Re: [PATCH 04/15] block: add an API to atomically update queue limits

[Christoph Hellwig <hch@lst.de>]

[switching Cc lists to be more relevant]

Hi Christian,

this is another case where the block layer previously warned about
somethign that it enforces now.  Switch the dma_boundary in
sata_dwc_sht to 16k (patch below) will "fix" this, but it directly
contradicts the comment next to that field.

---
diff --git a/drivers/ata/sata_dwc_460ex.c b/drivers/ata/sata_dwc_460ex.c
index 52f5168e4db542..4c182694ba51d1 100644
--- a/drivers/ata/sata_dwc_460ex.c
+++ b/drivers/ata/sata_dwc_460ex.c
@@ -1089,8 +1089,10 @@ static const struct scsi_host_template sata_dwc_sht = {
 	 * boundary. If the block spans such a FIS boundary, there is a chance
 	 * that a DMA burst will cross that boundary -- this results in an
 	 * error in the host controller.
+	 *
+	 * XXX
 	 */
-	.dma_boundary		= 0x1fff /* ATA_DMA_BOUNDARY */,
+	.dma_boundary		= SZ_16K - 1,
 };
 
 static struct ata_port_operations sata_dwc_ops = {