Skip to contentSkip to conversation

gro: a couple of minor optimization

PATCH Applied Paolo Abeni 6 participants 9 messages Feb 4, 2022 – Feb 5, 2022

Subsystems (2)
Networking Network drivers
lore
Share & tools
lore

Paolo Abeni

author maintainer
#0 lore raw

cover gro: a couple of minor optimization

This series collects a couple of small optimizations for the GRO engine, reducing slightly the number of cycles for dev_gro_receive(). The delta is within noise range in tput tests, but with big TCP coming every cycle saved from the GRO engine will count - I hope ;)

v1 -> v2:
- a few cleanup suggested from Alexander(s)
- moved away the more controversial 3rd patch

Paolo Abeni (2):
net: gro: avoid re-computing truesize twice on recycle
net: gro: minor optimization for dev_gro_receive()

 include/net/gro.h | 52 +++++++++++++++++++++++++----------------------
 net/core/gro.c    | 16 ++++-----------
 2 files changed, 32 insertions(+), 36 deletions(-)
Signature · 1 line
2.34.1
author maintainer
#1 lore raw

PATCH 1/2 net: gro: avoid re-computing truesize twice on recycle

After commit 5e10da5385d2 ("skbuff: allow 'slow_gro' for skb carring sock reference") and commit af352460b465 ("net: fix GRO skb truesize update") the truesize of the skb with stolen head is properly updated by the GRO engine, we don't need anymore resetting it at recycle time.

v1 -> v2:
- clarify the commit message (Alexander)

Signed-off-by Paolo Abeni <[email protected]>
---
 net/core/gro.c | 1 -
 1 file changed, 1 deletion(-)
net/core/gro.c +0 -1
index a11b286d1495..d43d42215bdb 100644
@@ -634,7 +634,6 @@ static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
 
 	skb->encapsulation = 0;
 	skb_shinfo(skb)->gso_type = 0;
-	skb->truesize = SKB_TRUESIZE(skb_end_offset(skb));
 	if (unlikely(skb->slow_gro)) {
 		skb_orphan(skb);
 		skb_ext_reset(skb);
Signature · 1 line
2.34.1
author maintainer
#3 lore raw

PATCH 2/2 net: gro: minor optimization for dev_gro_receive()

In reply to Paolo Abeni

While inspecting some perf report, I noticed that the compiler emits suboptimal code for the napi CB initialization, fetching and storing multiple times the memory for flags bitfield. This is with gcc 10.3.1, but I observed the same with older compiler versions.

We can help the compiler to do a nicer work clearing several fields at once using an u32 alias. The generated code is quite smaller, with the same number of conditional.

Before:
objdump -t net/core/gro.o | grep " F .text"
0000000000000bb0 l F .text 0000000000000357 dev_gro_receive

After:
0000000000000bb0 l F .text 000000000000033c dev_gro_receive

v1 -> v2:
- use struct_group (Alexander and Alex)

RFC -> v1:
- use __struct_group to delimit the zeroed area (Alexander)

Signed-off-by Paolo Abeni <[email protected]>
---
 include/net/gro.h | 52 +++++++++++++++++++++++++----------------------
 net/core/gro.c    | 15 ++++----------
 2 files changed, 32 insertions(+), 35 deletions(-)
include/net/gro.h +28 -24
index 8f75802d50fd..a765fedda5c4 100644
@@ -29,46 +29,50 @@ struct napi_gro_cb {
 	/* Number of segments aggregated. */
 	u16	count;
 
-	/* Start offset for remote checksum offload */
-	u16	gro_remcsum_start;
+	/* Used in ipv6_gro_receive() and foo-over-udp */
+	u16	proto;
 
 	/* jiffies when first packet was created/queued */
 	unsigned long age;
 
-	/* Used in ipv6_gro_receive() and foo-over-udp */
-	u16	proto;
+	/* portion of the cb set to zero at every gro iteration */
+	struct_group(zeroed,
+
+		/* Start offset for remote checksum offload */
+		u16	gro_remcsum_start;
 
-	/* This is non-zero if the packet may be of the same flow. */
-	u8	same_flow:1;
+		/* This is non-zero if the packet may be of the same flow. */
+		u8	same_flow:1;
 
-	/* Used in tunnel GRO receive */
-	u8	encap_mark:1;
+		/* Used in tunnel GRO receive */
+		u8	encap_mark:1;
 
-	/* GRO checksum is valid */
-	u8	csum_valid:1;
+		/* GRO checksum is valid */
+		u8	csum_valid:1;
 
-	/* Number of checksums via CHECKSUM_UNNECESSARY */
-	u8	csum_cnt:3;
+		/* Number of checksums via CHECKSUM_UNNECESSARY */
+		u8	csum_cnt:3;
 
