diff options
author | Ahmed Zaki <ahmed.zaki@intel.com> | 2025-02-24 16:22:22 -0700 |
---|---|---|
committer | Jakub Kicinski <kuba@kernel.org> | 2025-02-26 19:51:37 -0800 |
commit | bd7c00605ee0cf0bf27764d5da0b948d8004229e (patch) | |
tree | 50d5835e8d090080945ebbf3d1cb07f5b155ccfc /net/core/dev.c | |
parent | 28d68d396a1cd21591e8c6d74afbde33a7ea107e (diff) |
net: move aRFS rmap management and CPU affinity to core
A common task for most drivers is to remember the user-set CPU affinity
to its IRQs. On each netdev reset, the driver should re-assign the user's
settings to the IRQs. Unify this task across all drivers by moving the CPU
affinity to napi->config.
However, to move the CPU affinity to core, we also need to move aRFS
rmap management since aRFS uses its own IRQ notifiers.
For the aRFS, add a new netdev flag "rx_cpu_rmap_auto". Drivers supporting
aRFS should set the flag via netif_enable_cpu_rmap() and core will allocate
and manage the aRFS rmaps. Freeing the rmap is also done by core when the
netdev is freed. For better IRQ affinity management, move the IRQ rmap
notifier inside the napi_struct and add new notify.notify and
notify.release functions: netif_irq_cpu_rmap_notify() and
netif_napi_affinity_release().
Now we have the aRFS rmap management in core, add CPU affinity mask to
napi_config. To delegate the CPU affinity management to the core, drivers
must:
1 - set the new netdev flag "irq_affinity_auto":
netif_enable_irq_affinity(netdev)
2 - create the napi with persistent config:
netif_napi_add_config()
3 - bind an IRQ to the napi instance: netif_napi_set_irq()
the core will then make sure to use re-assign affinity to the napi's
IRQ.
The default IRQ mask is set to one cpu starting from the closest NUMA.
Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-2-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'net/core/dev.c')
-rw-r--r-- | net/core/dev.c | 169 |
1 files changed, 169 insertions, 0 deletions
diff --git a/net/core/dev.c b/net/core/dev.c index 3f525278a871..b0ab0169f507 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6943,11 +6943,175 @@ void netif_queue_set_napi(struct net_device *dev, unsigned int queue_index, } EXPORT_SYMBOL(netif_queue_set_napi); +static void +netif_napi_irq_notify(struct irq_affinity_notify *notify, + const cpumask_t *mask) +{ + struct napi_struct *napi = + container_of(notify, struct napi_struct, notify); +#ifdef CONFIG_RFS_ACCEL + struct cpu_rmap *rmap = napi->dev->rx_cpu_rmap; + int err; +#endif + + if (napi->config && napi->dev->irq_affinity_auto) + cpumask_copy(&napi->config->affinity_mask, mask); + +#ifdef CONFIG_RFS_ACCEL + if (napi->dev->rx_cpu_rmap_auto) { + err = cpu_rmap_update(rmap, napi->napi_rmap_idx, mask); + if (err) + netdev_warn(napi->dev, "RMAP update failed (%d)\n", + err); + } +#endif +} + +#ifdef CONFIG_RFS_ACCEL +static void netif_napi_affinity_release(struct kref *ref) +{ + struct napi_struct *napi = + container_of(ref, struct napi_struct, notify.kref); + struct cpu_rmap *rmap = napi->dev->rx_cpu_rmap; + + netdev_assert_locked(napi->dev); + WARN_ON(test_and_clear_bit(NAPI_STATE_HAS_NOTIFIER, + &napi->state)); + + if (!napi->dev->rx_cpu_rmap_auto) + return; + rmap->obj[napi->napi_rmap_idx] = NULL; + napi->napi_rmap_idx = -1; + cpu_rmap_put(rmap); +} + +int netif_enable_cpu_rmap(struct net_device *dev, unsigned int num_irqs) +{ + if (dev->rx_cpu_rmap_auto) + return 0; + + dev->rx_cpu_rmap = alloc_irq_cpu_rmap(num_irqs); + if (!dev->rx_cpu_rmap) + return -ENOMEM; + + dev->rx_cpu_rmap_auto = true; + return 0; +} +EXPORT_SYMBOL(netif_enable_cpu_rmap); + +static void netif_del_cpu_rmap(struct net_device *dev) +{ + struct cpu_rmap *rmap = dev->rx_cpu_rmap; + + if (!dev->rx_cpu_rmap_auto) + return; + + /* Free the rmap */ + cpu_rmap_put(rmap); + dev->rx_cpu_rmap = NULL; + dev->rx_cpu_rmap_auto = false; +} + +#else +static void netif_napi_affinity_release(struct kref *ref) +{ +} + +int netif_enable_cpu_rmap(struct net_device *dev, unsigned int num_irqs) +{ + return 0; +} +EXPORT_SYMBOL(netif_enable_cpu_rmap); + +static void netif_del_cpu_rmap(struct net_device *dev) +{ +} +#endif + +void netif_set_affinity_auto(struct net_device *dev) +{ + unsigned int i, maxqs, numa; + + maxqs = max(dev->num_tx_queues, dev->num_rx_queues); + numa = dev_to_node(&dev->dev); + + for (i = 0; i < maxqs; i++) + cpumask_set_cpu(cpumask_local_spread(i, numa), + &dev->napi_config[i].affinity_mask); + + dev->irq_affinity_auto = true; +} +EXPORT_SYMBOL(netif_set_affinity_auto); + +void netif_napi_set_irq_locked(struct napi_struct *napi, int irq) +{ + int rc; + + netdev_assert_locked_or_invisible(napi->dev); + + if (napi->irq == irq) + return; + + /* Remove existing resources */ + if (test_and_clear_bit(NAPI_STATE_HAS_NOTIFIER, &napi->state)) + irq_set_affinity_notifier(napi->irq, NULL); + + napi->irq = irq; + if (irq < 0 || + (!napi->dev->rx_cpu_rmap_auto && !napi->dev->irq_affinity_auto)) + return; + + /* Abort for buggy drivers */ + if (napi->dev->irq_affinity_auto && WARN_ON_ONCE(!napi->config)) + return; + +#ifdef CONFIG_RFS_ACCEL + if (napi->dev->rx_cpu_rmap_auto) { + rc = cpu_rmap_add(napi->dev->rx_cpu_rmap, napi); + if (rc < 0) + return; + + cpu_rmap_get(napi->dev->rx_cpu_rmap); + napi->napi_rmap_idx = rc; + } +#endif + + /* Use core IRQ notifier */ + napi->notify.notify = netif_napi_irq_notify; + napi->notify.release = netif_napi_affinity_release; + rc = irq_set_affinity_notifier(irq, &napi->notify); + if (rc) { + netdev_warn(napi->dev, "Unable to set IRQ notifier (%d)\n", + rc); + goto put_rmap; + } + + set_bit(NAPI_STATE_HAS_NOTIFIER, &napi->state); + return; + +put_rmap: +#ifdef CONFIG_RFS_ACCEL + if (napi->dev->rx_cpu_rmap_auto) { + cpu_rmap_put(napi->dev->rx_cpu_rmap); + napi->dev->rx_cpu_rmap->obj[napi->napi_rmap_idx] = NULL; + napi->napi_rmap_idx = -1; + } +#endif + napi->notify.notify = NULL; + napi->notify.release = NULL; +} +EXPORT_SYMBOL(netif_napi_set_irq_locked); + static void napi_restore_config(struct napi_struct *n) { n->defer_hard_irqs = n->config->defer_hard_irqs; n->gro_flush_timeout = n->config->gro_flush_timeout; n->irq_suspend_timeout = n->config->irq_suspend_timeout; + + if (n->dev->irq_affinity_auto && + test_bit(NAPI_STATE_HAS_NOTIFIER, &n->state)) + irq_set_affinity(n->irq, &n->config->affinity_mask); + /* a NAPI ID might be stored in the config, if so use it. if not, use * napi_hash_add to generate one for us. */ @@ -7168,6 +7332,9 @@ void __netif_napi_del_locked(struct napi_struct *napi) /* Make sure NAPI is disabled (or was never enabled). */ WARN_ON(!test_bit(NAPI_STATE_SCHED, &napi->state)); + if (test_and_clear_bit(NAPI_STATE_HAS_NOTIFIER, &napi->state)) + irq_set_affinity_notifier(napi->irq, NULL); + if (napi->config) { napi->index = -1; napi->config = NULL; @@ -11720,6 +11887,8 @@ void free_netdev(struct net_device *dev) netdev_napi_exit(dev); + netif_del_cpu_rmap(dev); + ref_tracker_dir_exit(&dev->refcnt_tracker); #ifdef CONFIG_PCPU_DEV_REFCNT free_percpu(dev->pcpu_refcnt); |