linux-kernel - [PATCH] ASoC: soc-component: don't report of_xlate_dai

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-next>] [day] [month] [year] [list]

Message-Id: <20200723142020.1338740-1-jbrunet@baylibre.com>
Date:   Thu, 23 Jul 2020 16:20:20 +0200
From:   Jerome Brunet <jbrunet@...libre.com>
To:     Mark Brown <broonie@...nel.org>,
        Liam Girdwood <lgirdwood@...il.com>
Cc:     Jerome Brunet <jbrunet@...libre.com>, alsa-devel@...a-project.org,
        linux-kernel@...r.kernel.org,
        Kuninori Morimoto <kuninori.morimoto.gx@...esas.com>
Subject: [PATCH] ASoC: soc-component: don't report of_xlate_dai_name failures

With commit e2329eeba45f ("ASoC: soc-component: add soc_component_err()")
every error different for ENOTSUPP or EPROBE_DEFER will log an error.

However, as explained in snd_soc_get_dai_name(), this callback may error
to indicate that the DAI is not matched by the component tested. If the
device provides other components, those may still match. Logging an error
in this case is misleading.

Don't use soc_component_ret() in snd_soc_component_of_xlate_dai_name()
to avoid spamming the log.

Fixes: e2329eeba45f ("ASoC: soc-component: add soc_component_err()")
Cc: Kuninori Morimoto <kuninori.morimoto.gx@...esas.com>
Signed-off-by: Jerome Brunet <jbrunet@...libre.com>
---

 If we wish to report an error when no name is found for dai phandle,
 I think it would be more approriate to it with with "struct device *"
 of the card, which is the one actually failing to get what it needs.

 These user of snd_soc_get_dai_name() are already reporting the error:
 * sound/soc/samsung/tm2_wm5110.c
 * sound/soc/rockchip/rk3288_hdmi_analog.c

 These could be silently failing:
 * Simple cards
 * Amlogic's cards
 * Samsumg odroid card

 sound/soc/soc-component.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/sound/soc/soc-component.c b/sound/soc/soc-component.c
index 9565a0dd7cb6..fb6479b22d26 100644
--- a/sound/soc/soc-component.c
+++ b/sound/soc/soc-component.c
@@ -344,13 +344,15 @@ int snd_soc_component_of_xlate_dai_name(struct snd_soc_component *component,
 					struct of_phandle_args *args,
 					const char **dai_name)
 {
-	int ret = -ENOTSUPP;
-
 	if (component->driver->of_xlate_dai_name)
-		ret = component->driver->of_xlate_dai_name(component,
-							   args, dai_name);
-
-	return soc_component_ret(component, ret);
+		return component->driver->of_xlate_dai_name(component,
+							    args, dai_name);
+	/*
+	 * Don't use soc_component_ret here because we may not want to report
+	 * the error just yet. If a device has more than one component, the
+	 * first may not match and we don't want spam the log with this.
+	 */
+	return -ENOTSUPP;
 }
 
 void snd_soc_component_setup_regmap(struct snd_soc_component *component)
-- 
2.25.4