From: Axel Lin <axel.lin@gmail.com>
To: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Cc: linux-kernel <linux-kernel@vger.kernel.org>,
Liam Girdwood <lrg@slimlogic.co.uk>,
Mark Brown <broonie@opensource.wolfsonmicro.com>,
Jaroslav Kysela <perex@perex.cz>, Takashi Iwai <tiwai@suse.de>,
Tejun Heo <tj@kernel.org>,
alsa-devel@alsa-project.org
Subject: Re: [PATCH 7/12] wm8940: fix resource reclaim in wm8940_register error path
Date: Thu, 15 Jul 2010 15:42:07 +0800 [thread overview]
Message-ID: <AANLkTil_buBStfVy_L9iTgPrxosmSRFhNU1MbaXWJfMx@mail.gmail.com> (raw)
In-Reply-To: <Pine.LNX.4.64.1007150855200.5310@axis700.grange>
Hi Guennadi,
2010/7/15 Guennadi Liakhovetski <g.liakhovetski@gmx.de>:
> Hi Axel
>
> On Thu, 15 Jul 2010, Axel Lin wrote:
>
>> >>From 44c93254e5158e3086c4707148d24746067f6b14 Mon Sep 17 00:00:00 2001
>> From: Axel Lin <axel.lin@gmail.com>
>> Date: Thu, 15 Jul 2010 10:13:34 +0800
>> Subject: [PATCH] wm8940: fix resource reclaim in wm8940_register error path
>>
>> This patch fixes the error path in wm8940_register to properly free resources.
>>
>> Signed-off-by: Axel Lin <axel.lin@gmail.com>
>
> Ok, these bugs are real, but let's fix them properly, i.e., release
> resources at the same level, where they have been acquired:
OK.
I'll send a revised patch now.
I change the patch titile to:
[PATCH v2 7/12] wm8940: fix a memory leak if wm8940_register return error
>
> static int register_codec(struct drv_priv *data)
> {
> int ret = snd_soc_register_codec();
> if (ret < 0)
> return ret;
>
> ret = try();
> if (failed)
> goto err;
>
> ...
>
> err:
> snd_soc_unregister_codec();
> }
>
> static int probe()
> {
> int ret;
> struct drv_priv *data = kzalloc(sizeof(*data), GFP_KERNEL);
> if (!data)
> return -ENOMEM;
>
> ret = register_codec(data);
> if (ret < 0)
> goto err;
>
> ...
>
> err:
> kfree(data);
> return ret;
> }
>
> That is, kfree() belongs to probe() and not to register_codec. This, of
> course, concerns all aour patches, at least those of them, that I have
> seen.
I agree that it is good to release resources at the same level,
where they have been acquired.
But I prefer to follow the maintainer/author's coding style.
For the changes in other drivers, I'd like to see Liam and Mark's comments.
Or if the driver maintainers request it, I can fix it.
Regards,
Axel
>
> Thanks
> Guennadi
>
>> ---
>> sound/soc/codecs/wm8940.c | 22 ++++++++++++++--------
>> 1 files changed, 14 insertions(+), 8 deletions(-)
>>
>> diff --git a/sound/soc/codecs/wm8940.c b/sound/soc/codecs/wm8940.c
>> index e3c4bbf..9673e6f 100644
>> --- a/sound/soc/codecs/wm8940.c
>> +++ b/sound/soc/codecs/wm8940.c
>> @@ -766,7 +766,8 @@ static int wm8940_register(struct wm8940_priv *wm8940,
>> u16 reg;
>> if (wm8940_codec) {
>> dev_err(codec->dev, "Another WM8940 is registered\n");
>> - return -EINVAL;
>> + ret = -EINVAL;
>> + goto err;
>> }
>>
>> INIT_LIST_HEAD(&codec->dapm_widgets);
>> @@ -785,7 +786,7 @@ static int wm8940_register(struct wm8940_priv *wm8940,
>> ret = snd_soc_codec_set_cache_io(codec, 8, 16, control);
>> if (ret < 0) {
>> dev_err(codec->dev, "Failed to set cache I/O: %d\n", ret);
>> - return ret;
>> + goto err;
>> }
>>
>> memcpy(codec->reg_cache, wm8940_reg_defaults,
>> @@ -794,7 +795,7 @@ static int wm8940_register(struct wm8940_priv *wm8940,
>> ret = wm8940_reset(codec);
>> if (ret < 0) {
>> dev_err(codec->dev, "Failed to issue reset\n");
>> - return ret;
>> + goto err;
>> }
>>
>> wm8940_dai.dev = codec->dev;
>> @@ -803,7 +804,7 @@ static int wm8940_register(struct wm8940_priv *wm8940,
>>
>> ret = snd_soc_write(codec, WM8940_POWER1, 0x180);
