[PATCH v1 0/4] PCI: dwc: Drop or replace unused linux/gpio.h

[PATCH v1 0/4] PCI: dwc: Drop or replace unused linux/gpio.h

[Andy Shevchenko]

The <linux/gpio.h> is a legacy header that is going to die.
Make sure that PCI DesignWare drivers do not include it.

Compile tested on x86_64.

Andy Shevchenko (4):
  PCI: amd-mdb: Use the right GPIO header
  PCI: designware-plat: Drop unused include
  PCI: fu740: Drop unused include
  PCI: visconti: Drop unused include

 drivers/pci/controller/dwc/pcie-amd-mdb.c         | 2 +-
 drivers/pci/controller/dwc/pcie-designware-plat.c | 1 -
 drivers/pci/controller/dwc/pcie-fu740.c           | 1 -
 drivers/pci/controller/dwc/pcie-visconti.c        | 1 -
 4 files changed, 1 insertion(+), 4 deletions(-)

-- 
2.50.1

[PATCH v1 1/4] PCI: amd-mdb: Use the right GPIO header

[Andy Shevchenko]

The driver includes the legacy GPIO header <linux/gpio.h> but does not
use any symbols from it and actually wants <linux/gpio/consumer.h>,
so fix this up.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/pci/controller/dwc/pcie-amd-mdb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/controller/dwc/pcie-amd-mdb.c b/drivers/pci/controller/dwc/pcie-amd-mdb.c
index 7e50e11fbffd..88208e967201 100644
--- a/drivers/pci/controller/dwc/pcie-amd-mdb.c
+++ b/drivers/pci/controller/dwc/pcie-amd-mdb.c
@@ -7,7 +7,7 @@
 
 #include <linux/clk.h>
 #include <linux/delay.h>
-#include <linux/gpio.h>
+#include <linux/gpio/consumer.h>
 #include <linux/interrupt.h>
 #include <linux/irqdomain.h>
 #include <linux/kernel.h>
-- 
2.50.1

[PATCH v1 2/4] PCI: designware-plat: Drop unused include

[Andy Shevchenko]

This file does not use the symbols from the legacy
<linux/gpio.h> header, so let's drop it.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/pci/controller/dwc/pcie-designware-plat.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/pci/controller/dwc/pcie-designware-plat.c b/drivers/pci/controller/dwc/pcie-designware-plat.c
index d103ab759c4e..7d896752d43c 100644
--- a/drivers/pci/controller/dwc/pcie-designware-plat.c
+++ b/drivers/pci/controller/dwc/pcie-designware-plat.c
@@ -8,7 +8,6 @@
  */
 #include <linux/clk.h>
 #include <linux/delay.h>
-#include <linux/gpio.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
-- 
2.50.1

[PATCH v1 3/4] PCI: fu740: Drop unused include

[Andy Shevchenko]

This file does not use the symbols from the legacy
<linux/gpio.h> header, so let's drop it.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/pci/controller/dwc/pcie-fu740.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/pci/controller/dwc/pcie-fu740.c b/drivers/pci/controller/dwc/pcie-fu740.c
index 66367252032b..d0a34f680397 100644
--- a/drivers/pci/controller/dwc/pcie-fu740.c
+++ b/drivers/pci/controller/dwc/pcie-fu740.c
@@ -13,7 +13,6 @@
 
 #include <linux/clk.h>
 #include <linux/delay.h>
-#include <linux/gpio.h>
 #include <linux/gpio/consumer.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
-- 
2.50.1

[PATCH v1 4/4] PCI: visconti: Drop unused include

[Andy Shevchenko]

This file does not use the symbols from the legacy
<linux/gpio.h> header, so let's drop it.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/pci/controller/dwc/pcie-visconti.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/pci/controller/dwc/pcie-visconti.c b/drivers/pci/controller/dwc/pcie-visconti.c
index cdeac6177143..f21775c3b3a1 100644
--- a/drivers/pci/controller/dwc/pcie-visconti.c
+++ b/drivers/pci/controller/dwc/pcie-visconti.c
@@ -10,7 +10,6 @@
 
 #include <linux/clk.h>
 #include <linux/delay.h>
-#include <linux/gpio.h>
 #include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/iopoll.h>
-- 
2.50.1

Re: [PATCH v1 2/4] PCI: designware-plat: Drop unused include

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] In `dw_plat_pcie_probe()`, `pci_epc_init_notify()` is incorrectly called on the error path if `dw_pcie_ep_init_registers()` fails, falsely notifying EPF drivers that the EPC is ready.
--

commit e1010c1890e8eeb7c02639fb9293500993fe3d21
Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

PCI: designware-plat: Drop unused include
    
This commit removes the unused <linux/gpio.h> header from the
pcie-designware-plat.c file since its symbols are not used.

> diff --git a/drivers/pci/controller/dwc/pcie-designware-plat.c b/drivers/pci/controller/dwc/pcie-designware-plat.c
> index d103ab759c4ee..7d896752d43cd 100644
> --- a/drivers/pci/controller/dwc/pcie-designware-plat.c
> +++ b/drivers/pci/controller/dwc/pcie-designware-plat.c

[ ... ]

While reviewing this file, I noticed a pre-existing regression in the
error handling path of dw_plat_pcie_probe():

drivers/pci/controller/dwc/pcie-designware-plat.c:dw_plat_pcie_probe() {
        ...
        ret = dw_pcie_ep_init_registers(&pci->ep);
        if (ret) {
                dev_err(dev, "Failed to initialize DWC endpoint registers\n");
                dw_pcie_ep_deinit(&pci->ep);
        }

        pci_epc_init_notify(pci->ep.epc);
        ...
}

If dw_pcie_ep_init_registers() fails, execution falls through and
unconditionally calls pci_epc_init_notify(pci->ep.epc) rather than returning
an error.

Will this falsely notify all registered endpoint functions that the
endpoint controller was successfully initialized?

Endpoint function drivers react to the epc_init event by attempting
to allocate memory, set BARs, and configure hardware registers. Because
the memory windows are destroyed by dw_pcie_ep_deinit() when it sets
num_windows to 0, could these operations lead to failed memory allocations
or synchronous external aborts when writing to inaccessible registers?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260506084858.867884-1-andriy.shevchenko@linux.intel.com?part=2

Re: [PATCH v1 2/4] PCI: designware-plat: Drop unused include

[Andy Shevchenko]

On Wed, May 06, 2026 at 10:16:51AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [High] In `dw_plat_pcie_probe()`, `pci_epc_init_notify()` is incorrectly called on the error path if `dw_pcie_ep_init_registers()` fails, falsely notifying EPF drivers that the EPC is ready.
> --

Nice, but completely unrelated to the scope of this mini-series.
Whoever knows better the code and wants to address that, please do.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v1 0/4] PCI: dwc: Drop or replace unused linux/gpio.h

[Manivannan Sadhasivam]

On Wed, 06 May 2026 10:47:23 +0200, Andy Shevchenko wrote:
> The <linux/gpio.h> is a legacy header that is going to die.
> Make sure that PCI DesignWare drivers do not include it.
> 
> Compile tested on x86_64.
> 
> Andy Shevchenko (4):
>   PCI: amd-mdb: Use the right GPIO header
>   PCI: designware-plat: Drop unused include
>   PCI: fu740: Drop unused include
>   PCI: visconti: Drop unused include
> 
> [...]

Applied, thanks!

[1/4] PCI: amd-mdb: Use the right GPIO header
      commit: 8d846a0a6f95cd717df892b077a1c990a86b29fa
[2/4] PCI: designware-plat: Drop unused include
      commit: 800c861cfcbedd033bdffa65c00188fefdcf0767
[3/4] PCI: fu740: Drop unused include
      commit: 71d007f6fb17f519c2fe51d35e0bee5f602169e6
[4/4] PCI: visconti: Drop unused include
      commit: 7cc21269ef74269e423d64defc940808ccbf9774

Best regards,
-- 
Manivannan Sadhasivam <mani@kernel.org>