linux-ide.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
To: Alan <alan@lxorguk.ukuu.org.uk>
Cc: linux-ide@vger.kernel.org
Subject: Re: [RFC] libata IORDY handling
Date: Mon, 29 Jan 2007 17:28:19 +0300	[thread overview]
Message-ID: <45BE0483.4070409@ru.mvista.com> (raw)
In-Reply-To: <20070129133423.75aa208e@localhost.localdomain>

[-- Attachment #1: Type: text/plain, Size: 764 bytes --]

Hello.

Alan wrote:

>>    It's sad to say but there's another bug in this function (even a 
>>regression from drivers/ide/pci/siimage.c) -- the 16-bit IORDY is not enabled 
>>when setting PIO mode (there's code that twiddles IORDY enable but that's 
>>actually only for *taskfile* accesses, 16-bit IORDY is controlled by the same 
>>PCI  config. registers 80h/84h that enable DMA/UDMA transfer on SiI 680).

> Fixed - also fixed clearing the bits when going into PIO mode from DMA,
> which fixes a potential DMA changedown bug.

     Yeah, this is an issue in siimage.c...

> Also redone the iordy stuff as your emails reminded me that it needed
> finishing off and sorting out.

     Atttaching my (already obsolete) patch, just in case...

> Alan

MBR, Sergei


[-- Attachment #2: sil680_iordy-fixes.patch --]
[-- Type: text/x-patch, Size: 1895 bytes --]

Index: linux-2.6/drivers/ata/pata_sil680.c
===================================================================
--- linux-2.6.orig/drivers/ata/pata_sil680.c
+++ linux-2.6/drivers/ata/pata_sil680.c
@@ -139,9 +139,12 @@ static void sil680_set_piomode(struct at
 
 	unsigned long tfaddr = sil680_selreg(ap, 0x02);
 	unsigned long addr = sil680_seldev(ap, adev, 0x04);
+	unsigned long addr_mask = 0x80 + 4 * ap->port_no;
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int port_shift = adev->devno * 4;
 	int pio = adev->pio_mode - XFER_PIO_0;
 	int lowest_pio = pio;
+	u8  mode;
 	u16 reg;
 
 	struct ata_device *pair = ata_dev_pair(adev);
@@ -152,11 +155,31 @@ static void sil680_set_piomode(struct at
 	pci_write_config_word(pdev, addr, speed_p[pio]);
 	pci_write_config_word(pdev, tfaddr, speed_t[lowest_pio]);
 
-	pci_read_config_word(pdev, tfaddr-2, &reg);
-	reg &= ~0x0200;			/* Clear IORDY */
+	/*
+	 * Let's see if we need to enable IORDY checking on data transfer...
+	 *
+	 * NOTE: Selecting any DMA mode also enables IORDY  checkign,
+	 * so we must first check if not already in DMA mode beforehand
+	 * to avoid disabling it...
+	 */
+	pci_read_config_byte(pdev, addr_mask, &mode);
+	if (ata_pio_need_iordy(adev) && ((mode >> port_shift) & 0x03) == 0x00)
+		mode |= 0x01 << port_shift;
+	else if (((mode >> port_shift) & 0x03) == 0x01)
+		mode &= ~(0x03 << port_shift);
+	pci_write_config_byte(pdev, addr_mask, mode);
+
+	/*
+	 * Let's see if we need to enable IORDY checking on taskfile.
+	 *
+	 * NOTE: We can only disable IORDY if both drives agree to it.
+	 */
+	pci_read_config_word(pdev, tfaddr - 2, &reg);
 	if (ata_pio_need_iordy(adev))
-		reg |= 0x0200;		/* Enable IORDY */
-	pci_write_config_word(pdev, tfaddr-2, reg);
+		reg |=  0x0200;
+	else if (pair == NULL || !ata_pio_need_iordy(pair))
+		reg &= ~0x0200;
+	pci_write_config_word(pdev, tfaddr - 2, reg);
 }
 
 /**


  reply	other threads:[~2007-01-29 14:28 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-01-28 17:24 [PATCH] pata_sil680: PIO1 taskfile transfers oveclocking fix Sergei Shtylyov
2007-01-28 18:22 ` [RFC] libata IORDY handling Sergei Shtylyov
2007-01-28 19:03   ` Alan
2007-01-28 19:00     ` Sergei Shtylyov
2007-01-29 13:34   ` Alan
2007-01-29 14:28     ` Sergei Shtylyov [this message]
2007-01-29 15:45       ` Alan
2007-01-28 18:41 ` [PATCH] pata_sil680: PIO1 taskfile transfers oveclocking fix Alan

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=45BE0483.4070409@ru.mvista.com \
    --to=sshtylyov@ru.mvista.com \
    --cc=alan@lxorguk.ukuu.org.uk \
    --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 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).