| Message ID | 20180112003723.19208-3-saeedm@mellanox.com |
|---|---|
| State | Accepted, archived |
| Delegated to: | David Miller |
| Headers | show |
| Series | [net,01/11] {net,ib}/mlx5: Don't disable local loopback multicast traffic when needed | expand |
Hi Saeed, > mlx5_get_vector_affinity used to call pci_irq_get_affinity and after > reverting the patch that sets the device affinity via PCI_IRQ_AFFINITY > API, calling pci_irq_get_affinity becomes useless and it breaks RDMA > mlx5 users. To fix this, this patch provides an alternative way to > retrieve IRQ vector affinity using legacy IRQ API, following > smp_affinity read procfs implementation. > > Fixes: 231243c82793 ("Revert mlx5: move affinity hints assignments to generic code") > Fixes: a435393acafb ("mlx5: move affinity hints assignments to generic code") > Cc: Sagi Grimberg <sagi@grimberg.me> > Cc: Qing Huang <qing.huang@oracle.com> > Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> I didn't have time to test it out thus far, I assume you tested it though, so looks good to me, Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 1f509d072026..a0610427e168 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -36,6 +36,7 @@ #include <linux/kernel.h> #include <linux/completion.h> #include <linux/pci.h> +#include <linux/irq.h> #include <linux/spinlock_types.h> #include <linux/semaphore.h> #include <linux/slab.h> @@ -1231,7 +1232,23 @@ enum { static inline const struct cpumask * mlx5_get_vector_affinity(struct mlx5_core_dev *dev, int vector) { - return pci_irq_get_affinity(dev->pdev, MLX5_EQ_VEC_COMP_BASE + vector); + const struct cpumask *mask; + struct irq_desc *desc; + unsigned int irq; + int eqn; + int err; + + err = mlx5_vector2eqn(dev, vector, &eqn, &irq); + if (err) + return NULL; + + desc = irq_to_desc(irq); +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + mask = irq_data_get_effective_affinity_mask(&desc->irq_data); +#else + mask = desc->irq_common_data.affinity; +#endif + return mask; } #endif /* MLX5_DRIVER_H */
mlx5_get_vector_affinity used to call pci_irq_get_affinity and after reverting the patch that sets the device affinity via PCI_IRQ_AFFINITY API, calling pci_irq_get_affinity becomes useless and it breaks RDMA mlx5 users. To fix this, this patch provides an alternative way to retrieve IRQ vector affinity using legacy IRQ API, following smp_affinity read procfs implementation. Fixes: 231243c82793 ("Revert mlx5: move affinity hints assignments to generic code") Fixes: a435393acafb ("mlx5: move affinity hints assignments to generic code") Cc: Sagi Grimberg <sagi@grimberg.me> Cc: Qing Huang <qing.huang@oracle.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> --- include/linux/mlx5/driver.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-)