Re: [PATCH 2/2] ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device

From: ilkka@os.amperecomputing.com
To: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Ilkka Koskinen <ilkka@os.amperecomputing.com>,
	lorenzo.pieralisi@arm.com, guohanjun@huawei.com,
	sudeep.holla@arm.com, rafael@kernel.org, lenb@kernel.org,
	robert.moore@intel.com, linux-acpi@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, devel@acpica.org,
	patches@amperecomputing.com, scott@os.amperecomputing.com,
	darren@os.amperecomputing.com
Subject: Re: [PATCH 2/2] ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
Date: Fri, 10 Dec 2021 00:32:09 -0800 (PST)	[thread overview]
Message-ID: <alpine.DEB.2.22.394.2112092359001.2491@ubuntu200401> (raw)
In-Reply-To: <YbIwfzND2R4WyQO7@shell.armlinux.org.uk>

Hi,

Thanks for reviewing the patch

On Thu, 9 Dec 2021, Russell King (Oracle) wrote:
> Hi,
>
> On Thu, Dec 02, 2021 at 06:43:11PM -0800, Ilkka Koskinen wrote:
>> +static int __init agdi_init(void)
>> +{
>> +	int ret;
>> +	acpi_status status;
>> +	struct acpi_table_agdi *agdi_table;
>> +	struct agdi_data *pdata;
>> +	struct platform_device *pdev;
>> +
>> +	if (acpi_disabled)
>> +		return 0;
>> +
>> +	status = acpi_get_table(ACPI_SIG_AGDI, 0,
>> +				(struct acpi_table_header **) &agdi_table);
>> +	if (ACPI_FAILURE(status))
>> +		return -ENODEV;
>> +
>> +	pdata = kzalloc(sizeof(*pdata), GFP_ATOMIC);
>
> Why does this need to be GFP_ATOMIC? Also, struct agdi_data is a single
> int in size, why do you need to kzalloc() it?

There's no reason for that. It should obviously be GFP_KERNEL

>
>> +	if (!pdata) {
>> +		ret = -ENOMEM;
>> +		goto err_put_table;
>> +	}
>> +
>> +	if (agdi_table->flags & ACPI_AGDI_SIGNALING_MODE) {
>> +		pr_warn("Interrupt signaling is not supported");
>> +		ret = -ENODEV;
>> +		goto err_free_pdata;
>> +	}
>> +
>> +	pdata->sdei_event = agdi_table->sdei_event;
>> +
>> +	pdev = platform_device_register_data(NULL, "agdi", 0, pdata, sizeof(*pdata));
>
> platform_device_register_data() uses kmemdup() internally with the
> platform data, meaning it takes a copy of the platform data. There is
> no need for the pdata allocation to persist past this point. Hence,
> given that it is a single int in size, you may as well put
> "struct agdi_data" on the stack.

I somehow missed kmemdup() part. I remove the allocation and move pdata to 
the stack instead.

Thanks,
Ilkka

>
> Thanks.
>
> -- 
> RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
> FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
>