Skip to contentSkip to conversation

can: grcan: move napi_enable() from under spin lock

PATCH Applied Jakub Kicinski 5 participants 5 messages Jan 11, 2025 – Jan 14, 2025

Subsystems (3)
lore
Share & tools
lore

Jakub Kicinski

author maintainer
#0 lore raw

PATCH 1/1 can: grcan: move napi_enable() from under spin lock

I don't see any reason why napi_enable() needs to be under the lock, only reason I could think of is if the IRQ also took this lock but it doesn't. napi_enable() will soon need to sleep.

---
Marc, if this is correct is it okay for me to take via net-next directly? I have a bunch of patches which depend on it.

---
 drivers/net/can/grcan.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/can/grcan.c +2 -1
index cdf0ec9fa7f3..21a61b86f67d 100644
@@ -1073,9 +1073,10 @@ static int grcan_open(struct net_device *dev)
 	if (err)
 		goto exit_close_candev;
 
+	napi_enable(&priv->napi);
+
 	spin_lock_irqsave(&priv->lock, flags);
 
-	napi_enable(&priv->napi);
 	grcan_start(dev);
 	if (!(priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY))
 		netif_start_queue(dev);
Signature · 1 line
2.47.1
Acked-by on v1 1/1
#1 lore raw
Quoted 5 lines · Jakub Kicinski

I don't see any reason why napi_enable() needs to be under the lock, only reason I could think of is if the IRQ also took this lock but it doesn't. napi_enable() will soon need to sleep.

Quoted 3 lines

---
Marc, if this is correct is it okay for me to take via net-next directly? I have a bunch of patches which depend on it.

Even if the question is not addressed to me, I am personally fine if this directly goes into net-next.

Quoted 13 lines
---
 drivers/net/can/grcan.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/can/grcan.c +0 -0
index cdf0ec9fa7f3..21a61b86f67d 100644
@@ -1073,9 +1073,10 @@ static int grcan_open(struct net_device *dev)
 	if (err)

drivers/net/can/grcan.c static int grcan_open(struct net_device *dev) · line 1082

Quoted 10 lines
drivers/net/can/grcan.c +2 -1
 		goto exit_close_candev;
 
+	napi_enable(&priv->napi);
+
 	spin_lock_irqsave(&priv->lock, flags);
 
-	napi_enable(&priv->napi);
 	grcan_start(dev);
 	if (!(priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY))
 		netif_start_queue(dev);

Yours sincerely,
Vincent Mailhol

Reviewed-by on v1 1/1
#2 lore raw
In reply to Jakub Kicinski

Jakub Kicinski <[email protected]> :

Quoted 3 lines

I don't see any reason why napi_enable() needs to be under the lock, only reason I could think of is if the IRQ also took this lock but it doesn't. napi_enable() will soon need to sleep.

Anything that depends on the napi handler being run may also behave differently because of 'priv->resetting = false;' and 'priv->closing = false;' also done under the lock after napi_enable in the original version.

Both priv->closing and priv->resetting are always accessed with lock held so it's fine.

(nothing to chew in grcan_start either)

Signature · 1 line
Ueimor
maintainer Acked-by on v1 1/1
#3 lore raw
In reply to Jakub Kicinski
Quoted 8 lines · Jakub Kicinski

I don't see any reason why napi_enable() needs to be under the lock, only reason I could think of is if the IRQ also took this lock but it doesn't. napi_enable() will soon need to sleep.

---
Marc, if this is correct is it okay for me to take via net-next directly? I have a bunch of patches which depend on it.

Yes, please take it via net-next directly.

Marc

Signature · 4 lines
Pengutronix e.K.                 | Marc Kleine-Budde          |
Embedded Linux                   | https://www.pengutronix.de |
Vertretung Nürnberg              | Phone: +49-5121-206917-129 |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-9   |
#4 lore raw
In reply to Jakub Kicinski
This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 10 lines · On Fri, 10 Jan 2025 18:47:42 -0800 you wrote:

I don't see any reason why napi_enable() needs to be under the lock, only reason I could think of is if the IRQ also took this lock but it doesn't. napi_enable() will soon need to sleep.

---
Marc, if this is correct is it okay for me to take via net-next directly? I have a bunch of patches which depend on it.

[...]

Here is the summary with links:
- [net-next] can: grcan: move napi_enable() from under spin lock

    https://git.kernel.org/netdev/net-next/c/7c125d5b767b

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html