Commit e1cc9de8 authored by Alan Cox's avatar Alan Cox Committed by Jeff Garzik

libata-sff: Fix documentation

Code moved to ioread/iowrite but the comment didn't
Also note a posting issue
Signed-off-by: default avatarAlan Cox <alan@redhat.com>
Signed-off-by: default avatarJeff Garzik <jeff@garzik.org>
parent 0e3dbc01
......@@ -297,7 +297,7 @@ void ata_bmdma_start (struct ata_queued_cmd *qc)
dmactl = ioread8(ap->ioaddr.bmdma_addr + ATA_DMA_CMD);
iowrite8(dmactl | ATA_DMA_START, ap->ioaddr.bmdma_addr + ATA_DMA_CMD);
/* Strictly, one may wish to issue a readb() here, to
/* Strictly, one may wish to issue an ioread8() here, to
* flush the mmio write. However, control also passes
* to the hardware at this point, and it will interrupt
* us when we are to resume control. So, in effect,
......@@ -307,6 +307,9 @@ void ata_bmdma_start (struct ata_queued_cmd *qc)
* is expected, so I think it is best to not add a readb()
* without first all the MMIO ATA cards/mobos.
* Or maybe I'm just being paranoid.
*
* FIXME: The posting of this write means I/O starts are
* unneccessarily delayed for MMIO
*/
}
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment