linux-kernel - Re: [PATCH] snd: hda: realtek: fix incorrect IS

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <87tt88oo05.wl-tiwai@suse.de>
Date: Tue, 04 Mar 2025 15:43:06 +0100
From: Takashi Iwai <tiwai@...e.de>
To: Arnd Bergmann <arnd@...nel.org>
Cc: Jaroslav Kysela <perex@...ex.cz>,
	Takashi Iwai <tiwai@...e.com>,
	Kailang Yang <kailang@...ltek.com>,
	Arnd Bergmann <arnd@...db.de>,
	Simon Trimmer <simont@...nsource.cirrus.com>,
	Joshua Grisham <josh@...huagrisham.com>,
	linux-sound@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] snd: hda: realtek: fix incorrect IS_REACHABLE() usage

On Tue, 04 Mar 2025 15:25:55 +0100,
Arnd Bergmann wrote:
> 
> From: Arnd Bergmann <arnd@...db.de>
> 
> The alternative path leads to a build error after a recent change:
> 
> sound/pci/hda/patch_realtek.c: In function 'alc233_fixup_lenovo_low_en_micmute_led':
> include/linux/stddef.h:9:14: error: called object is not a function or function pointer
>     9 | #define NULL ((void *)0)
>       |              ^
> sound/pci/hda/patch_realtek.c:5041:49: note: in expansion of macro 'NULL'
>  5041 | #define alc233_fixup_lenovo_line2_mic_hotkey    NULL
>       |                                                 ^~~~
> sound/pci/hda/patch_realtek.c:5063:9: note: in expansion of macro 'alc233_fixup_lenovo_line2_mic_hotkey'
>  5063 |         alc233_fixup_lenovo_line2_mic_hotkey(codec, fix, action);
> 
> Using IS_REACHABLE() is somewhat questionable here anyway since it
> leads to the input code not working when the HDA driver is builtin
> but input is in a loadable module. Replace this with a hard compile-time
> dependency on CONFIG_INPUT. In practice this won't chance much
> other than solve the compiler error because it is rare to require
> sound output but no input support.
> 
> Fixes: f603b159231b ("ALSA: hda/realtek - add supported Mic Mute LED for Lenovo platform")
> Signed-off-by: Arnd Bergmann <arnd@...db.de>

Thanks for the patch.  Indeed it's a very corner case, but I still
hesitate to add a kconfig dependency.  Can we take an alternative fix
to define the proper dummy functions like below instead?


Takashi

-- 8< --
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -5038,8 +5038,21 @@ static void alc233_fixup_lenovo_line2_mic_hotkey(struct hda_codec *codec,
 	}
 }
 #else /* INPUT */
-#define alc280_fixup_hp_gpio2_mic_hotkey	NULL
-#define alc233_fixup_lenovo_line2_mic_hotkey	NULL
+static void alc280_fixup_hp_gpio2_mic_hotkey(struct hda_codec *codec,
+					     const struct hda_fixup *fix,
+					     int action)
+{
+	if (action == HDA_FIXUP_ACT_PRE_PROBE)
+		codec_info(codec, "Input is disabled, skipping quirk\n");
+}
+
+static void alc233_fixup_lenovo_line2_mic_hotkey(struct hda_codec *codec,
+						 const struct hda_fixup *fix,
+						 int action)
+{
+	if (action == HDA_FIXUP_ACT_PRE_PROBE)
+		codec_info(codec, "Input is disabled, skipping quirk\n");
+}
 #endif /* INPUT */
 
 static void alc269_fixup_hp_line1_mic1_led(struct hda_codec *codec,