ALSA: fix building with CONFIG_SND_PROC_FS disabled

PATCH In review Arnd Bergmann 3 participants 3 messages May 29, 2015 Sound

lore

Share & tools

lore

Arnd Bergmann

author

2015-05-29 08:47 UTC #0 lore raw

PATCH 1/1 ALSA: fix building with CONFIG_SND_PROC_FS disabled

A recent change to split the procfs support for ALSA turned out to be incomplete, leading to build errors:

sound/core/seq/seq_info.c:61:12: error: redefinition of 'snd_seq_info_init' sound/core/seq/seq_device.c: In function 'seq_dev_proc_init':
sound/core/seq/seq_device.c:276:2: error: 'info_entry' undeclared (first use in this function) sound/core/seq/seq_device.c: In function 'alsa_seq_device_exit':
sound/core/seq/seq_device.c:309:22: error: 'info_entry' undeclared (first use in this function)

This replaces some more instances of CONFIG_PROC_FS with CONFIG_SND_PROC_FS, as the original patch started. It is enough to avoid all randconfig regressions I saw on ARM because of the CONFIG_SND_PROC_FS introduction.

Signed-off-by Arnd Bergmann <[email protected]>

Fixes cd6a65036f0e ("ALSA: replace CONFIG_PROC_FS with CONFIG_SND_PROC_FS")

---

I assume this was already reported. If not, please apply my patch or fold it into the original commit.

sound/core/seq/Makefile +1 -1

index b29ffe835205..b65fa5a1943b 100644
@@ -7,7 +7,7 @@ snd-seq-device-objs := seq_device.o
 snd-seq-objs := seq.o seq_lock.o seq_clientmgr.o seq_memory.o seq_queue.o \
                 seq_fifo.o seq_prioq.o seq_timer.o \
                 seq_system.o seq_ports.o
-snd-seq-$(CONFIG_PROC_FS) += seq_info.o
+snd-seq-$(CONFIG_SND_PROC_FS) += seq_info.o
 snd-seq-midi-objs := seq_midi.o
 snd-seq-midi-emul-objs := seq_midi_emul.o
 snd-seq-midi-event-objs := seq_midi_event.o

sound/core/seq/seq_device.c +2 -2

index 288af56a43cd..c4acf17e9f5e 100644
@@ -272,7 +272,7 @@ EXPORT_SYMBOL_GPL(snd_seq_driver_unregister);
 
 static int __init seq_dev_proc_init(void)
 {
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_SND_PROC_FS
 	info_entry = snd_info_create_module_entry(THIS_MODULE, "drivers",
 						  snd_seq_root);
 	if (info_entry == NULL)

@@ -305,7 +305,7 @@ static void __exit alsa_seq_device_exit(void)
 #ifdef CONFIG_MODULES
 	cancel_work_sync(&autoload_work);
 #endif
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_SND_PROC_FS
 	snd_info_free_entry(info_entry);
 #endif
 	bus_unregister(&snd_seq_bus_type);

Jie, Yang

2015-05-29 09:03 UTC #1 lore raw

Quoted 24 lines

-----Original Message-----
From: Arnd Bergmann [mailto:arnd at arndb.de]
Sent: Friday, May 29, 2015 4:47 PM
To: Takashi Iwai
Cc: broonie at kernel.org; alsa-devel at alsa-project.org; linux-arm- kernel at lists.infradead.org; Jie, Yang
Subject: [PATCH] ALSA: fix building with CONFIG_SND_PROC_FS disabled
A recent change to split the procfs support for ALSA turned out to be incomplete, leading to build errors:
sound/core/seq/seq_info.c:61:12: error: redefinition of 'snd_seq_info_init' sound/core/seq/seq_device.c: In function 'seq_dev_proc_init':
sound/core/seq/seq_device.c:276:2: error: 'info_entry' undeclared (first use in this function)
sound/core/seq/seq_device.c: In function 'alsa_seq_device_exit':
sound/core/seq/seq_device.c:309:22: error: 'info_entry' undeclared (first use in this function)
This replaces some more instances of CONFIG_PROC_FS with CONFIG_SND_PROC_FS, as the original patch started. It is enough to avoid all randconfig regressions I saw on ARM because of the CONFIG_SND_PROC_FS introduction.

Quoted 7 lines

Signed-off-by Arnd Bergmann <[email protected]>
Fixes cd6a65036f0e ("ALSA: replace CONFIG_PROC_FS with
CONFIG_SND_PROC_FS")
---
I assume this was already reported. If not, please apply my patch or fold it into the original commit.

Thank you, Seems this already fixed by Takashi's latest series?

