devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Suman Tripathi <stripathi@apm.com>
To: olof@lixom.net, tj@kernel.org, arnd@arndb.de
Cc: linux-scsi@vger.kernel.org, linux-ide@vger.kernel.org,
	devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	ddutile@redhat.com, jcm@redhat.com, patches@apm.com,
	Suman Tripathi <stripathi@apm.com>, Loc Ho <lho@apm.com>
Subject: [PATCH v5 2/2] ata: Fix the dma state machine lockup for the IDENTIFY DEVICE PIO mode command.
Date: Wed,  2 Jul 2014 15:45:08 +0530	[thread overview]
Message-ID: <1404296108-16233-3-git-send-email-stripathi@apm.com> (raw)
In-Reply-To: <1404296108-16233-1-git-send-email-stripathi@apm.com>

This patch fixes the dma state machine lockup due to the processing
of IDENTIFY DEVICE PIO mode command. The X-Gene AHCI controller
has an errata in which it cannot clear the BSY bit after the PIO setup
FIS. The dma state machine enters CMFatalErrorUpdate state and locks
up. This patch also removes the dma restart workaround from the read_id
function as the read_id function is only called by libata layer for
ATA_INTERNAL commands. But for some cases eg: PORT MULTIPLIER
and udev, the framework will enumerate using SCSI commands and it will
not call read_id function.

Signed-off-by: Loc Ho <lho@apm.com>
Signed-off-by: Suman Tripathi <stripathi@apm.com>
---
 drivers/ata/ahci_xgene.c | 43 ++++++++++++++++++++++++++++++-------------
 1 file changed, 30 insertions(+), 13 deletions(-)

diff --git a/drivers/ata/ahci_xgene.c b/drivers/ata/ahci_xgene.c
index 77c89bf..5b9bddf 100644
--- a/drivers/ata/ahci_xgene.c
+++ b/drivers/ata/ahci_xgene.c
@@ -78,6 +78,7 @@
 struct xgene_ahci_context {
 	struct ahci_host_priv *hpriv;
 	struct device *dev;
+	u8 last_cmd[MAX_AHCI_CHN_PERCTR]; /* tracking the last command issued*/
 	void __iomem *csr_core;		/* Core CSR address of IP */
 	void __iomem *csr_diag;		/* Diag CSR address of IP */
 	void __iomem *csr_axi;		/* AXI CSR address of IP */
@@ -98,20 +99,45 @@ static int xgene_ahci_init_memram(struct xgene_ahci_context *ctx)
 }

 /**
+ * xgene_ahci_qc_issue - Issue commands to the device
+ * @qc: Command to issue
+ *
+ * Due to Hardware errata for IDENTIFY DEVICE command, the controller cannot
+ * clear the BSY bit after receiving the PIO setup FIS. This results in the dma
+ * state machine goes into the CMFatalErrorUpdate state and locks up. By
+ * restarting the dma engine, it removes the controller out of lock up state.
+ */
+static unsigned int xgene_ahci_qc_issue(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
+	struct xgene_ahci_context *ctx = hpriv->plat_data;
+	int rc = 0;
+
+	if (unlikely(ctx->last_cmd[ap->port_no] == ATA_CMD_ID_ATA))
+		ahci_restart_engine(ap);
+
+	rc = ahci_qc_issue(qc);
+
+	/* Save the last command issued */
+	ctx->last_cmd[ap->port_no] = qc->tf.command;
+
+	return rc;
+}
+
+/**
  * xgene_ahci_read_id - Read ID data from the specified device
  * @dev: device
  * @tf: proposed taskfile
  * @id: data buffer
  *
  * This custom read ID function is required due to the fact that the HW
- * does not support DEVSLP and the controller state machine may get stuck
- * after processing the ID query command.
+ * does not support DEVSLP.
  */
 static unsigned int xgene_ahci_read_id(struct ata_device *dev,
 				       struct ata_taskfile *tf, u16 *id)
 {
 	u32 err_mask;
-	void __iomem *port_mmio = ahci_port_base(dev->link->ap);

 	err_mask = ata_do_dev_read_id(dev, tf, id);
 	if (err_mask)
@@ -133,16 +159,6 @@ static unsigned int xgene_ahci_read_id(struct ata_device *dev,
 	 */
 	id[ATA_ID_FEATURE_SUPP] &= ~(1 << 8);

-	/*
-	 * Due to HW errata, restart the port if no other command active.
-	 * Otherwise the controller may get stuck.
-	 */
-	if (!readl(port_mmio + PORT_CMD_ISSUE)) {
-		writel(PORT_CMD_FIS_RX, port_mmio + PORT_CMD);
-		readl(port_mmio + PORT_CMD);	/* Force a barrier */
-		writel(PORT_CMD_FIS_RX | PORT_CMD_START, port_mmio + PORT_CMD);
-		readl(port_mmio + PORT_CMD);	/* Force a barrier */
-	}
 	return 0;
 }

@@ -300,6 +316,7 @@ static struct ata_port_operations xgene_ahci_ops = {
 	.host_stop = xgene_ahci_host_stop,
 	.hardreset = xgene_ahci_hardreset,
 	.read_id = xgene_ahci_read_id,
+	.qc_issue = xgene_ahci_qc_issue,
 };

 static const struct ata_port_info xgene_ahci_port_info = {
--
1.8.2.1


      parent reply	other threads:[~2014-07-02 10:15 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-02 10:15 [PATCH v5 0/2] ata: Fix the dma state machine lockup for APM X-Gene SoC Suman Tripathi
2014-07-02 10:15 ` [PATCH v5 1/2] libahci: Implement the function ahci_restart_engine to restart the port dma engine Suman Tripathi
2014-07-03 14:03   ` Tejun Heo
2014-07-02 10:15 ` Suman Tripathi [this message]

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=1404296108-16233-3-git-send-email-stripathi@apm.com \
    --to=stripathi@apm.com \
    --cc=arnd@arndb.de \
    --cc=ddutile@redhat.com \
    --cc=devicetree@vger.kernel.org \
    --cc=jcm@redhat.com \
    --cc=lho@apm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-ide@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=olof@lixom.net \
    --cc=patches@apm.com \
    --cc=tj@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).