Received-SPF: pass (google.com: domain of linux-kernel+bounces-161222-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Subject: Re: [PATCH V3 1/2] iommu/arm-smmu-v3: Add support for ECMDQ register
 mode
To: Tanmay Jagdale <tanmay@marvell.com>, will@kernel.org,
 robin.murphy@arm.com, joro@8bytes.org, nicolinc@nvidia.com,
 mshavit@google.com, baolu.lu@linux.intel.com, thunder.leizhen@huawei.com,
 set_pte_at@outlook.com, smostafa@google.com
Cc: sgoutham@marvell.com, gcherian@marvell.com, jcm@jonmasters.org,
 linux-arm-kernel@lists.infradead.org, iommu@lists.linux.dev,
 linux-kernel@vger.kernel.org
References: <20240425144152.52352-1-tanmay@marvell.com>
 <20240425144152.52352-2-tanmay@marvell.com>
From: "Leizhen (ThunderTown)" <thunder.leizhen@huaweicloud.com>
Message-ID: <85f8d4df-4e4b-0460-4e34-d38588f18639@huaweicloud.com>
Date: Sun, 28 Apr 2024 10:19:26 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101
 Thunderbird/60.7.0
Precedence: bulk
MIME-Version: 1.0
In-Reply-To: <20240425144152.52352-2-tanmay@marvell.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit


On 2024/4/25 22:41, Tanmay Jagdale wrote:
> +static int arm_smmu_ecmdq_probe(struct arm_smmu_device *smmu)
> +{
> +	int ret, cpu;
> +	u32 i, nump, numq, gap;
> +	u32 reg, shift_increment;
> +	u64 offset;
> +	void __iomem *cp_regs, *cp_base;
> +
> +	/* IDR6 */
> +	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR6);
> +	nump = 1 << FIELD_GET(IDR6_LOG2NUMP, reg);
> +	numq = 1 << FIELD_GET(IDR6_LOG2NUMQ, reg);
> +	smmu->nr_ecmdq = nump * numq;
> +	gap = ECMDQ_CP_RRESET_SIZE >> FIELD_GET(IDR6_LOG2NUMQ, reg);
> +
> +	cp_regs = ioremap(smmu->iobase + ARM_SMMU_ECMDQ_CP_BASE, PAGE_SIZE);
> +	if (!cp_regs)
> +		return -ENOMEM;
> +
> +	for (i = 0; i < nump; i++) {
> +		u64 val, pre_addr = 0;

This is my mistake. The 'pre_addr' should be defined out of this 'for' loop. Or, you
can remove all 'pre_addr' related statements, commercially available chips that have
been tested cannot have such errors.

> +
> +		val = readq_relaxed(cp_regs + 32 * i);
> +		if (!(val & ECMDQ_CP_PRESET)) {
> +			iounmap(cp_regs);
> +			dev_err(smmu->dev, "ecmdq control page %u is memory mode\n", i);
> +			return -EFAULT;
> +		}
> +

                                                     <---------------------------------------------
> +		if (i && ((val & ECMDQ_CP_ADDR) != (pre_addr + ECMDQ_CP_RRESET_SIZE))) {           |
> +			iounmap(cp_regs);                                                          |
> +			dev_err(smmu->dev, "ecmdq_cp memory region is not contiguous\n");          |
> +			return -EFAULT;                                                            |
> +		}                                                                                  |
> +                                                                                                |
> +		pre_addr = val & ECMDQ_CP_ADDR;                                                    |
                                                      <--------------------------------------------   feel free remove these
> +	}

-- 
Regards,
  Zhen Lei