>> if (ret < 0)
>> - return ret;
>> + goto err;
>>
>> if (!pdata)
>> dev_warn(codec->dev, "No platform data supplied\n");
>> @@ -811,7 +812,7 @@ static int wm8940_register(struct wm8940_priv *wm8940,
>> reg = snd_soc_read(codec, WM8940_OUTPUTCTL);
>> ret = snd_soc_write(codec, WM8940_OUTPUTCTL, reg | pdata->vroi);
>> if (ret < 0)
>> - return ret;
>> + goto err;
>> }
>>
>>
>> @@ -820,17 +821,22 @@ static int wm8940_register(struct wm8940_priv *wm8940,
>> ret = snd_soc_register_codec(codec);
>> if (ret) {
>> dev_err(codec->dev, "Failed to register codec: %d\n", ret);
>> - return ret;
>> + goto err;
>> }
>>
>> ret = snd_soc_register_dai(&wm8940_dai);
>> if (ret) {
>> dev_err(codec->dev, "Failed to register DAI: %d\n", ret);
>> - snd_soc_unregister_codec(codec);
>> - return ret;
>> + goto err_codec;
>> }
>>
>> return 0;
>> +
>> +err_codec:
>> + snd_soc_unregister_codec(codec);
>> +err:
>> + kfree(wm8940);
>> + return ret;
>> }
>>
>> static void wm8940_unregister(struct wm8940_priv *wm8940)
>> --
>> 1.5.4.3
>>
>>
>>
>
> ---
> Guennadi Liakhovetski, Ph.D.
> Freelance Open-Source Software Developer
> http://www.open-technology.de/
>
next prev parent reply other threads:[~2010-07-15 7:42 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-15 2:49 [PATCH 0/12] sound/alsa/soc/codec: fix memory leak and resource relaim in error path Axel Lin
2010-07-15 2:50 ` [PATCH 1/12] ad1836: fix a memory leak if another ad1836 is registered Axel Lin
2010-07-20 10:31 ` Barry Song
2010-07-15 2:52 ` [PATCH 2/12] ak4642: fix a memory leak if failed to initialise AK4642 Axel Lin
2010-07-15 5:53 ` Kuninori Morimoto
2010-07-15 6:19 ` Axel Lin
2010-07-15 6:29 ` Kuninori Morimoto
2010-07-15 2:53 ` [PATCH 3/12] da7210: fix a memory leak if failed to initialise da7210 audio codec Axel Lin
2010-07-15 2:56 ` [PATCH 4/12] wm8523: fix resource reclaim in wm8523_register error path Axel Lin
2010-07-15 2:57 ` [PATCH 5/12] wm8711: fix a memory leak if another WM8711 is registered Axel Lin
2010-07-15 2:59 ` [PATCH 6/12] wm8904: fix resource reclaim in wm8904_register error path Axel Lin
2010-07-15 3:01 ` [PATCH 7/12] wm8940: fix resource reclaim in wm8940_register " Axel Lin
2010-07-15 7:04 ` Guennadi Liakhovetski
2010-07-15 7:42 ` Axel Lin [this message]
2010-07-15 7:59 ` Guennadi Liakhovetski
2010-07-15 8:54 ` Liam Girdwood
2010-07-15 7:48 ` [PATCH v2 7/12] wm8940: fix a memory leak if wm8940_register return error Axel Lin
2010-07-15 3:03 ` [PATCH 8/12] wm8955: fix resource reclaim in wm8955_register error path Axel Lin
2010-07-15 3:06 ` [PATCH 9/12] wm8961: fix resource reclaim in wm8961_register " Axel Lin
2010-07-15 3:07 ` [PATCH 10/12] wm8974: fix a memory leak if another WM8974 is registered Axel Lin
2010-07-15 3:08 ` [PATCH 11/12] wm8978: fix a memory leak if another WM8978 " Axel Lin
2010-07-15 8:37 ` [PATCH v2 " Axel Lin
2010-07-15 8:53 ` Guennadi Liakhovetski
2010-07-15 3:11 ` [PATCH 12/12] wm9081: fix resource reclaim in wm9081_register error path Axel Lin
2010-07-15 8:59 ` [PATCH 0/12] sound/alsa/soc/codec: fix memory leak and resource relaim in " Mark Brown
2010-07-19 1:41 ` Axel Lin
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=AANLkTil_buBStfVy_L9iTgPrxosmSRFhNU1MbaXWJfMx@mail.gmail.com \
--to=axel.lin@gmail.com \
--cc=alsa-devel@alsa-project.org \
--cc=broonie@opensource.wolfsonmicro.com \
--cc=g.liakhovetski@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=lrg@slimlogic.co.uk \
--cc=perex@perex.cz \
--cc=tiwai@suse.de \
--cc=tj@kernel.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;
as well as URLs for NNTP newsgroup(s).