netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx
@ 2023-12-21 14:07 Eric Dumazet
  2023-12-23 16:16 ` Simon Horman
  2024-01-02 12:40 ` patchwork-bot+netdevbpf
  0 siblings, 2 replies; 4+ messages in thread
From: Eric Dumazet @ 2023-12-21 14:07 UTC (permalink / raw)
  To: David S . Miller, Jakub Kicinski, Paolo Abeni
  Cc: Willem de Bruijn, netdev, eric.dumazet, Eric Dumazet, Coco Li,
	David Ahern

dev->gso_partial_features is read from tx fast path for GSO packets.

Move it to appropriate section to avoid a cache line miss.

Fixes: 43a71cd66b9c ("net-device: reorganize net_device fast path variables")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Coco Li <lixiaoyan@google.com>
Cc: David Ahern <dsahern@kernel.org>
---
 Documentation/networking/net_cachelines/net_device.rst | 2 +-
 include/linux/netdevice.h                              | 2 +-
 net/core/dev.c                                         | 3 ++-
 3 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/net_cachelines/net_device.rst b/Documentation/networking/net_cachelines/net_device.rst
index 6cab1b797739f57f15962dc0daccc2bd6aafeb97..2dd8d8f20da2558fddcc341f3a8a27da3c4a1796 100644
--- a/Documentation/networking/net_cachelines/net_device.rst
+++ b/Documentation/networking/net_cachelines/net_device.rst
@@ -38,7 +38,7 @@ netdev_features_t                   wanted_features
 netdev_features_t                   vlan_features                                                   
 netdev_features_t                   hw_enc_features         -                   -                   netif_skb_features
 netdev_features_t                   mpls_features                                                   
-netdev_features_t                   gso_partial_features                                            
+netdev_features_t                   gso_partial_features    read_mostly                             gso_features_check
 unsigned_int                        min_mtu                                                         
 unsigned_int                        max_mtu                                                         
 unsigned_short                      type                                                            
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 75c7725e5e4fdf59da55923cd803e084956b0fa0..5d1ec780122919c31e4215358d736aef3f8a0acd 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2114,6 +2114,7 @@ struct net_device {
 	const struct net_device_ops *netdev_ops;
 	const struct header_ops *header_ops;
 	struct netdev_queue	*_tx;
+	netdev_features_t	gso_partial_features;
 	unsigned int		real_num_tx_queues;
 	unsigned int		gso_max_size;
 	unsigned int		gso_ipv4_max_size;
@@ -2210,7 +2211,6 @@ struct net_device {
 	netdev_features_t	vlan_features;
 	netdev_features_t	hw_enc_features;
 	netdev_features_t	mpls_features;
-	netdev_features_t	gso_partial_features;
 
 	unsigned int		min_mtu;
 	unsigned int		max_mtu;
diff --git a/net/core/dev.c b/net/core/dev.c
index b875040783209e95afb92217a0a07ede42a2e425..0e18a802252b35948dad5d739c07d4a06b466eb8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11623,6 +11623,7 @@ static void __init net_dev_struct_check(void)
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, gso_max_size);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, gso_ipv4_max_size);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, gso_max_segs);
+	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, gso_partial_features);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, num_tc);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, mtu);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, needed_headroom);
@@ -11636,7 +11637,7 @@ static void __init net_dev_struct_check(void)
 #ifdef CONFIG_NET_XGRESS
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_tx, tcx_egress);
 #endif
-	CACHELINE_ASSERT_GROUP_SIZE(struct net_device, net_device_read_tx, 152);
+	CACHELINE_ASSERT_GROUP_SIZE(struct net_device, net_device_read_tx, 160);
 
 	/* TXRX read-mostly hotpath */
 	CACHELINE_ASSERT_GROUP_MEMBER(struct net_device, net_device_read_txrx, flags);
-- 
2.43.0.472.g3155946c3a-goog


