linux: pci: Fix PCI probing race condition

The Keystone PCI controller had a race condition hidden that made its
probing fail. Add a patch provided by TI (currently mainlined) that
fixes this race condition and returns PCI back to a funcional state.
Tested with a Hailo-8 NPU card.

While at it add a second patch suggested by TI that fixes "a 45 second
delay when no Endpoint Device is connected to the PCIe connector".

Signed-off-by: Diogo Ivo <[email protected]>

recipes-kernel/linux/files/patches-6.1/0095-PCI-keystone-Fix-race-condition-when-initializing-PH.patch

@@ -0,0 +1,54 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Siddharth Vadapalli <[email protected]>
+Date: Wed, 27 Sep 2023 09:48:45 +0530
+Subject: [PATCH] PCI: keystone: Fix race condition when initializing PHYs
+
+The PCI driver invokes the PHY APIs using the ks_pcie_enable_phy()
+function. The PHY in this case is the Serdes. It is possible that the
+PCI instance is configured for two lane operation across two different
+Serdes instances, using one lane of each Serdes.
+
+In such a configuration, if the reference clock for one Serdes is
+provided by the other Serdes, it results in a race condition. After the
+Serdes providing the reference clock is initialized by the PCI driver by
+invoking its PHY APIs, it is not guaranteed that this Serdes remains
+powered on long enough for the PHY APIs based initialization of the
+dependent Serdes. In such cases, the PLL of the dependent Serdes fails
+to lock due to the absence of the reference clock from the former Serdes
+which has been powered off by the PM Core.
+
+Fix this by obtaining reference to the PHYs before invoking the PHY
+initialization APIs and releasing reference after the initialization is
+complete.
+
+Link: https://lore.kernel.org/linux-pci/[email protected]
+Fixes: 49229238ab47 ("PCI: keystone: Cleanup PHY handling")
+Signed-off-by: Siddharth Vadapalli <[email protected]>
+Signed-off-by: Krzysztof Wilczyński <[email protected]>
+Acked-by: Ravi Gunasekaran <[email protected]>
+[Commit c12ca110c613 upstream]
+---
+ drivers/pci/controller/dwc/pci-keystone.c | 9 +++++++++
+ 1 file changed, 9 insertions(+)
+
+diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
+index fefb375a777a..a5dbe8c9b077 100644
+--- a/drivers/pci/controller/dwc/pci-keystone.c
++++ b/drivers/pci/controller/dwc/pci-keystone.c
+@@ -1367,7 +1367,16 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
+ 		goto err_link;
+ 	}
+
++	/* Obtain references to the PHYs */
++	for (i = 0; i < num_lanes; i++)
++		phy_pm_runtime_get_sync(ks_pcie->phy[i]);
++
+ 	ret = ks_pcie_enable_phy(ks_pcie);
++
++	/* Release references to the PHYs */
++	for (i = 0; i < num_lanes; i++)
++		phy_pm_runtime_put_sync(ks_pcie->phy[i]);
++
+ 	if (ret) {
+ 		dev_err(dev, "failed to enable phy\n");
+ 		goto err_link;

recipes-kernel/linux/files/patches-6.1/0096-PCI-keystone-Fix-pci_ops-for-AM654x-SoC.patch

@@ -0,0 +1,68 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Siddharth Vadapalli <[email protected]>
+Date: Mon, 22 Jan 2024 11:19:32 +0000
+Subject: [PATCH] PCI: keystone: Fix pci_ops for AM654x SoC
+
+In the process of converting .scan_bus() callbacks to .add_bus(), the
+ks_pcie_v3_65_scan_bus() function was changed to ks_pcie_v3_65_add_bus().
+The .scan_bus() method belonged to ks_pcie_host_ops which was specific
+to controller version 3.65a, while the .add_bus() method had been added
+to ks_pcie_ops which is shared between the controller versions 3.65a and
+4.90a. Neither the older ks_pcie_v3_65_scan_bus() method, nor the newer
+ks_pcie_v3_65_add_bus() method are applicable to the controller version
+4.90a which is present in AM654x SoCs.
+
+Thus, fix this by creating ks_pcie_am6_ops for the AM654x SoC which uses DW
+PCIe IP-core version 4.90a controller and omitting the .add_bus() method
+which is not applicable to the 4.90a controller. Update ks_pcie_host_init()
+accordingly in order to set the pci_ops to ks_pcie_am6_ops if the platform
+is AM654x SoC and to ks_pcie_ops otherwise, by making use of the "is_am6"
+flag.
+
+Fixes: 6ab15b5e7057 ("PCI: dwc: keystone: Convert .scan_bus() callback to use add_bus")
+Signed-off-by: Siddharth Vadapalli <[email protected]>
+Reviewed-by: Serge Semin <[email protected]>
+[Taken from https://lore.kernel.org/all/[email protected]/, manually edited]
+---
+ drivers/pci/controller/dwc/pci-keystone.c | 18 ++++++++++++------
+ 1 file changed, 12 insertions(+), 6 deletions(-)
+
+diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
+index a5dbe8c9b077..356bfdec00eb 100644
+--- a/drivers/pci/controller/dwc/pci-keystone.c
++++ b/drivers/pci/controller/dwc/pci-keystone.c
+@@ -581,6 +581,12 @@ static struct pci_ops ks_pcie_ops = {
+ 	.add_bus = ks_pcie_v3_65_add_bus,
+ };
+
++static struct pci_ops ks_pcie_am6_ops = {
++	.map_bus = dw_pcie_own_conf_map_bus,
++	.read = pci_generic_config_read,
++	.write = pci_generic_config_write,
++};
++
+ /**
+  * ks_pcie_link_up() - Check if link up
+  * @pci: A pointer to the dw_pcie structure which holds the DesignWare PCIe host
+@@ -948,15 +954,15 @@ static int __init ks_pcie_host_init(struct dw_pcie_rp *pp)
+ 	struct keystone_pcie *ks_pcie = to_keystone_pcie(pci);
+ 	int ret;
+
+-	pp->bridge->ops = &ks_pcie_ops;
+-
+-	if (!ks_pcie->is_am6) {
+-		pp->bridge->child_ops = &ks_child_pcie_ops;
+-		ret = ks_pcie_config_legacy_irq(ks_pcie);
++	if (ks_pcie->is_am6) {
++		pp->bridge->ops = &ks_pcie_am6_ops;
++		ret = ks_pcie_am654_config_legacy_irq(ks_pcie);
+ 		if (ret)
+ 			return ret;
+ 	} else {
+-		ret = ks_pcie_am654_config_legacy_irq(ks_pcie);
++		pp->bridge->ops = &ks_pcie_ops;
++		pp->bridge->child_ops = &ks_child_pcie_ops;
++		ret = ks_pcie_config_legacy_irq(ks_pcie);
+ 		if (ret)
+ 			return ret;
+ 	}