From: vladimir_zapolskiy@mentor.com (Vladimir Zapolskiy)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFC] Question about freeing of resource in 'atlas7_pinmux_probe()', in file 'drivers/pinctrl/sirf/pinctrl-atlas7.c'
Date: Mon, 19 Dec 2016 15:54:24 +0200 [thread overview]
Message-ID: <7123e174-cc22-4a76-337c-661ee87994df@mentor.com> (raw)
In-Reply-To: <612eedc5-0f28-7a38-abad-4b9573208875@wanadoo.fr>
Hi,
On 12/19/2016 08:19 AM, Marion & Christophe JAILLET wrote:
> Hi,
>
> while playing with coccinelle, a missing 'of_node_put()' triggered in
> 'atlas7_pinmux_probe()', in file 'drivers/pinctrl/sirf/pinctrl-atlas7.c'.
>
> /* The sd3 and sd9 shared all pins, and the function select by
> * SYS2PCI_SDIO9SEL register
> */
> sys2pci_np = of_find_node_by_name(NULL, "sys2pci");
> if (!sys2pci_np)
> return -EINVAL;
> ret = of_address_to_resource(sys2pci_np, 0, &res);
> if (ret) <------------- missing of_node_put(sys2pci_np);
> return ret;
> pmx->sys2pci_base = devm_ioremap_resource(&pdev->dev, &res);
> if (IS_ERR(pmx->sys2pci_base)) {
> of_node_put(sys2pci_np); <------------- added by commit
> 151b8c5ba1eb
> return -ENOMEM;
> }
>
> Looking at the history of this file, I found a recent commit that added
> another missing of_node_put (see above).
>
>
> Adding missing 'of_node_put()' in error handling paths is fine, but in
> this particular case, I was wondering if one was not also missing in the
> normal path?
Right, in this particular case 'of_node_put()' should be both on error
and normal paths.
>
> In such a case, I would revert 151b8c5ba1eb and propose something like:
> /* The sd3 and sd9 shared all pins, and the function select by
> * SYS2PCI_SDIO9SEL register
> */
> sys2pci_np = of_find_node_by_name(NULL, "sys2pci");
> if (!sys2pci_np)
> return -EINVAL;
> ret = of_address_to_resource(sys2pci_np, 0, &res);
> if (ret) {
> of_node_put(sys2pci_np);
> return ret;
> }
> of_node_put(sys2pci_np);
Functionally it looks good, I have two comments though.
1) you don't need to revert 151b8c5ba1eb, the commit is a proper fix
per se but incomplete, please add your change on top of it,
2) minimizing the lines of code by removing duplicates is always good,
so here a better and complete fix will be like the following one:
sys2pci_np = of_find_node_by_name(NULL, "sys2pci");
if (!sys2pci_np)
return -EINVAL;
ret = of_address_to_resource(sys2pci_np, 0, &res);
of_node_put(sys2pci_np);
if (ret)
return ret;
pmx->sys2pci_base = devm_ioremap_resource(&pdev->dev, &res);
if (IS_ERR(pmx->sys2pci_base))
return -ENOMEM;
>
> pmx->sys2pci_base = devm_ioremap_resource(&pdev->dev, &res);
> if (IS_ERR(pmx->sys2pci_base))
> return -ENOMEM;
--
With best wishes,
Vladimir
next prev parent reply other threads:[~2016-12-19 13:54 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-19 6:19 [RFC] Question about freeing of resource in 'atlas7_pinmux_probe()', in file 'drivers/pinctrl/sirf/pinctrl-atlas7.c' Marion & Christophe JAILLET
2016-12-19 13:54 ` Vladimir Zapolskiy [this message]
2016-12-20 5:44 ` Christophe JAILLET
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=7123e174-cc22-4a76-337c-661ee87994df@mentor.com \
--to=vladimir_zapolskiy@mentor.com \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox