linux-kernel - Re: [PATCH v2 2/2] ACPI: TAD: Improve runtime PM using guard macros

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20251020161122.00006179@huawei.com>
Date: Mon, 20 Oct 2025 16:11:22 +0100
From: Jonathan Cameron <jonathan.cameron@...wei.com>
To: "Rafael J. Wysocki" <rafael@...nel.org>
CC: Linux ACPI <linux-acpi@...r.kernel.org>, Linux PM
	<linux-pm@...r.kernel.org>, Takashi Iwai <tiwai@...e.de>, LKML
	<linux-kernel@...r.kernel.org>, Zhang Qilong <zhangqilong3@...wei.com>, Frank
 Li <Frank.Li@....com>, Dhruva Gole <d-gole@...com>, Mika Westerberg
	<mika.westerberg@...ux.intel.com>, Dan Williams <dan.j.williams@...el.com>
Subject: Re: [PATCH v2 2/2] ACPI: TAD: Improve runtime PM using guard macros

On Sat, 18 Oct 2025 14:24:42 +0200
"Rafael J. Wysocki" <rafael@...nel.org> wrote:

> From: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
> 
> Use guard pm_runtime_active_try to simplify runtime PM cleanup and
> implement runtime resume error handling in multiple places.
> 
> Also use guard pm_runtime_noresume to simplify acpi_tad_remove().
> 
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
> ---
>  drivers/acpi/acpi_tad.c |   57 +++++++++++++++++++++---------------------------
>  1 file changed, 25 insertions(+), 32 deletions(-)
> 
> --- a/drivers/acpi/acpi_tad.c
> +++ b/drivers/acpi/acpi_tad.c

> @@ -111,12 +110,11 @@ static int acpi_tad_get_real_time(struct
>  	acpi_status status;
>  	int ret = -EIO;
>  
> -	pm_runtime_get_sync(dev);
> +	ACQUIRE(pm_runtime_active_try, pm)(dev);
> +	if (ACQUIRE_ERR(pm_runtime_active_try, &pm))
> +		return -ENXIO;
>  
>  	status = acpi_evaluate_object(handle, "_GRT", NULL, &output);
> -
> -	pm_runtime_put_sync(dev);
> -
>  	if (ACPI_FAILURE(status))

Whilst it isn't actually a bug, this does run up against the guidance
in cleanup.h to avoid mixing gotos and cleanup.h usage in a single function.
That's partly a simplification to avoid having to explain the issues with
jumping past inline declarations.

If you want to follow that guidance, either you'd need to add a helper along the
lines of:

DEFINE_FREE(acpi_buffer, void *m ACPI_FREE(_T));
void *acpi_eval_grt(acpi_handle handle, struct acpi_buffer *output)
{
	acpi_status status = acpi_evaluate_object(handle, "_GRT", NULL, &output);

	if (ACPI_FAILURE(status)) {
		ACPI_FREE(output.pointer);
		return ERR_PTR(-EIO); //whatever error makse sense.
	}

	return output.pointer;
}

void *out_obj __free(acpi_buffer) = acpi_eval_grt(handle, &output);

Then can return directly at all error paths.  Not the nicest bit of
code though.

Or, factor out everthing after that allocation down
to the label as helper function and have direct returns in that.

or leave it all as is and hope Linus doesn't get grumpy about mix
and match (which lead to that guidance being so general in the first place!)

The rest look good to me.

Jonathan



>  		goto out_free;
>