On 10/31/2023 6:23 AM, Jeff Johnson wrote:
> On 10/30/2023 3:26 PM, Raj Kumar Bhagat wrote:
>> From: Karthikeyan Kathirvel <quic_kathirve@xxxxxxxxxxx>
>>
>> Most of the RX descriptors fields are currently not used in the
>> ath12k driver. Hence add support to selectively subscribe to the
>> required quad words (64 bits) within msdu_end and mpdu_start of
>> rx_desc.
>>
>> Add compact rx_desc structures and configure the bit mask for Rx TLVs
>> (msdu_end, mpdu_start, mpdu_end) via registers. With these registers
>> SW can configure to DMA the partial TLV struct to Rx buffer.
>>
>> Each TLV type has its own register to configure the mask value.
>> The mask value configured in register will indicate if a particular
>> QWORD has to be written to rx buffer or not i.e., if Nth bit is enabled
>> in the mask Nth QWORD will be written and it will not be written if the
>> bit is disabled in mask. While 0th bit indicates whether TLV tag will be
>> written or not.
>>
>> Advantages of Qword subscription of TLVs
>> - Avoid multiple cache-line misses as the all the required fields
>> of the TLV are within 128 bytes.
>> - Memory optimization as TLVs + DATA + SHINFO can fit in 2k buffer
>> even for 64 bit kernel.
>>
>> Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00188-QCAHKSWPL_SILICONZ-1
>> Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
>>
>> Signed-off-by: Karthikeyan Kathirvel <quic_kathirve@xxxxxxxxxxx>
>> Co-developed-by: Raj Kumar Bhagat <quic_rajkbhag@xxxxxxxxxxx>
>> Signed-off-by: Raj Kumar Bhagat <quic_rajkbhag@xxxxxxxxxxx>
>> ---
>> drivers/net/wireless/ath/ath12k/dp.c | 9 +
>> drivers/net/wireless/ath/ath12k/dp.h | 13 +
>> drivers/net/wireless/ath/ath12k/dp_rx.c | 16 +-
>> drivers/net/wireless/ath/ath12k/dp_tx.c | 20 ++
>> drivers/net/wireless/ath/ath12k/hal.c | 352 ++++++++++++++++++++++
>> drivers/net/wireless/ath/ath12k/hal.h | 3 +
>> drivers/net/wireless/ath/ath12k/rx_desc.h | 112 ++++++-
>> drivers/net/wireless/ath/ath12k/wmi.h | 2 +
>> 8 files changed, 519 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
>> index 80d7ce44d..faeef965e 100644
>> --- a/drivers/net/wireless/ath/ath12k/dp.c
>> +++ b/drivers/net/wireless/ath/ath12k/dp.c
>> @@ -1001,6 +1001,15 @@ void ath12k_dp_pdev_pre_alloc(struct ath12k_base *ab)
>> void ath12k_dp_hal_rx_desc_init(struct ath12k_base *ab)
>> {
>> + if (test_bit(WMI_TLV_SERVICE_WMSK_COMPACTION_RX_TLVS, ab->wmi_ab.svc_map) &&
>> + ab->hw_params->hal_ops->rxdma_ring_wmask_rx_mpdu_start &&
>> + ab->hw_params->hal_ops->rxdma_ring_wmask_rx_msdu_end) {
>> + /* RX TLVS compaction is supported, hence change the hal_rx_ops
>> + * based on device.
>> + */
>> + if (ab->hal_rx_ops == &hal_rx_qcn9274_ops)
>> + ab->hal_rx_ops = &hal_rx_qcn9274_compact_ops;
>
> I only have one comment on this patch.
>
> in order to avoid chipset-specific logic here suggest that there should be an abstraction.
>
> several ideas come to mind:
> 1) have a hal_ops callback to retrieve it
> 2) have a pointer to the compact ops in the hw_params
>
> since we are already using hal_ops to get the masks, suggest that is where we should get the pointer to the compact ops
>
Thanks Jeff for the suggestion, will implement hal_ops to retrieve
the corresponding compact ops.
>> + }
>> ab->hal.hal_desc_sz =
>> ab->hal_rx_ops->rx_desc_get_desc_size();
>> }
>