-	/* Free the skb? */
-	u8	free:2;
+		/* Free the skb? */
+		u8	free:2;
 #define NAPI_GRO_FREE		  1
 #define NAPI_GRO_FREE_STOLEN_HEAD 2
 
-	/* Used in foo-over-udp, set in udp[46]_gro_receive */
-	u8	is_ipv6:1;
+		/* Used in foo-over-udp, set in udp[46]_gro_receive */
+		u8	is_ipv6:1;
 
-	/* Used in GRE, set in fou/gue_gro_receive */
-	u8	is_fou:1;
+		/* Used in GRE, set in fou/gue_gro_receive */
+		u8	is_fou:1;
 
-	/* Used to determine if flush_id can be ignored */
-	u8	is_atomic:1;
+		/* Used to determine if flush_id can be ignored */
+		u8	is_atomic:1;
 
-	/* Number of gro_receive callbacks this packet already went through */
-	u8 recursion_counter:4;
+		/* Number of gro_receive callbacks this packet already went through */
+		u8 recursion_counter:4;
 
-	/* GRO is done by frag_list pointer chaining. */
-	u8	is_flist:1;
+		/* GRO is done by frag_list pointer chaining. */
+		u8	is_flist:1;
+	);
 
 	/* used to support CHECKSUM_COMPLETE for tunneling protocols */
 	__wsum	csum;
net/core/gro.c +4 -11
index d43d42215bdb..ee5e7e889d8b 100644
@@ -459,29 +459,22 @@ static enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff
 
 		skb_set_network_header(skb, skb_gro_offset(skb));
 		skb_reset_mac_len(skb);
-		NAPI_GRO_CB(skb)->same_flow = 0;
+		BUILD_BUG_ON(sizeof_field(struct napi_gro_cb, zeroed) != sizeof(u32));
+		BUILD_BUG_ON(!IS_ALIGNED(offsetof(struct napi_gro_cb, zeroed),
+					 sizeof(u32))); /* Avoid slow unaligned acc */
+		*(u32 *)&NAPI_GRO_CB(skb)->zeroed = 0;
 		NAPI_GRO_CB(skb)->flush = skb_is_gso(skb) || skb_has_frag_list(skb);
-		NAPI_GRO_CB(skb)->free = 0;
-		NAPI_GRO_CB(skb)->encap_mark = 0;
-		NAPI_GRO_CB(skb)->recursion_counter = 0;
-		NAPI_GRO_CB(skb)->is_fou = 0;
 		NAPI_GRO_CB(skb)->is_atomic = 1;
-		NAPI_GRO_CB(skb)->gro_remcsum_start = 0;
 
 		/* Setup for GRO checksum validation */
 		switch (skb->ip_summed) {
 		case CHECKSUM_COMPLETE:
 			NAPI_GRO_CB(skb)->csum = skb->csum;
 			NAPI_GRO_CB(skb)->csum_valid = 1;
-			NAPI_GRO_CB(skb)->csum_cnt = 0;
 			break;
 		case CHECKSUM_UNNECESSARY:
 			NAPI_GRO_CB(skb)->csum_cnt = skb->csum_level + 1;
-			NAPI_GRO_CB(skb)->csum_valid = 0;
 			break;
-		default:
-			NAPI_GRO_CB(skb)->csum_cnt = 0;
-			NAPI_GRO_CB(skb)->csum_valid = 0;
 		}
 
 		pp = INDIRECT_CALL_INET(ptype->callbacks.gro_receive,
Signature · 1 line
2.34.1
author maintainer
#4 lore raw
In reply to Paolo Abeni
Quoted 9 lines · Paolo Abeni

This series collects a couple of small optimizations for the GRO engine, reducing slightly the number of cycles for dev_gro_receive(). The delta is within noise range in tput tests, but with big TCP coming every cycle saved from the GRO engine will count - I hope ;)

v1 -> v2:
- a few cleanup suggested from Alexander(s)
- moved away the more controversial 3rd patch

Quoted 7 lines

Paolo Abeni (2):
net: gro: avoid re-computing truesize twice on recycle
net: gro: minor optimization for dev_gro_receive()

 include/net/gro.h | 52 +++++++++++++++++++++++++----------------------
 net/core/gro.c    | 16 ++++-----------
 2 files changed, 32 insertions(+), 36 deletions(-)

This is really a v2. Please let me know if you prefer a formal repost.

Thanks!

