All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tejun Heo <htejun@gmail.com>
To: jeff@garzik.org, linux-ide@vger.kernel.org,
	alan@lxorguk.ukuu.org.uk, liml@rtr.ca, albertl@mail.com,
	jens.axboe@oracle.com
Cc: Tejun Heo <htejun@gmail.com>
Subject: [PATCH 08/13] libata: change ATA_QCFLAG_DMAMAP semantics
Date: Wed, 28 Nov 2007 00:13:54 +0900	[thread overview]
Message-ID: <11961764403965-git-send-email-htejun@gmail.com> (raw)
In-Reply-To: <11961764391983-git-send-email-htejun@gmail.com>

ATA_QCFLAG_DMAMAP was a bit peculiar in that it got set during qc
initialization and cleared if DMA mapping wasn't necessary.  Make it
more straight forward by making the following changes.

* Don't set it during initialization.  Set it after DMA is actually
  mapped.

* Add BUG_ON() to guarantee that there is data to transfer if DMAMAP
  is set.  This always holds for the current code.  The BUG_ON() is
  for docummentation and sanity check.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c |   15 +++++++--------
 1 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 547ba30..ee608cd 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4484,7 +4484,6 @@ void ata_sg_clean(struct ata_queued_cmd *qc)
 	int dir = qc->dma_dir;
 	void *pad_buf = NULL;
 
-	WARN_ON(!(qc->flags & ATA_QCFLAG_DMAMAP));
 	WARN_ON(sg == NULL);
 
 	VPRINTK("unmapping %u sg elements\n", qc->n_elem);
@@ -4756,11 +4755,9 @@ void ata_noop_qc_prep(struct ata_queued_cmd *qc) { }
  *	LOCKING:
  *	spin_lock_irqsave(host lock)
  */
-
 void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
 		 unsigned int n_elem)
 {
-	qc->flags |= ATA_QCFLAG_DMAMAP;
 	qc->__sg = sg;
 	qc->n_elem = n_elem;
 	qc->orig_n_elem = n_elem;
@@ -4789,7 +4786,6 @@ static int ata_sg_setup(struct ata_queued_cmd *qc)
 	int n_elem, pre_n_elem, dir, trim_sg = 0;
 
 	VPRINTK("ENTER, ata%u\n", ap->print_id);
-	WARN_ON(!(qc->flags & ATA_QCFLAG_DMAMAP));
 
 	/* we must lengthen transfers to end on a 32-bit boundary */
 	qc->pad_len = lsg->length & 3;
@@ -4849,6 +4845,7 @@ static int ata_sg_setup(struct ata_queued_cmd *qc)
 
 skip_map:
 	qc->n_elem = n_elem;
+	qc->flags |= ATA_QCFLAG_DMAMAP;
 
 	return 0;
 }
@@ -5888,12 +5885,15 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 	qc->flags |= ATA_QCFLAG_ACTIVE;
 	ap->qc_active |= 1 << qc->tag;
 
+	/* We guarantee to LLDs that they will have at least one
+	 * non-zero sg if the command is a data command.
+	 */
+	BUG_ON(ata_is_data(prot) && (!qc->__sg || !qc->n_elem || !qc->nbytes));
+
 	if (ata_is_dma(prot) || (ata_is_pio(prot) &&
-				 (ap->flags & ATA_FLAG_PIO_DMA))) {
+				 (ap->flags & ATA_FLAG_PIO_DMA)))
 		if (ata_sg_setup(qc))
 			goto sg_err;
-	} else
-		qc->flags &= ATA_QCFLAG_DMAMAP;
 
 	/* if device is sleeping, schedule softreset and abort the link */
 	if (unlikely(qc->dev->flags & ATA_DFLAG_SLEEPING)) {
@@ -5911,7 +5911,6 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 	return;
 
 sg_err:
-	qc->flags &= ~ATA_QCFLAG_DMAMAP;
 	qc->err_mask |= AC_ERR_SYSTEM;
 err:
 	ata_qc_complete(qc);
-- 
1.5.2.4


  parent reply	other threads:[~2007-11-27 15:14 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-27 15:13 [PATCHSET] libata: improve ATAPI data transfer handling Tejun Heo
2007-11-27 15:13 ` [PATCH 01/13] libata: update atapi_eh_request_sense() such that lbam/lbah contains buffer size Tejun Heo
2007-11-27 15:13 ` [PATCH 02/13] cdrom: add more GPCMD_* constants Tejun Heo
2007-11-27 15:13 ` [PATCH 03/13] libata: rename ATA_PROT_ATAPI_* to ATAPI_PROT_* Tejun Heo
2007-11-27 15:13 ` [PATCH 04/13] libata: add ATAPI_* cmd types and implement atapi_cmd_type() Tejun Heo
2007-11-27 15:13 ` [PATCH 05/13] libata: improve ATAPI draining Tejun Heo
2007-11-29  6:28   ` Albert Lee
2007-11-29  7:26     ` Tejun Heo
2007-11-29 14:34       ` Tejun Heo
2007-11-27 15:13 ` [PATCH 06/13] libata: make atapi_request_sense() use sg Tejun Heo
2007-11-27 15:13 ` [PATCH 07/13] libata: kill non-sg DMA interface Tejun Heo
2007-11-27 15:13 ` Tejun Heo [this message]
2007-11-27 15:13 ` [PATCH 09/13] libata: convert to chained sg Tejun Heo
2007-11-27 15:13 ` [PATCH 10/13] libata: add qc->dma_nbytes Tejun Heo
2007-11-27 17:20   ` Alan Cox
2007-11-27 15:13 ` [PATCH 11/13] libata: implement ATAPI drain buffer Tejun Heo
2007-11-27 15:13 ` [PATCH 12/13] libata: implement ATAPI per-command-type DMA horkages Tejun Heo
2007-11-27 15:13 ` [PATCH 13/13] libata: use PIO for misc ATAPI commands Tejun Heo
2007-11-27 16:56   ` Alan Cox
2007-11-27 23:01     ` Tejun Heo
2007-11-27 23:31       ` Mark Lord
2007-11-27 23:34         ` Mark Lord
2007-11-27 23:37         ` Tejun Heo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=11961764403965-git-send-email-htejun@gmail.com \
    --to=htejun@gmail.com \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=albertl@mail.com \
    --cc=jeff@garzik.org \
    --cc=jens.axboe@oracle.com \
    --cc=liml@rtr.ca \
    --cc=linux-ide@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.