* [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).