PCI: Fix restoring BARs on BAR resize rollback path

BAR resize operation is implemented in the pci_resize_resource() and
pbus_reassign_bridge_resources() functions. pci_resize_resource() can be
called either from __resource_resize_store() from sysfs or directly by the
driver for the Endpoint Device.

The pci_resize_resource() requires that caller has released the device
resources that share the bridge window with the BAR to be resized as
otherwise the bridge window is pinned in place and cannot be changed.

pbus_reassign_bridge_resources() rolls back resources if the resize
operation fails, but rollback is performed only for the bridge windows.
Because releasing the device resources are done by the caller of the BAR
resize interface, these functions performing the BAR resize do not have
access to the device resources as they were before the resize.

pbus_reassign_bridge_resources() could try __pci_bridge_assign_resources()
after rolling back the bridge windows as they were, however, it will not
guarantee the resource are assigned due to differences in how FW and the
kernel assign the resources (alignment of the start address and tail).

To perform rollback robustly, the BAR resize interface has to be altered to
also release the device resources that share the bridge window with the BAR
to be resized.

Also, remove restoring from the entries failed list as saved list should
now contain both the bridge windows and device resources so the extra
restore is duplicated work.

Some drivers (currently only amdgpu) want to prevent releasing some
resources. Add exclude_bars param to pci_resize_resource() and make amdgpu
pass its register BAR (BAR 2 or 5), which should never be released during
resize operation. Normally 64-bit prefetchable resources do not share a
bridge window with the 32-bit only register BAR, but there are various
fallbacks in the resource assignment logic which may make the resources
share the bridge window in rare cases.

This change (together with the driver side changes) is to counter the
resource releases that had to be done to prevent resource tree corruption
in the ("PCI: Release assigned resource before restoring them") change. As
such, it likely restores functionality in cases where device resources were
released to avoid resource tree conflicts which appeared to be "working"
when such conflicts were not correctly detected by the kernel.

Reported-by: Simon Richter <Simon.Richter@hogyros.de>
Link: https://lore.kernel.org/linux-pci/f9a8c975-f5d3-4dd2-988e-4371a1433a60@hogyros.de/
Reported-by: Alex Bennée <alex.bennee@linaro.org>
Link: https://lore.kernel.org/linux-pci/874irqop6b.fsf@draig.linaro.org/
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
[bhelgaas: squash amdgpu BAR selection from
https://lore.kernel.org/r/20251114103053.13778-1-ilpo.jarvinen@linux.intel.com]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Tested-by: Alex Bennée <alex.bennee@linaro.org> # AVA, AMD GPU
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patch.msgid.link/20251113162628.5946-7-ilpo.jarvinen@linux.intel.com

1 files changed, 75 insertions, 25 deletions

diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index 51f5e5a80b54..7e268960954b 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -2420,18 +2420,16 @@ EXPORT_SYMBOL_GPL(pci_assign_unassigned_bridge_resources);
  * release it when possible. If the bridge window contains assigned
  * resources, it cannot be released.
  */
-int pbus_reassign_bridge_resources(struct pci_bus *bus, struct resource *res)
+static int pbus_reassign_bridge_resources(struct pci_bus *bus, struct resource *res,
+					  struct list_head *saved)
 {
 	unsigned long type = res->flags;
 	struct pci_dev_resource *dev_res;
 	struct pci_dev *bridge = NULL;
-	LIST_HEAD(saved);
 	LIST_HEAD(added);
 	LIST_HEAD(failed);
 	unsigned int i;
-	int ret;
-
-	down_read(&pci_bus_sem);
+	int ret = 0;
 
 	while (!pci_is_root_bus(bus)) {
 		bridge = bus->self;
@@ -2443,9 +2441,9 @@ int pbus_reassign_bridge_resources(struct pci_bus *bus, struct resource *res)
 
