi budget than the size of RX
BD ring.  You can't have more BDs than RX_BD_RING_LEN in backlog for
napi_poll to process.  Increase the RX_BD_RING_LEN if you want to
increase UCC_GETH_DEV_WEIGHT.  However please also provide the
performance comparison for this kind of change.  Thanks
Bring it up with David Miller. After my initial attempt to just 
increase
quoted
weight somewhat, he requested that I hardcoded it to 64. Just read the
whole thread.
If I don't increase weight somewhat, ping -f -l 3 almost halts the 
board.
quoted
Logging
in takes forever. These are my "performance numbers".
Faster response time is surely good.  But it might also mean CPU is
not fully loaded.  IMHO, throughput is a more important factor for
network devices.  When you try to optimize the driver, please also
consider the throughput change.  Thanks.
This particular change isn't about performance, it is about not
"bricking" the board during heavy traffic. Next step is to optimize
the driver.

 Jocke
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help