~Keyon

Quoted 34 lines

sound/core/seq/Makefile index renamed from sound/core/seq/Makefile +0 -0

b29ffe835205..b65fa5a1943b 100644

sound/core/seq/Makefile +1 -1

@@ -7,7 +7,7 @@ snd-seq-device-objs := seq_device.o  snd-seq-objs :=
 seq.o seq_lock.o seq_clientmgr.o seq_memory.o seq_queue.o \
                 seq_fifo.o seq_prioq.o seq_timer.o \
                 seq_system.o seq_ports.o
-snd-seq-$(CONFIG_PROC_FS) += seq_info.o
+snd-seq-$(CONFIG_SND_PROC_FS) += seq_info.o
 snd-seq-midi-objs := seq_midi.o
 snd-seq-midi-emul-objs := seq_midi_emul.o  snd-seq-midi-event-objs :=

seq_midi_event.o diff --git a/sound/core/seq/seq_device.c b/sound/core/seq/seq_device.c index 288af56a43cd..c4acf17e9f5e 100644

sound/core/seq/seq_device.c +1 -0

@@ -272,7 +272,7 @@ EXPORT_SYMBOL_GPL(snd_seq_driver_unregister);
 
 static int __init seq_dev_proc_init(void)  { -#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_SND_PROC_FS
 	info_entry = snd_info_create_module_entry(THIS_MODULE,
 "drivers",
 						  snd_seq_root);
 	if (info_entry == NULL)

@@ -305,7 +305,7 @@ static void __exit alsa_seq_device_exit(void)  #ifdef

CONFIG_MODULES

sound/core/seq/seq_device.c +1 -1

 	cancel_work_sync(&autoload_work);
 #endif
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_SND_PROC_FS
 	snd_info_free_entry(info_entry);
 #endif
 	bus_unregister(&snd_seq_bus_type);

Takashi Iwai

2015-05-29 09:31 UTC #2 lore raw

At Fri, 29 May 2015 09:03:35 +0000,

Quoted 30 lines · Yang

Quoted 30 lines
-----Original Message-----
From: Arnd Bergmann [mailto:[email protected]]
Sent: Friday, May 29, 2015 4:47 PM
To: Takashi Iwai
Cc: [email protected]; [email protected]; linux-arm- [email protected]; Jie, Yang
Subject: [PATCH] ALSA: fix building with CONFIG_SND_PROC_FS disabled
A recent change to split the procfs support for ALSA turned out to be incomplete, leading to build errors:
sound/core/seq/seq_info.c:61:12: error: redefinition of 'snd_seq_info_init' sound/core/seq/seq_device.c: In function 'seq_dev_proc_init':
sound/core/seq/seq_device.c:276:2: error: 'info_entry' undeclared (first use in this function)
sound/core/seq/seq_device.c: In function 'alsa_seq_device_exit':
sound/core/seq/seq_device.c:309:22: error: 'info_entry' undeclared (first use in this function)
This replaces some more instances of CONFIG_PROC_FS with CONFIG_SND_PROC_FS, as the original patch started. It is enough to avoid all randconfig regressions I saw on ARM because of the CONFIG_SND_PROC_FS introduction.
Signed-off-by Arnd Bergmann <[email protected]>
Fixes cd6a65036f0e ("ALSA: replace CONFIG_PROC_FS with
CONFIG_SND_PROC_FS")
---

Quoted 4 lines

Quoted 2 lines
I assume this was already reported. If not, please apply my patch or fold it into the original commit.
Thank you, Seems this already fixed by Takashi's latest series?

Yes.

Takashi

Quoted 38 lines

~Keyon

Quoted 34 lines

sound/core/seq/Makefile index renamed from sound/core/seq/Makefile +0 -0

b29ffe835205..b65fa5a1943b 100644

sound/core/seq/Makefile +1 -1

@@ -7,7 +7,7 @@ snd-seq-device-objs := seq_device.o  snd-seq-objs :=
 seq.o seq_lock.o seq_clientmgr.o seq_memory.o seq_queue.o \
                 seq_fifo.o seq_prioq.o seq_timer.o \
                 seq_system.o seq_ports.o
-snd-seq-$(CONFIG_PROC_FS) += seq_info.o
+snd-seq-$(CONFIG_SND_PROC_FS) += seq_info.o
 snd-seq-midi-objs := seq_midi.o
 snd-seq-midi-emul-objs := seq_midi_emul.o  snd-seq-midi-event-objs :=

seq_midi_event.o diff --git a/sound/core/seq/seq_device.c b/sound/core/seq/seq_device.c index 288af56a43cd..c4acf17e9f5e 100644

sound/core/seq/seq_device.c +1 -0

@@ -272,7 +272,7 @@ EXPORT_SYMBOL_GPL(snd_seq_driver_unregister);
 
 static int __init seq_dev_proc_init(void)  { -#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_SND_PROC_FS
 	info_entry = snd_info_create_module_entry(THIS_MODULE,
 "drivers",
 						  snd_seq_root);
 	if (info_entry == NULL)

@@ -305,7 +305,7 @@ static void __exit alsa_seq_device_exit(void)  #ifdef

CONFIG_MODULES

sound/core/seq/seq_device.c +1 -1

 	cancel_work_sync(&autoload_work);
 #endif
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_SND_PROC_FS
 	snd_info_free_entry(info_entry);
 #endif
 	bus_unregister(&snd_seq_bus_type);