[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <aYKl9RqkISSV3Fn7@smile.fi.intel.com>
Date: Wed, 4 Feb 2026 03:50:45 +0200
From: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
To: Andi Shyti <andi.shyti@...nel.org>
Cc: Artem Shimko <a.shimko.dev@...il.com>, p.zabel@...gutronix.de,
mika.westerberg@...ux.intel.com, jsd@...ihalf.com,
linux-i2c@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 1/2] i2c: designware-platdrv: simplify reset control
On Wed, Feb 04, 2026 at 02:43:30AM +0100, Andi Shyti wrote:
> On Fri, Jan 30, 2026 at 02:10:36PM +0300, Artem Shimko wrote:
> > The current implementation uses separate calls to acquire and deassert
> > reset control, requiring manual error handling for the deassertion
> > operation. This can be simplified using the dedicated devm function that
> > combines both operations.
> >
> > Replace devm_reset_control_get_optional_exclusive() with
> > devm_reset_control_get_optional_exclusive_deasserted(), which handles both
> > reset acquisition and deassertion in a single call as well as
> > reset_control_put() which is called automatically on driver detach. This
> > eliminates the need for explicit deassertion and its associated error
> > checking while maintaining the same functional behavior through automatic
> > resource management.
...
> > - if (ret) {
> > - dev_err_probe(device, ret, "failed to probe lock support\n");
> > - goto exit_reset;
> > - }
> > + if (ret)
> > + return dev_err_probe(device, ret, "failed to probe lock support\n");
>
> the dev_err_probe() changes are not mentioned in the commit log.
> Can you please split this patch in two, one for the
> devm_reset_control and one for the dev_err_probe?
I believe they can go together.
But can be split as well, first we replace goto by return ret followed by
the joining the dev_err_probe(). In any case the commit message might be
improved, yes.
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists