Skip to contentSkip to conversation

pata_winbond: fix module init

PATCH In review Dan Carpenter 4 participants 6 messages Aug 3, 2010 – Aug 17, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 pata_winbond: fix module init

winbond_init_one() returns 0 on success and an error code on failure. In the original code, we always just released the regions and return -ENODEV. This was introduce in 2006 by 7e45b0e5f6f6 "[PATCH] libata:
Winbond support".

---
Compile tested.

drivers/ata/pata_winbond.c +3 -2
index 6d8619b..b34b594 100644
@@ -248,9 +248,10 @@ static __init int winbond_init(void)
 
 			if (request_region(port, 2, "pata_winbond")) {
 				ret = winbond_init_one(port);
-				if (ret <= 0)
+				if (ret < 0)
 					release_region(port, 2);
-				else ct+= ret;
+				else
+					ct++;
 			}
 		}
 	}
#1 lore raw

Hi Dan,

Quoted 4 lines · Dan Carpenter

winbond_init_one() returns 0 on success and an error code on failure. In the original code, we always just released the regions and return -ENODEV. This was introduce in 2006 by 7e45b0e5f6f6 "[PATCH] libata:
Winbond support".

You might like to pick up

	http://patchwork.ozlabs.org/patch/39358/

instead..

Thanks.
--
Bartlomiej Zolnierkiewicz

#2 lore raw
Quoted 9 lines · Bartlomiej Zolnierkiewicz

Hi Dan,

Quoted 4 lines · Dan Carpenter

winbond_init_one() returns 0 on success and an error code on failure. In the original code, we always just released the regions and return -ENODEV. This was introduce in 2006 by 7e45b0e5f6f6 "[PATCH] libata:
Winbond support".

Quoted 6 lines

You might like to pick up

	http://patchwork.ozlabs.org/patch/39358/

instead..

I don't have the hardware, but yes, probably it's better to replace this driver since it has never been able to load.

regards,
dan carpenter

#3 lore raw
Quoted 12 lines · On Tue, 17 Aug 2010 08:29:40 +0200 Dan Carpenter <[email protected]> wrote:
Quoted 11 lines · Bartlomiej Zolnierkiewicz

Hi Dan,

Quoted 4 lines · Dan Carpenter

winbond_init_one() returns 0 on success and an error code on failure. In the original code, we always just released the regions and return -ENODEV. This was introduce in 2006 by 7e45b0e5f6f6 "[PATCH] libata:
Winbond support".

You might like to pick up

Quoted 7 lines
Quoted 4 lines
	http://patchwork.ozlabs.org/patch/39358/

instead..

I don't have the hardware, but yes, probably it's better to replace this driver since it has never been able to load.

It was obsoleted into pata_legacy before it was ever finalised - unfortunately I could then not get Jeff to delete it.

#4 lore raw
Quoted 17 lines · Alan Cox
Quoted 14 lines · On Tue, 17 Aug 2010 08:29:40 +0200 Dan Carpenter<[email protected]> wrote:
Quoted 13 lines · Bartlomiej Zolnierkiewicz

Hi Dan,

Quoted 4 lines · Dan Carpenter

winbond_init_one() returns 0 on success and an error code on failure. In the original code, we always just released the regions and return -ENODEV. This was introduce in 2006 by 7e45b0e5f6f6 "[PATCH] libata:
Winbond support".

You might like to pick up

	http://patchwork.ozlabs.org/patch/39358/
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

instead..

I don't have the hardware, but yes, probably it's better to replace this driver since it has never been able to load.

It was obsoleted into pata_legacy before it was ever finalised - unfortunately I could then not get Jeff to delete it.

Easy enough to delete now...

#5 lore raw
In reply to Bartlomiej Zolnierkiewicz
Quoted 4 lines · Bartlomiej Zolnierkiewicz

Hi Dan,

On Tuesday 03 August 2010 05:27:47 pm Dan Carpenter wrote:

Quoted 8 lines
Quoted 4 lines

winbond_init_one() returns 0 on success and an error code on failure. In the original code, we always just released the regions and return -ENODEV. This was introduce in 2006 by 7e45b0e5f6f6 "[PATCH] libata:
Winbond support".

You might like to pick up

	http://patchwork.ozlabs.org/patch/39358/

Picked it up...