[PATCH net-next 0/2] enetc: trivial PTP one-step TX timestamping cleanups

[PATCH net-next 0/2] enetc: trivial PTP one-step TX timestamping cleanups

[Vladimir Oltean]

These are two cleanup patches for some inconsistencies I noticed in the
driver's TX ring cleanup function.

Vladimir Oltean (2):
  net: enetc: remove local "priv" variable in enetc_clean_tx_ring()
  net: enetc: use the skb variable directly in enetc_clean_tx_ring()

 drivers/net/ethernet/freescale/enetc/enetc.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

-- 
2.25.1

[PATCH net-next 1/2] net: enetc: remove local "priv" variable in enetc_clean_tx_ring()

[Vladimir Oltean]

The "priv" variable is needed in the "check_writeback" scope since
commit d39823121911 ("enetc: add hardware timestamping support").

Since commit 7294380c5211 ("enetc: support PTP Sync packet one-step
timestamping"), we also need "priv" in the larger function scope.

So the local variable from the "if" block scope is not needed, and
actually shadows the other one. Delete it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
 drivers/net/ethernet/freescale/enetc/enetc.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 8e31fe15bf3c..082b94e482ce 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -813,10 +813,7 @@ static bool enetc_clean_tx_ring(struct enetc_bdr *tx_ring, int napi_budget)
 		bool is_eof = tx_swbd->is_eof;
 
 		if (unlikely(tx_swbd->check_wb)) {
-			struct enetc_ndev_priv *priv = netdev_priv(ndev);
-			union enetc_tx_bd *txbd;
-
-			txbd = ENETC_TXBD(*tx_ring, i);
+			union enetc_tx_bd *txbd = ENETC_TXBD(*tx_ring, i);
 
 			if (txbd->flags & ENETC_TXBD_FLAGS_W &&
 			    tx_swbd->do_twostep_tstamp) {
-- 
2.25.1

[PATCH net-next 2/2] net: enetc: use the skb variable directly in enetc_clean_tx_ring()

[Vladimir Oltean]

The code checks whether the skb had one-step TX timestamping enabled, in
order to schedule the work item for emptying the priv->tx_skbs queue.

That code checks for "tx_swbd->skb" directly, when we already had a skb
retrieved using enetc_tx_swbd_get_skb(tx_swbd) - a TX software BD can
also hold an XDP_TX packet or an XDP frame. But since the direct tx_swbd
dereference is in an "if" block guarded by the non-NULL quality of
"skb", accessing "tx_swbd->skb" directly is not wrong, just confusing.

Just use the local variable named "skb".

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
 drivers/net/ethernet/freescale/enetc/enetc.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 082b94e482ce..504e12554079 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -831,8 +831,7 @@ static bool enetc_clean_tx_ring(struct enetc_bdr *tx_ring, int napi_budget)
 		if (xdp_frame) {
 			xdp_return_frame(xdp_frame);
 		} else if (skb) {
-			if (unlikely(tx_swbd->skb->cb[0] &
-				     ENETC_F_TX_ONESTEP_SYNC_TSTAMP)) {
+			if (unlikely(skb->cb[0] & ENETC_F_TX_ONESTEP_SYNC_TSTAMP)) {
 				/* Start work to release lock for next one-step
 				 * timestamping packet. And send one skb in
 				 * tx_skbs queue if has.
-- 
2.25.1

RE: [PATCH net-next 0/2] enetc: trivial PTP one-step TX timestamping cleanups

[Claudiu Manoil]

> -----Original Message-----
> From: Vladimir Oltean <vladimir.oltean@nxp.com>
> Sent: Wednesday, October 20, 2021 8:42 PM
[...]
> Subject: [PATCH net-next 0/2] enetc: trivial PTP one-step TX timestamping
> cleanups
> 
> These are two cleanup patches for some inconsistencies I noticed in the
> driver's TX ring cleanup function.
> 
> Vladimir Oltean (2):
>   net: enetc: remove local "priv" variable in enetc_clean_tx_ring()
>   net: enetc: use the skb variable directly in enetc_clean_tx_ring()
> 
>  drivers/net/ethernet/freescale/enetc/enetc.c | 8 ++------
>  1 file changed, 2 insertions(+), 6 deletions(-)
> 
> --

Reviewed-by: <Claudiu Manoil <claudiu.manoil@nxp.com>

Re: [PATCH net-next 0/2] enetc: trivial PTP one-step TX timestamping cleanups

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (master)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 20 Oct 2021 20:42:18 +0300 you wrote:
> These are two cleanup patches for some inconsistencies I noticed in the
> driver's TX ring cleanup function.
> 
> Vladimir Oltean (2):
>   net: enetc: remove local "priv" variable in enetc_clean_tx_ring()
>   net: enetc: use the skb variable directly in enetc_clean_tx_ring()
> 
> [...]

Here is the summary with links:
  - [net-next,1/2] net: enetc: remove local "priv" variable in enetc_clean_tx_ring()
    https://git.kernel.org/netdev/net-next/c/ae77bdbc2fc6
  - [net-next,2/2] net: enetc: use the skb variable directly in enetc_clean_tx_ring()
    https://git.kernel.org/netdev/net-next/c/520661495409

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html