diff mbox series

[v4,3/4] PCI: dwc: Reduce DT reads by allocating host bridge via DWC glue driver

Message ID 20250207-ecam_v4-v4-3-94b5d5ec5017@oss.qualcomm.com
State New
Headers show
Series PCI: dwc: Add ECAM support with iATU configuration | expand

Commit Message

Krishna Chaitanya Chundru Feb. 6, 2025, 11:28 p.m. UTC
dw_pcie_ecam_supported() needs to read bus-range to find the maximum
bus range value. The devm_pci_alloc_host_bridge() is already reading
bus range and storing it in host bridge.If devm_pci_alloc_host_bridge()
moved to start of the controller probe, the dt reading can be avoided
and use values stored in the host bridge.

Allow DWC glue drivers to allocate the host bridge, avoiding redundant
device tree reads primarily in dw_pcie_ecam_supported().

Suggested-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Krishna Chaitanya Chundru <krishna.chundru@oss.qualcomm.com>
---
 drivers/pci/controller/dwc/pcie-designware-host.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

Comments

Manivannan Sadhasivam March 5, 2025, 6:18 p.m. UTC | #1
On Fri, Feb 07, 2025 at 04:58:58AM +0530, Krishna Chaitanya Chundru wrote:
> dw_pcie_ecam_supported() needs to read bus-range to find the maximum
> bus range value. The devm_pci_alloc_host_bridge() is already reading
> bus range and storing it in host bridge.If devm_pci_alloc_host_bridge()
> moved to start of the controller probe, the dt reading can be avoided
> and use values stored in the host bridge.
> 
> Allow DWC glue drivers to allocate the host bridge, avoiding redundant
> device tree reads primarily in dw_pcie_ecam_supported().
> 
> Suggested-by: Bjorn Helgaas <bhelgaas@google.com>
> Signed-off-by: Krishna Chaitanya Chundru <krishna.chundru@oss.qualcomm.com>
> ---
>  drivers/pci/controller/dwc/pcie-designware-host.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c
> index 826ff9338646..a18cb1e411e4 100644
> --- a/drivers/pci/controller/dwc/pcie-designware-host.c
> +++ b/drivers/pci/controller/dwc/pcie-designware-host.c
> @@ -484,8 +484,8 @@ int dw_pcie_host_init(struct dw_pcie_rp *pp)
>  	struct device *dev = pci->dev;
>  	struct device_node *np = dev->of_node;
>  	struct platform_device *pdev = to_platform_device(dev);
> +	struct pci_host_bridge *bridge = pp->bridge;
>  	struct resource_entry *win;
> -	struct pci_host_bridge *bridge;
>  	struct resource *res;
>  	int ret;
>  
> @@ -527,7 +527,12 @@ int dw_pcie_host_init(struct dw_pcie_rp *pp)
>  	if (!bridge)
>  		return -ENOMEM;
>  
> -	pp->bridge = bridge;
> +	if (!pp->bridge) {

'pp->bridge' is getting dereferenced above as I indicated in patch 1.

> +		bridge = devm_pci_alloc_host_bridge(dev, 0);
> +		if (!bridge)
> +			return -ENOMEM;
> +		pp->bridge = bridge;

There is already a previous devm_pci_alloc_host_bridge() call before this and
you are just duplicating the code here.

- Mani
diff mbox series

Patch

diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c
index 826ff9338646..a18cb1e411e4 100644
--- a/drivers/pci/controller/dwc/pcie-designware-host.c
+++ b/drivers/pci/controller/dwc/pcie-designware-host.c
@@ -484,8 +484,8 @@  int dw_pcie_host_init(struct dw_pcie_rp *pp)
 	struct device *dev = pci->dev;
 	struct device_node *np = dev->of_node;
 	struct platform_device *pdev = to_platform_device(dev);
+	struct pci_host_bridge *bridge = pp->bridge;
 	struct resource_entry *win;
-	struct pci_host_bridge *bridge;
 	struct resource *res;
 	int ret;
 
@@ -527,7 +527,12 @@  int dw_pcie_host_init(struct dw_pcie_rp *pp)
 	if (!bridge)
 		return -ENOMEM;
 
-	pp->bridge = bridge;
+	if (!pp->bridge) {
+		bridge = devm_pci_alloc_host_bridge(dev, 0);
+		if (!bridge)
+			return -ENOMEM;
+		pp->bridge = bridge;
+	}
 
 	/* Get the I/O range from DT */
 	win = resource_list_first_type(&bridge->windows, IORESOURCE_IO);