Re: [PATCH v4 08/13] wifi: ath12k: add AHB driver support for IPQ5332

[Raj Kumar Bhagat <quic_rajkbhag@quicinc.com>]

On 12/10/2024 8:17 PM, Krzysztof Kozlowski wrote:
> On 10/12/2024 08:41, Raj Kumar Bhagat wrote:
>> From: Balamurugan S <quic_bselvara@quicinc.com>
>>
>> Add Initial Ath12k AHB driver support for IPQ5332. IPQ5332 is AHB
>> based IEEE802.11be 2 GHz 2x2 WiFi device.
>>
>> Tested-on: IPQ5332 hw1.0 AHB WLAN.WBE.1.3.1-00130-QCAHKSWPL_SILICONZ-1
>> Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00210-QCAHKSWPL_SILICONZ-1
>>
>> Signed-off-by: Balamurugan S <quic_bselvara@quicinc.com>
>> Co-developed-by: P Praneesh <quic_ppranees@quicinc.com>
>> Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
>> Co-developed-by: Raj Kumar Bhagat <quic_rajkbhag@quicinc.com>
>> Signed-off-by: Raj Kumar Bhagat <quic_rajkbhag@quicinc.com>
>> ---
>>  drivers/net/wireless/ath/ath12k/ahb.c  | 878 +++++++++++++++++++++++++
>>  drivers/net/wireless/ath/ath12k/ahb.h  |  37 ++
>>  drivers/net/wireless/ath/ath12k/core.h |   4 +
>>  drivers/net/wireless/ath/ath12k/hal.h  |   1 +
>>  drivers/net/wireless/ath/ath12k/hw.h   |   1 +
>>  5 files changed, 921 insertions(+)
>>  create mode 100644 drivers/net/wireless/ath/ath12k/ahb.c
>>  create mode 100644 drivers/net/wireless/ath/ath12k/ahb.h
>>
>> diff --git a/drivers/net/wireless/ath/ath12k/ahb.c b/drivers/net/wireless/ath/ath12k/ahb.c
>> new file mode 100644
>> index 000000000000..fcd949faea9f
>> --- /dev/null
>> +++ b/drivers/net/wireless/ath/ath12k/ahb.c
>> @@ -0,0 +1,878 @@
>> +// SPDX-License-Identifier: BSD-3-Clause-Clear
>> +/*
>> + * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
>> + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved.
>> + */
>> +
>> +#include <linux/dma-mapping.h>
>> +#include <linux/iommu.h>
>> +#include <linux/module.h>
>> +#include <linux/of.h>
>> +#include <linux/of_address.h>
> 
> You don't use this header.
> 
>> +#include <linux/of_device.h>
>> +#include <linux/platform_device.h>
>> +#include <linux/remoteproc.h>
> 
> More headers here look unused.
> 

sure, will remove unused headers in next version.

>> +#include <linux/soc/qcom/smem.h>
>> +#include <linux/soc/qcom/smem_state.h>
>> +#include "ahb.h"
>> +#include "debug.h"
>> +#include "hif.h"
>> +
>> +static const struct of_device_id ath12k_ahb_of_match[] = {
>> +	{ .compatible = "qcom,ipq5332-wifi",
>> +	  .data = (void *)ATH12K_HW_IPQ5332_HW10,
>> +	},
>> +	{ }
>> +};
>> +
>> +MODULE_DEVICE_TABLE(of, ath12k_ahb_of_match);
> 
> This goes to driver struct declaration. It is really unusual to see it
> in some other places.
> 
> 
>> +
>> +#define ATH12K_IRQ_CE0_OFFSET 4
>> +
> 
> 
> ...
> 
>> +
>> +static int ath12k_ahb_probe(struct platform_device *pdev)
>> +{
>> +	struct ath12k_base *ab;
>> +	const struct ath12k_hif_ops *hif_ops;
>> +	struct device_node *mem_node;
>> +	enum ath12k_hw_rev hw_rev;
>> +	u32 addr;
>> +	int ret;
>> +
>> +	hw_rev = ath12k_ahb_get_hw_rev(pdev);
>> +	switch (hw_rev) {
>> +	case ATH12K_HW_IPQ5332_HW10:
>> +		hif_ops = &ath12k_ahb_hif_ops_ipq5332;
>> +		break;
>> +	default:
>> +		dev_err(&pdev->dev, "Unsupported device type %d\n", hw_rev);
> 
> You already print once in ath12k_ahb_get_hw_rev() and none of these can
> happen. No need to print impossible conditions at all. No need to print
> impossible things twice.
> 

sure, will drop this print.

>> +		return -EOPNOTSUPP;
>> +	}
>> +
> 
>> +	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
>> +	if (ret) {
>> +		dev_err(&pdev->dev, "Failed to set 32-bit coherent dma\n");
>> +		return ret;
>> +	}
>> +
>> +	ab = ath12k_core_alloc(&pdev->dev, sizeof(struct ath12k_ahb),
>> +			       ATH12K_BUS_AHB);
>> +	if (!ab) {
>> +		dev_err(&pdev->dev, "failed to allocate ath12k base\n");
>> +		return -ENOMEM;
>> +	}
>> +
>> +	ab->hif.ops = hif_ops;
>> +	ab->pdev = pdev;
>> +	ab->hw_rev = hw_rev;
>> +	platform_set_drvdata(pdev, ab);
>> +
>> +	/* Set fixed_mem_region to true for platforms that support fixed memory
>> +	 * reservation from DT. If memory is reserved from DT for FW, ath12k driver
>> +	 * need not to allocate memory.
>> +	 */
>> +	if (!of_property_read_u32(ab->dev->of_node, "memory-region", &addr)) {
>> +		set_bit(ATH12K_FLAG_FIXED_MEM_REGION, &ab->dev_flags);
>> +
>> +		/* If the platform supports fixed memory, then it should define/
>> +		 * reserve MLO global memory in DT to support Multi Link Operation
>> +		 * (IEEE 802.11be).
>> +		 * If MLO global memory is not reserved in fixed memory mode, then
>> +		 * MLO cannot be supported.
>> +		 */
>> +		mem_node = of_find_node_by_name(NULL, "mlo_global_mem_0");
> 
> NAK. Incorrect name, not conforming to coding style. Undocumented ABI.
> 
> Drop all calls to of_find_node_by_name() or any other stuff like this.
> 

Sure, will remove all calls to - of_find_node_by_name()

>> +		if (!mem_node)
>> +			ab->single_chip_mlo_supp = false;
>> +	}
>> +
>> +	ret = ath12k_core_pre_init(ab);
>> +	if (ret)
>> +		goto err_core_free;
> 
> Best regards,
> Krzysztof