| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20241021223551.508030-1-jdamato@fastly.com>
Date: Mon, 21 Oct 2024 22:35:51 +0000
From: Joe Damato <jdamato@...tly.com>
To: netdev@...r.kernel.org
Cc: Joe Damato <jdamato@...tly.com>,
Tony Nguyen <anthony.l.nguyen@...el.com>,
Przemek Kitszel <przemyslaw.kitszel@...el.com>,
Andrew Lunn <andrew+netdev@...n.ch>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
intel-wired-lan@...ts.osuosl.org (moderated list:INTEL ETHERNET DRIVERS),
linux-kernel@...r.kernel.org (open list)
Subject: [PATCH intel-next] ice: Add support for persistent NAPI config
Use netif_napi_add_config to assign persistent per-NAPI config when
initializing NAPIs. This preserves NAPI config settings when queue
counts are adjusted.
Tested with an E810-2CQDA2 NIC.
Begin by setting the queue count to 4:
$ sudo ethtool -L eth4 combined 4
Check the queue settings:
$ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
--dump napi-get --json='{"ifindex": 4}'
[{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8452,
'ifindex': 4,
'irq': 2782},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8451,
'ifindex': 4,
'irq': 2781},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8450,
'ifindex': 4,
'irq': 2780},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8449,
'ifindex': 4,
'irq': 2779}]
Now, set the queue with NAPI ID 8451 to have a gro-flush-timeout of
1111:
$ sudo ./tools/net/ynl/cli.py \
--spec Documentation/netlink/specs/netdev.yaml \
--do napi-set --json='{"id": 8451, "gro-flush-timeout": 1111}'
None
Check that worked:
$ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
--dump napi-get --json='{"ifindex": 4}'
[{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8452,
'ifindex': 4,
'irq': 2782},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 1111,
'id': 8451,
'ifindex': 4,
'irq': 2781},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8450,
'ifindex': 4,
'irq': 2780},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8449,
'ifindex': 4,
'irq': 2779}]
Now reduce the queue count to 2, which would destroy the queue with NAPI
ID 8451:
$ sudo ethtool -L eth4 combined 2
Check the queue settings, noting that NAPI ID 8451 is gone:
$ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
--dump napi-get --json='{"ifindex": 4}'
[{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8450,
'ifindex': 4,
'irq': 2780},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8449,
'ifindex': 4,
'irq': 2779}]
Now, increase the number of queues back to 4:
$ sudo ethtool -L eth4 combined 4
Dump the settings, expecting to see the same NAPI IDs as above and for
NAPI ID 8451 to have its gro-flush-timeout set to 1111:
$ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
--dump napi-get --json='{"ifindex": 4}'
[{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8452,
'ifindex': 4,
'irq': 2782},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 1111,
'id': 8451,
'ifindex': 4,
'irq': 2781},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8450,
'ifindex': 4,
'irq': 2780},
{'defer-hard-irqs': 0,
'gro-flush-timeout': 0,
'id': 8449,
'ifindex': 4,
'irq': 2779}]
Signed-off-by: Joe Damato <jdamato@...tly.com>
---
drivers/net/ethernet/intel/ice/ice_base.c | 3 ++-
drivers/net/ethernet/intel/ice/ice_lib.c | 6 ++++--
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index 3a8e156d7d86..82a9cd4ec7ae 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -156,7 +156,8 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, u16 v_idx)
* handler here (i.e. resume, reset/rebuild, etc.)
*/
if (vsi->netdev)
- netif_napi_add(vsi->netdev, &q_vector->napi, ice_napi_poll);
+ netif_napi_add_config(vsi->netdev, &q_vector->napi,
+ ice_napi_poll, v_idx);
out:
/* tie q_vector and VSI together */
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index d4e74f96a8ad..a7d45a8ce7ac 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2777,8 +2777,10 @@ void ice_napi_add(struct ice_vsi *vsi)
return;
ice_for_each_q_vector(vsi, v_idx)
- netif_napi_add(vsi->netdev, &vsi->q_vectors[v_idx]->napi,
- ice_napi_poll);
+ netif_napi_add_config(vsi->netdev,
+ &vsi->q_vectors[v_idx]->napi,
+ ice_napi_poll,
+ v_idx);
}
/**
base-commit: 6f07cd8301706b661776074ddc97c991d107cc91
--
2.25.1
Powered by blists - more mailing lists