linux-next: manual merge of the bpf-next tree with the net-next tree

Stephen Rothwell posted 1 patch 2 years, 6 months ago
There is a newer version of this series
linux-next: manual merge of the bpf-next tree with the net-next tree
Posted by Stephen Rothwell 2 years, 6 months ago
Hi all,

Today's linux-next merge of the bpf-next tree got a conflict in:

  drivers/net/ethernet/intel/ice/ice_xsk.c

between commit:

  675f176b4dcc ("Merge ra.kernel.org:/pub/scm/linux/kernel/git/netdev/net")

from the net-next tree and commit:

  aa1d3faf71a6 ("ice: Robustify cleaning/completing XDP Tx buffers")

from the bpf-next tree.

I fixed it up (I guessed - see below) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/net/ethernet/intel/ice/ice_xsk.c
index b2d96ae5668c,917c75e530ca..000000000000
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@@ -629,29 -613,28 +629,30 @@@ static void ice_clean_xdp_irq_zc(struc
  
  	last_rs = xdp_ring->next_to_use ? xdp_ring->next_to_use - 1 : cnt - 1;
  	tx_desc = ICE_TX_DESC(xdp_ring, last_rs);
 -	if (tx_desc->cmd_type_offset_bsz &
 -	    cpu_to_le64(ICE_TX_DESC_DTYPE_DESC_DONE)) {
 +	if ((tx_desc->cmd_type_offset_bsz &
 +	    cpu_to_le64(ICE_TX_DESC_DTYPE_DESC_DONE))) {
  		if (last_rs >= ntc)
 -			xsk_frames = last_rs - ntc + 1;
 +			completed_frames = last_rs - ntc + 1;
  		else
 -			xsk_frames = last_rs + cnt - ntc + 1;
 +			completed_frames = last_rs + cnt - ntc + 1;
  	}
  
 -	if (!xsk_frames)
 +	if (!completed_frames)
  		return;
  
 -	if (likely(!xdp_ring->xdp_tx_active))
 +	if (likely(!xdp_ring->xdp_tx_active)) {
 +		xsk_frames = completed_frames;
  		goto skip;
 +	}
  
  	ntc = xdp_ring->next_to_clean;
 -	for (i = 0; i < xsk_frames; i++) {
 +	for (i = 0; i < completed_frames; i++) {
  		tx_buf = &xdp_ring->tx_buf[ntc];
  
- 		if (tx_buf->raw_buf) {
+ 		if (tx_buf->type == ICE_TX_BUF_XSK_TX) {
+ 			tx_buf->type = ICE_TX_BUF_EMPTY;
 -			xsk_buff_free(tx_buf->xdp);
 -			xdp_ring->xdp_tx_active--;
 +			ice_clean_xdp_tx_buf(xdp_ring, tx_buf);
 +			tx_buf->raw_buf = NULL;
  		} else {
  			xsk_frames++;
  		}
Re: linux-next: manual merge of the bpf-next tree with the net-next tree
Posted by Alexander Lobakin 2 years, 6 months ago
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 20 Feb 2023 10:00:56 +1100

> Hi all,
> 
> Today's linux-next merge of the bpf-next tree got a conflict in:
> 
>   drivers/net/ethernet/intel/ice/ice_xsk.c
> 
> between commit:
> 
>   675f176b4dcc ("Merge ra.kernel.org:/pub/scm/linux/kernel/git/netdev/net")
> 
> from the net-next tree and commit:
> 
>   aa1d3faf71a6 ("ice: Robustify cleaning/completing XDP Tx buffers")
> 
> from the bpf-next tree.
> 
> I fixed it up (I guessed - see below) and can carry the fix as
> necessary. This is now fixed as far as linux-next is concerned, but any
> non trivial conflicts should be mentioned to your upstream maintainer
> when your tree is submitted for merging.  You may also want to consider
> cooperating with the maintainer of the conflicting tree to minimise any
> particularly complex conflicts.
> 

Hi,

the correct fix will be as follows (at least my Git yielded me this when
merging net-next with bpf-next).
Strange movements from Git, but basically the resolution is to pick the
latest ice_xsk.c from bpf-next and then manually carry Larysa's fix[0]
for @xsk_frames (she initially did it for bpf-next, so it's easier for
us to resolve this conflict).
Much thanks for noticing and notifying.

[0]
https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/?id=1f090494170ea298530cf1285fb8d078e355b4c0

Thanks,
Olek
---
diff --cc drivers/net/ethernet/intel/ice/ice_xsk.c
index b2d96ae5668c,917c75e530ca..1fcfa07c205b
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@@ -629,29 -613,28 +614,30 @@@ static void ice_clean_xdp_irq_zc(struc

  	last_rs = xdp_ring->next_to_use ? xdp_ring->next_to_use - 1 : cnt - 1;
  	tx_desc = ICE_TX_DESC(xdp_ring, last_rs);
- 	if ((tx_desc->cmd_type_offset_bsz &
- 	    cpu_to_le64(ICE_TX_DESC_DTYPE_DESC_DONE))) {
+ 	if (tx_desc->cmd_type_offset_bsz &
+ 	    cpu_to_le64(ICE_TX_DESC_DTYPE_DESC_DONE)) {
  		if (last_rs >= ntc)
 -			xsk_frames = last_rs - ntc + 1;
 +			completed_frames = last_rs - ntc + 1;
  		else
 -			xsk_frames = last_rs + cnt - ntc + 1;
 +			completed_frames = last_rs + cnt - ntc + 1;
  	}

 -	if (!xsk_frames)
 +	if (!completed_frames)
  		return;

 -	if (likely(!xdp_ring->xdp_tx_active))
 +	if (likely(!xdp_ring->xdp_tx_active)) {
 +		xsk_frames = completed_frames;
  		goto skip;
 +	}

  	ntc = xdp_ring->next_to_clean;
 -	for (i = 0; i < xsk_frames; i++) {
 +	for (i = 0; i < completed_frames; i++) {
  		tx_buf = &xdp_ring->tx_buf[ntc];

- 		if (tx_buf->raw_buf) {
- 			ice_clean_xdp_tx_buf(xdp_ring, tx_buf);
- 			tx_buf->raw_buf = NULL;
+ 		if (tx_buf->type == ICE_TX_BUF_XSK_TX) {
+ 			tx_buf->type = ICE_TX_BUF_EMPTY;
+ 			xsk_buff_free(tx_buf->xdp);
+ 			xdp_ring->xdp_tx_active--;
  		} else {
  			xsk_frames++;
  		}