Re: [PATCH v5 04/22] ice: Register auxiliary device to provide RDMA
From: Leon Romanovsky <leon@kernel.org>
Date: 2021-05-19 11:31:14
Also in:
linux-rdma
On Fri, May 14, 2021 at 09:11:56AM -0500, Shiraz Saleem wrote:
quoted hunk ↗ jump to hunk
From: Dave Ertman <david.m.ertman@intel.com> Register ice client auxiliary RDMA device on the auxiliary bus per PCIe device function for the auxiliary driver (irdma) to attach to. It allows to realize a single RDMA driver (irdma) capable of working with multiple netdev drivers over multi-generation Intel HW supporting RDMA. There is no load ordering dependencies between ice and irdma. Signed-off-by: Dave Ertman <david.m.ertman@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Shiraz Saleem <redacted> --- drivers/net/ethernet/intel/Kconfig | 1 + drivers/net/ethernet/intel/ice/ice.h | 8 +++- drivers/net/ethernet/intel/ice/ice_idc.c | 71 ++++++++++++++++++++++++++++++- drivers/net/ethernet/intel/ice/ice_main.c | 11 ++++- 4 files changed, 87 insertions(+), 4 deletions(-)diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index c1d1556..d8a12da 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig@@ -294,6 +294,7 @@ config ICE tristate "Intel(R) Ethernet Connection E800 Series Support" default n depends on PCI_MSI + select AUXILIARY_BUS select DIMLIB select NET_DEVLINK select PLDMFWdiff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 225f8a5..228055e 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h@@ -34,6 +34,7 @@ #include <linux/if_bridge.h> #include <linux/ctype.h> #include <linux/bpf.h> +#include <linux/auxiliary_bus.h> #include <linux/avf/virtchnl.h> #include <linux/cpu_rmap.h> #include <linux/dim.h>@@ -647,6 +648,8 @@ static inline void ice_clear_sriov_cap(struct ice_pf *pf) void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size); int ice_schedule_reset(struct ice_pf *pf, enum ice_reset_req reset); void ice_print_link_msg(struct ice_vsi *vsi, bool isup); +int ice_plug_aux_dev(struct ice_pf *pf); +void ice_unplug_aux_dev(struct ice_pf *pf); int ice_init_rdma(struct ice_pf *pf); const char *ice_stat_str(enum ice_status stat_err); const char *ice_aq_str(enum ice_aq_err aq_err);@@ -678,8 +681,10 @@ int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, */ static inline void ice_set_rdma_cap(struct ice_pf *pf) { - if (pf->hw.func_caps.common_cap.rdma && pf->num_rdma_msix) + if (pf->hw.func_caps.common_cap.rdma && pf->num_rdma_msix) { set_bit(ICE_FLAG_RDMA_ENA, pf->flags); + ice_plug_aux_dev(pf); + } } /**@@ -688,6 +693,7 @@ static inline void ice_set_rdma_cap(struct ice_pf *pf) */ static inline void ice_clear_rdma_cap(struct ice_pf *pf) { + ice_unplug_aux_dev(pf); clear_bit(ICE_FLAG_RDMA_ENA, pf->flags); } #endif /* _ICE_H_ */diff --git a/drivers/net/ethernet/intel/ice/ice_idc.c b/drivers/net/ethernet/intel/ice/ice_idc.c index ffca0d5..e7bb8f6 100644 --- a/drivers/net/ethernet/intel/ice/ice_idc.c +++ b/drivers/net/ethernet/intel/ice/ice_idc.c@@ -255,6 +255,71 @@ static int ice_reserve_rdma_qvector(struct ice_pf *pf) } /** + * ice_adev_release - function to be mapped to AUX dev's release op + * @dev: pointer to device to free + */ +static void ice_adev_release(struct device *dev) +{ + struct iidc_auxiliary_dev *iadev; + + iadev = container_of(dev, struct iidc_auxiliary_dev, adev.dev); + kfree(iadev); +} + +/** + * ice_plug_aux_dev - allocate and register AUX device + * @pf: pointer to pf struct + */ +int ice_plug_aux_dev(struct ice_pf *pf) +{ + struct iidc_auxiliary_dev *iadev; + struct auxiliary_device *adev; + int ret; + + iadev = kzalloc(sizeof(*iadev), GFP_KERNEL); + if (!iadev) + return -ENOMEM; + + adev = &iadev->adev; + pf->adev = adev; + iadev->pf = pf; + + adev->id = pf->aux_idx; + adev->dev.release = ice_adev_release; + adev->dev.parent = &pf->pdev->dev; + adev->name = IIDC_RDMA_ROCE_NAME;
You declared IIDC_RDMA_ROCE_NAME as intel_rdma_roce, so it will create extremely awful device name, something like irdma.intel_rdma_roce.0 I would say that "intel" and "rdma" can be probably dropped. Thanks