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx
  2023-12-21 14:07 [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx Eric Dumazet
@ 2023-12-23 16:16 ` Simon Horman
  2024-01-02 15:50   ` Eric Dumazet
  2024-01-02 12:40 ` patchwork-bot+netdevbpf
  1 sibling, 1 reply; 4+ messages in thread
From: Simon Horman @ 2023-12-23 16:16 UTC (permalink / raw)
  To: Eric Dumazet
  Cc: David S . Miller, Jakub Kicinski, Paolo Abeni, Willem de Bruijn,
	netdev, eric.dumazet, Coco Li, David Ahern

On Thu, Dec 21, 2023 at 02:07:47PM +0000, Eric Dumazet wrote:
> dev->gso_partial_features is read from tx fast path for GSO packets.
> 
> Move it to appropriate section to avoid a cache line miss.
> 
> Fixes: 43a71cd66b9c ("net-device: reorganize net_device fast path variables")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Coco Li <lixiaoyan@google.com>
> Cc: David Ahern <dsahern@kernel.org>

Thanks Eric,

FWIIW, this change looks good to me.

Reviewed-by: Simon Horman <horms@kernel.org>

I have a follow-up question below.

...

> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> index 75c7725e5e4fdf59da55923cd803e084956b0fa0..5d1ec780122919c31e4215358d736aef3f8a0acd 100644
> --- a/include/linux/netdevice.h
> +++ b/include/linux/netdevice.h
> @@ -2114,6 +2114,7 @@ struct net_device {
>  	const struct net_device_ops *netdev_ops;
>  	const struct header_ops *header_ops;
>  	struct netdev_queue	*_tx;
> +	netdev_features_t	gso_partial_features;
>  	unsigned int		real_num_tx_queues;
>  	unsigned int		gso_max_size;
>  	unsigned int		gso_ipv4_max_size;

While looking at this I came to wonder if it would
be worth adding a 16bit pad a little below this hunk
so that tc_to_txq sits on it's own cacheline.

I'm unsure if the access pattern of tc_to_txq makes this worthwhile.
But if so it would be a simple tweak.

With such a change in place, on top of your patch, the diff of pahole output
on x86_64 is:

@@ -7432,10 +7432,9 @@
        s16                        num_tc;               /*    54     2 */
        unsigned int               mtu;                  /*    56     4 */
        short unsigned int         needed_headroom;      /*    60     2 */
-       struct netdev_tc_txq       tc_to_txq[16];        /*    62    64 */
-
-       /* XXX 2 bytes hole, try to pack */
-
+       u16                        pad1;                 /*    62     2 */
+       /* --- cacheline 1 boundary (64 bytes) --- */
+       struct netdev_tc_txq       tc_to_txq[16];        /*    64    64 */
        /* --- cacheline 2 boundary (128 bytes) --- */
        struct xps_dev_maps *      xps_maps[2];          /*   128    16 */
        struct nf_hook_entries *   nf_hooks_egress;      /*   144     8 */

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx
  2023-12-21 14:07 [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx Eric Dumazet
  2023-12-23 16:16 ` Simon Horman
@ 2024-01-02 12:40 ` patchwork-bot+netdevbpf
  1 sibling, 0 replies; 4+ messages in thread
From: patchwork-bot+netdevbpf @ 2024-01-02 12:40 UTC (permalink / raw)
  To: Eric Dumazet
  Cc: davem, kuba, pabeni, willemb, netdev, eric.dumazet, lixiaoyan,
	dsahern

Hello:

This patch was applied to netdev/net-next.git (main)
by David S. Miller <davem@davemloft.net>:

On Thu, 21 Dec 2023 14:07:47 +0000 you wrote:
> dev->gso_partial_features is read from tx fast path for GSO packets.
> 
> Move it to appropriate section to avoid a cache line miss.
> 
> Fixes: 43a71cd66b9c ("net-device: reorganize net_device fast path variables")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Coco Li <lixiaoyan@google.com>
> Cc: David Ahern <dsahern@kernel.org>
> 
> [...]

