[PATCH 4/4] libata: odd length data handling for __atapi_pio_bytes()

[Albert Lee <albertcc@tw.ibm.com>]

[-- Attachment #1: Type: text/plain, Size: 424 bytes --]

Hi Jeff,

Resend the patch #4.

Problem:
- When odd length data is transfered, the last byte of data is lost.

Changes:
- When an odd length data transfer is seen, round it up to the next even length.
   The same as atapi_input_bytes() in the ide-iops does.
   This may overrun the odd length buffer by one byte.

- Rename 'dwords' to 'words' in ata_pio_data_xfer()


Albert

Signed-off-by: Albert Lee <albertcc@tw.ibm.com>

[-- Attachment #2: 04_odd_data.diff --]
[-- Type: text/plain, Size: 953 bytes --]

--- 13_atapi_pio_trailing_data_handling/drivers/scsi/libata-core.c	2005-06-09 11:05:10.000000000 +0800
+++ 14_atapi_pio_odd_handling/drivers/scsi/libata-core.c	2005-06-10 15:08:24.000000000 +0800
@@ -2491,7 +2491,7 @@
 			       unsigned int buflen, int write_data)
 {
 	unsigned int i;
-	unsigned int words = buflen >> 1;
+	unsigned int words = (buflen+1) >> 1;
 	u16 *buf16 = (u16 *) buf;
 	void __iomem *mmio = (void __iomem *)ap->ioaddr.data_addr;
 
@@ -2507,12 +2507,12 @@
 static void ata_pio_data_xfer(struct ata_port *ap, unsigned char *buf,
 			      unsigned int buflen, int write_data)
 {
-	unsigned int dwords = buflen >> 1;
+	unsigned int words = (buflen+1) >> 1;
 
 	if (write_data)
-		outsw(ap->ioaddr.data_addr, buf, dwords);
+		outsw(ap->ioaddr.data_addr, buf, words);
 	else
-		insw(ap->ioaddr.data_addr, buf, dwords);
+		insw(ap->ioaddr.data_addr, buf, words);
 }
 
 static void ata_data_xfer(struct ata_port *ap, unsigned char *buf,