[PATCH 5.4 061/138] Revert "PCI/PM: Assume ports without DLL Link Active train links in 100 ms"

From: Greg Kroah-Hartman
Date: Mon Jul 27 2020 - 10:15:26 EST


From: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>

[ Upstream commit d08c30d7a0d1826f771f16cde32bd86e48401791 ]

This reverts commit ec411e02b7a2e785a4ed9ed283207cd14f48699d.

Patrick reported that this commit broke hybrid graphics on a ThinkPad X1
Extreme 2nd with Intel UHD Graphics 630 and NVIDIA GeForce GTX 1650 Mobile:

nouveau 0000:01:00.0: fifo: PBDMA0: 01000000 [] ch 0 [00ff992000 DRM] subc 0 mthd 0008 data 00000000

Karol reported that this commit broke Nouveau firmware loading on a Lenovo
P1G2 with Intel UHD Graphics 630 and NVIDIA TU117GLM [Quadro T1000 Mobile]:

nouveau 0000:01:00.0: acr: AHESASC binary failed

In both cases, reverting ec411e02b7a2 solved the problem. Unfortunately,
this revert will reintroduce the "Thunderbolt bridges take long time to
resume from D3cold" problem:
https://bugzilla.kernel.org/show_bug.cgi?id=206837

Link: https://lore.kernel.org/r/CAErSpo5sTeK_my1dEhWp7aHD0xOp87+oHYWkTjbL7ALgDbXo-Q@xxxxxxxxxxxxxx
Link: https://lore.kernel.org/r/CACO55tsAEa5GXw5oeJPG=mcn+qxNvspXreJYWDJGZBy5v82JDA@xxxxxxxxxxxxxx
Link: https://bugzilla.kernel.org/show_bug.cgi?id=208597
Reported-by: Patrick Volkerding <volkerdi@xxxxxxxxx>
Reported-by: Karol Herbst <kherbst@xxxxxxxxxx>
Fixes: ec411e02b7a2 ("PCI/PM: Assume ports without DLL Link Active train links in 100 ms")
Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/pci/pci.c | 30 +++++++++---------------------
1 file changed, 9 insertions(+), 21 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 08f7b1ed8c62e..b1b2c8ddbc927 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -4610,8 +4610,7 @@ static int pci_pm_reset(struct pci_dev *dev, int probe)
* pcie_wait_for_link_delay - Wait until link is active or inactive
* @pdev: Bridge device
* @active: waiting for active or inactive?
- * @delay: Delay to wait after link has become active (in ms). Specify %0
- * for no delay.
+ * @delay: Delay to wait after link has become active (in ms)
*
* Use this to wait till link becomes active or inactive.
*/
@@ -4652,7 +4651,7 @@ static bool pcie_wait_for_link_delay(struct pci_dev *pdev, bool active,
msleep(10);
timeout -= 10;
}
- if (active && ret && delay)
+ if (active && ret)
msleep(delay);
else if (ret != active)
pci_info(pdev, "Data Link Layer Link Active not %s in 1000 msec\n",
@@ -4773,28 +4772,17 @@ void pci_bridge_wait_for_secondary_bus(struct pci_dev *dev)
if (!pcie_downstream_port(dev))
return;

- /*
- * Per PCIe r5.0, sec 6.6.1, for downstream ports that support
- * speeds > 5 GT/s, we must wait for link training to complete
- * before the mandatory delay.
- *
- * We can only tell when link training completes via DLL Link
- * Active, which is required for downstream ports that support
- * speeds > 5 GT/s (sec 7.5.3.6). Unfortunately some common
- * devices do not implement Link Active reporting even when it's
- * required, so we'll check for that directly instead of checking
- * the supported link speed. We assume devices without Link Active
- * reporting can train in 100 ms regardless of speed.
- */
- if (dev->link_active_reporting) {
- pci_dbg(dev, "waiting for link to train\n");
- if (!pcie_wait_for_link_delay(dev, true, 0)) {
+ if (pcie_get_speed_cap(dev) <= PCIE_SPEED_5_0GT) {
+ pci_dbg(dev, "waiting %d ms for downstream link\n", delay);
+ msleep(delay);
+ } else {
+ pci_dbg(dev, "waiting %d ms for downstream link, after activation\n",
+ delay);
+ if (!pcie_wait_for_link_delay(dev, true, delay)) {
/* Did not train, no need to wait any further */
return;
}
}
- pci_dbg(child, "waiting %d ms to become accessible\n", delay);
- msleep(delay);

if (!pci_device_is_present(child)) {
pci_dbg(child, "waiting additional %d ms to become accessible\n", delay);
--
2.25.1