pata_cs5520: remove incorrect warning message

PATCH In review Bartlomiej Zolnierkiewicz 2 participants 4 messages Oct 11, 2011 – Oct 12, 2011 libata

lore

Share & tools

lore

Bartlomiej Zolnierkiewicz

author

2011-10-11 18:00 UTC #0 lore raw

PATCH 1/1 pata_cs5520: remove incorrect warning message

From: Bartlomiej Zolnierkiewicz <[email protected]>
Subject: [PATCH] pata_cs5520: remove incorrect warning message

Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

---
 drivers/ata/pata_cs5520.c |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

drivers/ata/pata_cs5520.c +1 -3

Index: b/drivers/ata/pata_cs5520.c
===================================================================
@@ -148,10 +148,8 @@ static int __devinit cs5520_init_one(str
 	if (pcicfg & 2)
 		ppi[1] = &pi;
 
-	if ((pcicfg & 0x40) == 0) {
-		dev_warn(&pdev->dev, "DMA mode disabled. Enabling.\n");
+	if ((pcicfg & 0x40) == 0)
 		pci_write_config_byte(pdev, 0x60, pcicfg | 0x40);
-	}
 
 	pi.mwdma_mask = id->driver_data;

Alan Cox

2011-10-11 19:56 UTC #1 lore raw

Quoted 4 lines · On Tue, 11 Oct 2011 20:00:44 +0200 Bartlomiej Zolnierkiewicz…

From: Bartlomiej Zolnierkiewicz <[email protected]>
Subject: [PATCH] pata_cs5520: remove incorrect warning message
Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

Nothing incorrect about it. It's rather useful information for debugging CS5520 problems. Users should never see it - and I've not seen it occuring on any standard VSA releases.

Do you have any examples of it confusing users or even showing up ?

Alan

Bartlomiej Zolnierkiewicz

author

2011-10-12 15:28 UTC #2 lore raw

Quoted 6 lines · Alan Cox

Quoted 3 lines · On Tue, 11 Oct 2011 20:00:44 +0200 Bartlomiej Zolnierkiewicz…
From: Bartlomiej Zolnierkiewicz <[email protected]>
Subject: [PATCH] pata_cs5520: remove incorrect warning message

Quoted 7 lines

Quoted 1 line
Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>
Nothing incorrect about it. It's rather useful information for debugging CS5520 problems. Users should never see it - and I've not seen it occuring on any standard VSA releases.
Do you have any examples of it confusing users or even showing up ?

This driver dooen't even use DMA mode so warning about forcing the setting may be puzzling. In addition we do the same thing cs5520_reinit_one() but without warning the user this time. Thus removal of the warning makes it easier for *_reinit_one() replacement by ->init_host method later (patch is ready for this).

Alan Cox

2011-10-12 17:50 UTC #3 lore raw

O> This driver dooen't even use DMA mode so warning about forcing the setting

Quoted 4 lines

may be puzzling. In addition we do the same thing cs5520_reinit_one() but without warning the user this time. Thus removal of the warning makes it easier for *_reinit_one() replacement by ->init_host method later (patch is ready for this).

The re-init case won't find it mis-set. And the CS5520 is not a normal controller even for DMA so I'd prefer it left as is.