From: Tejun Heo <htejun@gmail.com>
Date: Fri, 31 Mar 2006 16:38:18 +0000 (+0900)
Subject: [PATCH] libata: make ata_set_mode() handle no-device case properly
X-Git-Tag: v2.6.18-rc1~1079^2~161
X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=4f65977df0b9a667fdcd85b95d457a220c94113f;p=karo-tx-linux.git

[PATCH] libata: make ata_set_mode() handle no-device case properly

Make ata_set_mode() return without doing anything if there is no
device on the port.  This is in preparation for ata_bus_probe()
changes.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index c10c550da38b..913e378eca7f 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -1823,7 +1823,7 @@ static void ata_host_set_dma(struct ata_port *ap)
  */
 static void ata_set_mode(struct ata_port *ap)
 {
-	int i, rc, used_dma = 0;
+	int i, rc, used_dma = 0, found = 0;
 
 	/* step 1: calculate xfer_mask */
 	for (i = 0; i < ATA_MAX_DEVICES; i++) {
@@ -1842,9 +1842,12 @@ static void ata_set_mode(struct ata_port *ap)
 		dev->pio_mode = ata_xfer_mask2mode(pio_mask);
 		dev->dma_mode = ata_xfer_mask2mode(dma_mask);
 
+		found = 1;
 		if (dev->dma_mode)
 			used_dma = 1;
 	}
+	if (!found)
+		return;
 
 	/* step 2: always set host PIO timings */
 	rc = ata_host_set_pio(ap);