 		/* Ignore BARs which are still in use */
 		if (!res->child) {
-			ret = add_to_list(&saved, bridge, res, 0, 0);
+			ret = add_to_list(saved, bridge, res, 0, 0);
 			if (ret)
-				goto cleanup;
+				return ret;
 
 			pci_release_resource(bridge, i);
 		} else {
@@ -2468,34 +2466,78 @@ int pbus_reassign_bridge_resources(struct pci_bus *bus, struct resource *res)
 		free_list(&added);
 
 	if (!list_empty(&failed)) {
-		if (pci_required_resource_failed(&failed, type)) {
+		if (pci_required_resource_failed(&failed, type))
 			ret = -ENOSPC;
-			goto cleanup;
-		}
-		/* Only resources with unrelated types failed (again) */
 		free_list(&failed);
+		if (ret)
+			return ret;
+
+		/* Only resources with unrelated types failed (again) */
 	}
 
-	list_for_each_entry(dev_res, &saved, list) {
+	list_for_each_entry(dev_res, saved, list) {
 		struct pci_dev *dev = dev_res->dev;
 
 		/* Skip the bridge we just assigned resources for */
 		if (bridge == dev)
 			continue;
 
+		if (!dev->subordinate)
+			continue;
+
 		pci_setup_bridge(dev->subordinate);
 	}
 
-	free_list(&saved);
-	up_read(&pci_bus_sem);
 	return 0;
+}
 
-cleanup:
-	/* Restore size and flags */
-	list_for_each_entry(dev_res, &failed, list)
-		restore_dev_resource(dev_res);
-	free_list(&failed);
+int pci_do_resource_release_and_resize(struct pci_dev *pdev, int resno, int size,
+				       int exclude_bars)
+{
+	struct resource *res = pci_resource_n(pdev, resno);
+	struct pci_dev_resource *dev_res;
+	struct pci_bus *bus = pdev->bus;
+	struct resource *b_win, *r;
+	LIST_HEAD(saved);
+	unsigned int i;
+	int ret = 0;
+
+	b_win = pbus_select_window(bus, res);
+	if (!b_win)
+		return -EINVAL;
+
+	pci_dev_for_each_resource(pdev, r, i) {
+		if (i >= PCI_BRIDGE_RESOURCES)
+			break;
+
+		if (exclude_bars & BIT(i))
+			continue;
 
+		if (b_win != pbus_select_window(bus, r))
+			continue;
+
+		ret = add_to_list(&saved, pdev, r, 0, 0);
+		if (ret)
+			goto restore;
+		pci_release_resource(pdev, i);
+	}
+
+	pci_resize_resource_set_size(pdev, resno, size);
+
+	if (!bus->self)
+		goto out;
+
+	down_read(&pci_bus_sem);
+	ret = pbus_reassign_bridge_resources(bus, res, &saved);
+	if (ret)
+		goto restore;
+
+out:
+	up_read(&pci_bus_sem);
+	free_list(&saved);
+	return ret;
+
+restore:
 	/* Revert to the old configuration */
 	list_for_each_entry(dev_res, &saved, list) {
 		struct resource *res = dev_res->res;
@@ -2510,13 +2552,21 @@ cleanup:
 
 		restore_dev_resource(dev_res);
 
-		pci_claim_resource(dev, i);
-		pci_setup_bridge(dev->subordinate);
-	}
-	up_read(&pci_bus_sem);
-	free_list(&saved);
+		ret = pci_claim_resource(dev, i);
+		if (ret)
+			continue;
 
-	return ret;
+		if (i < PCI_BRIDGE_RESOURCES) {
+			const char *res_name = pci_resource_name(dev, i);
+
+			pci_update_resource(dev, i);
+			pci_info(dev, "%s %pR: old value restored\n",
+				 res_name, res);
+		}
+		if (dev->subordinate)
+			pci_setup_bridge(dev->subordinate);
+	}
+	goto out;
 }
 
 void pci_assign_unassigned_bus_resources(struct pci_bus *bus)