driverdev-devel.linuxdriverproject.org archive mirror
 help / color / mirror / Atom feed
From: Colin Ian King <colin.king@canonical.com>
To: Johan Hovold <johan@kernel.org>, Alex Elder <elder@kernel.org>,
	"Greg Kroah-Hartman <gregkh"@linuxfoundation.org,
	Vaibhav Agarwal <vaibhav.sr@gmail.com>,
	Mark Greer <mgreer@animalcreek.com>,
	greybus-dev@lists.linaro.org,
	"devel@driverdev.osuosl.org" <devel@driverdev.osuosl.org>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: issue with uninitialized value used in a comparison in gbcodec_mixer_dapm_ctl_put
Date: Thu, 30 Jul 2020 17:02:22 +0100	[thread overview]
Message-ID: <7da6363c-9423-2b9f-029a-395cc8a932d7@canonical.com> (raw)

Hi,

Static analysis with Coverity has detected an uninitialized value being
used in a comparison.  The error was detected on a recent change to
drivers/staging/greybus/audio_topology.c however the issue actually
dates back to the original commit:

commit 6339d2322c47f4b8ebabf9daf0130328ed72648b
Author: Vaibhav Agarwal <vaibhav.agarwal@linaro.org>
Date:   Wed Jan 13 14:07:51 2016 -0700

    greybus: audio: Add topology parser for GB codec

The analysis is as follows:

425 static int gbcodec_mixer_dapm_ctl_put(struct snd_kcontrol *kcontrol,
426                                      struct snd_ctl_elem_value
*ucontrol)
427 {
428        int ret, wi, max, connect;
429        unsigned int mask, val;
430        struct gb_audio_ctl_elem_info *info;
431        struct gbaudio_ctl_pvt *data;

   1. var_decl: Declaring variable gbvalue without initializer.
432        struct gb_audio_ctl_elem_value gbvalue;
433        struct gbaudio_module_info *module;
434        struct snd_soc_dapm_widget_list *wlist =
snd_kcontrol_chip(kcontrol);
435        struct snd_soc_dapm_widget *widget = wlist->widgets[0];
436        struct device *codec_dev = widget->dapm->dev;
437        struct gbaudio_codec_info *gb = dev_get_drvdata(codec_dev);
438        struct gb_bundle *bundle;
439

   2. Condition 0 /* __builtin_types_compatible_p() */, taking false branch.
   3. Condition 1 /* __builtin_types_compatible_p() */, taking true branch.
   4. Falling through to end of if statement.
   5. Condition !!branch, taking false branch.
   6. Condition ({...; !!branch;}), taking false branch.

440        dev_dbg(codec_dev, "Entered %s:%s\n", __func__,
kcontrol->id.name);
441        module = find_gb_module(gb, kcontrol->id.name);

   7. Condition !module, taking false branch.
442        if (!module)
443                return -EINVAL;
444
445        data = (struct gbaudio_ctl_pvt *)kcontrol->private_value;
446        info = (struct gb_audio_ctl_elem_info *)data->info;

   8. Condition 0 /* !!(!__builtin_types_compatible_p() &&
!__builtin_types_compatible_p()) */, taking false branch.
447        bundle = to_gb_bundle(module->dev);
448

   9. Condition data->vcount == 2, taking true branch.
449        if (data->vcount == 2)
450                dev_warn(widget->dapm->dev,
451                         "GB: Control '%s' is stereo, which is not
supported\n",
452                         kcontrol->id.name);
453
454        max = le32_to_cpu(info->value.integer.max);
455        mask = (1 << fls(max)) - 1;
456        val = ucontrol->value.integer.value[0] & mask;

   10. Condition !!val, taking true branch.
457        connect = !!val;
458
459        /* update ucontrol */

Uninitialized scalar variable (UNINIT)
   11. uninit_use: Using uninitialized value gbvalue.value.integer_value[0].
460        if (gbvalue.value.integer_value[0] != val) {

The gbvalue.value.integer_value[0] read is bogus since gbvalue was
declared on the stack but was not initialized.  There seems to be no
where that sets this data. I'm assuming most of the time that the
comparison works because the garbage value is different from val and so
the code in the if stanza is executed.

Anyhow, I'm unsure what the original intent of the code was, so I've not
attempted to fix this.

Colin


_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

             reply	other threads:[~2020-07-30 16:02 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-30 16:02 Colin Ian King [this message]
2020-07-31  5:08 ` issue with uninitialized value used in a comparison in gbcodec_mixer_dapm_ctl_put Vaibhav Agarwal
2020-08-05 13:35 ` Alex Elder
2020-08-06 16:17   ` Vaibhav Agarwal

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=7da6363c-9423-2b9f-029a-395cc8a932d7@canonical.com \
    --to=colin.king@canonical.com \
    --cc="Greg Kroah-Hartman <gregkh"@linuxfoundation.org \
    --cc=devel@driverdev.osuosl.org \
    --cc=elder@kernel.org \
    --cc=greybus-dev@lists.linaro.org \
    --cc=johan@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgreer@animalcreek.com \
    --cc=vaibhav.sr@gmail.com \
    /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).