Re: [KJ] Fwd: I noticed these

Re: [KJ] Fwd: I noticed these

[Duncan Palmer]

[-- Attachment #1: Type: text/plain, Size: 794 bytes --]

(sorry for not replying to the original post - I've just joined the list)

Randy Dunlap wrote:

> 1. Psmouse and input event drivers don't have module alias autoload 
> support. I think it should be possible to do this similar to how 
> misc_device works.

I've just been having a look at this, and have attached an example
patch (untested, and only partial) - does this look like the right
approach? The various input event drivers each use a range of minor
numbers, so if people think I'm doing the right thing so far, I'll
take these into account when I finish my changes.

2. /proc interface in /dev/rtc doesn't seem to have MODULE_OWNER ref 
counting.

No idea about this. SET_MODULE_OWNER is defined to a no-op, and I
can't find any info as to its purpose. Can anyone enlighten me?

Dunk

[-- Attachment #2: diff --]
[-- Type: application/octet-stream, Size: 1798 bytes --]

diff -u linux-2.6.11-rc2-mm1.orig/include/linux/input.h linux-2.6.11-rc2-mm1/include/linux/input.h
--- linux-2.6.11-rc2-mm1.orig/include/linux/input.h	2005-01-28 13:01:08.000000000 +1100
+++ linux-2.6.11-rc2-mm1/include/linux/input.h	2005-02-01 11:32:40.000000000 +1100
@@ -1012,5 +1012,9 @@
 
 extern struct class_simple *input_class;
 
+#define MODULE_ALIAS_INPUT(minorbase)  \
+	MODULE_ALIAS("char-major-" __stringify(INPUT_MAJOR) "-" \
+            __stringify(minorbase))
+
 #endif
 #endif
diff -u linux-2.6.11-rc2-mm1.orig/drivers/input/input.c linux-2.6.11-rc2-mm1/drivers/input/input.c
--- linux-2.6.11-rc2-mm1.orig/drivers/input/input.c	2005-01-28 13:00:29.000000000 +1100
+++ linux-2.6.11-rc2-mm1/drivers/input/input.c	2005-02-01 11:40:44.000000000 +1100
@@ -536,8 +536,13 @@
 	struct file_operations *old_fops, *new_fops = NULL;
 	int err;
 
-	/* No load-on-demand here? */
-	if (!handler || !(new_fops = fops_get(handler->fops)))
+	if (!handler)
+    {
+        request_module("char-major-%d-%d", INPUT_MAJOR, iminor(inode));
+		/* The module will register itself upon loading */
+		handler = input_table[iminor(inode) >> 5];
+    }
+    if (!handler || !(new_fops = fops_get(handler->fops)))
 		return -ENODEV;
 
 	/*
diff -u linux-2.6.11-rc2-mm1.orig/drivers/input/mousedev.c linux-2.6.11-rc2-mm1/drivers/input/mousedev.c
--- linux-2.6.11-rc2-mm1.orig/drivers/input/mousedev.c	2005-01-28 13:00:30.000000000 +1100
+++ linux-2.6.11-rc2-mm1/drivers/input/mousedev.c	2005-02-01 11:32:53.000000000 +1100
@@ -32,6 +32,7 @@
 MODULE_AUTHOR("Vojtech Pavlik <vojtech@ucw.cz>");
 MODULE_DESCRIPTION("Mouse (ExplorerPS/2) device interfaces");
 MODULE_LICENSE("GPL");
+MODULE_ALIAS_INPUT(MOUSEDEV_MINOR_BASE);
 
 #ifndef CONFIG_INPUT_MOUSEDEV_SCREEN_X
 #define CONFIG_INPUT_MOUSEDEV_SCREEN_X	1024

[-- Attachment #3: Type: text/plain, Size: 167 bytes --]

_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors

[KJ] Fwd: I noticed these

[Randy.Dunlap]

anyone interested?

-------- Original Message --------

Could you look into, or forward to Kernel Janitor list.

1. Psmouse and input event drivers don't have module alias autoload 
support.
    I think it should be possible to do this similar to how 
misc_device works.

2. /proc interface in /dev/rtc doesn't seem to have MODULE_OWNER ref 
counting.

---
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors