linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 0/2] VMD subdevice secondary bus resets
@ 2020-09-28  1:05 Jon Derrick
  2020-09-28  1:05 ` [PATCH 1/2] PCI: vmd: Reset the VMD subdevice domain on probe Jon Derrick
                   ` (2 more replies)
  0 siblings, 3 replies; 5+ messages in thread
From: Jon Derrick @ 2020-09-28  1:05 UTC (permalink / raw)
  To: linux-pci
  Cc: Lorenzo Pieralisi, Bjorn Helgaas, Andrzej Jakowski, Dave Fugate,
	Jon Derrick

This set adds some resets for VMD. It's very common code but doesn't
seem to fit well anywhere that can also be exported if VMD is built as a
module.

Jon Derrick (2):
  PCI: vmd: Reset the VMD subdevice domain on probe
  PCI: Add a reset quirk for VMD

 drivers/pci/controller/vmd.c | 32 ++++++++++++++++++++++++
 drivers/pci/quirks.c         | 48 ++++++++++++++++++++++++++++++++++++
 2 files changed, 80 insertions(+)

-- 
2.18.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH 1/2] PCI: vmd: Reset the VMD subdevice domain on probe
  2020-09-28  1:05 [PATCH 0/2] VMD subdevice secondary bus resets Jon Derrick
@ 2020-09-28  1:05 ` Jon Derrick
  2020-09-28  1:05 ` [PATCH 2/2] PCI: Add a reset quirk for VMD Jon Derrick
  2020-11-19 11:37 ` [PATCH 0/2] VMD subdevice secondary bus resets Lorenzo Pieralisi
  2 siblings, 0 replies; 5+ messages in thread
From: Jon Derrick @ 2020-09-28  1:05 UTC (permalink / raw)
  To: linux-pci
  Cc: Lorenzo Pieralisi, Bjorn Helgaas, Andrzej Jakowski, Dave Fugate,
	Jon Derrick

The VMD subdevice domain resource requirements may have changed
in-between module loads. Generic PCI resource assignment code may rely
on existing resource configuration rather than the VMD preference of
re-examining the domain. Add a Secondary Bus Reset to the VMD subdevice
domain during driver attachment to clear the PCI config space of the
subdevices.

Signed-off-by: Jon Derrick <jonathan.derrick@intel.com>
---
 drivers/pci/controller/vmd.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c
index 50b3520d261d..676acff3622f 100644
--- a/drivers/pci/controller/vmd.c
+++ b/drivers/pci/controller/vmd.c
@@ -445,6 +445,36 @@ static struct pci_ops vmd_ops = {
 	.write		= vmd_pci_write,
 };
 
+static void vmd_domain_reset_sbr(struct vmd_dev *vmd)
+{
+	char __iomem *base;
+	int rp;
+	u16 ctl;
+
+	/*
+	 * Subdevice config space is mapped linearly using 4k config space
+	 * increments. Use increments of 0x8000 to locate root ports devices.
+	 */
+	for (rp = 0; rp < 4; rp++) {
+		base = vmd->cfgbar + rp * 0x8000;
+		if (readl(base + PCI_COMMAND) == 0xFFFFFFFF)
+			continue;
+
+		/* pci_reset_secondary_bus() */
+		ctl = readw(base + PCI_BRIDGE_CONTROL);
+		ctl |= PCI_BRIDGE_CTL_BUS_RESET;
+		writew(ctl, base + PCI_BRIDGE_CONTROL);
+		readw(base + PCI_BRIDGE_CONTROL);
+		msleep(2);
+
+		ctl &= ~PCI_BRIDGE_CTL_BUS_RESET;
+		writew(ctl, base + PCI_BRIDGE_CONTROL);
+		readw(base + PCI_BRIDGE_CONTROL);
+	}
+
+	ssleep(1);
+}
+
 static void vmd_attach_resources(struct vmd_dev *vmd)
 {
 	vmd->dev->resource[VMD_MEMBAR1].child = &vmd->resources[1];
@@ -784,6 +814,7 @@ static int vmd_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!vmd->cfgbar)
 		return -ENOMEM;
 
+	vmd_domain_reset_sbr(dev);
 	pci_set_master(dev);
 	if (dma_set_mask_and_coherent(&dev->dev, DMA_BIT_MASK(64)) &&
 	    dma_set_mask_and_coherent(&dev->dev, DMA_BIT_MASK(32)))
-- 
2.18.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [PATCH 2/2] PCI: Add a reset quirk for VMD
  2020-09-28  1:05 [PATCH 0/2] VMD subdevice secondary bus resets Jon Derrick
  2020-09-28  1:05 ` [PATCH 1/2] PCI: vmd: Reset the VMD subdevice domain on probe Jon Derrick
@ 2020-09-28  1:05 ` Jon Derrick
  2020-11-19 11:37 ` [PATCH 0/2] VMD subdevice secondary bus resets Lorenzo Pieralisi
  2 siblings, 0 replies; 5+ messages in thread
From: Jon Derrick @ 2020-09-28  1:05 UTC (permalink / raw)
  To: linux-pci
  Cc: Lorenzo Pieralisi, Bjorn Helgaas, Andrzej Jakowski, Dave Fugate,
	Jon Derrick

VMD domains should be reset in-between special attachment such as VFIO
users. VMD does not offer a reset however the subdevice domain itself
can be reset starting at the Root Bus. Add a Secondary Bus Reset on each
of the individual root port devices immediately downstream of the VMD
root bus.

Signed-off-by: Jon Derrick <jonathan.derrick@intel.com>
---
 drivers/pci/controller/vmd.c |  3 ++-
 drivers/pci/quirks.c         | 48 ++++++++++++++++++++++++++++++++++++
 2 files changed, 50 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c
index 676acff3622f..fdc1a206f73e 100644
--- a/drivers/pci/controller/vmd.c
+++ b/drivers/pci/controller/vmd.c
@@ -14,6 +14,7 @@
 #include <linux/srcu.h>
 #include <linux/rculist.h>
 #include <linux/rcupdate.h>
+#include <linux/delay.h>
 
 #include <asm/irqdomain.h>
 #include <asm/device.h>
@@ -814,7 +815,7 @@ static int vmd_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!vmd->cfgbar)
 		return -ENOMEM;
 
-	vmd_domain_reset_sbr(dev);
+	vmd_domain_reset_sbr(vmd);
 	pci_set_master(dev);
 	if (dma_set_mask_and_coherent(&dev->dev, DMA_BIT_MASK(64)) &&
 	    dma_set_mask_and_coherent(&dev->dev, DMA_BIT_MASK(32)))
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 8f409d442bc0..f6a9c2b2625a 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -3801,6 +3801,49 @@ static int reset_ivb_igd(struct pci_dev *dev, int probe)
 	return 0;
 }
 
+/* Issues SBR to VMD domain to clear PCI configuration */
+static int reset_vmd_sbr(struct pci_dev *dev, int probe)
+{
+	char __iomem *cfgbar, *base;
+	int rp;
+	u16 ctl;
+
+	if (probe)
+		return 0;
+
+	if (dev->dev.driver)
+		return 0;
+
+	cfgbar = pci_iomap(dev, 0, 0);
+	if (!cfgbar)
+		return -ENOMEM;
+
+	/*
+	 * Subdevice config space is mapped linearly using 4k config space
+	 * increments. Use increments of 0x8000 to locate root ports devices.
+	 */
+	for (rp = 0; rp < 4; rp++) {
+		base = cfgbar + rp * 0x8000;
+		if (readl(base + PCI_COMMAND) == 0xFFFFFFFF)
+			continue;
+
+		/* pci_reset_secondary_bus() */
+		ctl = readw(base + PCI_BRIDGE_CONTROL);
+		ctl |= PCI_BRIDGE_CTL_BUS_RESET;
+		writew(ctl, base + PCI_BRIDGE_CONTROL);
+		readw(base + PCI_BRIDGE_CONTROL);
+		msleep(2);
+
+		ctl &= ~PCI_BRIDGE_CTL_BUS_RESET;
+		writew(ctl, base + PCI_BRIDGE_CONTROL);
+		readw(base + PCI_BRIDGE_CONTROL);
+	}
+
+	ssleep(1);
+	pci_iounmap(dev, cfgbar);
+	return 0;
+}
+
 /* Device-specific reset method for Chelsio T4-based adapters */
 static int reset_chelsio_generic_dev(struct pci_dev *dev, int probe)
 {
@@ -3976,6 +4019,11 @@ static const struct pci_dev_reset_methods pci_dev_reset_methods[] = {
 		reset_ivb_igd },
 	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IVB_M2_VGA,
 		reset_ivb_igd },
+	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_VMD_201D, reset_vmd_sbr },
+	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_VMD_28C0, reset_vmd_sbr },
+	{ PCI_VENDOR_ID_INTEL, 0x467f, reset_vmd_sbr },
+	{ PCI_VENDOR_ID_INTEL, 0x4c3d, reset_vmd_sbr },
+	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_VMD_9A0B, reset_vmd_sbr },
 	{ PCI_VENDOR_ID_SAMSUNG, 0xa804, nvme_disable_and_flr },
 	{ PCI_VENDOR_ID_INTEL, 0x0953, delay_250ms_after_flr },
 	{ PCI_VENDOR_ID_CHELSIO, PCI_ANY_ID,
-- 
2.18.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH 0/2] VMD subdevice secondary bus resets
  2020-09-28  1:05 [PATCH 0/2] VMD subdevice secondary bus resets Jon Derrick
  2020-09-28  1:05 ` [PATCH 1/2] PCI: vmd: Reset the VMD subdevice domain on probe Jon Derrick
  2020-09-28  1:05 ` [PATCH 2/2] PCI: Add a reset quirk for VMD Jon Derrick
@ 2020-11-19 11:37 ` Lorenzo Pieralisi
  2020-11-19 18:45   ` Derrick, Jonathan
  2 siblings, 1 reply; 5+ messages in thread
From: Lorenzo Pieralisi @ 2020-11-19 11:37 UTC (permalink / raw)
  To: Jon Derrick; +Cc: linux-pci, Bjorn Helgaas, Andrzej Jakowski, Dave Fugate

