Hello.

Bartlomiej Zolnierkiewicz wrote:

On Saturday 21 June 2008, Sergei Shtylyov wrote:
  

quoted

Hello.

Bartlomiej Zolnierkiewicz wrote:
    

quoted

Export sff_dma_ops and then remove ide_setup_dma().

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

      

Acked-by: Sergei Shtylyov <redacted>

quoted

quoted

Index: b/drivers/ide/arm/palm_bk3710.c
===================================================================

--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c

@@ -318,15 +318,14 @@ static u8 __devinit palm_bk3710_cable_de
 static int __devinit palm_bk3710_init_dma(ide_hwif_t *hwif,
 					  const struct ide_port_info *d)
 {
-	unsigned long base =
-		hwif->io_ports.data_addr - IDE_PALM_ATA_PRI_REG_OFFSET;
-
 	printk(KERN_INFO "    %s: MMIO-DMA\n", hwif->name);
 
 	if (ide_allocate_dma_engine(hwif))
 		return -1;
 
-	ide_setup_dma(hwif, base);
+	hwif->dma_base = hwif->io_ports.data_addr - IDE_PALM_ATA_PRI_REG_OFFSET;
  
      

   That's er... too complex. Actually, 'base' has the value for 
'dma_base' at that moment...
    

Well, the above place is the only user of 'base' so isn't the variable
kind of superfluous?
  

   Oops, sorry. I somehow thought this change was in other context (was 
looking at the older version of driver recently).

Thanks,
Bart
  

MBR, Sergei