Disable ASPM on various devices

Disable ASPM on various devices

[Matthew Garrett]

If the firmware has given us control of PCIe capabilities then it's valid
for an operating system to configure ASPM more aggressively than the
firmware did. A small number of devices object to this and exhibit various
failure modes. Windows provides a mechanism to disable ASPM in the driver,
indicated by the Needs=PciASPMOptOut statement in the .inf file. Trawling
through Windows drivers has indicated the following set of hardware that
disables ASPM in Windows but doesn't currently disable it in Linux. It makes
sense for us to mimic Windows in this situation.

[PATCH 1/6] net/ethernet/atl1c: Disable ASPM on various chipsets

[Matthew Garrett]

The Windows driver disables ASPM support for various chipsets supported
by atl1c. This adds the same set of logic to the Linux driver. ASPM is
disabled on l1c, l2c, l2cb and l2cb2 devices except for those in Toshiba
or Lenovo devices. Data taken from
http://www.atheros.cz/atheros-inf-file.php?inf=199&chipset=51&system=6

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: netdev@vger.kernel.org
Cc: jcliburn@gmail.com
Cc: chris.snook@gmail.com
---
 drivers/net/ethernet/atheros/atl1c/atl1c_main.c |   15 +++++++++++++++
 1 files changed, 15 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 02c7ed8..d91dabd 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -19,6 +19,7 @@
  * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
+#include <linux/pci-aspm.h>
 #include "atl1c.h"
 
 #define ATL1C_DRV_VERSION "1.0.1.0-NAPI"
@@ -2652,6 +2653,20 @@ static int __devinit atl1c_probe(struct pci_dev *pdev,
 
 	int err = 0;
 
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_ATTANSIC_L1C:
+	case PCI_DEVICE_ID_ATTANSIC_L2C:
+	case PCI_DEVICE_ID_ATHEROS_L2C_B:
+	case PCI_DEVICE_ID_ATHEROS_L2C_B2:
+		if (pdev->subsystem_vendor == PCI_VENDOR_ID_TOSHIBA ||
+		    pdev->subsystem_vendor == PCI_VENDOR_ID_LENOVO)
+			break;
+		pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S |
+				       PCIE_LINK_STATE_L1 |
+				       PCIE_LINK_STATE_CLKPM);
+		break;
+	}
+
 	/* enable device (incl. PCI PM wakeup and hotplug setup) */
 	err = pci_enable_device_mem(pdev);
 	if (err) {
-- 
1.7.7.1

[PATCH 2/6] net/ethernet/atl1e: Disable ASPM

[Matthew Garrett]

http://www.atheros.cz/atheros-inf-file.php?inf=209&chipset=45&system=6
indicates that ASPM is disabled on all L1E hardware. Duplicate for sanity.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: netdev@vger.kernel.org
Cc: jcliburn@gmail.com
Cc: chris.snook@gmail.com
---
 drivers/net/ethernet/atheros/atl1e/atl1e_main.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
index 95483bc..fc74dd1 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -2264,6 +2264,9 @@ static int __devinit atl1e_probe(struct pci_dev *pdev,
 
 	int err = 0;
 
+	pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1 |
+			       PCIE_LINK_STATE_CLKPM);
+
 	err = pci_enable_device(pdev);
 	if (err) {
 		dev_err(&pdev->dev, "cannot enable PCI device\n");
-- 
1.7.7.1

[PATCH 3/6] net/ethernet/jme: Disable ASPM

[Matthew Garrett]

http://driveragent.com/archive/30421/7-0-14 indicates that ASPM is
disabled on the 250 and 260. Duplicate for sanity.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: netdev@vger.kernel.org
Cc: cooldavid@cooldavid.org
---
 drivers/net/ethernet/jme.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c
index 7becff1..b28a873 100644
--- a/drivers/net/ethernet/jme.c
+++ b/drivers/net/ethernet/jme.c
@@ -2860,6 +2860,9 @@ jme_init_one(struct pci_dev *pdev,
 	/*
 	 * set up PCI device basics
 	 */
+	pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1 |
+			       PCIE_LINK_STATE_CLKPM);
+
 	rc = pci_enable_device(pdev);
 	if (rc) {
 		pr_err("Cannot enable PCI device\n");
-- 
1.7.7.1

[PATCH 4/6] block/cciss: Disable ASPM

[Matthew Garrett]

The Windows driver .inf disables ASPM on all cciss devices. Do the same.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: mike.miller@hp.com
Cc: iss_storagedev@hp.com
Cc: axboe@kernel.dk
---
 drivers/block/cciss.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index 486f94e..7e43ad3 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -4319,6 +4319,10 @@ static int __devinit cciss_pci_init(ctlr_info_t *h)
 		dev_warn(&h->pdev->dev, "controller appears to be disabled\n");
 		return -ENODEV;
 	}
+
+	pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S |
+			       PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM);
+
 	err = pci_enable_device(h->pdev);
 	if (err) {
 		dev_warn(&h->pdev->dev, "Unable to Enable PCI device\n");
-- 
1.7.7.1

