sata_mv: correct S/G table limits
Jeff Garzik [Tue, 9 Oct 2007 17:51:57 +0000 (13:51 -0400)]
The recent mv_fill_sg() rewrite, to fix a data corruption problem
related to IOMMU virtual merging, forgot to account for the
potentially-increased size of the scatter/gather table after its run.

Additionally, the DMA boundary is reduced from 0xffffffff to 0xffff
to more closely match the needs of mv_fill_sg().

Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

drivers/ata/sata_mv.c

index 1a82e22..cb7dec9 100644 (file)
@@ -313,7 +313,10 @@ enum {
 #define IS_GEN_IIE(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_IIE)
 
 enum {
-       MV_DMA_BOUNDARY         = 0xffffffffU,
+       /* DMA boundary 0xffff is required by the s/g splitting
+        * we need on /length/ in mv_fill-sg().
+        */
+       MV_DMA_BOUNDARY         = 0xffffU,
 
        /* mask of register bits containing lower 32 bits
         * of EDMA request queue DMA address
@@ -448,7 +451,7 @@ static struct scsi_host_template mv5_sht = {
        .queuecommand           = ata_scsi_queuecmd,
        .can_queue              = ATA_DEF_QUEUE,
        .this_id                = ATA_SHT_THIS_ID,
-       .sg_tablesize           = MV_MAX_SG_CT,
+       .sg_tablesize           = MV_MAX_SG_CT / 2,
        .cmd_per_lun            = ATA_SHT_CMD_PER_LUN,
        .emulated               = ATA_SHT_EMULATED,
        .use_clustering         = 1,
@@ -466,7 +469,7 @@ static struct scsi_host_template mv6_sht = {
        .queuecommand           = ata_scsi_queuecmd,
        .can_queue              = ATA_DEF_QUEUE,
        .this_id                = ATA_SHT_THIS_ID,
-       .sg_tablesize           = MV_MAX_SG_CT,
+       .sg_tablesize           = MV_MAX_SG_CT / 2,
        .cmd_per_lun            = ATA_SHT_CMD_PER_LUN,
        .emulated               = ATA_SHT_EMULATED,
        .use_clustering         = 1,