On Sun, Sep 27, 2020 at 09:05:55PM -0400, Jon Derrick wrote:
> This set adds some resets for VMD. It's very common code but doesn't
> seem to fit well anywhere that can also be exported if VMD is built as a
> module.
> 
> Jon Derrick (2):
>   PCI: vmd: Reset the VMD subdevice domain on probe
>   PCI: Add a reset quirk for VMD
> 
>  drivers/pci/controller/vmd.c | 32 ++++++++++++++++++++++++
>  drivers/pci/quirks.c         | 48 ++++++++++++++++++++++++++++++++++++
>  2 files changed, 80 insertions(+)

I can queue it up but I need Bjorn's ACK on patch (2).

Lorenzo

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 0/2] VMD subdevice secondary bus resets
  2020-11-19 11:37 ` [PATCH 0/2] VMD subdevice secondary bus resets Lorenzo Pieralisi
@ 2020-11-19 18:45   ` Derrick, Jonathan
  0 siblings, 0 replies; 5+ messages in thread
From: Derrick, Jonathan @ 2020-11-19 18:45 UTC (permalink / raw)
  To: lorenzo.pieralisi; +Cc: linux-pci, andrzej.jakowski, helgaas, Fugate, David

On Thu, 2020-11-19 at 11:37 +0000, Lorenzo Pieralisi wrote:
> On Sun, Sep 27, 2020 at 09:05:55PM -0400, Jon Derrick wrote:
> > This set adds some resets for VMD. It's very common code but doesn't
> > seem to fit well anywhere that can also be exported if VMD is built as a
> > module.
> > 
> > Jon Derrick (2):
> >   PCI: vmd: Reset the VMD subdevice domain on probe
> >   PCI: Add a reset quirk for VMD
> > 
> >  drivers/pci/controller/vmd.c | 32 ++++++++++++++++++++++++
> >  drivers/pci/quirks.c         | 48 ++++++++++++++++++++++++++++++++++++
> >  2 files changed, 80 insertions(+)
> 
> I can queue it up but I need Bjorn's ACK on patch (2).
> 
> Lorenzo

I just noticed 2/2 fixes something in 1/2, so I will send a v2 for this
set.

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2020-11-19 18:45 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-09-28  1:05 [PATCH 0/2] VMD subdevice secondary bus resets Jon Derrick
2020-09-28  1:05 ` [PATCH 1/2] PCI: vmd: Reset the VMD subdevice domain on probe Jon Derrick
2020-09-28  1:05 ` [PATCH 2/2] PCI: Add a reset quirk for VMD Jon Derrick
2020-11-19 11:37 ` [PATCH 0/2] VMD subdevice secondary bus resets Lorenzo Pieralisi
2020-11-19 18:45   ` Derrick, Jonathan

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).