Paolo

Reviewed-by on the v1 series
#6 lore raw
In reply to Paolo Abeni

From: Paolo Abeni <[email protected]>
Date: Fri, 4 Feb 2022 12:28:35 +0100

Quoted 5 lines

This series collects a couple of small optimizations for the GRO engine, reducing slightly the number of cycles for dev_gro_receive(). The delta is within noise range in tput tests, but with big TCP coming every cycle saved from the GRO engine will count - I hope ;)

Quoted 7 lines

v1 -> v2:
- a few cleanup suggested from Alexander(s)
- moved away the more controversial 3rd patch

Paolo Abeni (2):
net: gro: avoid re-computing truesize twice on recycle
net: gro: minor optimization for dev_gro_receive()

Looks nice to me now, thanks!
For the series:

Reviewed-by Alexander Lobakin <[email protected]>
Quoted 7 lines
 include/net/gro.h | 52 +++++++++++++++++++++++++----------------------
 net/core/gro.c    | 16 ++++-----------
 2 files changed, 32 insertions(+), 36 deletions(-)
Signature · 1 line
2.34.1

Al

Reviewed-by on the v1 series
#7 lore raw
In reply to Paolo Abeni
Quoted 10 lines · Paolo Abeni

This series collects a couple of small optimizations for the GRO engine, reducing slightly the number of cycles for dev_gro_receive(). The delta is within noise range in tput tests, but with big TCP coming every cycle saved from the GRO engine will count - I hope ;)

v1 -> v2:
- a few cleanup suggested from Alexander(s)
- moved away the more controversial 3rd patch

Quoted 7 lines

Paolo Abeni (2):
net: gro: avoid re-computing truesize twice on recycle
net: gro: minor optimization for dev_gro_receive()

 include/net/gro.h | 52 +++++++++++++++++++++++++----------------------
 net/core/gro.c    | 16 ++++-----------
 2 files changed, 32 insertions(+), 36 deletions(-)

This addresses the concern I had.

Reviewed-by Alexander Duyck <[email protected]>
Reviewed-by on v1 1/2
#2 lore raw

Re: [PATCH net-next 1/2] net: gro: avoid re-computing truesize twice on recycle

In reply to Paolo Abeni
Quoted 7 lines · Paolo Abeni

After commit 5e10da5385d2 ("skbuff: allow 'slow_gro' for skb carring sock reference") and commit af352460b465 ("net: fix GRO skb truesize update") the truesize of the skb with stolen head is properly updated by the GRO engine, we don't need anymore resetting it at recycle time.

Quoted 5 lines

v1 -> v2:
- clarify the commit message (Alexander)

Signed-off-by Paolo Abeni <[email protected]>

---

Reviewed-by Eric Dumazet <[email protected]>
maintainer
#5 lore raw
In reply to Paolo Abeni

From: Paolo Abeni <[email protected]>
Date: Fri, 04 Feb 2022 12:34:03 +0100

Quoted 14 lines · Paolo Abeni
Quoted 13 lines

This series collects a couple of small optimizations for the GRO engine, reducing slightly the number of cycles for dev_gro_receive(). The delta is within noise range in tput tests, but with big TCP coming every cycle saved from the GRO engine will count - I hope ;)

v1 -> v2:
- a few cleanup suggested from Alexander(s)
- moved away the more controversial 3rd patch

Paolo Abeni (2):
net: gro: avoid re-computing truesize twice on recycle
net: gro: minor optimization for dev_gro_receive()

Quoted 5 lines
Quoted 3 lines
 include/net/gro.h | 52 +++++++++++++++++++++++++----------------------
 net/core/gro.c    | 16 ++++-----------
 2 files changed, 32 insertions(+), 36 deletions(-)

This is really a v2. Please let me know if you prefer a formal repost.

Not necessary.

Thank you.

CI Applied
#8 lore raw
In reply to Paolo Abeni
This series was applied to netdev/net-next.git (master) by David S. Miller <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (master) by David S. Miller <[email protected]>:

Quoted 10 lines · On Fri, 4 Feb 2022 12:28:35 +0100 you wrote:

This series collects a couple of small optimizations for the GRO engine, reducing slightly the number of cycles for dev_gro_receive(). The delta is within noise range in tput tests, but with big TCP coming every cycle saved from the GRO engine will count - I hope ;)

v1 -> v2:
- a few cleanup suggested from Alexander(s)
- moved away the more controversial 3rd patch

[...]

Here is the summary with links:
- [net-next,1/2] net: gro: avoid re-computing truesize twice on recycle

    https://git.kernel.org/netdev/net-next/c/7881453e4adf
  - [net-next,2/2] net: gro: minor optimization for dev_gro_receive()
    https://git.kernel.org/netdev/net-next/c/de5a1f3ce4c8

You are awesome, thank you!

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