On 8/10/2017 5:52 PM, Bjorn Helgaas wrote:

On Tue, Aug 08, 2017 at 08:57:24PM -0400, Sinan Kaya wrote:

quoted

Sporadic reset issues have been observed with Intel 750 NVMe drive by
writing to the reset file in sysfs in a loop. The sequence of events
observed is as follows:

- perform a Function Level Reset (FLR)
- sleep up to 1000ms total
- read ~0 from PCI_COMMAND
- warn that the device didn't return from FLR
- touch the device before it's ready

What's the driver-visible or user-visible effect of touching the
device before it's ready?  This sequence is sort of like the joke
without the punch line :)

The issue was first reported as a sporadic failure to assign NVMe
physical function to the guest machine and observing NVMe drive
initialization failures in the guest machine. I then repeated the problem
by creating a reset loop.

"NVMe (non-volatile memory) is not passing through to Virtual machine. 
NVMe is listed using lspci, however, the device is not listed using lsblk and 
fdisk -l. dmesg from virtual machine shows

root at null-8cfdf006971f:~# dmesg | grep nvme
[ 1.648842] nvme nvme0: pci function 0000:02:01.0
[ 1.650176] nvme 0000:02:01.0: enabling device (0000 -> 0002)
[ 2.547091] [<ffff000000a4c8e0>] nvme_irq [nvme]
[ 70.795558] nvme nvme0: I/O 203 QID 0 timeout, disable controller
[ 70.904369] nvme nvme0: Removing after probe failure status: -4"

I can include this into the commit message. 

quoted

An endpoint is allowed to issue Configuration Request Retry Status (CRS)
following a FLR request to indicate that it is not ready to accept new
requests. CRS is defined in PCIe r3.1, sec 2.3.1. Request Handling Rules
and CRS usage in FLR context is mentioned in PCIe r3.1a, sec 6.6.2.
Function-Level Reset.

You reference both PCIe r3.1 and r3.1a.  Is there something new in
r3.1a in this area?  If not, just reference r3.1 for both cases.

I can use r3.1. I have r3.1a. I'm not sure if the chapter numbers match or not.

quoted

A CRS indication will only be given if the address to be read is vendor ID
register. pci_bus_read_dev_vendor_id() knows how to deal with CRS returned
0xFFFF0001 value and will continue polling until a value other than
0xFFFF0001 is returned within a given timeout.

Try to discover device presence via CRS first. If device is not found, fall
through to old behavior.

Signed-off-by: Sinan Kaya <redacted>
---
 drivers/pci/pci.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index af0cc34..4366299 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c

@@ -3821,17 +3821,39 @@ static void pci_flr_wait(struct pci_dev *dev)
 {
 	int i = 0;
 	u32 id;
+	bool ret;
+
+	/*
+	 * Don't touch the HW before waiting 100ms. HW has to finish
+	 * within 100ms according to PCI Express Base Specification
+	 * Revision 3.1 Section 6.6.2: Function-Level Reset (FLR).
+	 */
+	msleep(100);
+
+	/*
+	 * See if we can find a device via CRS first. Physical functions
+	 * return from here if found, Virtual functions fall through as
+	 * they return ~0 on vendor id read once CRS is completed.
+	 */
+	ret = pci_bus_read_dev_vendor_id(dev->bus, dev->devfn, &id,
+					 60000);
+	if (ret)
+		return;

Alex was concerned that pci_bus_read_dev_vendor_id() could return
false ("no device here") with an ID value of ~0 for a functional VF
[1].

I think that is indeed possible:

  - a VF that is ready will return 0xffff as both Vendor ID and Device
    ID (SR-IOV r1.1, sec 3.4.1.1, .2), so the very first read in
    pci_bus_read_dev_vendor_id() would see 0xffffffff and return
    "false".

  - a VF that needs more time will return CRS and we'll loop in
    pci_bus_read_dev_vendor_id() until it becomes ready, and we'll
    return "true".

Falling into the code below for the "false" case probably will work,
but it's a little bit ugly because (a) we're using two mechanisms to
figure out when the device is ready for config requests, and (b) we
have to worry about VFs both in pci_bus_read_dev_vendor_id() and here
in the caller.

OK, I'm open to improving the code.

Here's an idea to make pci_bus_read_dev_vendor_id() work for both VFs
and PFs.  It can't distinguish the 0xffffffff from a VF vs one from a
non-existent device, but the caller might be able to pass that
information in, e.g., when we're enumerating and don't know whether
the device exists, we don't have a pci_dev and would use this:

How about creating a pci_bus_wait_crs() function with the loop in 
pci_bus_read_dev_vendor_id() and calling it from both places?

  pci_bus_read_dev_vendor_id(bus, devfn, &l, 60000, 0)

While places where we do have a pci_dev and expect the device to exist
(e.g., waiting after a reset), would do this:

  pci_bus_read_dev_vendor_id(bus, devfn, &l, 60000, dev->is_virtfn)

And we would skip the 0xffffffff check for VFs, e.g.,

  bool pci_bus_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *l,
                                  int crs_timeout, int is_vf)
  {
    if (pci_bus_read_config_dword(bus, devfn, PCI_VENDOR_ID, l))
      return false;

    if (!is_vf &&
        (*l == 0xffffffff || *l == 0x00000000 ||
         *l == 0x0000ffff || *l == 0xffff0000))
        return false;

    while ((*l & 0xffff) == 0x0001) {
      ...
      if (pci_bus_read_config_dword(bus, devfn, PCI_VENDOR_ID, l))
	return false;
    }

    return true;
  }

Would that work?

I don't know if this would be the best solution.  This is a messy
area.  We're relying on the host bridge to fabricate the 0xffff data
for non-existent devices, and most (maybe all) do that, but I don't
think that's actually in the spec.

AFAIK, returning 0xFFFFFFFF is a very typical practice. I don't know
the spec reference either.

quoted

+	pci_read_config_dword(dev, PCI_COMMAND, &id);
+	if (id != ~0)
+		return;
 
 	do {
 		msleep(100);
 		pci_read_config_dword(dev, PCI_COMMAND, &id);
-	} while (i++ < 10 && id == ~0);
+	} while (i++ < 9 && id == ~0);

 	if (id == ~0)
 		dev_warn(&dev->dev, "Failed to return from FLR\n");
 	else if (i > 1)
 		dev_info(&dev->dev, "Required additional %dms to return from FLR\n",
-			 (i - 1) * 100);
+			 i * 100);
 }
 
 /**

[1] http://lkml.kernel.org/r/20170221135138.791ba4e2 at t450s.home

-- 
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.