Skip to contentSkip to conversation

2.5.5-pre1 kernel - compile as modules?

DISCUSSION Duncan Sands 5 participants 8 messages Feb 14, 2002 – Feb 15, 2002

lore
Share & tools
lore
#0 lore raw

I just gave the 2.5.5-pre1 kernel a whirl. Just to be different, I compiled all sound support into the kernel rather than as modules. On startup I get:

Feb 14 15:43:55 baldrick kernel: Advanced Linux Sound Architecture Driver Version 0.9.0beta10.
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-0, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-1, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-2, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-3, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-4, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-5, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-6, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-7, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: PCI: Found IRQ 5 for device 00:0a.0
Feb 14 15:43:55 baldrick kernel: ALSA device list:
Feb 14 15:43:55 baldrick kernel:   #0: Sound Fusion CS46xx at 0xda800000/0xda000000, irq 5

And then later there are all sorts of problems, such as

cat /proc/asound/cards

claims there is no such device.

Must sound support be compiled as modules?

Thanks,

Duncan.

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

#1 lore raw
Quoted 8 lines · Duncan Sands

I just gave the 2.5.5-pre1 kernel a whirl. Just to be different, I compiled all sound support into the kernel rather than as modules. On startup I get:

Feb 14 15:43:55 baldrick kernel: Advanced Linux Sound Architecture Driver Version 0.9.0beta10.
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-0, errno = 2
Feb 14 15:43:55 baldrick last message repeated 7 times
Feb 14 15:43:55 baldrick kernel: kmod: failed to exec /sbin/modprobe -s -k snd-card-1, errno = 2

I cut a bit out here...

Quoted 5 lines

And then later there are all sorts of problems, such as

cat /proc/asound/cards

Quoted 8 lines

claims there is no such device.

Must sound support be compiled as modules?

Thanks,

Duncan.

I have been backporting the 2.5 patch to 2.4.18-preX and compiling drivers into the kernel only for hardware which I have installed on the machine.

So far this has worked (at least partially - see below) with yamaha YMF-744
sound blaster live
sound blaster 1371

I also have compiled for the Delta 1010 (ICE) but don't have all the utilities so I don't know if it is really working (might check it out a little more this weekend)

I've mostly used the mixer (aumix) and PCM devices (playing MP3 files) with success.

Because this seemed to work I did not even look to see if there were any error messages generated. The only thing I did before trying to use the devices was to look in /proc.

If anyone has suggestions for how to easily test more devices quickly feel free to post some instructions and I'll post the results here. (I have some other HW I could test as well. gravis and others)

Actually if anyone would care to help in putting together an automated test suite... I could justify spending time to work on it.

Regards,

Curtis

Signature · 11 lines
Curtis Veit                    email: [email protected]
Lineo Inc.                     Where Open Meets Smart
390 South 400 West             phone: 801.426.5001 
Lindon, Utah, USA  84042       fax:   801.852.8071



_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel
#3 lore raw

did you insmod the snd.o ???

Quoted 10 lines · On 15-Feb-02 Duncan Sands wrote:

Actually, the problem is deeper: even if I compile everything as modules, I still get "no such device" when trying to read from /proc/asound entries. I will look into this further myself.

Duncan.

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Guenther Sohler

NewLogic Technologies AG
Millennium Park 6
A-6890 Lustenau

                                           Phone: +43-5577-62000-507

E-Mail: [email protected] Fax: +43-5577-62000-988

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

#4 lore raw
Quoted 1 line · Guenther Sohler

did you insmod the snd.o ???

Yes. I have /proc/asound with all entries. They just aren't readable. The soundcard is detected when I modprobe the card module (nonetheless /proc/asound/cards is not readable:
no such device). I have been using the CVS and other versions of alsa with no problems, so this is presumably something to do with the kernel sound patch.

Thanks for thinking about this,

Duncan.

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

#5 lore raw

Could this be a devfs (device filesystem) problem ?
With devfs, devices appear as if by magic in the /dev directory. The old system had all devices having entries in /dev whether present or not.

Cheers
James

Quoted 19 lines

-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of Duncan Sands
Sent: 15 February 2002 08:34
To: Guenther Sohler
Cc: [email protected]; [email protected]
Subject: Re: [Alsa-devel] 2.5.5-pre1 kernel - compile as modules?

Quoted 1 line · Guenther Sohler

did you insmod the snd.o ???

Yes. I have /proc/asound with all entries. They just aren't readable. The soundcard is detected when I modprobe the card module (nonetheless /proc/asound/cards is not readable:
no such device). I have been using the CVS and other versions of alsa with no problems, so this is presumably something to do with the kernel sound patch.

Quoted 8 lines

Thanks for thinking about this,

Duncan.

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

#6 lore raw
In reply to Duncan Sands
Quoted 3 lines · Duncan Sands

Actually, the problem is deeper: even if I compile everything as modules, I still get "no such device" when trying to read from /proc/asound entries. I will look into this further myself.

It's problem with recent 2.5 kernel. Try the attached patch.

Jaroslav

-----
Jaroslav Kysela <[email protected]>
Linux Kernel Sound Maintainer
ALSA Project http://www.alsa-project.org
SuSE Linux http://www.suse.com

Index: info.c
===================================================================
RCS file: /cvsroot/alsa/alsa-kernel/core/info.c,v
retrieving revision 1.7
retrieving revision 1.8

info.c +5 -5
diff -u -r1.7 -r1.8
@@ -278,7 +278,7 @@
 	int mode, err;
 
 	down(&info_mutex);
-	p = (struct proc_dir_entry *) inode->u.generic_ip;
+	p = PDE(inode);
 	entry = p == NULL ? NULL : (snd_info_entry_t *)p->data;
 	if (entry == NULL) {
 		up(&info_mutex);
@@ -541,7 +541,7 @@
 static int snd_info_card_readlink(struct dentry *dentry,
 				  char *buffer, int buflen)
 {
-        char *s = ((struct proc_dir_entry *) dentry->d_inode->u.generic_ip)->data;
+        char *s = PDE(dentry->d_inode)->data;
 #ifndef LINUX_2_2
 	return vfs_readlink(dentry, buffer, buflen, s);
 #else
@@ -562,7 +562,7 @@
 static int snd_info_card_followlink(struct dentry *dentry,
 				    struct nameidata *nd)
 {
-        char *s = ((struct proc_dir_entry *) dentry->d_inode->u.generic_ip)->data;
+        char *s = PDE(dentry->d_inode)->data;
         return vfs_follow_link(nd, s);
 }
 #else
@@ -570,7 +570,7 @@
 					       struct dentry *base,
 					       unsigned int follow)
 {
-	char *s = ((struct proc_dir_entry *) dentry->d_inode->u.generic_ip)->data;
+	char *s = PDE(dentry->d_inode)->data;
 	return lookup_dentry(s, base, follow);
 }
 #endif
@@ -845,7 +845,7 @@
 		return;
 	}
 	MOD_INC_USE_COUNT;
-	de = (struct proc_dir_entry *) inode->u.generic_ip;
+	de = PDE(inode);
 	if (de == NULL)
 		return;
 	entry = (snd_info_entry_t *) de->data;
#7 lore raw
Quoted 6 lines · Jaroslav Kysela
Quoted 3 lines · Duncan Sands

Actually, the problem is deeper: even if I compile everything as modules, I still get "no such device" when trying to read from /proc/asound entries. I will look into this further myself.

It's problem with recent 2.5 kernel. Try the attached patch.

Yes, that works. Thanks!

Duncan.

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel