diff mbox series

[v3,net-next,1/2] net: mana: Allow tso_max_size to go up-to GSO_MAX_SIZE

Message ID 1739763746-28637-1-git-send-email-shradhagupta@linux.microsoft.com (mailing list archive)
State Accepted
Commit 27315836f4bcc8e4879d50dfc1fa6eb41e7952ef
Delegated to: Netdev Maintainers
Headers show
Series Enable Big TCP for MANA devices | expand

Checks

Context Check Description
netdev/series_format success Posting correctly formatted
netdev/tree_selection success Clearly marked for net-next
netdev/ynl success Generated files up to date; no warnings/errors; no diff in generated;
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 0 this patch: 0
netdev/build_tools success No tools touched, skip
netdev/cc_maintainers success CCed 11 of 11 maintainers
netdev/build_clang success Errors and warnings before: 1 this patch: 1
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 0 this patch: 0
netdev/checkpatch success total: 0 errors, 0 warnings, 0 checks, 17 lines checked
netdev/build_clang_rust success No Rust files in patch. Skipping build
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0
netdev/contest success net-next-2025-02-18--00-00 (tests: 891)

Commit Message

Shradha Gupta Feb. 17, 2025, 3:42 a.m. UTC
Allow the max aggregated pkt size to go up-to GSO_MAX_SIZE for MANA NIC.
This patch only increases the max allowable gso/gro pkt size for MANA
devices and does not change the defaults.
Following are the perf benefits by increasing the pkt aggregate size from
legacy gso_max_size value(64K) to newer one(up-to 511K

IPv4 tests
for i in {1..10}; do netperf -t TCP_RR  -H 10.0.0.5 -p50000 -- -r80000,80000
-O MIN_LATENCY,P90_LATENCY,P99_LATENCY,THROUGHPUT|tail -1; done

min	p90	p99	Throughput		gso_max_size
93	171	194	6594.25
97	154	180	7183.74
95	165	189	6927.86
96	165	188	6976.04
93	154	185	7338.05			64K
93	168	189	6938.03
94	169	189	6784.93
92	166	189	7117.56
94	179	191	6678.44
95	157	183	7277.81

min	p90	p99	Throughput
93	134	146	8448.75
95	134	140	8396.54
94	137	148	8204.12
94	137	148	8244.41
94	128	139	8666.52			80K
94	141	153	8116.86
94	138	149	8163.92
92	135	142	8362.72
92	134	142	8497.57
93	136	148	8393.23

IPv6 Tests
for i in {1..10}; do netperf -t TCP_RR  -H fd00:9013:cadd::4 -p50000 --
-r80000,80000 -O MIN_LATENCY,P90_LATENCY,P99_LATENCY,THROUGHPUT|tail -1; done

min	p90	p99	Throughput		gso_max_size
108	165	170	6673.2
101	169	189	6451.69
101	165	169	6737.65
102	167	175	6614.64
101	178	189	6247.13			64K
107	163	169	6678.63
106	176	187	6350.86
100	164	169	6617.36
102	163	170	6849.21
102	168	175	6605.7

min	p90	p99	Throughput
108	155	166	7183			
110	154	163	7268.87			
109	152	159	7434.35			
107	145	157	7569.15			
107	149	164	7496.17			80K
110	154	159	7245.85			
108	156	162	7266.24			
109	145	158	7526.66			
106	145	151	7785.75			
111	148	157	7246.65			


Tested on azure env with Accelerated Networking enabled and disabled.

Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
---
 Changes in v3
 * Add ipv6_hopopt_jumbo_remove() while sending Big TCP packets
---
 Changes in v2
 * Instead of using 'tcp segment' throughout the patch used more accurate
   term 'aggregated pkt size'
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 5 +++++
 1 file changed, 5 insertions(+)
diff mbox series

Patch

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index aa1e47233fe5..3b0fb4d95cf7 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -256,6 +256,9 @@  netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	if (skb_cow_head(skb, MANA_HEADROOM))
 		goto tx_drop_count;
 
+	if (unlikely(ipv6_hopopt_jumbo_remove(skb)))
+		goto tx_drop_count;
+
 	txq = &apc->tx_qp[txq_idx].txq;
 	gdma_sq = txq->gdma_sq;
 	cq = &apc->tx_qp[txq_idx].tx_cq;
@@ -2873,6 +2876,8 @@  static int mana_probe_port(struct mana_context *ac, int port_idx,
 	ndev->dev_port = port_idx;
 	SET_NETDEV_DEV(ndev, gc->dev);
 
+	netif_set_tso_max_size(ndev, GSO_MAX_SIZE);
+
 	netif_carrier_off(ndev);
 
 	netdev_rss_key_fill(apc->hashkey, MANA_HASH_KEY_SIZE);