Re: [PATCH 2.6.20-rc3] sata_promise: ATAPI support

On Sun, 07 Jan 2007 20:50:48 -0500, Jeff Garzik wrote:

quoted

With this patch ATAPI will work on SATA ports on second-generation
chips, and on the first-generation PATA-only 20619. ATAPI on the
2057x' PATA port will work automatically when #promise-sata-pata
is updated. ATAPI on the 2037x' PATA port is enabled by a followup
patch for the #promise-sata-pata branch.

Signed-off-by: Mikael Pettersson <redacted>

No objections, minor revision requests follow.

* general comment:  please combine the 'make first gen SATAPI work' 
patch with this one.

Will do.

quoted

+static unsigned int pdc_wait_for_drq(struct ata_port *ap)
+{ 
+	void __iomem *port_mmio = (void __iomem *) ap->ioaddr.cmd_addr;
+	unsigned int i;
+	unsigned int status;
+
+	/* Following pdc-ultra's WaitForDrq() we loop here until BSY
+	 * is clear and DRQ is set in altstatus. We could possibly call
+	 * ata_busy_wait() and loop until DRQ is set, but since we don't
+	 * know how much time a call to ata_busy_wait() took, we don't
+	 * know when to time out the outer loop.
+	 */
+	for(i = 0; i < 1000; ++i) {
+		status = readb(port_mmio + 0x38); /* altstatus */
+		if (status == 0xFF)
+			break;
+		if (status & ATA_BUSY)
+			;
+		else if (status & (ATA_DRQ | ATA_ERR))
+			break;
+		mdelay(1);
+	}

This really hurts to do this with spinlocks held.  Long term, ponder 
ways to move this to a kernel thread that can sleep [if it takes too long?]

On the TODO list for a later update.

quoted

+static void pdc_issue_atapi_pkt_cmd(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	void __iomem *port_mmio = (void __iomem *) ap->ioaddr.cmd_addr;
+	void __iomem *host_mmio = ap->host->mmio_base;
+	unsigned int nbytes;
+	unsigned int tmp;
+
+	/* disable INTA here, it will be re-enable when CAM use SEQ 0 for packets */

cut-n-pasted comment?  This is not CAM :)

Indeed :-) Will fix.

quoted

+	writeb(0x00, port_mmio + PDC_CTLSTAT); /* that the drive INT pass to SEQ 0*/
+	writeb(0x20, host_mmio + 0); /* but mask SEQ 0 INT */

please create named constants for these magic numbers

Will do (for this and the other places you pointed out).

/Mikael