* [PATCH net-next-26] cxgb4vf: Ingress Queue Entry Size needs to be 64 bytes
@ 2010-12-09 19:38 Casey Leedom
2010-12-09 21:52 ` Casey Leedom
0 siblings, 1 reply; 3+ messages in thread
From: Casey Leedom @ 2010-12-09 19:38 UTC (permalink / raw)
To: netdev; +Cc: davem, Casey Leedom
Was using L1_CACHE_BYTES for the Ingress Queue Entry Size but it really
needs to be 64 bytes in order to support the largest message sizes.
Signed-off-by: Casey Leedom <leedom@chelsio.com>
---
drivers/net/cxgb4vf/cxgb4vf_main.c | 15 +++++++++++----
1 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/drivers/net/cxgb4vf/cxgb4vf_main.c b/drivers/net/cxgb4vf/cxgb4vf_main.c
index d887a76..6bf464a 100644
--- a/drivers/net/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/cxgb4vf/cxgb4vf_main.c
@@ -2269,6 +2269,7 @@ static void __devinit cfg_queues(struct adapter *adapter)
{
struct sge *s = &adapter->sge;
int q10g, n10g, qidx, pidx, qs;
+ size_t iqe_size;
/*
* We should not be called till we know how many Queue Sets we can
@@ -2313,6 +2314,13 @@ static void __devinit cfg_queues(struct adapter *adapter)
s->ethqsets = qidx;
/*
+ * The Ingress Queue Entry Size for our various Response Queues needs
+ * to be big enough to accommodate the largest message we can receive
+ * from the chip/firmware; which is 64 bytes ...
+ */
+ iqe_size = 64;
+
+ /*
* Set up default Queue Set parameters ... Start off with the
* shortest interrupt holdoff timer.
*/
@@ -2320,7 +2328,7 @@ static void __devinit cfg_queues(struct adapter *adapter)
struct sge_eth_rxq *rxq = &s->ethrxq[qs];
struct sge_eth_txq *txq = &s->ethtxq[qs];
- init_rspq(&rxq->rspq, 0, 0, 1024, L1_CACHE_BYTES);
+ init_rspq(&rxq->rspq, 0, 0, 1024, iqe_size);
rxq->fl.size = 72;
txq->q.size = 1024;
}
@@ -2329,8 +2337,7 @@ static void __devinit cfg_queues(struct adapter *adapter)
* The firmware event queue is used for link state changes and
* notifications of TX DMA completions.
*/
- init_rspq(&s->fw_evtq, SGE_TIMER_RSTRT_CNTR, 0, 512,
- L1_CACHE_BYTES);
+ init_rspq(&s->fw_evtq, SGE_TIMER_RSTRT_CNTR, 0, 512, iqe_size);
/*
* The forwarded interrupt queue is used when we're in MSI interrupt
@@ -2346,7 +2353,7 @@ static void __devinit cfg_queues(struct adapter *adapter)
* any time ...
*/
init_rspq(&s->intrq, SGE_TIMER_RSTRT_CNTR, 0, MSIX_ENTRIES + 1,
- L1_CACHE_BYTES);
+ iqe_size);
}
/*
--
1.7.0.4
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH net-next-26] cxgb4vf: Ingress Queue Entry Size needs to be 64 bytes
2010-12-09 19:38 [PATCH net-next-26] cxgb4vf: Ingress Queue Entry Size needs to be 64 bytes Casey Leedom
@ 2010-12-09 21:52 ` Casey Leedom
2010-12-10 23:49 ` David Miller
0 siblings, 1 reply; 3+ messages in thread
From: Casey Leedom @ 2010-12-09 21:52 UTC (permalink / raw)
To: netdev; +Cc: davem
Urk. Sorry about that: the subject prefix should have read "net-26" since
this is a fairly critical bug fix. Let me know if you'd like me to resubmit the
patch.
Casey
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH net-next-26] cxgb4vf: Ingress Queue Entry Size needs to be 64 bytes
2010-12-09 21:52 ` Casey Leedom
@ 2010-12-10 23:49 ` David Miller
0 siblings, 0 replies; 3+ messages in thread
From: David Miller @ 2010-12-10 23:49 UTC (permalink / raw)
To: leedom; +Cc: netdev
From: Casey Leedom <leedom@chelsio.com>
Date: Thu, 9 Dec 2010 13:52:40 -0800
> Urk. Sorry about that: the subject prefix should have read "net-26" since
> this is a fairly critical bug fix. Let me know if you'd like me to resubmit the
> patch.
No need to resubmit, applied, thanks.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2010-12-10 23:49 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-12-09 19:38 [PATCH net-next-26] cxgb4vf: Ingress Queue Entry Size needs to be 64 bytes Casey Leedom
2010-12-09 21:52 ` Casey Leedom
2010-12-10 23:49 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).