[PATCH #upstream] ahci/sata_vsc: drop intx manipulation on msi enable

[PATCH #upstream] ahci/sata_vsc: drop intx manipulation on msi enable

[Tejun Heo]

There's no need to turn off intx explicitly on msi enable.  This is
automatically handled by pci.  Drop it.

Note that, in sata_vsc's case, it was completely redundant; however,
for ahci, it might be needed on machines if the BIOS turns intx off
during boot.  However, there's no evidence of such behavior for ahci
and the only such case seems to be ICH5 PATA according to ata_piix.
Also, given the way ahci operates, it's highly unlikely BIOS ever
disables IRQ for the controller.  However, as this change has slight
possibility of introducing failure, please schedule it for #upstream.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
There are similar places in sata_sis and sata_uli.  I have no idea
whether they're there because they're needed or just copied over from
other drivers.  Eh...

Thanks.

 drivers/ata/ahci.c     |    4 ++--
 drivers/ata/sata_vsc.c |    3 +--
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 7f701cb..18fac60 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -2616,8 +2616,8 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (board_id == board_ahci_sb700 && pdev->revision >= 0x40)
 		hpriv->flags &= ~AHCI_HFLAG_IGN_SERR_INTERNAL;
 
-	if ((hpriv->flags & AHCI_HFLAG_NO_MSI) || pci_enable_msi(pdev))
-		pci_intx(pdev, 1);
+	if (!(hpriv->flags & AHCI_HFLAG_NO_MSI))
+		pci_enable_msi(pdev);
 
 	/* save initial config */
 	ahci_save_initial_config(pdev, hpriv);
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index c57cdff..398328a 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -411,8 +411,7 @@ static int __devinit vsc_sata_init_one(struct pci_dev *pdev,
 	if (cls == 0x00)
 		pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE, 0x80);
 
-	if (pci_enable_msi(pdev) == 0)
-		pci_intx(pdev, 0);
+	pci_enable_msi(pdev);
 
 	/*
 	 * Config offset 0x98 is "Extended Control and Status Register 0"

Re: [PATCH #upstream] ahci/sata_vsc: drop intx manipulation on msi enable

[Tejun Heo]

Tejun Heo wrote:
> There's no need to turn off intx explicitly on msi enable.  This is
> automatically handled by pci.  Drop it.
> 
> Note that, in sata_vsc's case, it was completely redundant; however,
> for ahci, it might be needed on machines if the BIOS turns intx off
> during boot.  However, there's no evidence of such behavior for ahci
> and the only such case seems to be ICH5 PATA according to ata_piix.
> Also, given the way ahci operates, it's highly unlikely BIOS ever
> disables IRQ for the controller.  However, as this change has slight
> possibility of introducing failure, please schedule it for #upstream.
> 
> Signed-off-by: Tejun Heo <tj@kernel.org>

Ping.

-- 
tejun

Re: [PATCH #upstream] ahci/sata_vsc: drop intx manipulation on msi enable

[Jeff Garzik]

Tejun Heo wrote:
> There's no need to turn off intx explicitly on msi enable.  This is
> automatically handled by pci.  Drop it.
> 
> Note that, in sata_vsc's case, it was completely redundant; however,
> for ahci, it might be needed on machines if the BIOS turns intx off
> during boot.  However, there's no evidence of such behavior for ahci
> and the only such case seems to be ICH5 PATA according to ata_piix.
> Also, given the way ahci operates, it's highly unlikely BIOS ever
> disables IRQ for the controller.  However, as this change has slight
> possibility of introducing failure, please schedule it for #upstream.
> 
> Signed-off-by: Tejun Heo <tj@kernel.org>
> ---
> There are similar places in sata_sis and sata_uli.  I have no idea
> whether they're there because they're needed or just copied over from
> other drivers.  Eh...

Most of them were just copied, IIRC.

as for this patch:  applied the ahci portion.

For sata_vsc, ISTR there being a hardware-specific reason why the 
MSI+INTX manipulation was done in that fashion.  Needs a bit of 
research, or at least, real world testing results for the MSI and !MSI 
cases...

	Jeff