From: Nazim Amirul <muhammad.nazim.amirul.nazle.asmade@altera.com>

Calling hrtimer_start() on an already-active txtimer is unnecessary
and expensive. Skip the restart if the timer is already active by
adding an hrtimer_active() check before hrtimer_start().

Previously, each packet reset the timer to tx_coal_timer in the future,
acting as a sliding window that delayed NAPI under burst traffic. With
this change, an already-active timer is left to fire sooner, scheduling
NAPI within tx_coal_timer of the first packet and freeing TX descriptors
earlier.

There is no race concern: hrtimer_start() is internally serialized and
safe to call on an active timer. In the event of a race between
hrtimer_active() and hrtimer_start(), the worst case is calling
hrtimer_start() on an already-active timer, which is identical to the
pre-patch behaviour.

Performance on Cyclone V with dwmac-socfpga (iperf3 -u -b 0 -l 64):
  Before: ~45200 pps
  After:  ~52300 pps (~15% improvement)

Additionally, ~10% improvement in UDP throughput observed on Agilex5,
with hrtimer CPU usage reduced from ~8% to ~0.6%.

Signed-off-by: Rohan G Thomas <rohan.g.thomas@altera.com>
Tested-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Nazim Amirul <muhammad.nazim.amirul.nazle.asmade@altera.com>
---
Changes in v3:
 - Corrected commit message to accurately describe timer behaviour change:
   timer now fires sooner under burst traffic, not "unchanged" (Andrew Lunn)
 - Added Reviewed-by from Jacob Keller

Changes in v2:
 - Expanded commit message to address race condition concern and clarify
   tx_coal_timer semantics (Andrew Lunn)
 - Added performance numbers to commit message (Andrew Lunn)
 - Added Agilex5 performance data with hrtimer CPU usage improvement
 - Added Tested-by and Reviewed-by from Maxime Chevallier
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions()

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 3591755ea30b..35da51c26248 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -3341,12 +3341,14 @@ static void stmmac_tx_timer_arm(struct stmmac_priv *priv, u32 queue)
 	 * Try to cancel any timer if napi is scheduled, timer will be armed
 	 * again in the next scheduled napi.
 	 */
-	if (unlikely(!napi_is_scheduled(napi)))
-		hrtimer_start(&tx_q->txtimer,
-			      STMMAC_COAL_TIMER(tx_coal_timer),
-			      HRTIMER_MODE_REL);
-	else
+	if (unlikely(!napi_is_scheduled(napi))) {
+		if (unlikely(!(hrtimer_active(&tx_q->txtimer))))
+			hrtimer_start(&tx_q->txtimer,
+				      STMMAC_COAL_TIMER(tx_coal_timer),
+				      HRTIMER_MODE_REL);
+	} else {
 		hrtimer_try_to_cancel(&tx_q->txtimer);
+	}
 }
 
 /**
-- 
2.43.7

On Thu, 28 May 2026 23:46:59 -0700
muhammad.nazim.amirul.nazle.asmade@altera.com wrote:
> There is no race concern: hrtimer_start() is internally serialized and
> safe to call on an active timer. In the event of a race between
> hrtimer_active() and hrtimer_start(), the worst case is calling
> hrtimer_start() on an already-active timer, which is identical to the
> pre-patch behaviour.

I mean.. I wouldn't say that this is not racy. But I think you're not
making it much worse than it already was :S

Could you experiment with deleting all this local driver logic and just
relying on NAPI's built-in defer IRQ mechanism?

On Thu, May 28, 2026 at 11:46:59PM -0700, muhammad.nazim.amirul.nazle.asmade@altera.com wrote:
> From: Nazim Amirul <muhammad.nazim.amirul.nazle.asmade@altera.com>
> 
> Calling hrtimer_start() on an already-active txtimer is unnecessary
> and expensive. Skip the restart if the timer is already active by
> adding an hrtimer_active() check before hrtimer_start().
> 
> Previously, each packet reset the timer to tx_coal_timer in the future,
> acting as a sliding window that delayed NAPI under burst traffic. With
> this change, an already-active timer is left to fire sooner, scheduling
> NAPI within tx_coal_timer of the first packet and freeing TX descriptors
> earlier.
> 
> There is no race concern: hrtimer_start() is internally serialized and
> safe to call on an active timer. In the event of a race between
> hrtimer_active() and hrtimer_start(), the worst case is calling
> hrtimer_start() on an already-active timer, which is identical to the
> pre-patch behaviour.
> 
> Performance on Cyclone V with dwmac-socfpga (iperf3 -u -b 0 -l 64):
>   Before: ~45200 pps
>   After:  ~52300 pps (~15% improvement)
> 
> Additionally, ~10% improvement in UDP throughput observed on Agilex5,
> with hrtimer CPU usage reduced from ~8% to ~0.6%.
> 
> Signed-off-by: Rohan G Thomas <rohan.g.thomas@altera.com>
> Tested-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Signed-off-by: Nazim Amirul <muhammad.nazim.amirul.nazle.asmade@altera.com>

Reviewed-by: Andrew Lunn <andrew@lunn.ch>

    Andrew