Skip to contentSkip to conversation

ide: fix drive_is_ready() for non-PCI hosts and CONFIG_IDEPCI_SHARE_IRQ=y

PATCH In review Bartlomiej Zolnierkiewicz 3 participants 6 messages Oct 25, 2007 – Oct 26, 2007 libata

lore
Share & tools
lore

Bartlomiej Zolnierkiewicz

author
#0 lore raw

PATCH 2/3 ide: fix drive_is_ready() for non-PCI hosts and CONFIG_IDEPCI_SHARE_IRQ=y

Need to check if the host is a PCI one before reading IDE_ALTSTATUS_REG.

Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>
---
 drivers/ide/ide-iops.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ide/ide-iops.c +1 -1
Index: b/drivers/ide/ide-iops.c
===================================================================
@@ -455,7 +455,7 @@ int drive_is_ready (ide_drive_t *drive)
 	 * an interrupt with another pci card/device.  We make no assumptions
 	 * about possible isa-pnp and pci-pnp issues yet.
 	 */
-	if (IDE_CONTROL_REG)
+	if (hwif->pci_dev && IDE_CONTROL_REG)
 		stat = hwif->INB(IDE_ALTSTATUS_REG);
 	else
 #endif /* CONFIG_IDEPCI_SHARE_IRQ */
#1 lore raw
Quoted 4 lines · On Fri, 26 Oct 2007 01:36:37 +0200 Bartlomiej Zolnierkiewicz…

Need to check if the host is a PCI one before reading IDE_ALTSTATUS_REG.

Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

Umm why ? The altstatus register goes back to ST-506 and the original IBM XT hard disk controller card ?

author
#2 lore raw
Quoted 10 lines · Alan Cox
Quoted 4 lines · On Fri, 26 Oct 2007 01:36:37 +0200 Bartlomiej Zolnierkiewicz…

Need to check if the host is a PCI one before reading IDE_ALTSTATUS_REG.

Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

Umm why ? The altstatus register goes back to ST-506 and the original IBM XT hard disk controller card ?

Thought that there have to be some (non-obvious) reason behing the fact that IDE_ALTSTATUS_REG reading was covered by #ifdef CONFIG_IDEPCI_SHARE_IRQ.

However if this is the case patch #2/3 can be dumped and we can go straight to patch #3/3.

Thanks,
Bart

#3 lore raw
Quoted 7 lines · On Fri, 26 Oct 2007 02:30:52 +0200 Bartlomiej Zolnierkiewicz…
Quoted 6 lines · Alan Cox
Quoted 3 lines · On Fri, 26 Oct 2007 01:36:37 +0200 Bartlomiej Zolnierkiewicz…

Need to check if the host is a PCI one before reading IDE_ALTSTATUS_REG.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

Umm why ? The altstatus register goes back to ST-506 and the original IBM XT hard disk controller card ?

Thought that there have to be some (non-obvious) reason behing the fact that IDE_ALTSTATUS_REG reading was covered by #ifdef CONFIG_IDEPCI_SHARE_IRQ.

Only that reading altstatus is done to get the state without clearing the IRQ which a status read would do.

Alan

#4 lore raw
In reply to Alan Cox
Quoted 9 lines · Alan Cox
Quoted 3 lines · On Fri, 26 Oct 2007 01:36:37 +0200 Bartlomiej Zolnierkiewicz…

Need to check if the host is a PCI one before reading IDE_ALTSTATUS_REG.

Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

Umm why ? The altstatus register goes back to ST-506 and the original IBM XT hard disk controller card ?

I'm pretty sure there are at least a few controllers out there where the AltStatus/Control register block is simply non-existent.

Jeff

#5 lore raw
Quoted 6 lines · On Thu, 25 Oct 2007 20:42:16 -0400 Jeff Garzik <[email protected]> wrote:
Quoted 5 lines · Alan Cox
Quoted 2 lines · On Fri, 26 Oct 2007 01:36:37 +0200 Bartlomiej Zolnierkiewicz…

Need to check if the host is a PCI one before reading IDE_ALTSTATUS_REG.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Bartlomiej Zolnierkiewicz <[email protected]>

Umm why ? The altstatus register goes back to ST-506 and the original IBM XT hard disk controller card ?

I'm pretty sure there are at least a few controllers out there where the AltStatus/Control register block is simply non-existent.

One or two of the commode Amiga controllers did this. Makes error recovery rather interesting. The register itself is however present even on things like the WD10xx devices.

Alan