libata/sff: Use ops->bmdma_stop instead of ata_bmdma_stop()
PATCH 4/5 libata/sff: Use ops->bmdma_stop instead of ata_bmdma_stop()
In libata-sff, ata_sff_post_internal_cmd() directly calls ata_bmdma_stop() instead of ap->ops->bmdma_stop(). This can be a problem for controllers that use their own bmdma_stop for which the generic sff one isn't suitable
Signed-off-by Benjamin Herrenschmidt <[email protected]>
---
drivers/ata/libata-sff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ata/libata-sff.c +1 -1
@@ -2384,7 +2384,7 @@ void ata_sff_post_internal_cmd(struct at ap->hsm_task_state =HSM_ST_IDLE ;if (ap->ioaddr.bmdma_addr) -ata_bmdma_stop (qc); + ap->ops->bmdma_stop (qc);spin_unlock_irqrestore (ap->lock, flags); }
Quoted 13 lines · Benjamin Herrenschmidt
In libata-sff,
ata_sff_post_internal_cmd()directly callsata_bmdma_stop()instead ofap->ops->bmdma_stop(). This can be a problem for controllers that use their ownbmdma_stopfor which the generic sff one isn't suitableSigned-off-by Benjamin Herrenschmidt<[email protected]>---
drivers/ata/libata-sff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ata/libata-sff.c void ata_sff_post_internal_cmd(struct at · line 2389
Quoted 7 lines
drivers/ata/libata-sff.c +1
-1@@ -2384,7 +2384,7 @@ void ata_sff_post_internal_cmd(struct at ap->hsm_task_state =HSM_ST_IDLE ;if (ap->ioaddr.bmdma_addr) -ata_bmdma_stop (qc); + ap->ops->bmdma_stop (qc);spin_unlock_irqrestore (ap->lock, flags);
applied
As we discussed, feel free to carry this in your tree as well. git should be able to sort it out.
Acked-by Jeff Garzik <[email protected]>