[PATCH 5/6] scsi/hpsa: Disable ASPM

[Matthew Garrett]

The Windows driver .inf disables ASPM on hpsa devices. Do the same.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: scameron@beardog.cce.hp.com
Cc: iss_storagedev@hp.com
Cc: linux-scsi@vger.kernel.org
---
 drivers/scsi/hpsa.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index e76107b..fc2f4c4 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -3922,6 +3922,10 @@ static int __devinit hpsa_pci_init(struct ctlr_info *h)
 		dev_warn(&h->pdev->dev, "controller appears to be disabled\n");
 		return -ENODEV;
 	}
+
+	pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S |
+			       PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM);
+
 	err = pci_enable_device(h->pdev);
 	if (err) {
 		dev_warn(&h->pdev->dev, "unable to enable PCI device\n");
-- 
1.7.7.1

[PATCH 6/6] infiniband/nes: Disable ASPM

[Matthew Garrett]

The Windows driver .inf disables ASPM on hpsa devices. Do the same.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: faisal.latif@intel.com
Cc: linux-rdma@vger.kernel.org
---
 drivers/infiniband/hw/nes/nes.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c
index 5965b3d..2737b7a 100644
--- a/drivers/infiniband/hw/nes/nes.c
+++ b/drivers/infiniband/hw/nes/nes.c
@@ -463,6 +463,9 @@ static int __devinit nes_probe(struct pci_dev *pcidev, const struct pci_device_i
 	printk(KERN_INFO PFX "NetEffect RNIC driver v%s loading. (%s)\n",
 			DRV_VERSION, pci_name(pcidev));
 
+	pci_disable_link_state(pcidev, PCIE_LINK_STATE_L0S |
+			       PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM);
+
 	ret = pci_enable_device(pcidev);
 	if (ret) {
 		printk(KERN_ERR PFX "Unable to enable PCI device. (%s)\n", pci_name(pcidev));
-- 
1.7.7.1

Re: [PATCH 6/6] infiniband/nes: Disable ASPM

[Roland Dreier]

On Fri, Nov 11, 2011 at 8:05 AM, Matthew Garrett <mjg-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> The Windows driver .inf disables ASPM on hpsa devices. Do the same.

This isn't a patch to the hpsa driver... cut and paste in the description,
or misfire on the patch?

Faisal, any info on whether this is needed/desired for nes devices
based on Intel's knowledge of the PCIe core used?

 - R.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 6/6] infiniband/nes: Disable ASPM

[Roland Dreier]

On Fri, Nov 11, 2011 at 8:05 AM, Matthew Garrett <mjg@redhat.com> wrote:
> The Windows driver .inf disables ASPM on hpsa devices. Do the same.

This isn't a patch to the hpsa driver... cut and paste in the description,
or misfire on the patch?

Faisal, any info on whether this is needed/desired for nes devices
based on Intel's knowledge of the PCIe core used?

 - R.

Re: [PATCH 6/6] infiniband/nes: Disable ASPM

[Matthew Garrett]

On Fri, Nov 11, 2011 at 09:31:44AM -0800, Roland Dreier wrote:
> On Fri, Nov 11, 2011 at 8:05 AM, Matthew Garrett <mjg-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> > The Windows driver .inf disables ASPM on hpsa devices. Do the same.
> 
> This isn't a patch to the hpsa driver... cut and paste in the description,
> or misfire on the patch?

Sorry, yes, cut and paste in the description. It should say nes.

-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 6/6] infiniband/nes: Disable ASPM

[Matthew Garrett]

On Fri, Nov 11, 2011 at 09:31:44AM -0800, Roland Dreier wrote:
> On Fri, Nov 11, 2011 at 8:05 AM, Matthew Garrett <mjg@redhat.com> wrote:
> > The Windows driver .inf disables ASPM on hpsa devices. Do the same.
> 
> This isn't a patch to the hpsa driver... cut and paste in the description,
> or misfire on the patch?

Sorry, yes, cut and paste in the description. It should say nes.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: [PATCH 6/6] infiniband/nes: Disable ASPM

[Roland Dreier]

> Sorry, yes, cut and paste in the description. It should say nes.

OK.  Just got a vacation message from Faisal that he's out til
next week, let's see if Intel has a comment on this.

Thanks,
  Roland
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 6/6] infiniband/nes: Disable ASPM

[Roland Dreier]

> Sorry, yes, cut and paste in the description. It should say nes.

OK.  Just got a vacation message from Faisal that he's out til
next week, let's see if Intel has a comment on this.

Thanks,
  Roland