On Wed, May 07, 2025 at 08:58:39AM -0700, Haiyang Zhang wrote:
> To collaborate with hardware servicing events, upon receiving the special
> EQE notification from the HW channel, remove the devices on this bus.
> Then, after a waiting period based on the device specs, rescan the parent
> bus to recover the devices.
>
> Signed-off-by: Haiyang Zhang <haiya...@microsoft.com>
> ---
> .../net/ethernet/microsoft/mana/gdma_main.c | 61 +++++++++++++++++++
> include/net/mana/gdma.h | 5 +-
> 2 files changed, 65 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c
> b/drivers/net/ethernet/microsoft/mana/gdma_main.c
> index 4ffaf7588885..aa2ccf4d0ec6 100644
> --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
> +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
> @@ -352,11 +352,52 @@ void mana_gd_ring_cq(struct gdma_queue *cq, u8 arm_bit)
> }
> EXPORT_SYMBOL_NS(mana_gd_ring_cq, "NET_MANA");
>
> +#define MANA_SERVICE_PERIOD 10
> +
> +struct mana_serv_work {
> + struct work_struct serv_work;
> + struct pci_dev *pdev;
> +};
> +
> +static void mana_serv_func(struct work_struct *w)
> +{
> + struct mana_serv_work *mns_wk = container_of(w, struct mana_serv_work,
> serv_work);
> + struct pci_dev *pdev = mns_wk->pdev;
> + struct pci_bus *bus, *parent;
> +
> + if (!pdev)
> + goto out;
> +
> + bus = pdev->bus;
> + if (!bus) {
> + dev_err(&pdev->dev, "MANA service: no bus\n");
> + goto out;
> + }
> +
> + parent = bus->parent;
> + if (!parent) {
> + dev_err(&pdev->dev, "MANA service: no parent bus\n");
> + goto out;
> + }
> +
> + pci_stop_and_remove_bus_device_locked(bus->self);
> +
> + msleep(MANA_SERVICE_PERIOD * 1000);
> +
> + pci_lock_rescan_remove();
> + pci_rescan_bus(parent);
> + pci_unlock_rescan_remove();
> +
> +out:
> + kfree(mns_wk);
Shouldn't gc->in_service be set to false again?
> +}
> +
> static void mana_gd_process_eqe(struct gdma_queue *eq)
> {
> u32 head = eq->head % (eq->queue_size / GDMA_EQE_SIZE);
> struct gdma_context *gc = eq->gdma_dev->gdma_context;
> struct gdma_eqe *eq_eqe_ptr = eq->queue_mem_ptr;
> + struct mana_serv_work *mns_wk;
> union gdma_eqe_info eqe_info;
> enum gdma_eqe_type type;
> struct gdma_event event;
> @@ -400,6 +441,26 @@ static void mana_gd_process_eqe(struct gdma_queue *eq)
> eq->eq.callback(eq->eq.context, eq, &event);
> break;
>
> + case GDMA_EQE_HWC_FPGA_RECONFIG:
> + case GDMA_EQE_HWC_SOCMANA_CRASH:
may be we also add a log(dev_dbg) to indicate if the servicing is for
FPGA reconfig or socmana crash.
> + if (gc->in_service) {
> + dev_info(gc->dev, "Already in service\n");
> + break;
> + }
> +
> + mns_wk = kzalloc(sizeof(*mns_wk), GFP_ATOMIC);
> + if (!mns_wk) {
> + dev_err(gc->dev, "Fail to alloc mana_serv_work\n");
> + break;
> + }
> +
> + dev_info(gc->dev, "Start MANA service\n");
> + gc->in_service = true;
> + mns_wk->pdev = to_pci_dev(gc->dev);
> + INIT_WORK(&mns_wk->serv_work, mana_serv_func);
> + schedule_work(&mns_wk->serv_work);
> + break;
> +
> default:
> break;
> }
> diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
> index 228603bf03f2..13cfbcf67815 100644
> --- a/include/net/mana/gdma.h
> +++ b/include/net/mana/gdma.h
> @@ -58,8 +58,9 @@ enum gdma_eqe_type {
> GDMA_EQE_HWC_INIT_EQ_ID_DB = 129,
> GDMA_EQE_HWC_INIT_DATA = 130,
> GDMA_EQE_HWC_INIT_DONE = 131,
> - GDMA_EQE_HWC_SOC_RECONFIG = 132,
> + GDMA_EQE_HWC_FPGA_RECONFIG = 132,
> GDMA_EQE_HWC_SOC_RECONFIG_DATA = 133,
> + GDMA_EQE_HWC_SOCMANA_CRASH = 135,
> GDMA_EQE_RNIC_QP_FATAL = 176,
> };
>
> @@ -388,6 +389,8 @@ struct gdma_context {
> u32 test_event_eq_id;
>
> bool is_pf;
> + bool in_service;
> +
> phys_addr_t bar0_pa;
> void __iomem *bar0_va;
> void __iomem *shm_base;
> --
> 2.34.1