Here is the summary with links:
  - [net-next] net-device: move gso_partial_features to net_device_read_tx
    https://git.kernel.org/netdev/net-next/c/993498e537af

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



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx
  2023-12-23 16:16 ` Simon Horman
@ 2024-01-02 15:50   ` Eric Dumazet
  0 siblings, 0 replies; 4+ messages in thread
From: Eric Dumazet @ 2024-01-02 15:50 UTC (permalink / raw)
  To: Simon Horman
  Cc: David S . Miller, Jakub Kicinski, Paolo Abeni, Willem de Bruijn,
	netdev, eric.dumazet, Coco Li, David Ahern

On Sat, Dec 23, 2023 at 5:16 PM Simon Horman <horms@kernel.org> wrote:
>
> On Thu, Dec 21, 2023 at 02:07:47PM +0000, Eric Dumazet wrote:
> > dev->gso_partial_features is read from tx fast path for GSO packets.
> >
> > Move it to appropriate section to avoid a cache line miss.
> >
> > Fixes: 43a71cd66b9c ("net-device: reorganize net_device fast path variables")
> > Signed-off-by: Eric Dumazet <edumazet@google.com>
> > Cc: Coco Li <lixiaoyan@google.com>
> > Cc: David Ahern <dsahern@kernel.org>
>
> Thanks Eric,
>
> FWIIW, this change looks good to me.
>
> Reviewed-by: Simon Horman <horms@kernel.org>
>
> I have a follow-up question below.
>
> ...
>
> > diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> > index 75c7725e5e4fdf59da55923cd803e084956b0fa0..5d1ec780122919c31e4215358d736aef3f8a0acd 100644
> > --- a/include/linux/netdevice.h
> > +++ b/include/linux/netdevice.h
> > @@ -2114,6 +2114,7 @@ struct net_device {
> >       const struct net_device_ops *netdev_ops;
> >       const struct header_ops *header_ops;
> >       struct netdev_queue     *_tx;
> > +     netdev_features_t       gso_partial_features;
> >       unsigned int            real_num_tx_queues;
> >       unsigned int            gso_max_size;
> >       unsigned int            gso_ipv4_max_size;
>
> While looking at this I came to wonder if it would
> be worth adding a 16bit pad a little below this hunk
> so that tc_to_txq sits on it's own cacheline.
>

Hi Simon, thanks for the suggestion.

I am still working on struct net_device layout (after being off for
last ~10 days), I will try to address your feedback soon.


> I'm unsure if the access pattern of tc_to_txq makes this worthwhile.
> But if so it would be a simple tweak.
>
> With such a change in place, on top of your patch, the diff of pahole output
> on x86_64 is:
>
> @@ -7432,10 +7432,9 @@
>         s16                        num_tc;               /*    54     2 */
>         unsigned int               mtu;                  /*    56     4 */
>         short unsigned int         needed_headroom;      /*    60     2 */
> -       struct netdev_tc_txq       tc_to_txq[16];        /*    62    64 */
> -
> -       /* XXX 2 bytes hole, try to pack */
> -
> +       u16                        pad1;                 /*    62     2 */
> +       /* --- cacheline 1 boundary (64 bytes) --- */
> +       struct netdev_tc_txq       tc_to_txq[16];        /*    64    64 */
>         /* --- cacheline 2 boundary (128 bytes) --- */
>         struct xps_dev_maps *      xps_maps[2];          /*   128    16 */
>         struct nf_hook_entries *   nf_hooks_egress;      /*   144     8 */

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2024-01-02 15:50 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-12-21 14:07 [PATCH net-next] net-device: move gso_partial_features to net_device_read_tx Eric Dumazet
2023-12-23 16:16 ` Simon Horman
2024-01-02 15:50   ` Eric Dumazet
2024-01-02 12:40 ` patchwork-bot+netdevbpf

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).