2.6.x libata fixes

Please pull from the 'upstream-fixes' branch of rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git

to obtain the two fixes described in the attached diffstat/changelog/patch.

 drivers/scsi/libata-scsi.c |    1 +
 drivers/scsi/sata_sx4.c    |    2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

commit 42517438f9c1011a03e49a542cba32ac5a80dd8e

Author: Tejun Heo <[email protected]>
Date:   Wed Aug 10 13:38:27 2005 -0400

libata: fix EH-related lockup by properly cleaning EH command list

Yet another hack due to the fact that libata is the only user of SCSI's ->eh_strategy_handler() hook.

commit fae009847c9ea3d668bbee21ce1d76764eca5039

Author: Tejun Heo <[email protected]>
Date:   Sun Aug 7 14:53:40 2005 +0900

[PATCH] sata: fix sata_sx4 dma_prep to not use sg->length

sata_sx4 directly references sg->length to calculate total_len in
pdc20621_dma_prep(). This is incorrect as dma_map_sg() could have merged multiple sg's into one and, in such case, sg->length doesn't reflect true size of the entry. This patch makes it use sg_dma_len(sg).

@@ -385,6 +385,7 @@ int ata_scsi_error(struct Scsi_Host *hos
 	 * appropriate place
 	 */
 	host->host_failed--;
+	INIT_LIST_HEAD(&host->eh_cmd_q);
 
 	DPRINTK("EXIT\n");
 	return 0;

@@ -468,7 +468,7 @@ static void pdc20621_dma_prep(struct ata
 	for (i = 0; i < last; i++) {
 		buf[idx++] = cpu_to_le32(sg_dma_address(&sg[i]));
 		buf[idx++] = cpu_to_le32(sg_dma_len(&sg[i]));
-		total_len += sg[i].length;
+		total_len += sg_dma_len(&sg[i]);
 	}
 	buf[idx - 1] |= cpu_to_le32(ATA_PRD_EOT);
 	sgt_len = idx * 4;