netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH]bnx2x: tx_has_work should not wait for FW
@ 2009-01-26 16:55 Eilon Greenstein
  2009-01-26 20:36 ` David Miller
  0 siblings, 1 reply; 2+ messages in thread
From: Eilon Greenstein @ 2009-01-26 16:55 UTC (permalink / raw)
  To: David Miller; +Cc: netdev, Vladislav Zolotarov

The current tx_has_work waited until all packets sent by the driver are marked
as completed by the FW. This is too greedy and it causes the bnx2x_poll to spin
in vain. The driver should only check that all packets FW already completed are
freed - only in unload flow the driver should make sure that transmit queue is
empty

Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
 drivers/net/bnx2x_main.c |   17 ++++++++++++-----
 1 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index 4b84f6c..d3e7775 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -57,8 +57,8 @@
 #include "bnx2x.h"
 #include "bnx2x_init.h"

-#define DRV_MODULE_VERSION	"1.45.25"
-#define DRV_MODULE_RELDATE	"2009/01/22"
+#define DRV_MODULE_VERSION	"1.45.26"
+#define DRV_MODULE_RELDATE	"2009/01/26"
 #define BNX2X_BC_VER		0x040200

 /* Time in jiffies before concluding the transmitter is hung */
@@ -740,8 +740,15 @@ static inline int bnx2x_has_tx_work(struct bnx2x_fastpath *fp)
 	/* Tell compiler that status block fields can change */
 	barrier();
 	tx_cons_sb = le16_to_cpu(*fp->tx_cons_sb);
-	return ((fp->tx_pkt_prod != tx_cons_sb) ||
-		(fp->tx_pkt_prod != fp->tx_pkt_cons));
+	return (fp->tx_pkt_cons != tx_cons_sb);
+}
+
+static inline int bnx2x_has_tx_work_unload(struct bnx2x_fastpath *fp)
+{
+	/* Tell compiler that consumer and producer can change */
+	barrier();
+	return (fp->tx_pkt_prod != fp->tx_pkt_cons);
+
 }

 /* free skb in the packet ring at pos idx
@@ -6729,7 +6736,7 @@ static int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode)

 		cnt = 1000;
 		smp_rmb();
-		while (bnx2x_has_tx_work(fp)) {
+		while (bnx2x_has_tx_work_unload(fp)) {

 			bnx2x_tx_int(fp, 1000);
 			if (!cnt) {
--
1.5.4.3





^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH]bnx2x: tx_has_work should not wait for FW
  2009-01-26 16:55 [PATCH]bnx2x: tx_has_work should not wait for FW Eilon Greenstein
@ 2009-01-26 20:36 ` David Miller
  0 siblings, 0 replies; 2+ messages in thread
From: David Miller @ 2009-01-26 20:36 UTC (permalink / raw)
  To: eilong; +Cc: netdev, vladz

From: "Eilon Greenstein" <eilong@broadcom.com>
Date: Mon, 26 Jan 2009 18:55:56 +0200

> The current tx_has_work waited until all packets sent by the driver are marked
> as completed by the FW. This is too greedy and it causes the bnx2x_poll to spin
> in vain. The driver should only check that all packets FW already completed are
> freed - only in unload flow the driver should make sure that transmit queue is
> empty
> 
> Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
> Signed-off-by: Eilon Greenstein <eilong@broadcom.com>

Applied, thanks.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2009-01-26 20:36 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-01-26 16:55 [PATCH]bnx2x: tx_has_work should not wait for FW Eilon Greenstein
2009-01-26 20:36 ` 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).