[Skiboot] [PATCH 2/3] npu2-opencapi: Detect if link trained in degraded mode

[Andrew Donnellan]

On 23/11/18 7:54 pm, Frederic Barrat wrote:
> There's no status readily available to tell the effective link
> width. Instead, we have to look at the individual status of each lane,
> on the transmit and receive direction. All relevant information is in
> the ODL status register.
> 
> Signed-off-by: Frederic Barrat <fbarrat at linux.ibm.com>

Reviewed-by: Andrew Donnellan <andrew.donnellan at au1.ibm.com>

> ---
>   hw/npu2-opencapi.c  | 50 ++++++++++++++++++++++++++++-----------------
>   include/npu2-regs.h |  2 ++
>   2 files changed, 33 insertions(+), 19 deletions(-)
> 
> diff --git a/hw/npu2-opencapi.c b/hw/npu2-opencapi.c
> index c771fae8..da86b46c 100644
> --- a/hw/npu2-opencapi.c
> +++ b/hw/npu2-opencapi.c
> @@ -1013,32 +1013,43 @@ static int64_t npu2_opencapi_get_presence_state(struct pci_slot __unused *slot,
>   	return OPAL_SUCCESS;
>   }
>   
> +static enum OpalShpcLinkState get_link_width(uint64_t odl_status)
> +{
> +	uint64_t tx_lanes, rx_lanes, state;
> +
> +	/*
> +	 * On P9, the 'trained mode' field of the ODL status is
> +	 * hard-coded to x8 and is useless for us. We need to look at
> +	 * the status of the individual lanes.
> +	 * The link trains at x8, x4 or not at all.
> +	 */
> +	state = GETFIELD(OB_ODL_STATUS_TRAINING_STATE_MACHINE, odl_status);
> +	if (state != OCAPI_LINK_STATE_TRAINED)
> +		return OPAL_SHPC_LINK_DOWN;
> +
> +	rx_lanes = GETFIELD(OB_ODL_STATUS_RX_TRAINED_LANES, odl_status);
> +	tx_lanes = GETFIELD(OB_ODL_STATUS_TX_TRAINED_LANES, odl_status);
> +	if ((rx_lanes != 0xFF) || (tx_lanes != 0xFF))
> +		return OPAL_SHPC_LINK_UP_x4;
> +	else
> +		return OPAL_SHPC_LINK_UP_x8;
> +}
> +
>   static int64_t npu2_opencapi_get_link_state(struct pci_slot *slot, uint8_t *val)
>   {
>   	struct npu2_dev *dev = phb_to_npu2_dev_ocapi(slot->phb);
>   	uint64_t reg;
> -	int64_t link_width, training_status, rc = OPAL_SUCCESS;
>   
>   	reg = get_odl_status(dev->npu->chip_id, dev->brick_index);
> -	link_width = GETFIELD(OB_ODL_STATUS_TRAINED_MODE, reg);
> -	training_status = GETFIELD(OB_ODL_STATUS_TRAINING_STATE_MACHINE, reg);
> -
> -	if (training_status != OCAPI_LINK_STATE_TRAINED) {
> -		*val = OPAL_SHPC_LINK_DOWN;
> -		return OPAL_SUCCESS;
> -	}
> +	*val = get_link_width(reg);
> +	return OPAL_SUCCESS;
> +}
>   
> -	switch (link_width) {
> -	case 0b0001:
> -		*val = OPAL_SHPC_LINK_UP_x4;
> -		break;
> -	case 0b0010:
> -		*val = OPAL_SHPC_LINK_UP_x8;
> -		break;
> -	default:
> -		rc = OPAL_HARDWARE;
> -	}
> -	return rc;
> +static void check_trained_link(struct npu2_dev *dev, uint64_t odl_status)
> +{
> +	if (get_link_width(odl_status) != OPAL_SHPC_LINK_UP_x8)
> +		OCAPIERR(dev, "Link trained in degraded mode (%016llx)\n",
> +			odl_status);
>   }
>   
>   static int64_t npu2_opencapi_retry_state(struct pci_slot *slot,
> @@ -1089,6 +1100,7 @@ static int64_t npu2_opencapi_poll_link(struct pci_slot *slot)
>   			OCAPI_LINK_STATE_TRAINED) {
>   			OCAPIINF(dev, "link trained in %lld ms\n",
>   				OCAPI_LINK_TRAINING_TIMEOUT - slot->retries);
> +			check_trained_link(dev, reg);
>   			pci_slot_set_state(slot, OCAPI_SLOT_LINK_TRAINED);
>   			return pci_slot_set_sm_timeout(slot, msecs_to_tb(1));
>   		}
> diff --git a/include/npu2-regs.h b/include/npu2-regs.h
> index 165e0b79..c5096dd1 100644
> --- a/include/npu2-regs.h
> +++ b/include/npu2-regs.h
> @@ -741,6 +741,8 @@ void npu2_scom_write(uint64_t gcid, uint64_t scom_base,
>   #define OB3_ODL0_STATUS				0xC01082C
>   #define OB3_ODL1_STATUS				0xC01082D
>   #define   OB_ODL_STATUS_TRAINED_MODE		PPC_BITMASK(0,3)
> +#define   OB_ODL_STATUS_RX_TRAINED_LANES	PPC_BITMASK(16, 23)
> +#define   OB_ODL_STATUS_TX_TRAINED_LANES	PPC_BITMASK(24, 31)
>   #define   OB_ODL_STATUS_TRAINING_STATE_MACHINE	PPC_BITMASK(49, 51)
>   
>   #define OB0_ODL0_TRAINING_STATUS		0x901082E
> 

-- 
Andrew Donnellan              OzLabs, ADL Canberra
andrew.donnellan at au1.ibm.com  IBM Australia Limited